Wolfgang 

Schneider

Memory
Development from
Early Childhood
Through Emerging
Adulthood
Memory Development from Early Childhood
Through Emerging Adulthood
Wolfgang Schneider

Memory Development from

Early Childhood Through
Emerging Adulthood

13
Wolfgang Schneider
Department of Psychology
University of Würzburg
Würzburg
Germany

ISBN 978-3-319-09610-0 ISBN 978-3-319-09611-7  (eBook)

DOI 10.1007/978-3-319-09611-7

Library of Congress Control Number: 2014945774

Springer Cham Heidelberg New York Dordrecht London

© Springer International Publishing Switzerland 2015

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or
information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed. Exempted from this legal reservation are brief excerpts
in connection with reviews or scholarly analysis or material supplied specifically for the purpose of
being entered and executed on a computer system, for exclusive use by the purchaser of the work.
Duplication of this publication or parts thereof is permitted only under the provisions of the Copyright
Law of the Publisher’s location, in its current version, and permission for use must always be obtained
from Springer. Permissions for use may be obtained through RightsLink at the Copyright Clearance
Center. Violations are liable to prosecution under the respective Copyright Law.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
While the advice and information in this book are believed to be true and accurate at the date of
publication, neither the authors nor the editors nor the publisher can accept any legal responsibility for
any errors or omissions that may be made. The publisher makes no warranty, express or implied, with
respect to the material contained herein.

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

This book is dedicated to the memory
of Michael Pressley, Thomas O. Nelson,
and Franz E.Weinert
Supported by a grant from Volkswagen
Foundation and Thyssen Foundation
Preface

It took a long time for me to write this book, and there were times when I thought
it would never happen. My good friend and colleague Michael Pressley and
I first published a volume on memory development between two and twenty in
1989, followed by a second edition in 1997. In the preface to the latter volume,
we announced a third edition within the next years. This did not happen for two
reasons. First, both of us were busy with different issues during the time we had
scheduled for an update of our memory book, and we found it difficult to refocus
our efforts on aspects of memory development. Second and, much more impor-
tant, Mike became seriously ill during this time and passed away much too early in
2006. This tragic and unexpected event not only shocked me but also forced me to
postpone the plan to write another book on memory development.
It took another 5 years before I saw a chance to reactivate this plan. At that
time, Volkwagen Foundation and Thyssen Foundation offered grants for Opus
Magnum writings, that is, comprehensive books on important themes in the
humanities and social sciences. I realized that this could be a nice opportunity to
write a book on memory development, and submitted a proposal. I was delighted
to learn a few months later that my proposal was accepted, and that I was awarded
the Opus Magnum grant by the two foundations. This grant freed me of teaching
and administrative duties for a year and definitively facilitated writing a new book
on memory development. I am indebted to Volkswagen Foundation and Thyssen
Foundation for this generous support.
If I see it correctly, this is my last book on memory development. It completes
a process started in the mid-1970s when my academic mentor and teacher Franz
Weinert hired me as a research assistant. My first job was to come up with a list
of publications on memory development written between the mid-1960s and mid-
1970s. Although the resulting list was brief, it contained a number of studies that
I found remarkable. I am very grateful to Franz Weinert for not only stimulating
my interest in the topic but also for providing continuous support during the fol-
lowing decades. Another turning point was meeting John Flavell at a conference
in Heidelberg organized by Franz Weinert. I was fortunate to receive a grant from
Volkswagen Foundation and to spend an academic year with John at Stanford

vii
viii Preface

University in the early 1980s. John has influenced my thinking on the develop-
ment of memory and metamemory enormeously, challenging me to think about
memory differently than I would have otherwise. It was during this time that I met
Michael Pressley and John Borkowski who stimulated my thinking and research
on memory development for a long time.
Many other people deserve thanks for their contributions to my previous work
and the writing of this book. My ideas about the development of memory and
metacognition as well as about the memory-metamemory relationships have been
expanded by discussions with my good friends and colleagues Patricia Bauer,
Harry Bahrick, David Bjorklund, Stephen Ceci, Marcus Hasselhorn, Robert Kail,
Asher Koriat, Joachim Körkel, Beth Kurtz-Costes, Ulman Lindenberger, Thomas
Nelson, Peter Ornstein, Mitchell Rabinowitz, Elaine Reese, Claudia Roebers,
Harriet Salatas-Waters, Robert Siegler, and Beate Sodian. I am particularly grate-
ful to Elaine, Dave, Rob, and Peter for reading drafts of the manuscript and com-
menting on my chapters.
Given that I am not a native speaker of English, I benefited enormously from
Jane Zagorski’s thorough editing of the book chapters and her didactic qualities.
I learned a lot about the subtleties of English by her thoughtful examples show-
ing that several of my formally correct expressions did not sound right to a native
speaker. My thanks also go to my coworker Michaela Pirkner who spent many
hours with formally editing the manuscript and carefully checking the reference
list. Last but not least I am very grateful to my wife Elisabeth for her never end-
ing support, feedback, and encouragement throughout this project. I doubt that I
would not have been able to finish it without her consistent help.
I dedicate this book to the memory of Michael Pressley, Thomas O. Nelson,
and Franz E. Weinert, three outstanding scientists who contributed considerably
to the development of my knowledge about memory and metacognition, and who
passed away much too early. While they will be unable to read and comment on
the book, I hope that the readers of this volume find it useful for their purposes.
This book is intended for a number of audiences. It is meant as a coherent over-
view of recent trends in research on memory development for advanced under-
graduate and graduate students, as well as for professionals. It is also a volume for
my peers in that I try to provide clear stances on many of the major issues of the
day. I hope that I wrote a book that summarizes the field well in a fashion that is
interesting. Let me know what you think.

Wolfgang Schneider
Contents

1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Organization of This Book. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

2 A Brief History of Memory Development Research . . . . . . . . . . . . . . 9

1880–1935. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Evidence from Child Psychology Studies. . . . . . . . . . . . . . . . . . . . . 10
Findings from Early Experimental Studies . . . . . . . . . . . . . . . . . . . . 11
Children’s Eyewitness Memory. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Assessment of “General” Memory Development . . . . . . . . . . . . . . . 15
1936–1965. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
German Research. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
American Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Russian Research. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Transition to the Modern Era . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

3 The Development of Implicit Memory . . . . . . . . . . . . . . . . . . . . . . . . . 25

Development of Perceptual Priming. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
The Development of Conceptual Priming. . . . . . . . . . . . . . . . . . . . . . . . . 30
Interactions Between Implicit and Explicit Memory . . . . . . . . . . . . . . . . 32
Implicit Memory in Clinical Populations. . . . . . . . . . . . . . . . . . . . . . . . . 33
The Impact of Advances in Neuroscience. . . . . . . . . . . . . . . . . . . . . . . . . 34
Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

4 Memory Development During the Infant and Toddler Years. . . . . . . 39

A Short History. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Short-Term/Working Memory in Infancy. . . . . . . . . . . . . . . . . . . . . . . . . 42
Memory Tasks Used with Preverbal Infants. . . . . . . . . . . . . . . . . . . . . . . 43
Hide-and-Seek Tasks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
Novelty-Preference Paradigms. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

ix
x Contents

The Development of Long-Term Memory in Infants and Toddlers . . . . . 54

Conditioning Paradigms. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Deferred Imitation Tasks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
Does the Age of 9 Months Indicate an Important Transition
Period?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
Effects of Memory Reinstatement and Reactivation. . . . . . . . . . . . . 61
The Impact of Language Acquisition on Toddlers’ LTM. . . . . . . . . . 63
Determinants of Young Children’s Long-Term Recall. . . . . . . . . . . . 65
Kinds of Memory: Implicit and Explicit Memory in Infancy. . . . . . 67
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72

5 The Development of Event and Autobiographical Memory

in Childhood and Adolescence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
Memory for Routine Events (Script-Based Memory). . . . . . . . . . . . . . . . 78
Developmental Differences in Event Recall . . . . . . . . . . . . . . . . . . . 79
Developmental Differences in Event Recall Format. . . . . . . . . . . . . 80
Memories for Single Instances of Routine Events. . . . . . . . . . . . . . . . . . 82
Memory for Fixed and Variable Items. . . . . . . . . . . . . . . . . . . . . . . . 83
The Schema-Confirmation-Deployment Model
as an Explanatory Approach. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
The Importance of Memory Cues . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
Memories of Novel and Salient Events. . . . . . . . . . . . . . . . . . . . . . . . . . . 87
Memory of Novel Events. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
Memory for Salient Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
Effects of Experience and Prior Knowledge . . . . . . . . . . . . . . . . . . . 90
Social–Cultural Contexts of Event Memories: Effects of Maternal
Reminiscing Style. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Conversations About Past Events. . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Conversations During Events. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
Relevance of Social–Cultural Context: Reminiscing
and Socioemotional Development. . . . . . . . . . . . . . . . . . . . . . . . . . . 97
How Long Do Autobiographical Memories Last?. . . . . . . . . . . . . . . 99
Infantile Amnesia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Empirical Findings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
Explanations of Infantile Amnesia and Their Evaluation . . . . . . . . . 103
Eyewitness Memory. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
How Long Do Eyewitness Memories Last?. . . . . . . . . . . . . . . . . . . . 111
The Role of Metacognitive and Background Knowledge . . . . . . . . . 113
Importance of the Salience of Events. . . . . . . . . . . . . . . . . . . . . . . . . 115
Age Differences in Suggestibility . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
Source Memory Problems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
Developmental Trends in False Memory Creation. . . . . . . . . . . . . . . 121
Familiarity and Recollection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
Contents xi

Facilitating Young Children’s Performance in Eyewitness

Situations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128

6 The Role of Basic Memory Capacities and Working Memory. . . . . . 131

The Development of Short-Term Memory Capacity . . . . . . . . . . . . . . . . 133
The Impact of Strategic Variables on Memory Span
Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
The Impact of Item-Ordering Ability on Memory Span
Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
Information Processing Speed and Short-Term Memory Capacity . . . . . 139
Developmental Changes in Information Processing Speed. . . . . . . . 139
Relations Between Information Processing Speed and
Memory Span. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
Capacity, Working Memory, and Cognitive Development:
Theoretical Approaches. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
Neo-Piagetian Models. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
Information Processing and Working Memory Models. . . . . . . . . . . 148
Empirical Tests of the Various Theoretical Working Memory
Models. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
Empirical Tests of Pascual-Leone’s Model . . . . . . . . . . . . . . . . . . . . 154
Tests of Case’s Resource-Sharing Model or Working Memory
and its Competitors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
Empirical Tests of the Baddeley’s Model . . . . . . . . . . . . . . . . . . . . . 160
Neuropsychological Evidence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
Evidence of the Episodic Buffer . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
Evaluation of Cowan’s Working Memory Model . . . . . . . . . . . . . . . 176
Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180

7 The Development of Encoding and Retrieval Strategies. . . . . . . . . . . 183

Evidence of Strategic Behavior in Young Children. . . . . . . . . . . . . . . . . . 186
Evidence of Early Retrieval Strategies. . . . . . . . . . . . . . . . . . . . . . . . 186
Evidence of Early Encoding Strategies. . . . . . . . . . . . . . . . . . . . . . . 189
What Develops Earlier: Encoding or Retrieval Strategies?. . . . . . . . 190
Encoding Strategies During the Elementary
and Middle-School Years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
Rehearsal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
Organizational Strategies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
Elaboration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
Strategies for Remembering Complex Materials. . . . . . . . . . . . . . . . 210
Retrieval Strategies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
Developmental Trends in the Use of External and Internal
Retrieval Cues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
xii Contents

Age Differences in the Flexibility of Retrieval Strategies. . . . . . . . . 214

Multiple and Variable Strategy Use. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
Factors Affecting Children’s Strategy Use and Development. . . . . . 218
Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229

8 Effects of the Knowledge Base on Memory Development. . . . . . . . . . 231

Conceptualizations of Knowledge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
Effects of the Knowledge Base on Memory. . . . . . . . . . . . . . . . . . . . . . . 234
Item-Specific Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234
Effortless Activation of Semantic Memory Relations. . . . . . . . . . . . 236
The Knowledge Base and Strategy Use. . . . . . . . . . . . . . . . . . . . . . . 237
Evidence That World Knowledge Mediates Memory and Learning. . . . . 238
Evidence for the Limited Effects of Domain Knowledge
on Subsequent Learning. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
Interrelations Between the Knowledge Base and Strategies. . . . . . . 241
Evidence for Strong Effects of the Knowledge Base on Subsequent
Learning and Memory: The Case of Expertise. . . . . . . . . . . . . . . . . . . . . 246
Intelligence, Knowledge, and Memory Performance. . . . . . . . . . . . . . . . 249
Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252

9 The Development of Metamemory . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255

Conceptualizations of Metacognitive Knowledge . . . . . . . . . . . . . . . . . . 256
Children’s “Theory of Mind” and Metacognitive Knowledge. . . . . . 260
Assessment of Metamemory. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262
Measures of Declarative Metamemory . . . . . . . . . . . . . . . . . . . . . . . 262
Measurement Problems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
Measures of Procedural Metamemory. . . . . . . . . . . . . . . . . . . . . . . . 264
Measurement Problems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266
The Development of Metamemory in Children and Adolescents. . . . . . . 268
Precursors of Metamemory: Knowledge of “Mental Verbs”
and the Acquisition of a “Theory of Mind”. . . . . . . . . . . . . . . . . . . . 268
Development of Declarative Metamemory from Early
Childhood to Adolescence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
The Development of Monitoring and Self-Control. . . . . . . . . . . . . . 275
Ease-of-Learning (EOL) Judgments . . . . . . . . . . . . . . . . . . . . . . . . . 276
Learning (JOLs). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278
Knowing (FOK) Judgments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281
Retrieval Monitoring: The Case of Confidence
Judgments (CJs). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283
The Relation Between Monitoring and Control Processes
in Children and Adolescents. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
Metamemory–Memory Relations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
Evidence from Correlational Studies. . . . . . . . . . . . . . . . . . . . . . . . . 295
Evidence from Training Studies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298
Evidence from Multivariate Analyses . . . . . . . . . . . . . . . . . . . . . . . . 299
Contents xiii

The Importance of Metacognition for Education. . . . . . . . . . . . . . . . . . . 302

Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307

10 Conclusions and Future Directions. . . . . . . . . . . . . . . . . . . . . . . . . . . . 309

Development of Implicit Memory. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310
Infancy and Early Childhood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311
Event Memory and Autobiographical Memory . . . . . . . . . . . . . . . . . . . . 313
Development of Memory Capacity and Working Memory. . . . . . . . . . . . 315
Development of Memory Strategies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316
The Impact of the Knowledge Base on Memory Development. . . . . . . . 318
The Development of Metamemory. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319
Future Directions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 320

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327

Index. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 381
Chapter 1
Introduction

Each of us has a memory. It is hard to overestimate the significance of memory in

our lives. Memory is a fundamental cognitive capacity. We use it with such ease
that it rarely occurs to us to marvel at our capacity for knowledge and the ways in
which we use it. Our memories define what we have accomplished in life, what
we know, and who we are. Only a few acts of cognition do not involve memory.
Problem-solving and decision-making processes all require the retrieval of previ-
ously stored information. Thus, memory is a prerequisite for the development of
thinking, and it is in turn affected by advances in intelligent information process-
ing. Memory is at the center of cognition, varying as a function of both basic-level
and higher-level processes and is involved in almost every act of complex cogni-
tion (Schneider and Bjorklund 1998).
Although we prefer to talk about memory as a single entity in everyday conver-
sations, there is plenty of evidence from behavioral and neuropsychological stud-
ies that memory is not a unitary phenomenon but is rather a remarkably complex
construct. Information must be encoded and possibly related to other information
known to the individual. Since the “cognitive revolution” and the introduction of
information-processing approaches into modern cognitive psychology—in par-
ticular, multistore models of memory (e.g., Atkinson and Shiffrin 1968)—we
have assumed that the flow of information from the environment includes several
memory systems. More specifically, we assume that information from the exter-
nal world is initially represented in sensory registers and then passed on to the
short-term store (STS), where capacity is much smaller than in the sensory reg-
ister, but representations are more durable, lasting for seconds. It is our STS that
makes contact with the world and holds information long enough for us to eval-
uate it. Given the limited capacity of the STS, it is important that some cogni-
tive operation (e.g., a strategy) is applied to the information in the STS so that it
can be transferred to the long-term store (LTS), where presumably it is retained
indefinitely.

© Springer International Publishing Switzerland 2015 1

W. Schneider, Memory Development from Early Childhood Through
Emerging Adulthood, DOI 10.1007/978-3-319-09611-7_1
2 1 Introduction

How is information represented in the LTS? Behavioral and brain research car-
ried out to explore the general nature of memory has led us to postulate two major
forms or types of memory that differ in a number of ways, labeled declarative
or explicit and nondeclarative, implicit, or procedural memory (e.g., Tulving
1985, 2005). Declarative memory refers to memory for facts and events and rep-
resents the type of memory responsible for the relatively rapid learning and reten-
tion of information that can be described verbally. It comes in two subcategories,
namely episodic memory, which is literally memory for episodes that happened
in one’s life, and semantic memory, which refers to our knowledge of language,
rules, and concepts, thus supporting our general knowledge about the world.
Nondeclarative memory is related to perceptual and motor learning processes,
which typically do not require conscious awareness. It refers to procedural knowl-
edge, which is mainly unconscious. Numerous research findings indicate that
different areas of the brain are involved in declarative versus nondeclarative mem-
ories. Again, this supports the argument that memory is not a single phenomenon
but rather consists of a set of domain-specific mental operations. In particular, ani-
mal models of lesions and disease have permitted experimental control over the
location of lesions and information about the neural networks that subserve differ-
ent forms of human memory.
Overall, substantial progress has been made in our general understanding of
memory structures and processes. The aim of this book is to provide an integrative
overview of theory and research on memory development from early childhood
to late adolescence. This is no longer an easy task. When Michael Pressley and
I decided to write a book on memory development in the late 1980s (Schneider
and Pressley 1989), the research situation appeared complex but not as difficult
to grasp as it is now 25 years later. Although the history of research on memory
development will be presented in more detail in the next chapter, it seems to make
sense to highlight major developments that have occurred since the “cognitive rev-
olution” to illuminate the difficulty of the task of summarizing major trends in this
field.
The modern era of research on memory development began in the late 1960s
and was stimulated by a shift from behavioral theories to cognitive theories, a
shift that emphasized information-processing considerations. Behavioral theo-
ries assumed that the ability to remember depends on the formation of associa-
tions, that is, of bonds between stimuli and responses, with the strength of those
bonds (or habit strengths) determining the ability to remember. In comparison, the
emphasis of the cognitive approach to memory was considerably different from
that of the behavioral approach, focusing on the mental activities that intervene
between stimulus and response and also on developmental differences in these
activities.
It seems, at least in retrospect, that the lack of interest in memory development
before the early 1960s was due to the then widespread opinion that such stud-
ies were of marginal theoretical significance, given that general learning theory
was so popular in those days. Researchers were inclined to assume that the laws
of memory must be the same at all ages. Consequently, developmental studies
1 Introduction 3

offered little more than descriptive value. The situation changed dramatically
thereafter. Whereas in 1965, the keyword “memory” was not in the index of Child
Development Abstracts and Bibliography (as noted by Kail and Hagen 1977),
every issue of that outlet now includes abstracts for a number of memory studies.
When Wimmer (1976) described the state of the art of research on memory devel-
opment, he was able to do so within the framework of a review article. Since then,
there has been a dramatic increase in the amount of research and the approaches
taken by memory development researchers. Only two decades after the “research
spurt” on memory development, Schneider and Pressley (1989) concluded that the
literature had become extensive enough to pose serious problems of selection and
organization for reviewers and editors. Needless to say, the situation is even more
complicated four decades after the beginning of this “spurt.”
Several factors have contributed to this rapid development. First, the discov-
ery of Piaget by American developmental psychology encouraged the cognitive
“Zeitgeist” in developmental cognitive psychology and also illustrated the impor-
tance of reconstructive processes in children’s memorizing (Ornstein 1978).
Although the impact of this discovery was generally substantial, the “intelligence
hypothesis of memory development” that Piaget and Inhelder had introduced
into developmental psychology did not survive for very long. According to this
hypothesis, “the development of memory with age is the history of gradual organ-
izations closely dependent on the structuring activities of intelligence, though
regulated by a special mechanism, namely, the structuring of the past or of past
experiences” (Piaget and Inhelder 1973, p. 380). Undoubtedly, reliable intercor-
relations between intelligence and memory have been demonstrated in numerous
studies. However, they were not substantial in most cases and thus did not con-
firm Piaget and Inhelder’s basic assumption. Nonetheless, it seems important to
note that the translation of Piagetian research into English fundamentally changed
the way scientists thought about development. Researchers interested in cogni-
tive development realized that children of different ages had qualitatively dif-
ferent capacities to represent the world and could not be conceived of as “little
adults” in their thinking (Bauer and Fivush 2014; Liben and Bowman 2014). This
change in perspective proved to be helpful and generated interesting studies on
memory development by neo-Piagetian researchers such as Case and Pascuale-
Leone, yielding new insights into the nature of memory capacity and its changes
over time.
Moreover, the introduction of relatively new tools of experimental develop-
mental psychology in the area of memory research contributed to this positive
development. In 1971, John Flavell, in a symposium held at the biennial meet-
ings of the Society for Research in Child Development, asked the question “What
is memory development the development of?” (p. 272). This now famous quote
initiated a new research program that changed the focus of theoretical think-
ing, shifting the emphasis to the interplay between the development of cogni-
tive resources in general and the functions of memory in particular. As will be
shown in more detail below, this research program as it was initiated in the mid-
1970s assessed the relevance of basic memory capacity, memory strategies, and
4 1 Introduction

different forms of knowledge (metamemory, the knowledge base). The result was
a great number of findings based on (mostly) cross-sectional studies, which were
summarized in a number of influential books (e.g., Brainerd and Pressley 1985;
Kail 1979; Kail and Hagen 1977; Ornstein 1978). Articles on the development
of memory turned out to be a routine feature of most issues of most American
developmental journals, and developmental articles also appeared regularly in the
mainstream adult memory journals, thus increasing the popularity of this research
program. In Europe, the popularity of memory development research was further
stimulated by a series of international conferences initiated by Franz Weinert and
colleagues during the 1980s and early 1990s. These events were stimulated by the
belief that more emphasis should be given to the analysis of universal changes
and individual differences in memory development. They not only yielded a num-
ber of edited volumes on the issue (e.g., Schneider and Weinert 1990; Weinert and
Kluwe 1987; Weinert and Perlmutter 1988; Weinert and Schneider 1995) but also
initiated many international collaborative research activities. Many of us still rave
about the stimulating and exciting discussions on various memory development
issues during scientific meetings at Castle Ringberg near Lake Tegernsee in the
Bavarian Alps.
Given this rapid increase in research activities since the mid-1970s, there is
no doubt that memory development has been one of the most studied topics in
all of cognitive development during the past four decades. Until the early 1990s,
the predominant view shared by most developmental scientists was that develop-
mental changes in memory performance were caused by four different sources,
namely basic memory capacities, memory strategies, domain-specific knowledge,
and metamemory (i.e., knowledge about memory). Moreover, these researchers
developed theoretical models that provided a structure for the interplay among
these sources of memory development (e.g., the Good Strategy User model;
Pressley et al. 1987). When completing the first edition of their book, after two
extremely productive decades of research on memory development, Schneider
and Pressley (1989) concluded that interest in the field was declining somewhat.
This turned out to be an incorrect judgment. When preparing a second edition of
their book less than a decade later, Schneider and Pressley (1997) realized that
their perceptions on that point were inadequate in that most of the issues cov-
ered in the first volume continued to be explored. Moreover, they also noticed
that there was much more in the field than the four sources of memory develop-
ment described above. Applied aspects of research on memory development had
become more popular, and this led them to include a new chapter on autobio-
graphical memory and eyewitness memory in the second edition of their book.
Schneider and Pressley decided against including research on infant memory in
their book because (a) at that time, research on infant memory seemed to be very
different from research on memory in later childhood, and (b) a fair representa-
tion of infant memory and its development would require many pages. This is
why they focused on the description of memory development during childhood
after the age of two.
Organization of This Book 5

Organization of This Book

How does the present volume differ from the two previous Schneider and Pressley
editions? First, I realized that our original belief shared with several developmental
scientists in the 1980s and 1990s—that memories of preverbal human infants do
not resemble those of older children and adults—is no longer tenable. In particu-
lar, seminal studies carried out by Patricia Bauer and Carolyn Rovee-Collier during
this period of time convinced us that this belief was wrong. Knowing how well
infants can remember is important not only because this research tells us about the
origins of the memory system, but it is also important for conceptualizing the long-
term effects of early experience (Miller 2014). Thus, the reader will find a chap-
ter on the development of infants’ short-term and long-term memory in the present
book, thus necessitating the change in the book’s title. There is no doubt that
research on infants and toddlers carried out during the past three decades has been
extremely creative, using innovative experimental designs as well as new tools and
techniques and providing evidence for unexpected mnemonic competencies dur-
ing this early period of life. As shown below, there is also controversial discussion
about the degree to which these surprising memory competencies are implicit or
represent conscious experiences, an issue treated in more detail in this chapter.
This interesting debate inspired me to include a new chapter on the develop-
ment of implicit memory in this book, thus allowing for a comparison of gen-
eral differences in the development of implicit and explicit memory. It has been
claimed in the relevant literature that implicit memory is present at the start of life
and that explicit memory develops considerably later (see Lloyd and Newcombe
2009; Lloyd and Miller 2014). Meanwhile, there has been an increasing acknowl-
edgment that explicit memory is already present in infancy (e.g., Bauer 2009)
and that developmental trends in implicit memory seem to differ for tasks assess-
ing perceptual versus conceptual priming. The main purpose of this chapter is to
explore the question of how much the development of implicit memory deviates
from the typical trends observed for explicit memory.
The remaining chapters refer to topics that were already discussed in our previ-
ous volumes. For instance, major historical developments are briefly summarized
(for extended reviews, see Miller 2014; Schneider 2000a). The chapter on history
addresses early research carried out before the “cognitive revolution” arrived in
the USA in the 1960s. The early studies reflected the influences of new thinking
about evolution and development and the launching of experimental psychology
(Baker-Ward and Ornstein 2014). They illustrated the original impact of Western
European and Soviet scientists on research issues such as the general development
of immediate (short-term) memory and the importance of individual differences
(e.g., gender, IQ). This early research also set the stage for the current interest in
applied issues such as children’s eyewitness competencies. The chapter does not
describe developmental trends in the modern era, which covers approximately the
last 50 years because most of the rest of this book is concerned with research con-
ducted since 1965 (see Miller 2014, for a review of modern history).
6 1 Introduction

Although the topics of the remaining chapters are very similar to those from the
Schneider and Pressley (1997) book, their contents differ considerably from those
of the earlier book. For instance, given the recent interest in memory development
research applied to practical issues, the chapter on autobiographical and event
memory not only summarizes the comprehensive literature on the development of
episodic memory as assessed in experimental settings but also discusses major out-
comes of research on children’s and adolescents’ eyewitness testimonies. Given that
this field has been particularly active and multifaceted during the past three dec-
ades, it is a really difficult and challenging task to summarize the basic insights.
This is also true for research trends in what Schneider and Pressley called
the major determinants of memory performance, that is, basic memory capaci-
ties (short-term and working memory), memory strategies, the knowledge base
(domain-specific knowledge), and metamemory. As can be inferred from a closer
look at the chapter on basic memory capacity, recent developments in the area
of working memory, and regarding theoretical models of working memory in
particular, have changed our views of typical developmental trends in memory
capacity and have yielded a much more complex pattern of findings than that
available when Michael Pressley and I summarized the literature nearly 20 years
ago. Undoubtedly, considerable advances have recently been made in this field of
research.
Although the area of memory strategy development is one of the classic ones,
going back to the early studies by John Flavell and colleagues in the 1960s and
intensively explored thereafter, the chapter on strategy development still contains
new and partially unexpected findings. This is mainly due to a shift from predomi-
nantly cross-sectional work to longitudinal studies and the use of multivariate sta-
tistical designs. Moreover, the recent implementation of large-scale longitudinal
studies that have explored the impact of the social context (parents, teachers) on
children’s strategy development has helped to identify mechanisms of memory
strategy development that seem to have long-lasting consequences for memory
performance (Bjorklund et al. 2009; Roebers 2014).
In comparison, the chapter on the knowledge base does not contain similarly
new evidence. The chapter on knowledge in the Schneider and Pressley (1989)
book was fairly short, which led us to expand this chapter considerably in the
second edition. This was possible because most evidence on the importance of
domain-specific knowledge for the development of memory was published during
the time period between the early 1980s and late 1990s. Research carried out on
this issue during the past three decades has convincingly shown that the knowl-
edge base can have considerable effects on memory behavior and performance,
even though prior knowledge is not always used automatically. The chapter sum-
marizes previous and more recent evidence on knowledge base effects and also
illustrates complex interactions among basic memory capacities, memory strate-
gies, and metamemory, and this sometimes makes it difficult to disentangle the
effects of specific sources from those of other influences.
The chapter on the development of metamemory differs considerably from the
version in Schneider and Pressley’s (1997) book. This is mainly due to the fact
Organization of This Book 7

that a great number of comprehensive cross-sectional and longitudinal studies

have recently been conducted on various aspects of both declarative and proce-
dural metamemory. These studies have provided interesting insights into early
developmental trends and have also illustrated relations between strategy use
and the emergence of metacognitive knowledge, both in the laboratory and in the
classroom. As predicted by Schneider and Pressley (1997), recent studies have
complemented traditional developmental and experimental methods by imple-
menting modern multivariate approaches (e.g., causal modeling), which have
permitted the evaluation of more comprehensive models of the functioning of
memory and metamemory in children and adolescents. One of the most striking
features of this research is the increasing prominence of metacognition in other
areas, for instance, research on reading, mathematics, and science.
When writing this book, I had to accept that it is no longer possible to provide a
representative and comprehensive account of memory development within a single
monograph. In this regard, it is interesting to note that even the two-volume hand-
book recently edited by Bauer and Fivush (2014), including 44 chapters on differ-
ent aspects of memory development, covering about 1,000 pages, and presenting
the most comprehensive recent assessment of the topic, does not include all theo-
retically relevant aspects of memory development. The present overview focuses
on major trends in the development of verbal memory from early childhood to
late adolescence. Preverbal memory is considered in the chapter on infancy, and
visuospatial memory development is mainly addressed in the chapter on working
memory. Those readers interested in the development of visuospatial memory are
referred to the recent review by Lourenco and Frick (2014).
Although I originally planned to include a chapter on the brain basis of mem-
ory development, a field that has attracted much interest in recent years (see
Bachevalier 2014; Bauer 2009; Klingberg 2013), I eventually decided against this
idea. Instead, I found it more interesting and appropriate to include sections on
brain–behavior relations primarily in the chapters on infant memory, event mem-
ory, and working memory, that is, in those chapters involving memory contexts
that have particularly benefitted from advances in developmental neuroscience.
Furthermore, I resisted the temptation to include a separate chapter on relevant
theories. There is no doubt that recent research on memory development has
profited from domain-general theories and models such as the Good-information
Processing model (Pressley et al. 1989), fuzzy-trace theory (Brainerd and Reyna
2014), or sociocultural theories of memory development (Nelson 2014). Again,
however, I thought it better to illustrate their merits in the relevant chapters.
One conclusion already drawn by Schneider and Pressley (1997) was that their
original doubts about the immediate future of memory development research
had dissipated and that there is good reason for optimism. I hope to convince the
reader of this book that this optimism is still justified and that the field is as active
as before. There is a lot of research yet to be done, and I am confident that this
summary of work to date will stimulate even more research on memory develop-
ment in the near future.
Chapter 2
A Brief History of Memory
Development Research

Most developmental scientists probably agree that memory development has been
one of the most-studied topics in all of cognitive development, and deservedly so.
In fact, an impressive number of scientific studies on this issue have been pub-
lished within the last four decades, stimulated by a shift away from behaviorism
theories toward considerations of information processing. The discovery of Piaget
by American developmental psychologists also supported the cognitive “Zeitgeist”
in developmental psychology (Ornstein 1978).
Whereas many people believe that scientific research on memory development
did not begin before the 1960s, in fact, the history of this research paradigm goes
back to the beginning of the experimental study of memory near the end of the nine-
teenth century. Accordingly, experimental studies of memory are as old as scien-
tific psychology. When Ebbinghaus (1885) was beginning his classic experiments
on memory and forgetting in 1879, Wundt had just founded the first psychological
laboratory in Leipzig, Germany. Although this is widely known, it is not equally
well-known that research on memory development also began at about that time.
Around the turn of the century, numerous studies were being conducted in Europe
to investigate developmental and individual differences in children’s memory. From
the beginning, these studies included examples of research conducted in the field on
the one hand and in the laboratory on the other. For example, both Darwin (1877)
and Preyer (1882) published naturalistic case studies of their own children’s devel-
opment, including the development of memory and other cognitive skills.
There were three rather independent lines of research that contributed to this
early trend. First, whereas the development of children did not attract much inter-
est before the end of the nineteenth century, carefully conducted case studies of
young children’s development (which also included systematic observations of
memory development in early childhood) received a lot of attention, leading to the
scientific foundation of child psychology in Germany. Examples of such observa-
tional diary studies on the “mind of the child” include the aforementioned research

© Springer International Publishing Switzerland 2015 9

W. Schneider, Memory Development from Early Childhood Through
Emerging Adulthood, DOI 10.1007/978-3-319-09611-7_2
10 2  A Brief History of Memory Development Research

by Darwin and Preyer as well as William and Clara Stern’s extremely detailed
diary on the development of their children (1913). Child psychology clearly flour-
ished during the time between 1890 and 1915, with the founding of more than
20 child psychology journals and university departments devoted to this topic
during those years (see Weinert and Weinert 1998). As a consequence of this new
research trend, several textbooks on child development were published early in the
twentieth century (e.g., Bühler 1918; Koffka 1921; Stern 1914; Werner 1926), all
of which included long chapters on memory development.
A second line of research was directly derived from memory experiments with
adults. Some of these studies explored whether findings obtained for adult popula-
tions could be easily generalized to children of different ages. For instance, using
laboratory-based methods, Jacobs (1887) reported age differences in digit span,
and Kirkpatrick (1894) observed developmental changes in free-recall perfor-
mance. Other investigations were less basic in nature and were driven by educa-
tional interests. These studies tested common (mis)conceptions held at that time;
for example, those children, because they practice their memory skills in school
almost every day, are better at remembering verbal material than adults. Also,
many of these studies examined the popular assumption that boys have better
memories than girls. As the issue of coeducation was at stake in Germany around
the turn of the century, this question was of high practical relevance.
The third line of research on children’s memory was even more applied, focus-
ing on children’s and adults’ testimonial competence. The prevailing legal attitude
had been one of the skepticism about the testimony of child witnesses. Nonetheless,
interest in children’s eyewitness memory competencies was particularly strong
in Germany and France, where systematic research on this issue flourished at the
beginning of the twentieth century (e.g., Stern 1910; Whipple 1909, 1911). Most
studies focused on children’s suggestibility, developing methodologies that are still
in use in modern research on the topic.
All three of these lines of research share the characteristic that they focused on
children’s episodic memory, that is, on memory for personally experienced events
that range from briefly presented stimuli to significant life events (cf. Baker-Ward
and Ornstein 2014; Schneider 2000a). In the next section, the major findings of
early research on memory development will be briefly summarized. Three differ-
ent time periods that clearly differed in their research focus will be distinguished.

1880–1935

Evidence from Child Psychology Studies

As noted above, Preyer (1882) careful observations of his son during the first 3
years of life represented the beginning of scientific child psychology. Preyer
voted for an explicit, systematic, and methodologically controlled observational
approach that allowed for the construction of an age-graded behavioral inventory.
His own inventory included numerous cognitive features such as sensorimotor
1880–1935 11

skills, perception, speech acquisition, learning, memory, and the development of

conscious thought. Given that many others adopted this observational approach, a
rich database on the course of (early) cognitive development was available around
the turn of the century. This new knowledge base was represented in several text-
books written by leading German child psychologists of that time. The empha-
sis of this research was on long-term memory (LTM), recognition, and imagery.
Statements about short-term memory (STM) were rare and not very specific. All
of the authors agreed that imagination is the result of memory processes, conceiv-
able as a replicable memory trace that is independent of actual perception but can
be influenced by knowledge structures. Most textbook writers also made a distinc-
tion between verbal and visual LTM.
One topic of interest was very young children’s recognition memory. All
researchers of that time agreed that “stimulus-bound” recognition memory can
already be observed in young infants, indicating that this skill develops very early
in life (but not before the age of 4 months). Its main function was understood
to be the categorization of input stimuli and the reduction of input complexity.
Recognition memory was assumed to improve considerably during the second year
of life, accompanied by a dramatic increase in retention span, which could last for
several months (cf. Bühler 1930). By contrast, “free” memory activities such as
recall were assumed to develop considerably later, beginning by about 4 years of
age. Most authors also agreed that visual memory functions were less deficient in
young children than verbal ones. Accordingly, these authors assumed that develop-
mental changes were more pronounced for verbal than for visual memory.
Several studies at that time investigated what is called “infantile amnesia”
today (see Offner 1924). That is, researchers were interested in determining the
age at which children begin to be able to permanently store memories of specific
events, a question that was examined using samples of older children and adults.
Early research by the French scientists Henri and Henri (1897) suggested that
memories could date back to the second year of life when the events in question
were particularly dramatic. However, subsequent studies did not confirm such an
early onset, indicating that the earliest events that participants could remember
occurred when they were 3 or 4 years old. This finding from retrospective analyses
squared well with the observation that children who had moved away from their
birthplace before the age of 4 could not remember any details about their original
homes (Bühler 1930; Offner 1924).

Findings from Early Experimental Studies

Although most observational child psychology studies were creative, they had
the major shortcoming of being fraught with methodological problems. Thus,
this research was quickly followed by studies that were conducted in the labora-
tory and often involved the manipulation of independent variables. The second
line of research outlined above was directly derived from general experimental
12 2  A Brief History of Memory Development Research

psychology and was later labeled “experimental memory psychology based on

children” by Schumann-Hengsteler (1995). Typically, these studies adopted
research paradigms developed in the tradition of Georg Elias Müller and his stu-
dent Hermann Ebbinghaus. The major goal was to describe changes in memory
performance across the life span. Children were treated as a special population
that was used to evaluate the generalizability of findings from studies with adults.
Large samples of children from different ages were included in most of these stud-
ies. In contrast to the child psychology studies summarized above, the majority of
these experimental studies focused on short-term (immediate) memory.
Ebbinghaus (1885, 1887) himself was concerned with the capacity of memory
span at different age levels, and he had a particular interest in identifying the devel-
opmental memory span curves for various types of materials. His findings were
based on children and young adults and indicated that considerable age effects could
be observed across various materials even though age differences were least pro-
nounced for meaningless words and syllables. Subsequent research (e.g., Lobsien
1902; Netschajeff 1900) confirmed this result in that memory span performance
increased with age for all types of items, and memory was better for objects and
labels than for sounds and abstract concepts. Other studies also demonstrated
that meaning played a significant role in determining the amount recalled (e.g.,
Netschajeff 1902). For instance, whether nonsense words were one or two syllables
in length had little impact on span. On the other hand, Binet and Henri (1894a, b)
found that preschoolers exhibited substantially better memory for sentences with
many words than for short lists of meaningless items. As a rule, the core verbal
units that suggested meaning were remembered best. These studies were the first to
emphasize the constructive aspects of immediate memory. As noted by Baker-Ward
and Ornstein (2014), Binet and Henri had anticipated the constructivism knowledge-
driven approaches to memory reflected in Bartlett’s (1932) work as well as in more
modern developmental studies of prose memory (e.g., Paris 1978).
In their reviews of the literature, Meumann (1907) and Offner (1924) regarded
these studies as particularly valuable because these studies highlighted the exist-
ence of several memory functions that did not develop in parallel. Thus, the same
9-year-old children who could produce about 30 % of a list of nonsense syl-
lables were more successful when the trial consisted of numerical series of the
same length (about 60 % recall) or meaningful word lists (about 70 % recall; e.g.,
Jacobs 1887; Lobsien 1902; Pohlmann 1906). These findings were replicated in
several studies (see Meumann 1907) and generally confirmed in experiments on
immediate memory span that used the “word-pair method,” the forerunner of the
paired-associate learning technique. For instance, Nagy (1930) conducted a study
of 700 schoolchildren between the ages of 7 and 19 and found that abstract word
pairs were harder to remember than concrete ones regardless of age.
Although there was broad agreement that children’s immediate memory is less
developed than that of adults, researchers disagreed about the course of memory
development from childhood to adulthood. Bühler (1930) assumed that develop-
mental gains are most pronounced in 10- to 12-year-old children and then again after
puberty. Experimental studies by Bourdon (1894) and Chamberlain (1915) suggested
1880–1935 13

considerable increases in memory capacity during the late elementary school years,
with only minor further improvements until early adulthood (see also Nagy 1930).
By contrast, Meumann (1907) own studies of children and adults showed rather slow
improvements in immediate memory until the age of 13, followed by rapid improve-
ments between the ages of 13 and 16. Peak performance was achieved by the age of
25 and followed by a period of stagnation and stability until the age of 46 (i.e., the
oldest participants in his study). Despite these differences in results, researchers have
agreed that memory development is characterized by nonlinear trends and that devel-
opmental patterns vary as a function of material and memory task.
Although studies on STM dominated the field at this time, there were also
investigations of LTM and forgetting. Interestingly, one of the first studies of
children’s and adults’ long-term retention and forgetting of verbal materials
(Radossawljewitsch 1907) was stimulated by criticisms of Ebbinghaus’ experi-
mental method. Meumann, Radossawljewitsch’s advisor, doubted Ebbinghaus
(1885) classic findings on forgetting curves because neither Meumann nor
his collaborators had been able to replicate them and because they were not
in accord with the experiences of everyday life and work. The participants in
Radossawljewitsch’s study (16 adults and 11 children between the ages of 7 and
13) learned nonsense syllables and meaningful poems and were tested on immedi-
ate memory and relearning after lapses of 5, 20 min, 1, 8 h, and subsequently after
2, 6, 14, 30, and 60 days. Compared to adults, children needed a very large num-
ber of repetitions to learn a series for the first time, but the children forgot less of
the material that was learned, and their rate of forgetting seemed less than that of
adults. Although the slopes of forgetting were different for children and adults, the
most important finding was that the forgetting curves obtained for the two groups
did not correspond to the curve obtained by Ebbinghaus. The discrepancy between
these two sets of data was great—Ebbinghaus who used himself as a subject forgot
more information in 1 h than adults in Radossawljewitsch’s study did in 8 h.
Subsequent studies by Vertes (1913, 1931) using the word-pair method men-
tioned above assessed LTM and forgetting in children and adolescents 6–18 years
of age. Again, there was no support for Ebbinghaus’ findings because forgetting
occurred at much slower rates. Although the curves were generally in accord with
the assumption that forgetting is a decelerating function of time, they were consid-
erably flatter than those reported by Ebbinghaus. Interestingly, children older than
10 years of age remembered more after 1 week than they had on the previous tests,
a phenomenon that is now called reminiscence.
All in all, the developmental studies on LTM and forgetting revealed that learn-
ing and forgetting rates differed as a function of age. Whereas it took children con-
siderably longer than adults to learn the stimulus lists, they did not forget at faster
rates. The fact that Ebbinghaus’ findings could not be replicated was assumed to
be largely due to the relatively rapid rate of self-presentation in that study and
the fact that Ebbinghaus was the subject of his studies and the experimenter at
the same time. An alternative explanation is that Ebbinghaus forgot so much so
quickly because he experienced a high amount of interference from the many lists
he had learned previously (i.e., proactive interference; Underwood 1957).
14 2  A Brief History of Memory Development Research

Children’s Eyewitness Memory

Although there was little interest in children’s testimonial competence in the

United States at the beginning of the twentieth century, systematic research on tes-
timonies by both adults and children flourished in Europe, particularly in Germany
and France (for a detailed review, see Ceci and Bruck 1993). Early reviews by
Whipple (1909, 1911) covered child’s suggestibility research in Europe, indicating
that young children are highly suggestible and capable of making serious errors in
testimonies, even when reporting on events of significant personal relevance.
One of the earliest studies on children’s testimonies was carried out by Binet
(1900). Binet assumed that suggestibility reflected the impact of two classes of fac-
tors. The first class was labeled “autosuggestion” and referred to the influence of a
prominent thought that develops within the individual without obvious external
cause. Autosuggestion was supposed to paralyze the critical senses and memories of
past events. The second class of factors was assumed to be external to the individual,
reflecting mental obedience to significant others, particularly adults. Although Binet’s
autosuggestion techniques are no longer of interest in modern studies, his research
paradigms for examining external forces are still used today. For instance, children
carefully examined a number of objects for a short time and were then tested on their
memory. Some of the children had to write down everything they had seen, whereas
others were asked direct (unbiased) or mildly leading questions. The major finding
was that free recall yielded the most accurate reports, followed by direct questioning.
Highly misleading questions resulted in the most inaccurate statements, regardless
of age. Binet ascribed children’s suggestibility to social factors, namely their desire
to comply with adult beliefs, thereby ignoring their own accurate memories. Most
important, Binet’s work also demonstrated that suggestions were not always incorpo-
rated into the memory record, thus demonstrating that children’s suggestibility was
not long-lasting as they were still able to realize their errors on subsequent occasions.
As emphasized by Ceci and Bruck (1993), Binet’s major contributions to the field
were that (1) he distinguished between errors of reporting caused by actual memory
changes versus those caused by social conformity and (2) his early research is of high
importance for the current debate regarding whether the original memory trace is
itself impaired or simply allowed to “coexist” with the traces produced by suggestion.
In Germany, early research on eyewitness testimony was stimulated by Clara and
William Stern’s observations of their own children (Stern 1910, 1914; Stern and
Stern 1913). They explored their children’s abilities to remember in tandem with
the children’s abilities to resist suggestion. Stern and Stern distinguished “memory
deceptions” from intentional lies, defining such deceptions as objectively incorrect
statements that were based on subjective confidence. The problems that young chil-
dren had with correctly recalling the circumstances and contents of previous expe-
riences were attributed to their poorly developed understanding of time. Preschool
and kindergarten children not only lack the ability to coordinate experienced events
with specific time markers, but they are also not interested in such relations. Given
that most questions in eyewitness testimony situations address the time issue, it does
1880–1935 15

not come as a surprise that children are prone to err under these circumstances. For
instance, the question of whether the child saw the policeman yesterday on her way
to kindergarten will immediately activate the image of the policeman and lead to a
positive response regardless of whether the policeman was seen today, yesterday, or
a week ago. According to Stern (1914), the major finding of research on young chil-
dren’s testimony is that spontaneous reports provide much more correct information
than specific questions even though children do not offer as much information spon-
taneously. In his view, interviewers need to know that specific questions can pro-
duce two different effects: A positive effect is that a child may remember a correct
detail that had not been offered spontaneously. On the negative side, a child may
generate answers that are based on diffuse memories of the event and that are mainly
incorrect. The latter possibility is highly probable during inquisitory assessments
particularly when questions are leading and suggestive. In such cases, the length of
the child’s path to the answer “yes” seems much shorter than the path to the answer
“no,” with the suggestibility of the child decreasing with increasing age.
In order to support his assumptions, Stern developed two types of experiments
that reflected two different paradigms and that are still in use today. In the first
experiment, participants were shown a picture and asked to study it for a short
period of time. They were then asked to recall what they had seen and were also
asked a series of direct (unbiased) and misleading questions. The second experi-
ment was developed to represent situations that were closer to real life. Here,
participants observed staged events and were then asked questions about details
from the scenario. Several of Stern’s observations seem still relevant today. For
instance, he cautioned against repeated questioning of the same event, claiming
that a person may better remember the answers he or she gave during the first
memory assessment than the actual events themselves (Stern 1910). Also, his
research indicated that although younger children were most suggestible to mis-
leading information, even adults could be misled by suggestive questions.
It seems fair to state that Stern and Stern’s (1909) exploration of children’s
memory and suggestibility were inspired by issues related to children as witnesses
in legal settings, anticipating many of the core research themes of recent work on
children’s eyewitness testimony. Thus, as noted by Ceci and Bruck (1993) as well
as by Ornstein and Elischberger (2004), early European work on eyewitness testi-
mony conducted by the Sterns and their colleagues foreshadowed a large propor-
tion of the findings that were to appear in modern literature.

Assessment of “General” Memory Development

Given the broad disagreement regarding the general course of memory develop-
ment described above, Brunswik et al. (1932) conducted a developmental study
that was aimed at providing a general description of STM and LTM in school-
age children and adolescents. This study also differed from earlier investigations
in that the issues addressed were directly derived from truly developmental theory;
16 2  A Brief History of Memory Development Research

that is, Charlotte and Karl Bühler’s doctrine of phases and stages (e.g., Bühler
1930). The research was stimulated by Charlotte Bühler and carried out at the then
famous Vienna department of psychology. The study was also unique compared
with previous research in that statistical significance tests were used, mainly due
to the assistance of Paul Lazarsfeld, who became internationally known for his
methodological expertise several years later. A large variety of memory tasks were
presented to a sample of about 700 participants, ranging from 6 to 18 years of age.
Tests involved STM and LTM for nonsense and real words, colors and numbers,
as well as memory for poems. Moreover, several nonverbal memory tasks such as
memory for motor actions and their correct sequence were included.
Undoubtedly, the study by Brunswik et al. (1932) represents a valuable con-
tribution to research on memory development. The use of more precise methods
and various learning materials gave rise to more specific hypotheses concerning
age differences in memory development. The disparate growth curves obtained
for different memory functions were consistent with the data in previous studies
(e.g., Netschajeff 1900, 1902; Offner 1924). The attempt to construct a curve of
the general development of immediate memory (“memory strength”) is particu-
larly interesting. The curve was based on scores of all participants and represented
an aggregation across all measures included in the study (see Fig. 2.1 in Schneider
and Pressley 1997, for a reconstruction of the graphical representation). The out-
comes were compatible with findings from other early studies in that linear and
steep increases in memory performance were found from 6 to 11 years of age and
that there was a plateau in performance during pre- and early adolescence.
One obvious problem with the study was that the authors tried hard to make
their findings compatible with the Bühlers’ perspective. Accordingly, they claimed
to have found support for Bühler’s position that memory development during the
early years is dominated by rote-associative processes (“mechanical learning”),
whereas the kinds of learning and memory predominant in older children and adults
are based on the creation of meaning (“logical memory”). A closer inspection of
findings revealed that the theoretical position was not entirely supported by the
data: In fact, younger children required more practice to learn nonsense syllables
than meaningful words with continuous improvement in the learning of nonsense
syllables up to age 18. Nonetheless, subsequent reports on the Brunswik et al. study
emphasized the qualitative shift from mechanical to logical memory (for instance,
see H. Werner’s comments on the findings by Brunswik et al. in the thoroughly
revised third edition of his textbook, which appeared in 1953). There is no doubt
that this perspective of memory development dominated the field in the forties and
fifties, particularly in Germany. However, the basic assumptions by Brunswik et al.
(1932) were subsequently questioned by Russian psychologists such as Rubinstein
(1973; originally published in 1940) and Smirnov (1948; cited in Smirnov 1973)
and later also empirically falsified by Fechner (1965), Smirnov (1973), and Weinert
(1962). For instance, Weinert found that 6-year-olds learned word pairs com-
posed of familiar words much more easily than they learned pairs consisting of
meaningless unfamiliar syllables. Smirnov’s work further indicated that the supe-
riority of logical memory over rote memory was more pronounced in elementary
1880–1935 17

schoolchildren than in adolescents. Thus, whereas the developmental curve regard-

ing improvements in memory performance was in accord with many findings of
the early period and also validated in subsequent investigations, the assumptions
concerning qualitative shifts in memory development favored by the Bühlers and
Brunswik and colleagues were not compatible with the existing database.

1936–1965

Overall, this time period represents the “dry middle years” (Ceci and Bruck 1993)
not only for research on children’s eyewitness memory but for research on mem-
ory development in general. However, the situation differed somewhat for the then
leading research communities as will be shown below.

German Research

The great progress made by German researchers in the early twentieth century came
to a halt as war exploded across Europe. Many leading child psychologists such as
the Bühlers, Koffka, the Sterns, and Heinz Werner left the country and began new
careers in the United States. Theoretical perspectives that predominated in the imme-
diate postwar period (i.e., Gestaltist, phenomenological) did not encourage analytical
research on memory development. As a consequence, no empirical studies on mem-
ory development were published in Germany between 1933 and 1961.

American Research

The situation was much different in America. Behaviorism dominated the field,
and theories of verbal learning were very popular. The verbal learning theorists
were not particularly interested in developmental issues, however, because their
primary concern was the identification of general laws (e.g., Goulet 1968; Keppel
1964). They were inclined to assume that the laws of memory are the same at all
ages. Consequently, Keppel (1964) stated in his review that the developmental
study is of little interest to the verbal learning theorist unless differential results
could be expected on theoretical grounds. In this sense, the conclusion made
by contemporary authors (Brainerd and Pressley 1985; Kail and Strauss 1984)
regarding a dearth of research on memory development prior to 1965 applies to
American developmental psychology between 1936 and 1965.
Most of the verbal learning studies with children were descriptive, and many
were conducted at a single age level, but there were a few notable exceptions.
For instance, Koppenaal et al. (1964) used a paired-associate learning paradigm
18 2  A Brief History of Memory Development Research

to test the developmental hypothesis that older children should experience greater
interference when tested on highly associative learning materials than younger
children. This assumption was based on the observation that verbal associative
strength is determined by children’s prior knowledge and that prior knowledge
is richer for older compared to younger children. Koppenaal et al.’s (1964) study
was conducted with preschoolers, kindergarteners, and third graders. The results
were consistent with the developmental hypothesis in that third graders showed
significant retroactive and proactive inhibition, which was not true for the younger
children.
Another important developmental study focusing on free recall was conducted
by Bousfield et al. (1958). These authors compared the degree of associative clus-
tering in free recall of third grade, fourth grade, and college students. The theoreti-
cal rationale was derived from Werner’s assumption that organizational principles
change during development such that perceptual-sensory categorization is ini-
tially dominant but is replaced later by a tendency to organize stimuli conceptu-
ally. An important aspect of this study was that it included measures of clustering
presumed to tap processing more directly than simply the amount of material
recalled. As a main result, it was demonstrated that both recall and conceptual
clustering increase as a function of age. Contrary to expectations, color clustering
was low in all age groups. Although the findings concerning color clustering were
not in accord with Werner’s theory, this did not necessarily contradict the theory
given that the age range was restricted and that younger children (e.g., preschool-
ers and kindergarten children) were not included in the sample.
Although developmental changes in memory were not the core concerns of North
American verbal learning researchers, the findings by Bousfield et al. (1958) and
Koppenaal et al. (1964) provided evidence for the importance of developmental dif-
ferences, thus falsifying the view that the processes mediating verbal learning and
memory at different age levels are qualitatively identical. In fact, Goulet’s (1968)
review of verbal learning in children published only 4 years after Keppel’s overview
differed considerably from the latter, pointing to the lack of correspondence between
data on children and adults. The few developmental studies carried out at the end of
the 1950s or the beginning of the 1960s already anticipated a great deal of work that
was conducted after 1965 on children’s paired-associate and list learning.

Russian Research

The research situation in Russia was clearly different from those in Germany and
the United States. Relative to the Western researchers of the day, Russian scientists
placed more emphasis on the development and evaluation of particular theoretical
positions (cf. Meacham 1977). For instance, Vygotsky’s position that highlighted
the importance of social origins for the development of higher mental functions
such as attention, memory, and volition was already influential. Russian develop-
mental researchers were particularly interested in the development of “logical”
1936–1965 19

(meaningful) memory compared to rote “mechanical” learning. They emphasized

the relevance of conscious, independent, and goal-oriented memory activity as the
causal origin of memory development (see reviews by Meacham 1977; Smirnov
and Zinchenko 1969). Two major concerns can be distinguished in these empirical
studies. One was the study of involuntary memory, which occurs when the ulti-
mate goal of the person is not memory but something else, often comprehension in
the Russian studies. In such cases, memory is involuntary in that it is a by-product
of comprehension. The second focus was voluntary memory, which was defined
as a product of activities that are driven by the goal to remember. Most Russian
researchers believed that superior (higher order) forms of memory develop on the
basis of the transition from natural involuntary memorizing to subject-controlled
and voluntary memory involving the use of mediating processes and cues.
A classic example of research on voluntary memory was provided by Istomina
(1948, cited in Istomina 1977). This study sought to determine the conditions
under which voluntary memorizing and recall first emerge. In a laboratory condi-
tion, the experimenter read five words to the children (preschoolers and kinder-
garteners), and the children had to recall the words after a short delay. In a game
condition, the children were asked to go on an errand to a store and buy five items
for the kindergarten. As a main result, children’s memory performance was much
better for the shopping task. Istomina concluded from this that recall differences
were due to different motivational incentives for the children. Whereas remember-
ing was an intrinsically important goal and had real meaning in the shopping situa-
tion, this was not true in the laboratory situation.
One typical feature of the Russian studies on voluntary memory was the focus on
learning from text. For example, Korman (1944, 1945, cited in Yendovitskaya 1971)
studied preschool children’s memory for connected material (fairy tales). Korman
was impressed by the children’s ability to recall the main events of the story cor-
rectly. Recall differences between 4-year-olds and 5-year-olds and 6-year-olds were
primarily quantitative. Subsequent research by Smirnov (1948; cited in Smirnov
1973) with schoolchildren focused on the impact of text structure on children’s
recall. Participants (second, fourth, and sixth graders) were presented two texts with
sentences organized in either a coherent fashion or randomly. After two attempts
to recall the two texts, children were instructed to break up the texts into pieces of
information that went together (done from memory). The most important finding
was that the youngest children were not able to reorganize the random text and that
only a small percentage of the older participants were able to do so. Interestingly,
although most of the children could indicate that random texts were harder to break
up than coherent texts, they were not consciously aware of the structural differences
between the texts. Smirnov concluded that meaningful grouping processes first pro-
ceed in an unconscious way before becoming conscious activities.
Several studies investigated the relation between voluntary and involuntary mem-
ory in children. Research by Leontjev, Smirnov, Zinchenko, and others revealed that
involuntary memorization instructions produced better recall than voluntary memo-
rization instructions for preschoolers, whereas this was not true for schoolchildren
and adults (for details, see Schneider and Pressley 1997). Correlational studies
20 2  A Brief History of Memory Development Research

undertaken by Istomina and others indicated that correlations between involuntary

and voluntary memory varied considerably with the age of the participants and the
type of task used. Aggregated across tasks, the mean correlation was 0.80 for pre-
schoolers but only 0.45 for kindergarteners. Mean correlations decreased steadily up
to Grade 8 (0.10) but increased again up to 0.25 for college students. The interindi-
vidual differences were explained in the sense that whereas the younger preschool
children behaved the same way when given involuntary and voluntary memorization
instructions, older children’s and adults’ memory behaviors differed considerably as
a function of experimental condition, with more memorizing activity and higher lev-
els of recall when voluntary memorization instructions were provided.
All in all, it seems fair to conclude that most of the research on memory devel-
opment conducted between 1936 and 1965 took place in Russia. In particular, the
outcome of this research improved our knowledge about preschoolers’ memory
activities, foreshadowing later investigations on children’s implicit and explicit
memory. The Russian finding suggested that preschool children generally do not
show intentional memory behaviors when instructions to remember are provided.
However, their memory could be improved by manipulations that increased their
meaningful processing of materials. They also indicated that the first signs of goal-
directed memory activities in the memorization of objects, pictures, and words
were observed among 5-year-olds, although children of this age usually did not
possess the means to achieve their memory goals.
Other major accomplishments of this period include the analysis of developmen-
tal differences in children’s text processing and recall and the study of processes that
influence the encoding and storage of verbal materials, such as repetition and self-
checking. Whereas the repertoire of memorizing methods seemed rather restricted
in kindergarteners and young schoolchildren, the number and quality of mediating
activities increased considerably between the second and fifth grades. In accordance
with earlier findings (e.g., Brunswik et al. 1932), memory development appeared to
be more dramatic for the elementary school years than for later periods.

Transition to the Modern Era

How does modern research on memory development differ from the histori-
cal approaches? One of the crucial differences concerns a shift from an emphasis
on describing developmental differences in memory to an emphasis on identifying
the underlying mechanisms of change. Another difference concerns the theoretical
framework used. Since the mid-sixties, research on memory development has been
influenced strongly by theoretical models derived from information processing and
neuroscience approaches (see the reviews by Bauer 2006; Kail 1990; Schneider and
Bjorklund 1998; Schneider and Pressley 1997). Memory researchers were clearly
affected by the “cognitive revolution” that was taking place in experimental psy-
chology around that time (see Miller et al. 1960; Neisser 1967). Developmental psy-
chologists began looking at changes in children’s thinking in terms of a computer
Transition to the Modern Era 21

metaphor. From this perspective, memory development can be seen as reflecting

either hardware (the capacity of memory systems and the speed with which infor-
mation can be processed) or software (e.g., use of strategies). From the mid-sixties
on, memory development has been conceptualized in terms of interacting informa-
tion processing components, with memory development as a product of develop-
ing components in ever more complex interactions. Developmental research was
strongly influenced by multistore memory models that distinguished between a sen-
sory register, a short-term store, and a long-term store (e.g., the model developed by
Atkinson and Shiffrin 1968). Regarding the contents of the long-term store, it was
proposed that information can be represented in two ways. Explicit or declarative
memory refers to our capacity for the conscious recollection of names, places, dates,
or events and comes in two types, episodic and semantic. Whereas episodic memory
concerns events and experiences that can be consciously retrieved, semantic memory
refers to our knowledge of language, rules, and concepts. By contrast, implicit or
procedural memory represents a variety of nonconscious abilities, including the
capacity to learn habits and skills and some forms of classical conditioning.
During the 1970s, an extensive literature on memory development in infants,
children, and the elderly began to accumulate. The first line of research was
already initiated by John Flavell in the mid-sixties, who investigated the develop-
ment of memory strategies, looking for “mediators” that influenced children’s ver-
bal memory performance (e.g., Flavell et al. 1966). When Flavell (1971) asked in
a symposium held at the biennial meetings of the Society for Research in Child
Development (SRCD), “What is memory development the development of?” he set
a new field in motion (see Bauer and Fivush 2014). Reports of memory develop-
ment research that had previously been limited to occasional papers and sympo-
sia at SRCD meetings occupied a solid chunk of SRCD programs from 1975 on.
By the late seventies, there was clear evidence for an active community of devel-
opmental memory researchers as documented in the volumes by Kail and Hagen
(1977) and Ornstein (1978). As noted by Miller (2014), hundreds of studies con-
ducted during the 1970s and 1980s investigated age differences in the spontaneous
production of memory strategies such as rehearsal and clustering and the impacts
of task and instructional differences on the outcomes. The development of strate-
gic competence in children was probably attractive to memory researchers because
it seemed at that time that there were qualitative differences in strategy use as a
function of time. Moreover, numerous cross-cultural studies explored the issue of
whether strategy development is universal, that is, not only observed in America
and Europe but also evident in non-Western samples (for reviews, see Rogoff and
Mistry 1985; Schneider and Pressley 1989). Most of these studies documented the
impact of schooling on memory development, demonstrating that strategy use was
either less frequently or not at all observed in nonschooled samples of children.
A second line of research that began in the mid-seventies and was again ini-
tiated by John Flavell focused on the developmental processes that underlie the
acquisition of memory strategies. The basic idea behind this approach was that
the application of memory strategies should lead to insight and knowledge about
the usefulness of such techniques and that children’s understanding of memory
22 2  A Brief History of Memory Development Research

processes (i.e., their metamemory) should in turn facilitate their strategy use
(Flavell and Wellman 1977). Although early studies on the relation between mem-
ory and metamemory were disappointing in that they yielded rather low correla-
tions, subsequent research based on more sophisticated theoretical models of the
relation and also on more advanced methods provided more promising results,
underlining the importance of general metacognitive knowledge for memory
development and noting that this metacognitive knowledge seemed to increase
with age. Subsequently, researchers also explored the impact of domain-specific
knowledge on memory development, demonstrating that advanced knowledge of a
subject domain (such as chess or physics) was closely related to superior strategy
use and better memory of materials in that particular domain (e.g., Chi 1978).
A third line of research that began in the early 1970s focused on STM and LTM
development in children and adolescents. This research investigated the development
of “basic processes,” that is, the development of encoding and retrieval processes
and their relative contributions to memory development. In the beginning, the domi-
nant view of STM was in line with the early work on memory span conducted at
the turn of the century and mainly based on the Atkinson and Shiffrin (1968) mod-
el’s assumptions that information is read into a limited-capacity short-term store,
in which it may be maintained by rehearsal and then forwarded to LTM. This con-
ception of STM was challenged and abandoned by Baddeley and Hitch (1974) who
replaced it with a fragmented conception of “working memory,” which consisted
of three components: a “central executive,” which can be conceived of as a control
center that selects and operates on various processes; a “phonological loop,” which
maintains verbal stimuli through subvocal rehearsal; and a “visuospatial sketchpad,”
a buffer responsible for the processing of nonverbal information. Several researchers
subsequently assumed that STM and working memory were separate and function-
ally distinct systems (see Case 1978; Dempster 1985). Whereas STM was conceived
of as a rather passive storage system, working memory was supposed to act as a sort
of central computing space where information being held in the short-term system
was transformed. Although the concept of STM was discussed controversially at that
time, there was general agreement that its capacity is limited, probably increasing
with age, and that information not further processed in STM will be lost.
Later on in the 1980s, another line of memory research that was not based on
experimental studies but on interviews in everyday environments was stimulated by
Katherine Nelson (1986) and her colleagues. The focus was on children’s “scripts,”
that is, their memory of specific recurring events such as meals, nighttime rituals, or
birthday parties. Research carried out in the 1980s and 1990s showed that even very
young children organize their recall of everyday routines in a script-like fashion. As
noted by Baker-Ward and Ornstein (2014), research on scripts led naturally to the
study of children’s autobiographical memory of events that had been experienced
only once. This kind of research has attracted a lot of attention among memory
researchers, and it is probably fair to state that autobiographical memory has become
the most active area of research on memory today (Miller 2014).
In sum, there is no doubt that the “cognitive revolution” of the 1960s as well
as Flavell’s pioneering work on memory strategy development led to a dramatic
Transition to the Modern Era 23

Number of Peer Reviewed Articles 2500

2000

1500

1000

500

5-Year Period

Fig. 2.1  Number of peer-reviewed journal articles on children’s memory by 5-year publication

period (slightly modified after Baker-Ward and Ornstein 2014, p. 50)

increase in research on children’s memory. There are several ways to demonstrate

that the interest in this research paradigm is still growing. For instance, Bauer
and Fivush (2014) refer to the fact that the two volumes of the 1983 Handbook
on Child Development, a required reading for all developmental psychologists,
had only one chapter on “learning, memory, and understanding” (Brown et al.
1983), whereas their own new handbook consists of 43 chapters reviewing differ-
ent aspects of memory development. Another demonstration comes from an analy-
sis of the PsycINFO abstracting and indexing database carried out by Baker-Ward
and Ornstein (2014), in which they identified all peer-reviewed English-language
articles published between 1950 and 2010 including the keywords “memory” and
“children (0–12).” The findings are summarized in Fig. 2.1. As can be seen in this
figure, the number of publications assessed in 5-year intervals increased consider-
ably from the mid-sixties on, with another rapid increase in production after 1990.
There is little doubt that the focus of research on memory development has shifted
over the years, from laboratory research on the impact of memory capacity, strate-
gies, and knowledge to more applied field research on autobiographical memory
and eyewitness memory. Laboratory research on memory development has con-
tinued to exist, with the predominant cross-sectional studies complemented by
longitudinal work. Compared with the situation at the beginning of the research
paradigm in the 1960s and 1970s, the situation today is much more complex, and
new research areas such as evolutionary perspectives, memory in atypical popula-
tions, and the study of the brain basis of memory have emerged.
Chapter 3
The Development of Implicit Memory

By definition, implicit memory is “memory without awareness” or memory for

information without conscious awareness that one is remembering (Jacoby 1991;
Schacter 1992). This term was first proposed by Graf and Schacter (1985) and refers
to an old idea, namely the notion that people can demonstrate the aftereffects of
experiences in their behavior without being able to consciously recollect the expe-
riences themselves (see also Roediger 1990). In experiments on memory, implicit
memory is revealed when previous experiences facilitate performance on a task that
does not require the intentional memorizing of these experiences (Schacter 1987).
Such a facilitation effect is typically referred to as priming. A similar concept was
introduced in scientific psychology in the 1930s by Soviet researchers who inves-
tigated “involuntary” or “incidental” memory, which was supposed to occur when
the ultimate goal of a person was not memory but something else, often comprehen-
sion (cf. Chap. 2 of this volume). Researchers such as Leontjev (1931) assumed that
superior forms of memory develop on the basis of the transition from natural and
involuntary memorizing to subject-controlled and voluntary memory.
Similarly, modern memory researchers believe that implicit or procedural (non-
declarative) memory refers to an evolutionarily older memory system and that it
can be contrasted with explicit declarative memory (see Bjorklund and Sellers
2014). Tulving (1985) referred to such implicit procedural knowledge as anoetic.
Associative learning principles as described in classical and operant conditioning
are good examples of procedural memory as they require no conscious awareness
but represent the enactment of a mental representation created from previous expe-
rience. As noted by Bjorklund and Sellers (2014), such memories do not require
the engagement of the self but only the application of a previously experienced
and then encoded connection between two stimuli.
In comparison, explicit memory refers to memory with awareness, or the con-
scious recollection of facts and events, and can be tested directly by measuring
recall and recognition. In explicit memory tasks, participants are typically aware

© Springer International Publishing Switzerland 2015 25

W. Schneider, Memory Development from Early Childhood Through
Emerging Adulthood, DOI 10.1007/978-3-319-09611-7_3
26 3  The Development of Implicit Memory

Memory

Declarative (explicit) Procedural (implicit)

episodic semantic

motorskills acquired classical perceptual verbal

autonomic conditioning learning priming
responses

Fig. 3.1  A proposed taxonomy of the memory system

that they are making a memory decision. Explicit memory tasks tap the episodic
and semantic components of the declarative memory system. For instance, peo-
ple refer to intentionally assessed memories when trying to remember which items
they should buy at the store today or when trying to identify the year in which they
acquired their driver’s license. Figure 3.1 provides a taxonomy of the memory sys-
tem suggested by Parkin (1997).
The current interest in implicit memory originated in neuropsychological
research on patients with amnesia. These patients typically display intact short-
term memory, but their long-term memory, particularly for new verbal material, is
considerably impaired. Interestingly, several studies have found that even though
such patients performed very poorly on explicit memory tests such as recall or rec-
ognition, they showed effects of priming (e.g., Graf et al. 1985, Exp. 1; Schacter
1992; Warrington and Weiskrantz 1968). Thus, there is evidence that the explicit
and implicit memory systems are neurologically distinct. Although patients with
damage to the hippocampus are unable to transfer new explicit information to their
long-term memories, they can retain implicitly acquired information.

Development of Perceptual Priming

Developmentally, it has been claimed that implicit memory is present from the
beginning of life and does not change much over the years (see reviews by Lloyd and
Miller 2014; Lloyd and Newcombe 2009; Parkin 1997; Rovee-Collier et al. 2001).
This conclusion was drawn from studies on perceptual priming. Most of these
Development of Perceptual Priming 27

developmental studies involved the use of fragmented pictures—perhaps pictures

of a dog—that children were asked to identify. This is very difficult to do initially,
but as more of the picture is completed, it becomes increasingly easier to identify
the objects in the pictures. After a series of such picture identification tasks has been
given, the children are provided with degraded pictures of both previously seen and
unseen objects. The typical finding is that repetition priming is observed. That is,
children identify fragmented pictures of previously seen pictures much faster than
fragmented pictures of previously unseen objects (see Ausley and Guttentag 1993).
Much of the initial work on the development of implicit memory applied para-
digms typically used by adults when working with children. In one of the first devel-
opmental studies on perceptual priming, Parkin and Streete (1988) exposed 3-, 5-,
and 7-year-old children plus adults to a series of degraded pictures. Participants were
required to name each picture, and if they failed, successively more informative pic-
ture versions were provided until the pictures could be identified. Some time later,
after retention intervals of 1 h, 2 h, or 2 weeks, participants were retested using the
same task materials. Retention was assessed by representing the pictures and meas-
uring the degree of savings in renaming pictures successfully (i.e., speed of identi-
fication for Trial 1 minus the speed of identification for Trial 2). Participants were
also given a test of their ability to recollect seeing the pictures before as a measure of
explicit memory. As a main result, Parkin and Streete found that the absolute amount
of savings increased with age. However, when the degree (proportion) of savings
was used as an index of implicit memory, priming was found to be age-invariant.
As noted by Parkin and Streete (1988), these findings had to be treated with caution
because older children and adults inevitably produce higher baseline performance
and thus have less potential to show priming effects. Findings regarding the explicit
memory (recollection) measure showed developmental differences such that savings
were higher for recognized pictures except in the youngest group in which recogni-
tion was at chance. Given that the youngest group showed high levels of savings,
one could conclude from these findings that implicit memory is effective before
explicit memory for the same information. However, firm conclusions about devel-
opmental trends in implicit memory could not be made because of concerns that
explicit memory may have contaminated the results.
In a subsequent study, Naito (1990) used a slightly different approach, study-
ing perceptual priming in three groups of children (6, 8, and 11 years) plus adults.
Participants had to answer either a physical or an elaborative question about a
series of words followed by either a word completion test or a free-recall test for
the words. There was a reliable priming effect regardless of age. Even though the
absolute number of completed words increased with age, the degree of priming in
proportional terms was approximately equal at each age level. This was in contrast
to the measure of explicit memory, which increased with age, with the effects being
most apparent for elaborative encoding. Explicit word recognition memory generally
deteriorated as the retention interval increased. By contrast, the priming advantage
on a fragment-completion task did not change as a function of the retention interval.
Findings regarding the developmental invariance of perceptual priming have
not always been consistent. For instance, Drummey and Newcombe (1995) used a
28 3  The Development of Implicit Memory

paradigm similar to that of Parkin and Streete (1988) with groups of 3- and 5-year-
old children as well as adults. Pictures were exposed in a more naturalistic setting
(pictures from a children’s book) using a technique by which the images of the
target items changed gradually from blurry to clear. The degree of priming, that
is, higher masking level for identification for studied versus unstudied pictures,
was fairly consistent for the two groups of children, although there was a trend for
5-year-olds to show larger priming effects. Surprisingly, adults showed lower lev-
els of priming than the 5-year-olds. The participants were also asked for recogni-
tion judgments, and this measure of explicit memory showed clear age effects.
Drummey and Newcombe interpreted their findings as supporting the assumption
that implicit memory is relatively robust across development but acknowledged the
need for caution in interpreting the priming effects, given that explicit memory may
have contaminated the implicit memory measures (which could explain the unex-
pected findings for adults). Also, given the increase in priming of 48 % between
3 and 5 years, one cannot comfortably exclude developmental trends in implicit
memory in early childhood (Parkin and Streete 1988). However, findings supporting
the developmental invariance hypothesis of perceptual priming were also reported
by Hayes and Hennessy (1996) who used an experimental paradigm similar to
Drummey and Newcombe’s with children who were 4, 5, and 10 years of age. Hayes
and Hennessy did not find any difference in the degree of priming between the three
age groups. A dissociation between implicit and explicit memory was also found in
this study such that recognition memory increased significantly from age 4 to 10.
Further evidence supporting the developmental invariance hypothesis came
from a study by Ellis et al. (1993) involving faces of classmates. Children from
three age groups (5, 8, and 11 years) first viewed pictures of classmates and made
decisions about expressions and gender. In the test phase, they were shown a
sequence of faces (half of which had been presented in the first phase) and asked
to decide as quickly as possible whether they knew the person or not. Findings
indicated that all three age groups showed priming in that recognition of previ-
ously exposed faces was faster than that of the familiar faces only shown in the
test phase. The authors were able to replicate this result in a second experiment
with 5-year-olds and adults. Given that there was a large effect of age on reaction
time but no interaction between age and the extent of priming, the authors con-
cluded that the observed priming effect is age-invariant (for confirming evidence,
see Lorsbach and Morris 1991; Newcombe and Fox 1994; Perrig and Perrig 1993).
Given that the studies described above were inconclusive to some extent
because the contaminating effects of explicit memory could not always be ruled
out, Russo et al. (1995) introduced additional design elements that represented
methodological improvements. In their experiment, two groups of children (4 and
6 years of age) and a group of young adults were exposed to a picture recogni-
tion procedure similar to the one used by Parkin and Streete (1988). The main dif-
ference between the two studies was that unlike Parkin and Streete, Russo et al.
(1995) did not present picture fragments during the study phase but presented only
complete objects instead. Accordingly, participants could not explicitly remember
the fragmented stimuli. Russo and colleagues calculated priming scores for items
Development of Perceptual Priming 29

that participants explicitly recalled having seen during the study phase and also
for items that participants failed to remember seeing. Using this procedure, Russo
et al. (1995) showed that there were no age differences in priming between the two
groups of children and the adults when the analysis was restricted to the pictures
that participants could not recall from the original learning episode. Thus, similar
levels of priming across age groups were found when explicit memory influences
were removed from the data. Subsequent studies (e.g., Billingsley et al. 2002;
Perez et al. 1998) also presented findings supporting the developmental invariance
hypothesis. For instance, Billingsley et al. (2002) showed that children and adults
performed similarly on two different implicit memory tests that required either the
generation of word categories or the identification of pictures.
The overall pattern of results described above supports the view that perceptual
priming is developmentally invariant (see Ausley and Guttentag 1993; Schneider
and Pressley 1997). However, summaries of subsequent reviews of the literature
have not been completely consistent. Although Parkin (1997) concluded that
implicit memory is operational at an early point in development and appears age-
invariant, he also pointed to methodological pitfalls such as the failure to take
age-related differences in baseline scores into account. For instance, Ellis et al.’s
(1993) and Lorsbach and Morris’s (1991) conclusion that priming effects in their
studies were age-invariant seems difficult to justify, given that there were large
age effects in reaction times. Thus, it is possible that potentially greater priming
effects in older participants may have been masked by a ceiling on reaction time
speeds. In Parkin’s view, only the study by Russo et al. (1995) met methodological
standards and gave good evidence of age invariance in implicit memory, but more
research is needed to support this assumption.
The review of the literature presented by Lloyd and Newcombe (2009) about
10 years later was based on a larger set of data and led the authors to conclude that
the available evidence continues to suggest that perceptual priming is relatively
stable throughout development. The only exception to this rule that they found
in the literature was Cycowicz et al.’s (2000) study, which reported evidence for
developmental improvements between the ages of 5 and 9 on a picture identifica-
tion task. According to Lloyd and Newcombe, this discrepancy in findings could
be attributed to the facts that the task used by Cycowicz et al. (2000) was harder
than that used in most other studies and that it puts participants under time pres-
sure. One general problem concerning the research on implicit memory noted by
Lloyd and Newcombe was that the available studies that tested the developmen-
tal invariance hypothesis did not include children younger than 3 years of age.
Although the findings of some studies suggest that priming effects already exist
in infancy (e.g., Rovee-Collier 1997; Webb and Nelson 2001), true developmental
invariance beginning with infancy has yet to be assessed.
Interestingly, the summary of research findings provided by Lloyd and Miller
(2014) only a few years later sounds slightly different. Although Lloyd and Miller
emphasized the fact that studies on perceptual priming carried out in the 1990s
showed little difference in priming effects between younger and older children, they
also referred to more recent research that suggested that such a conclusion may be
30 3  The Development of Implicit Memory

premature. Their chapter includes a section on evidence for developmental shifts in

perceptual priming, using the Cycowicz et al.’s (2000) study as an example for a meth-
odologically improved investigation of priming effects. As indicated above, the data
from this study suggest that whereas 5-year-old children are capable of performing
a difficult implicit memory task, they do not perform it with the same efficiency as
9-year-olds. Lloyd and Miller (2014) also referred to neuropsychology research that
suggested that the development of a memory skill may occur even when patterns of
results are comparable across age groups. Moreover, they emphasized the fact that
the timing of a study also matters, as shown in Lah et al.'s (2011) recent study, which
compared performance on a picture completion task in children who had a traumatic
brain injury (TBI) and control children. When the injury had occurred after age 6,
comparable amounts of perceptual priming were found in TBI and control children.
However, when the injury occurred prior to age 6, priming effects in TBI children
were reduced relative to the control group. Lloyd and Miller (2014) concluded that
more work is needed before one can firmly conclude that perceptual priming does not
develop meaningfully across the life span. On the other hand, they still acknowledged
that whether or not age invariance is demonstrated, and in contrast to many age-related
effects of explicit memory, it is clear that perceptual priming effects obtained in chil-
dren (at young ages and older) are at least similar to those of adults. Implicit memory
is clearly more functional in young children than the explicit memory system.

The Development of Conceptual Priming

Most developmental studies carried out in the 1990s focused on a single type of
implicit memory, that is, perceptual priming. As noted by Blaxton (1989), how-
ever, implicit memory might best be conceptualized as consisting of two com-
plementary processes, one perceptual and one conceptual. Whereas perceptual
priming relies on a physical overlap between study and test items, conceptual
priming focuses on faster access to meaning and semantic knowledge after expo-
sure to other items (Lloyd and Miller 2014). In order to draw firm conclusions
about the developmental invariance of priming, research on other types of implicit
memory such as conceptual priming is needed.
How is conceptual priming assessed? For instance, a conceptual measure of implicit
memory used in several studies provides participants with a list of category names and
requires them to produce the first exemplars that come to mind for each category. In
most developmental studies that have applied such a category production task, partici-
pants are asked to study a list of items belonging to different superordinate categories
(e.g., “fruit” or “furniture”). Whereas some of the category names used in the test phase
refer to instances presented at study (“old” categories), others do not (“new” catego-
ries). Effects of conceptual priming are inferred when more target exemplars are pro-
duced from the old than from the new categories. The typical finding from studies with
adults is that prior presentation of a category exemplar increases the likelihood that that
word will be named as an example of that category (Schumann-Hengsteler 1995).
The Development of Conceptual Priming 31

Given that the tests in such category production tasks emphasize the semantic
relations between studied and tested items and thus require conceptually driven
processing, one should expect age differences in conceptual priming. That is, older
children should show more priming because the semantic categories are more
meaningful to them than to younger children. Unfortunately, the empirical evi-
dence in this area is mixed.
In one of the earliest studies on conceptual priming (Greenbaum and Graf 1989),
4-, 5-, and 6-year-old children studied a list of items and were then asked to list
exemplars when cued with category names. There was no evidence of developmen-
tal changes, given that priming effects were similar across the three age groups.
Although the failure to find significant differences between the three groups could
be due to the low power of the experiment (12 children per group) and also to the
fact that only preschoolers were considered, a subsequent study by Mecklenbräuker
and Wippich (1995; Exp. 1) using a larger sample of 96 preschoolers and third and
fourth graders (32 children per age group) and using a more comprehensive cate-
gory production task basically replicated Greenbaum and Graf’s findings. Whereas
an explicit memory test yielded significant age-related effects, implicit memory was
comparable across age groups. Mecklenbräuker and Wippich confirmed this find-
ing in a second experiment with preschoolers and sixth graders, again demonstrating
similar amounts of conceptual priming across groups. Supporting evidence based on
a sample of schoolchildren and adults was reported by Billingsley et al. (2002), who
found equivalent priming across age with a similar category generation task. In the
study by Perez et al. (1998) already mentioned above, a direct comparison between
perceptual and conceptual priming was conducted and contrasted with perceptual as
well as conceptual memory of an explicit nature. A specific advantage of this design
was that two implicit and two explicit memory tests were available for each partici-
pant. As a main result, Perez et al. (1998) found that neither the perceptual nor the
conceptual implicit memory tests showed improvement with age, whereas signifi-
cant age effects were found for the two explicit memory measures.
However, several other studies reported different outcomes. For instance, Barry
(2007) compared perceptual and conceptual priming between a group of 87 second
graders and a group of 81 college students. In the perceptual priming condition, par-
ticipants made simple judgments about each word on a list (e.g., “Does the word
have an ‘I’ in it?”), whereas in the conceptual priming condition, participants were
given a sentence cue and asked to add the word that completed the sentence. In the
perceptual test condition, participants had to complete a word fragment, whereas in
the conceptual test condition, they were asked to generate exemplars from category
names. Supporting previous research, no age differences in perceptual priming were
found. However, age differences in priming were found in the conceptual priming
condition in which college students had significantly higher priming scores than sec-
ond graders. According to Barry (2007), these developmental dissociations in prim-
ing support the assumption that the processing requirements of conceptual implicit
memory are similar to those of explicit memory.
Findings supporting the view that the developmental patterns of perceptual and
conceptual implicit memory differ considerably were also reported by Perruchet
32 3  The Development of Implicit Memory

et al. (1995) and by Mecklenbräuker et al. (2003). Perruchet et al. carried out two
experiments with relatively large samples of second and fourth graders. Whereas the
outcome of their first experiment indicated an age trend in conceptual priming such
that older children showed larger amounts of priming than younger children, the sec-
ond experiment failed to replicate the age effect. In an attempt to explain the dis-
crepancy in findings, Perruchet et al. emphasized the fact that the two experiments
differed mainly in the choice of items. Target exemplars were chosen arbitrarily in
the first experiment, whereas they were chosen as a function of their typicality in the
second experiment. Apparently, differences in item typicality were important, with
age differences in measures of conceptual priming increasing when the exemplars
were atypical but not when they were typical exemplars of that category.
A subsequent study by Mecklenbräuker et al. (2003) tested this assumption in
more detail, systematically manipulating item typicality. The main assumption was
that categorical relations, that is, connections to the category name, should be rela-
tively strong for typical exemplars regardless of age. On the other hand, as pro-
posed in network models of semantic memory (cf. Bjorklund 1987), categorical
relations for atypical exemplars should be much weaker in younger than in older
children. In accordance with this assumption, Mecklenbräuker et al. obtained an
age-related increase in conceptual priming from kindergarten to advanced elemen-
tary school age when atypical exemplars of familiar taxonomic categories served
as the study items. As predicted, they did not find age effects in the conceptual
priming of typical items, thus corroborating the results by Perruchet and colleagues
(see also Murphy et al. 2003, for similar findings). As noted by Mecklenbräuker
et al., the developmental invariance in conceptual priming found in previous stud-
ies (e.g., Anooshian 1997; Greenbaum and Graf 1989; Perez et al. 1998) seemed to
be due to the fact that they used more or less typical items. Age trends in implicit
memory are expected to occur whenever the conceptual priming task requires an
advanced knowledge base. The unexpected finding of age-invariant conceptual
priming obtained in most studies can be explained by the predominance of famil-
iar semantic categories in those studies. Accordingly, although there is substantial
evidence for the age invariance of priming effects, performance on conceptual
priming tasks may change with age. However, these changes seem to be related
to changes in conceptual knowledge rather than caused by changes in the prim-
ing mechanism per se (see Lloyd and Newcombe 2009). More studies using novel
materials are needed to examine this issue further.

Interactions Between Implicit and Explicit Memory

Although implicit memory and explicit memory are usually conceived of as func-
tionally distinct and are conceptualized as separate systems, research on adults has
shown that there is evidence for interplay between the two systems. For instance,
Jacoby and Dallas (1981) found support for the assumption that the experience of
priming can serve as a clue for explicit memory decisions. Given that items that
Interactions Between Implicit and Explicit Memory 33

have been recently presented are easier to subsequently process than novel items,
this experience can lead to a feeling of greater processing fluency, which subse-
quently influences recognition memory decisions in adults. To date, only a few
developmental studies have explored the relation between implicit and explicit
memory (for a review, see Miller and Lloyd 2011). As shown by Drummey and
Newcombe (1995), the 3- and 5-year-olds in their sample did not use priming
information to guide (explicit) recognition decisions, whereas the adults in their
sample did. Drummey and Newcombe concluded that there is a link between per-
ceptual fluency and recognition memory that is not evident in young children, at
least not in children younger than 6 years of age. Subsequent research by Guttentag
and Dunn (2003) suggests that this link between processing fluency and recogni-
tion first appears in middle childhood. In their study, 4- and 8-year-old children and
adults first studied a list of pictures and then had to identify new and old pictures
on a recognition memory test. Although perceptual priming effects were found in
all groups, only the older children (as well as the adults) were more likely to rec-
ognize old items as having been presented earlier. Based on these findings, one is
inclined to assume that the link between processing fluency and recognition mem-
ory is established between 5 and 8 years of age. As noted by Lloyd and Newcombe
(2009), the understanding of the relation between prior exposure and ease of later
recognition is an example of metamemory, that is, knowledge about one’s memory.
As emphasized by Lloyd and Miller (2014), another piece of indirect evidence
that conceptual priming effects may occur on explicit memory tasks comes from
studies of false memory using the so-called Deese–Roediger–McDermott (DRM)
paradigm. In experiments exploring the false memory illusion, related lists of words
that center around a topic (e.g., bed–rest–dream) are presented. A critical theme
word (a “lure”) that ties the list together but was not presented in the list could be
“sleep.” It has been repeatedly shown that adults and older children will falsely
recall or recognize the critical lure at very high rates (for a review, see Brainerd and
Reyna 2014). However, younger children are less susceptible to such memory errors
(Howe 2006). One explanation for this pattern of results is that children do not
have the semantic organization that would cause the critical lure “sleep” to be more
familiar after the presentation of related words. It is only with the increased concep-
tual sophistication of a child’s memory that increased rates of memory errors occur.

Implicit Memory in Clinical Populations

As mentioned earlier, the initial contrast between explicit and implicit memory
came from studies that explored the memory problems of amnesic patients (e.g.,
Graf and Schacter 1985). More recent work conducted with populations of chil-
dren suffering from various developmental abnormalities indicates that implicit
memory remains intact despite impairments in explicit memory. For instance,
Brown et al. (2010) compared a group of 31 autistic children with 31 controls
(matched for sex, age, and IQ) on four different implicit learning tasks (i.e.,
34 3  The Development of Implicit Memory

contextual cuing, serial reaction time, artificial grammar learning, and probabilis-
tic classification learning) and an explicit learning task (paired-associate learning).
As a main result, it was demonstrated that the autistic children’s implicit learning
was intact across a number of tasks that differed in surface features, with equiva-
lent outcomes obtained for the two groups. On the other hand, the analysis of the
paired-associate learning data (explicit memory) revealed that the control group
outperformed the group of autistic children, suggesting that explicit learning may
be more problematic than implicit learning in the latter group. These findings con-
verge with other research that has demonstrated intact implicit memory in indi-
viduals with autism spectrum conditions (e.g., Barnes et al. 2008).
As noted by Lloyd and Miller (2014), research with clinical populations also
seems suitable for explaining the interaction between implicit memory and other cog-
nitive skills. Research focusing on children with attention-deficit/hyperactivity disor-
ders (e.g., ADHD) is a good example because it is possible to explore the interplay
between attention and memory by comparing ADHD children with age-matched con-
trols. For instance, Burden and Mitchell (2005) recruited children aged 7–14 years
with and without ADHD and conducted a study in which both perceptual and con-
ceptual implicit memory and explicit memory tests were given. ADHD and controls
performed similarly on tests of explicit memory (category-cued recall and recogni-
tion) and on perceptual aspects of implicit memory (word stem completion and pic-
ture fragment identification) as a function of age, retention interval, and stimulus
format (i.e., picture or word). However, there was no evidence of priming on a con-
ceptual implicit memory task (category exemplar generation) for boys with ADHD,
who performed worse than the control children on this measure. The authors con-
cluded that this type of conceptual memory task, which is likely mediated by frontal
cortex systems, may indicate that a unique memory deficit is associated with ADHD.
As noted by Lloyd and Miller (2014), the fact that different patterns of performance
demonstrated by ADHD children across different types of implicit memory tasks
makes it difficult to believe that these measures tap the underlying construct in a sim-
ilar way. Findings presented by Vicari (2001) for samples of children with Williams
syndrome (i.e., children with a mental disability coupled with strong language
skills), Down syndrome, and typically developing children underscore this point.
For instance, the results of this study indicated that children with Williams syndrome
showed normal implicit memory when perceptual priming tasks were used but were
impaired relative to children with Down syndrome and controls when implicit mem-
ory was measured with a skill-learning task. Accordingly, differences in the implicit
memory measures chosen by researchers can influence outcomes significantly.

The Impact of Advances in Neuroscience

The advances in neuroscience observed during the last two decades are extremely
helpful for understanding the neuroanatomy of memory systems and also con-
tribute to our understanding of differences between explicit and implicit memory
The Impact of Advances in Neuroscience 35

functioning (Lloyd and Newcombe 2009). Imaging techniques such as functional

magnetic resonance imagery (fMRI) or electrophysiological techniques such as
event-related potentials (ERPs) have been very useful for identifying brain struc-
tures activated by different types of memory tasks. Several studies have used ERPs
to differentiate explicit from implicit memory (e.g., Rugg et al. 1998) and to inves-
tigate the maturation of the memory system (for a review, see Cycowicz 2000).
A study by Newcombe and Fox (1994) assessed the neuroscience of percep-
tual priming based on a sample of 10-year-old schoolchildren. Participants were
presented with photographs of preschool classmates and unknown children for an
autobiographical recognition test. Skin conductance responses were assessed dur-
ing the recognition test. The main goals of the study were to determine whether
children could recognize their former classmates at above-chance levels and to
determine whether children who did not show high levels of recognition would
nevertheless show effects of their exposure to these faces. The hypothesis was
that skin conductance responses would be higher for familiar faces. Overall, par-
ticipants were only slightly above chance at recognizing their classmates in the
explicit memory task. However, they showed adequate implicit memory as indi-
cated by changes in skin conductance responses when confronted with classmates’
faces. Thus, participants showed evidence that they recognized their classmates’
faces regardless of recognition memory performance. These findings were later
replicated with a sample of 2- to 4-year-old children (Stormark 2004). Again, chil-
dren showed recognition differences in skin conductance and heart rate measures
even when there were no differences in explicit recognition judgments.
Similar patterns of findings were also demonstrated for 6-month-old infants
who watched a slideshow of upright and inverted women’s faces (Webb and
Nelson 2001). Some of these faces were repeated, thus allowing for an assess-
ment of perceptual priming. Despite a long list of items and a lag of up to 12 faces
before repetition, the ERPs that were recorded when they viewed repeated faces
showed greater negativity than the ones for novel faces, suggesting that the infants
demonstrated sensitivity to the previously experienced stimuli. These findings
can be conceived of as strong evidence for early perceptual priming. Lloyd and
Newcombe (2009) argued that the design of this study represents a gold standard
for research on the development of perceptual priming because the method can
be easily adapted for use across the life span and because it does not depend on
language skills or semantic knowledge. Future studies using such a methodologi-
cal approach across a wide age range should provide a strong and valid test of the
developmental invariance hypothesis.
So far, the findings of the few studies described above that used neurophysi-
ological measures when assessing perceptual priming have been consistent with
behavioral data that showed that perceptual priming is a stable type of implicit
memory that develops very early. However, research using neuropsychological
procedures also points to the problem that the concept of implicit memory is a
really broad one (see Lloyd and Miller 2014). For instance, tasks assessing dif-
ferent aspects of implicit memory are not always supported by the same physi-
ological architecture. Thus, priming has been linked with areas of the neocortex,
36 3  The Development of Implicit Memory

whereas classical conditioning is associated with both the cerebellum and the
amygdalae, and procedural (skill) learning has been linked with the striatum (see
Squire and Zola 1996). Lloyd and Miller (2014) correctly state that one advan-
tage offered by the neuroscience approach to the study of implicit memory is that
memory researchers can make more detailed distinctions between the different
subtypes of implicit memory.

Conclusions

The findings on the development of implicit memory presented in this chapter

strongly suggest that implicit memory is largely distinct from explicit memory,
operating through different processes in the brain. There is no doubt that research
in this area has been very active during the last decade. Whereas Parkin (1997)
based his review of the relevant literature on about a dozen studies, the most recent
review by Lloyd and Miller (2014) included more than 50 scientific articles. All
studies comparing the development of implicit and explicit memory have shown
clear-cut differences, with considerable age trends found for explicit but not for
implicit memory tests. It seems that the brain systems that mediate perceptual
and conceptual priming are fully developed early in life, a finding that is in clear
contrast to the continuous development of the explicit memory system (cf. Squire
et al. 1993). There is substantial evidence in cognitive neuroscience that perceptual
priming and conceptual priming do not depend on the medial–temporal and dien-
cephalic brain structures that mediate intentional declarative memory. According
to Reber (1993), implicit memory should develop early and should be relatively
robust against TBI because it appears to be underpinned by evolutionarily primi-
tive physiological structures such as the basal ganglia and the cerebellum. By con-
trast, explicit memory is based on evolutionarily more novel temporal structures
and thus should be more vulnerable to neurological damage.
There are considerable differences in the developmental patterns of implicit and
explicit memory development, indicating that the two memory systems are rela-
tively independent and distinct. However, studies have shown that they also inter-
act, with priming experiences influencing explicit memory decisions, particularly
in older children and adults.
Whereas most studies on age differences in perceptual priming have by and
large provided evidence that this type of implicit memory is developmentally invari-
ant, the evidence on conceptual priming is rather mixed. Although several studies
have reported dissociations between perceptual and conceptual priming, I am not
yet ready to agree with Lloyd and Miller’s conclusion that the results of experi-
ments on conceptual priming demonstrate consistent developmental trends. It is
quite surprising that so many developmental studies have not found age-related
changes in conceptual priming even though there are many theoretical reasons to
expect this to happen. A closer look at the various studies dealing with conceptual
priming development reveals that age-related effects were obtained only when the
Conclusions 37

tasks were particularly hard (e.g., when atypical category exemplars were chosen
instead of typical ones). It is interesting to note that studies focusing on effects of
aging on implicit memory have yielded conflicting results: Whereas several studies
have suggested a dissociation between conceptual and perceptual priming in older
adults (with impairments on conceptual but not on perceptual priming measures),
other studies did not find such age effects on conceptual priming tests (for a review,
see Dew et al. 2007). My view of the state of the art is in accord with Lloyd and
Newcombe’s (2009) view that some work on conceptual priming indicates age invar-
iance, whereas other work suggests developmental changes.
One important conclusion from more recent research on implicit memory is
that it constitutes a rather broad conceptual category. Many different tasks (e.g.,
priming, conditioning, sequence learning) can be said to be implicit in that they
demonstrate influences of past experience on memory without conscious aware-
ness playing a role. As emphasized by Lloyd and Miller (2014), implicit mem-
ory is thus an umbrella term, and the behavioral as well as physiological profiles
of different implicit memory measures are quite diverse. One of the promises of
recent neuroscience approaches is that their measures have the potential to over-
come the problems of many behavioral tools, allowing for more rigorous tests of
the developmental invariance hypothesis across a wide age range and also offer-
ing the option to explore developmental patterns in subtypes of implicit memory.
There is no doubt that future research on implicit memory will benefit from cogni-
tive neuroscientific studies that should be suitable for identifying different devel-
opmental trajectories for different kinds of implicit memory.
Chapter 4
Memory Development During the Infant
and Toddler Years

A Short History

As noted above, the classic studies on memory development focused on school

children, adolescents, and adults. The predominant view then was that preschool
children’s memories were poorly organized and generally unremarkable. As for
the memories of even younger children such as infants, it was widely believed that
infants lived in a “here and now” world that had no past and no future. However, a
few experiments (mostly case studies) conducted at the beginning of the twentieth
century had already noted that even very young children obviously remember past
experiences. For instance, Hunter (1917) used a delayed-reaction procedure with
his daughter when she was 13–16 months of age. She watched her father hide a
toy inside one of three small boxes with lids. Hunter then distracted his daughter
for some time (3–35 s). After that, the child was allowed to search for the toy,
and at the beginning of the experiment, she did so with some success when the
delay was less than 17 s. As noted by Hunter, his daughter’s best delays in subse-
quent assessments could be recorded as 20–24 s. Although Hunter’s study already
included an implicit measure of working memory (WM) capacity (i.e., the number
of seconds for which the toy was hidden), this finding did not receive much atten-
tion in the early days (cf. Reznick 2009, 2014; Rovee-Collier et al. 2001).
It took quite a while before research on infant memory got its momentum. One
reason for the delay was the popular belief that very young children lack memory
skills, a belief supported by Jean Piaget’s theory. Although Piaget (1954) devel-
oped a task that tested memory for location (the A-not-B task; see below) subse-
quently used in research with infants, his theoretical assumption was that children
younger than 2 years of age lack the symbolic means to represent information in
their minds. Accordingly, infants were thought to be unable to remember the past.
The distinction between implicit and explicit memory outlined above (see Chap. 3)
helps to explain why the assumption that very young infants do not remember much

© Springer International Publishing Switzerland 2015 39

W. Schneider, Memory Development from Early Childhood Through
Emerging Adulthood, DOI 10.1007/978-3-319-09611-7_4
40 4  Memory Development During the Infant and Toddler Years

has been held so tenaciously and has survived for almost 100 years. Methodologically,
it has been very difficult to elicit from infants the behaviors that clearly indicate the
functioning of declarative as opposed to implicit memory. The empirical study of
preverbal infants has been fraught with both procedural and interpretative problems,
mainly due to the fact that memory has to be inferred from nonverbal behavior alone
(Rovee-Collier et al. 2001). The questions of what preverbal infants remember and
when and how they remember it cannot be easily tackled by empirical science. As
these very young children cannot tell us what they remember, progress on this issue
has depended on the development of innovative new tools that elicit information from
infants and toddlers for whom verbal reporting in standard experiments is not an
option (Courage and Cowan 2009a). The problem of methodology was already men-
tioned in an early review of the literature by Daehler and Greco (1985), who focused
on memory development in very young children between 12 and 36 months of age.
Daehler and Greco explained the lack of information on memory development in that
age group by noting that it is difficult to find suitable methods to use with children
younger than 3 years of age.
This situation was certainly changed by technologies such as the
habituation/dishabituation or the preference-for-novelty procedures as well as
techniques adapted from operant conditioning and imitation paradigms such as
the mobile conjugate reinforcement procedure and deferred imitation (see below).
There is no doubt that the study of memory development in infants and very young
children has been very active since the 1970s. Although the first edited books on
memory development (Kail and Hagen 1977; Ornstein 1978) did not deal with
infant memory, four out of 12 chapters in Cowan’s (1997) book on the develop-
ment of memory in childhood focused on very young children, and a follow-up
version published about a decade later (Courage and Cowan 2009b) explicitly
referred to infancy in the book title with six out of 14 chapters discussing very
young children’s memory development. For a similar trend, see Oakes and Bauer
(2007) and the most recent book on children’s memory development edited by
Bauer and Fivush (2014). Newcombe and Crawley (2007) conducted a search of
the PsycINFO database for past decades beginning with 1901–1910 to examine
developmental trends in research on infant memory over the last 100 years. As
can be seen in Fig. 4.1, active interest in the issue did not begin before the 1970s.
From the 1980s on, the situation changed completely, showing an acceleration in
interest leading to more than 60 publications per year from the 1990s on.
Research on infant memory conducted by pioneers of the new era such as
Bauer, Diamond, Fagan, Fantz, Mandler, Meltzoff, Rose, Rovee-Collier, and their
research groups produced fascinating results, which will be described in more
detail below. Today, investigations of infant memory are very much a part of main-
stream cognitive developmental psychology. In fact, numerous methodological
advancements have stimulated so many interesting research questions that it is fair
to state that infant memory has been one of the most active areas in the field of
developmental memory research in the last three decades.
Most (but not all) researchers working on infant memory share the view that mem-
ory is composed of multiple systems that serve distinct functions and are character-
ized by different rules of operation. Whereas declarative or explicit memory involves
A Short History 41

70

60

50

40
Papers per year
30

20

10

0
1901 1921 1941 1961 1981 2001

Fig. 4.1  The number of papers per years on infant memory, as judged by a search of the
PsycINFO database, for past decades beginning with 1901–1910 (slightly modified after
Newcombe and Crawley 2007, p. 293)

the ability to explicitly recognize or recall names, episodes (events), dates, and so on,
nondeclarative or implicit memory refers to nonconscious abilities such as learning
habits and skills, priming, and classical conditioning. The distinction between these
two memory systems seems important for developmental scientists because declara-
tive and nondeclarative memory rely on different neural substrates that follow differ-
ent courses of development (cf. Bachevalier 2014; Bauer 2002, 2007; Nelson 1997;
Richmond and Nelson 2008). A variety of brain regions that seem related to implicit
memory such as the striatum and cerebellum are thought to develop early, supporting
the early emergence of this memory system. By contrast, declarative memory seems
to depend on a multicomponent neural network with several components developing
early such as the medial temporal lobe, and others such as the neocortex undergoing
an extended course of development. Several researchers exploring the development
of memory in infants have postulated that the explicit and implicit memory systems
mature at different rates during the infancy period (e.g., Bauer et al. 2011; Mandler
1990). More specifically, they claim that the more primitive implicit memory develops
first, followed by changes in explicit declarative memory, but this proposal has been
challenged by others (e.g., Rovee-Collier 1997). The issue will be taken up later in this
chapter when the empirical evidence for and against this claim will be weighed.
In this chapter, numerous research paradigms used to assess memory development
in very young children’s memory will be presented, beginning with classic techniques
and proceeding to more novel approaches. Although several paradigms discussed in
the literature may require not only the short-term memory (STM) system but may
also involve the long-term memory (LTM) system, other tasks and techniques have
avoided this methodological problem by definitively tapping the STM/WM capacity
of infants and toddlers. Findings based on such studies and concerning STM will be
discussed first, followed by a review of LTM development in infancy.
42 4  Memory Development During the Infant and Toddler Years

In a review of the literature provided more than 15 years ago, Schneider and
Bjorklund (1998) divided the findings into two broad categories, that is, memory
in early infancy, examining the memory abilities of infants across the first year, and
memory in later infancy, investigating memory in preverbal toddlers, from about
12–20 months of age. Schneider and Bjorklund based their decision on the belief that
the assessment of memory also differs between age groups. That is, their assump-
tion was that whereas young infants’ memory has been assessed primarily by means
of recognition or operant conditioning paradigms, memory in older infants has been
assessed primarily by means of cued-recall techniques, for instance, deferred imita-
tion. As will be shown below, such a categorization is no longer meaningful or tenable.

Short-Term/Working Memory in Infancy

What do we know about STM development in very young children? Since the begin-
ning of the modern era of research on infant memory, it has been obvious that very
young children are able to learn and benefit from past experience and thus provide
evidence that they have some kind of memory (Bauer 2002; Reznick 2009). Recent
findings demonstrate that from birth on, infants can remember things (e.g., faces, pic-
tures, objects) and that this kind of memory improves steadily during early childhood.
Interestingly, some infant researchers use the term “STM” to refer to this ability,
whereas others use the term “working memory” instead, even though these two sets
of findings are based on very similar experimental paradigms. As will be described
in more detail later, STM refers to rather passive information storage and retention
within very brief time periods measured in milliseconds and seconds, implying the
availability of information for a constrained period of time. In comparison, the term
WM addresses a more comprehensive broader memory system first described in the
model by Baddeley and Hitch (1974). Here, WM refers to processes that allow the
maintenance of task-relevant information during the performance of a task, extending
the ongoing representation of a stimulus beyond the termination of that stimulus for a
short duration of time (Reznick 2007, 2009). The WM concept refers to phonological
as well as visuospatial information processing and also includes executive attention
as a component that keeps the stored information activated.
The overlapping use of these terms has occurred for at least two reasons (see
Reznick 2014). First, WM is conceived of as a short-term phenomenon (e.g., being
able to find objects that are hidden in particular locations with short delays between
hiding and finding). Second, most infant researchers seem to believe that although
WM develops across several decades of life, it appears to be relatively undeveloped
during early infancy and does not increase considerably before the second half of the
first year (Pelphrey and Reznick 2003). There is reason to believe that many research
results that are interpreted as reflecting infant WM or infant STM are not necessarily
pure measures of either construct. In the following, I therefore adopt Reznick’s sug-
gestion to refer to “infant short-term working memory” (STWM) when summarizing
the evidence on developmental trends in this ability.
Memory Tasks Used with Preverbal Infants 43

Memory Tasks Used with Preverbal Infants

Hide-and-Seek Tasks
The procedure introduced by Hunter (1917) labeled the “delayed-reaction task”
and briefly illustrated above has been used in more or less this format in numer-
ous studies conducted during the last 60 years. This paradigm is now called the
delayed-response task and usually includes a sequence of trials with variations
in the number of locations as well as in the delay between hiding and seeking.
It is generally accepted as the canonical procedure for assessing STWM in pre-
verbal children (Goldman-Rakic 1987). A variation of this task uses an orienting
response rather than a reaching response. To claim that these tasks tap STM, long-
term storage must be ruled out. This is accomplished by testing for memory across
multiple trials in the same context with the focal representation varying across tri-
als, that is, by making sure that the location of hidden objects varies randomly
across trials (Reznick 2007).
A related task and actually one of the oldest techniques used to assess memory
in infants is the object search task (A-not-B). As noted above, the A-not-B task
was originally introduced by Piaget (1954) and was rediscovered as a method for
studying memory development in infants in the 1970s and 1980s (Diamond 1985;
Fox et al. 1979). The procedures used in the A-not-B task are almost identical to
those used in Hunter’s (1917) delayed-response task. In the standard version of
the task, infants are seated between two identical wells (A and B) where a small
object can be hidden. The object is first hidden in plain view of the infant in one of
the locations (A). The experimenter then simultaneously covers both wells. After
a short delay, the infant is permitted to reach for the object and usually searches
successfully for it in location A. This is repeated for a few trials, after which the
object is hidden in location B. After the hiding place is changed, infants often con-
tinue to reach for the object in location A despite having watched the experimenter
places it in well B. This behavior is called the A-not-B error. As noted by Bell
and Morasch (2007), the only difference between the A-not-B and the delayed-
response tasks concerns the rule for deciding in which of the two wells an attrac-
tive toy is to be hidden. That is, in the A-not-B task, the toy is hidden in the same
well until the infant correctly reaches for it for a prespecified number of times.
Then, the toy is hidden in the other well and the procedure is repeated. In the
delayed-response task, the hiding pattern varies according to a predefined schedule
regardless of whether or not the infant reaches for the object correctly.
Given the similarity between the two tasks, it does not come as a surprise that
infants’ performance on the delayed-response and A-not-B tasks is almost iden-
tical. For instance, Diamond (1990a) reported that infants demonstrated almost
identical developmental progressions in longitudinal assessments from 7.5 to
12 months of age. Figure 4.2 demonstrates the similarity in findings for infants in
this age range.
Using a variation on these hide-and-seek tasks, Reznick et al. (2004) explored
the onset of STWM in infancy, assessing memory capacity in 5- to 6-month-old
44 4  Memory Development During the Infant and Toddler Years

Fig. 4.2  Developmental 12
progression in the delay that
human infants can tolerate
10
in the response and A-not-B
tasks (redrawn from Diamond
1990b) (Rovee-Collier et al. 8

Delay (secs)
2001, p. 105)
6

4
AB

2 Delayed Response

0
7 8 9 10 11 12
Age in Months

infants. In this study, an examiner appeared in one of two windows, interacted

with the infant, and then disappeared behind a curtain. After 1–2 s, the curtain
was opened to reveal empty windows. The infant’s first gaze toward a window
was considered an attempt to locate the examiner, and the number of correct gazes
across a total of 12 trials was used as a measure of STWM. Reznick et al. (2004)
concluded that STWM for locations emerges in the middle of infants’ sixth month.
A review of research in this domain (Pelphrey and Reznick 2003) further revealed
that STWM capacity as measured in hide-and-seek tasks arises at about 6 months
of age and increases by seconds more or less linearly through 12 months (for sup-
porting evidence, see Diamond and Doar 1989; Rose et al. 2001).
The empirical evidence concerning the A-not-B task further suggests that the
frequency of the A-not-B error increases as a function of the delay and that older
infants tolerate increasingly longer delays between hiding and reaching. For
instance, Diamond (1985) found that the amount of delay necessary to yield the
A-not-B error increased with each successive month between 7 and 12 months of
age. Although Diamond assumed that developmental differences in the ability to
inhibit were partly responsible for the outcome, she acknowledged that her findings
might also reflect age changes in memory during this 6-month period (Diamond
et al. 1994). It should be noted, however, that minor variations in the task param-
eters produced considerable variability in findings (Diamond 1990a), thus seeming
to indicate that developmental improvement on object search tasks is due not only to
age-related changes in memory but also to other factors (Rovee-Collier et al. 2001).
A controversial issue discussed in the literature concerns the question of
whether hide-and-seek tasks that use visual behavior instead of reaching and those
using reaching as the critical variable yield similar results. For instance, Diamond
(1985) and Hofstadter and Reznick (1996) reported better performance on a
delayed-response task when infants were allowed to gaze toward a correct loca-
tion rather than reach for an object in that location. On the other hand, Matthews
Memory Tasks Used with Preverbal Infants 45

et al. (1996) who used a longitudinal design did not find different results for the
classic reaching A-not-B task and a visual-orientation version of the task in infants
from about 6.5–14 months of age. Similarly, Bell and Adams (1999), who also
used a longitudinal (within-subjects) design to investigate two cohorts of 8-month-
old infants, compared looking and reaching performance on an A-not-B task and
reported similar results for the two experimental conditions (see Pelphrey et al.
2004, for confirming evidence). Bell and Morasch (2007) argued that methodolog-
ical issues could have affected the discrepancy in findings. That is, given consider-
able interindividual differences in performance among same-age infants, it seems
important that research designs comparing looking performance relative to reach-
ing performance on hide-and-seek tasks should be based on within-subjects meth-
odologies rather than cross-sectional designs. In fact, most researchers who used a
within-subjects (longitudinal) design did not find differences in infant STWM on
looking and reaching versions of hide-and-seek tasks.
However, there are also data supporting the view that the age of the infant plays
a crucial role. Bell and Morasch (2007) conducted a longitudinal study to examine
the emergence of STWM in infants from 5 to 10 months of age. Infants were seen
monthly in the research laboratory for assessments with the looking and reach-
ing versions of the A-not-B task. As shown in Fig. 4.3, infants as a group dem-
onstrated better performance on the looking version from 5 to 8 months of age
and then showed comparable performance on both versions of the task thereafter.
One possible explanation for this trend is that, as long as infants are still develop-
ing gross motor abilities such as reaching, their performance on the looking ver-
sion of the task is better than performance on the reaching version. Around the
age of 8 months, however, when reaching behavior is more developed and skilled,

90

80

70
Percentage Correct

60

50

40

30
Look
20
Reach
10

0
5 6 7 8 9 10
Age in Months

Fig. 4.3  Comparison of infant performance on looking and reaching versions of the A-not-B

task on two measures. (Adapted from Bell 2006) (Bell and Morasch 2007, p. 41)
46 4  Memory Development During the Infant and Toddler Years

performance differences on the two versions of the task tend to disappear. This
interpretation is in accord with the hypothesis that the same cognitive skills (i.e.,
working memory, inhibitory control, and attention) are required for both versions
of the task (cf. Bell and Morasch 2007).
A variation of the delayed-response task in which infants do not need to
actively search for a hidden object is known as the violation-of-expectations
procedure. For this task, an object is hidden at a particular location, and after a
delay, the contents of several locations are revealed. Whereas the expected object
is present in the control condition, it is either not present or an unexpected object
is revealed in its place in the memory-challenge condition. In either case, greater
visual attention in the violation-of-expectations condition is interpreted as evi-
dence that the infant maintained the location and/or the identity of the object in
STWM (Reznick 2009). For instance, research by Baillargeon and colleagues
(e.g., Baillargeon and Graber 1988; Baillargeon et al. 1989) using this paradigm
demonstrated that infants show retention at a younger age and after longer delays
than on traditional hide-and-seek tasks. In these studies, 8-month-olds were pre-
sented with an object that was then placed behind one of two screens. After a 15-s
delay, a hand retrieved the object from behind the correct screen (possible event)
or from behind the incorrect screen (impossible event). Infants looked significantly
longer at the impossible event, suggesting that they remembered the location of
the original object and were surprised to see it retrieved from the other location.
The results of these studies indicated that infants still remembered the correct
location after 15 s and even after longer delays (up to 70 s).
In a more recent study, Luo et al. (2003) used a similar procedure with
5-month-olds. Infants looked longer at the unexpected event after a 3- or 4-min
delay, thus appearing to support the claim that infants can remember the hidden
box for several minutes. Baillargeon and her colleagues inferred from their find-
ings that random search errors on the original object search task cannot be attrib-
uted to inadequate memory mechanisms. When given a task that does not require a
manual search, 8-month-old infants remember trial-to-trial changes in an object’s
location for delays that are 10–20 times longer than those that produce A-not-B
errors on the traditional object search task. Baillargeon et al. (1989) speculated
that the demands of the A-not-B task involve infants’ problem-solving ability
rather than their memory.
Similar interpretation problems were reported for the delayed-nonmatching-to-
sample (DNMS) task. Originally, this paradigm was used in studies assessing the
effects of brain lesions on memory in monkeys (for details, see Rovee-Collier et al.
2001). Meanwhile, it has become a standard test of recognition memory in animals.
The original procedure is quite simple. At the start of a trial, a stimulus is presented
to an animal and then removed from view for some time. At the end of the delay,
the familiar sample stimulus and a novel stimulus are presented, and the animal is
rewarded for choosing the nonmatching novel stimulus. In order to be successful
across trials, the animal has to remember which stimulus it saw most recently.
The standard DNMS task used with human infants is procedurally identi-
cal to the tasks used with animals. Interestingly, studies using this paradigm (e.g.,
Memory Tasks Used with Preverbal Infants 47

Diamond et al. 1994) found that human infants do not succeed on this task dur-
ing their first year despite the fact that its procedure is similar to that of the visual
paired-comparison (VPC) task, which can be mastered by much younger infants
(see below). Most studies on the DNMS task have shown that infants do not reli-
ably choose the novel stimulus until midway through their second year of life (i.e.,
not until approximately 15–21 months of age) even when the delay between sample
trial and test trial is only 5–10 s. As noted by Rovee-Collier et al. (2001), the dis-
crepancy in findings is similar for animals, for which the acquisition of a DNMS
task takes considerably longer than that of a visual discrimination task. In mon-
keys, acquisition time is usually measured on a scale of weeks rather than days. As
in the object search task, seemingly minor variations in the task parameters affect
results considerably (for a review, see Diamond 1990b). Moreover, different statis-
tical success criteria may play a role. For instance, Overman et al. (1993) pointed
out that different levels of mastery indicate success on the VPC and DNMS tasks.
Whereas a statistically significant preference for novelty (i.e., significant deviations
from the chance level of 50 %) is usually taken as evidence for visual recognition
memory on VPC tasks, a much harder criterion (2 consecutive days of 87 % cor-
rect) typically defines success on the DNMS task. When Overman et al. (1993)
used the same criterion (i.e., a significant deviation from chance) for both the VPC
and DNMS tasks, even the youngest children in their sample (10- to 15-month-
olds) showed significant visual recognition on the DNMS task, and performance
was generally comparable across tasks. Although Overman et al. (1993) speculated
that performance on both tasks was mediated by the same neural mechanisms, they
also observed age-related performance differences, indicating that there were some
other cognitive demands inherent to the DNMS task not readily mastered by the
children. Considered in light of similar findings obtained with the object search
task, one is inclined to accept such a view and to conclude that factors other than
memory contribute to performance on the DNMS task (see Diamond 1995; Rovee-
Collier et al. 2001, for more detailed discussions). As noted by Bachevalier (2014),
the presence of a preference for novelty in early infancy (as measured by the VPC)
suggests that incidental recognition memory processes are already supported by the
early-developing medial temporal cortical areas. By contrast, the participation of
the hippocampus may emerge more slowly during infancy, illustrating a complex
pattern of development for recognition memory processes.
One problem with the version of the “violation-of-expectations” procedure
described above is that the infant may respond to the unexpected event because
information is stored in LTM rather than STWM per se. In a variation of this para-
digm, Káldy and Leslie (2005) used a procedure that clearly aimed to assess STWM
to explore whether infants are able to bind object identity information to an object
that changes location. After being familiarized with a disk and a triangle, infants
saw these two objects disappear behind two different screens. After a 2-s delay, the
screens were removed, revealing the two objects in unexpected (swapped) loca-
tions. During both the familiarization and test trials, the side of the triangle and the
disk alternated from trial to trial, preventing infants from using their LTM about
where a particular object usually was. Infants could succeed on this task only if they
48 4  Memory Development During the Infant and Toddler Years

constantly updated the contents of their STWM. As a main result, Káldy and Leslie
(2005) found that both 6.5- and 9-month-old infants looked longer at the unex-
pected outcome. Another interesting aspect of their findings was that the younger
and older infants solved the task in different ways. Because the objects were hidden
sequentially, the authors could test whether infants remembered both objects in the
sequence or only the last one they had seen. The results showed that the 6.5-month-
olds did not remember the shape of the first hidden object, whereas the 9-month-olds
did. Káldy and Leslie thus concluded that the older infants were able to bind shape
information to two object indexes that followed the objects as they were moved
to new locations. Apparently, this new ability emerges sometime between 7 and
9 months of age (see also Leslie and Káldy 2007).
As already noted above, the impact of individual differences in same-age
infants has been neglected in most infant memory studies. The research program
by Bell and colleagues constitutes an exception to this rule given that it focused
on individual differences in infant frontal lobe development, using EEG record-
ings as a marker of brain development (e.g., Bell 2001; Bell and Fox 1992). For
instance, Bell and Fox (1992) conducted a longitudinal study with infants from 7
to 12 months of age and reported that changes in baseline frontal EEG power val-
ues (reflecting the excitability of groups of neurons and considered to be a marker
of brain maturation) were associated with changes in spatial STWM. Infants tol-
erating long delays between the hiding of the object in the A-not-B task and the
actual manual search by 12 months of age showed changes from 7 to 12 months
of age in baseline frontal EEG power. Infants tolerating only short delays by
12 months did not. Moreover, successful reaching was associated not only with
age-related changes in frontal EEG power but also with consistently greater left
occipital EEG power relative to right occipital power. Bell and colleagues con-
cluded from their findings that both frontal and occipital power values are associ-
ated with spatial STWM performance, indicating that better memory performance
is related to greater brain maturation.

Novelty-Preference Paradigms

Most of what was originally known about infants’ memory abilities came from
studies analyzing infants’ looking patterns and were based on the assumption that
very young children tend to look longer at novel stimuli than at familiar ones. The
technique of VPC, first developed by Fantz (1956), involves briefly pre-exposing
infants to pairs of pictures of a stimulus, and then, after some period of familiari-
zation, presenting the familiar picture with a completely novel one. Recognition
is inferred if the percentage of total time spent looking at the new item exceeds
chance (50 %) and the longest test delay defines the duration of retention.
Using this procedure, memory for visual stimuli has been found for some new-
borns (Friedman 1972). Fagan (1970) was the first to introduce a delay between
infants’ pre-exposure and the paired-comparison test, searching for the longest delay
Memory Tasks Used with Preverbal Infants 49

after which infants still demonstrated a novelty preference. This delay, defining the
upper limit (UL) of infants’ visual recognition memory, was systematically varied in
subsequent studies. Fagan demonstrated that infants can retain briefly experienced vis-
ual information after substantial delays. In one study (Fagan 1974), 5- and 6-month-
old infants formed visual memories that lasted as long as 2 weeks. Related work using
the preference-for-novelty paradigm with even younger infants showed that infants as
young as 3 months can retain a memory for object motion across time intervals of
1 and 3 months (Bahrick and Pickens 1995). As shown by Rose, Gottfried, Melloy-
Carminar, and Bridger (1982), the number of seconds of familiarization required to
produce a novelty preference interacts with age, with younger infants (3- to 4-month-
olds) requiring more encoding time to show a novelty preference than older infants
(6- to 7-month-olds). For instance, after one training session, 3- to 4-month-old
infants typically show retention for 5–10 s, 6-month-old infants exhibit retention for
1 min, and 9- to 12-month-olds for 10 min (Rose et al. 2007). Retention intervals may
be even longer for faces. Even newborns exhibit facial recognition after 2 min, and
3-month-olds exhibit recognition after 24 h (Pascalis et al. 1998).
As noted by Daehler and Greco (1985), the assumption that infants prefer to
attend to novel stimuli needs some qualification. Several studies have shown that
infants less than 2 months of age actually prefer familiar rather than novel stimuli
(e.g., Weizmann et al. 1971), and this may indicate that an infant must have sufficient
opportunity to become familiar with a stimulus in order to display a preference for
another stimulus. This assumption was confirmed by a study by Hunter et al. (1982):
When 12-month-olds were given a rather long time to become familiar with toys,
they focused on novel toys more than familiar toys. However, when the familiariza-
tion time was brief, the opposite effect occurred. In other words, a preference for
novelty may be exhibited by infants of any age depending on whether sufficient time
has been allowed to encode the stimulus. Subsequent methodological improvements
such as the avoidance of fixed familiarization periods for all infants helped research-
ers to overcome this problem and to improve the validity of the basic paradigm.
The technique of paired comparison can be used to test infants’ reactions to pre-
viously encountered visual and auditory stimuli. The evidence for recognition over
longer intervals has been found to change in both domains: Whereas for shorter
periods (e.g., 1 day to 2 weeks), no preference was detected (i.e., the amount of
time spent looking at the familiar and novel stimuli did not differ), a preference for
familiar stimuli was manifested for longer time intervals. Such a pattern, which is
typically observed for visual stimuli, was also observed for auditory recognition of
familiar versus novel nursery rhymes (Bauer 2002; Spence 1996).
A variant of this technique, visual habituation, involves sequentially expos-
ing infants to numerous pictures of a stimulus until their attention to the stimulus
decreases (typically, a 50 % decrease in looking time). Then, they are tested with a
new stimulus. Recognition is inferred if looking at the new stimulus significantly
increases, indicating that the infant can discriminate the novel stimulus from the
earlier habituated stimulus. At this point, infants are tested for retention after dif-
ferent delays with either the habituation stimulus or a novel stimulus. When a
delay is introduced between the habituation trials and the “post-habituation” test
50 4  Memory Development During the Infant and Toddler Years

trial, spontaneous recovery as a reaction to the habituated stimulus increases with

the length of the delay interval. Forgetting of the habituated stimulus is assumed
when infants fixate on this stimulus for as long as they did during their first habitu-
ation trial. The longest test delay defines the duration of retention and is defined as
the limit of infants’ visual recognition memory.
Werner and Perlmutter (1979) summarized Stinson’s (1971) study, which used a
variation of the habituation procedure with 4-month-old infants and obtained a classic
forgetting curve function. The infants in this study learned to produce high-amplitude
sucking behavior that was reinforced by the brief presentation of the mother’s face
and voice. In accordance with expectations, the effectiveness of the reinforcer waned
with repeated presentations, leading to a decrease in responses. When the decrease
in responses reached a prespecified criterion level, delays of various duration were
imposed between the final reinforcer in the current trial and its presentation in the
next trial. Whereas responding remained low for short delays between training and
testing (between 0 and 15 s), it returned to its original level when the delay exceeded
30 s. Stinson thus inferred that forgetting was complete by a delay of 30 s or more.
Taken together, the numerous studies concerning newborns’ and infants’ pat-
terns of habituation and dishabituation indicate that basic memory activities such
as the association of a stimulus with a response and the distinction between old
and new stimuli (i.e., recognition processes) are especially dominant early in life.
Although experiments with newborns indicate that neonates can form at least a
brief memory of perceptual experiences, other research with very young infants
has demonstrated that their recognition of objects is surprisingly durable.
Moreover, there is evidence that infants’ rates of habituation and preferences
for novelty are related to subsequent childhood IQ. Infants who habituate faster
or show a stronger preference for novelty have higher childhood IQs than children
who habituate more slowly or show less of a novelty preference (e.g., Fagan 1984;
Rose et al. 1992). Two reviews of the relevant literature have demonstrated the
reliability of these findings, reporting average correlations between infant meas-
ures of habituation and/or preference for novelty ranging between 0.40 and 0.50
(Bornstein 1989; McCall and Carriger 1993). In addition, one other study found
that infant preferences for novelty were significantly correlated with a measure of
recognition memory at 36 months of age (Thompson et al. 1991). The latter find-
ing argues for a continuity of the cognitive mechanisms underlying memory func-
tioning between infancy and early childhood, such as the encoding, storage, and
categorization of stimuli as well as the abilities to compare a present stimulus with
a memory representation and to process novelty.
One more recent experimental variation of the preference-for-novelty paradigm
involving visual STM in infants was developed by Lisa Oakes and her colleagues
(see Oakes and Luck 2014, for a review). This research was based on a change-
detection paradigm and focused on how infants represent objects in STWM. Oakes
and colleagues assumed that most research on infant memory had used tasks that
almost certainly included LTM systems, and this would mean that firm conclusions
about the development of STWM would be difficult to draw. To avoid interpretation
problems, they developed a procedure that was based on Luck and Vogel’s (1997)
Memory Tasks Used with Preverbal Infants 51

change-detection paradigm for adults and that constituted a variant of the paired-com-
parison procedure described above. The first in a series of studies (Ross-Sheehy et al.
2003) sought to replicate the general developmental pattern reported by Rose et al.
(2001) who found that 5- to 7-month-old infants showed memory (novelty prefer-
ences) for only 1 or 2 items (small toys), whereas 12-month-old infants showed signif-
icant memory for 3 or 4 items. Rose et al.’s (2001) findings thus suggest that infants’
STWM is limited and undergoes significant change across the first year of life.
The change-detection paradigm used by Ross-Sheehy et al. (2003) seemed
well suited for studying the origins of STWM. In four experiments, infants with
approximate ages of 4, 6.5, 10, and 13 months were shown simultaneous displays
(streams) of colored squares on two separate computer monitors that blinked
on and off repeatedly (see Fig. 4.4). On one monitor, the colors of the squares
remained constant from cycle to cycle. On the other monitor, the color of one ran-
domly chosen square changed with each new presentation. The displays were pre-
sented for a brief period (500 ms) and were separated by a brief delay (250 ms).
It was assumed that given the choice of two similar displays, infants would look
longer at the display that imposed a greater information load (i.e., the display with
the changing streams) provided that they were able to form a memory of the colors
and to keep this memory active during the 250-ms delay. Memory capacity was
evaluated by varying the number of squares in the arrays (set size). Set size var-
ied between 1 and 6 items in one array. Ross-Sheehy et al. (2003) assumed that

Change No Change
(leftdisplay) (rightdisplay)

500 ms

250 ms
20 seconds

Fig. 4.4  Schematic representation of a trial used for Experiment 1 (set size 3). Trials for
Experiment 1 consisted of 1, 2, or 3 colored changing or nonchanging squares. Items are not
drawn to scale. (Adapted from Ross-Sheehy et al. 2003, p. 1810)
52 4  Memory Development During the Infant and Toddler Years

an infant would detect the color changes and show a preference for the changing
display if his or her memory capacity was greater than or equal to the set size.
Otherwise, they assumed that infants would be not able to detect the color changes
and would not exhibit a significant preference for changing displays. A change
preference score was calculated for each infant. The score was defined as the time
spent looking at the changing stream divided by the total time spent looking.
As the main result of their experiments, it was shown that 4- and 6.5-month-
olds showed a significant preference for the changing stream only at set size 1.
In comparison, 10- and 13-month-old infants significantly preferred the changing
stream for set sizes 1, 2, and 3. In one of the experiments, it was also shown that
10-month-olds could detect a change at set size 4 but not at set size 6. The authors
concluded that substantial developmental changes in visual STWM capacity were
revealed across the first year of life. It seems important to note that the develop-
mental changes in visual STWM capacity observed in these experiments occurred
at about the same time and were almost identical to those reported by Rose et al.
(2001). The findings for the 10- and 13-month-olds seemed particularly impressive
given that an adult’s capacity for objects of this type is only three to four objects in
a similar procedure (Luck and Vogel 1997).
In a follow-up study, Oakes et al. (2006) analyzed the development of feature
binding in infants of about the same age range, using the same change-detecting
task. In particular, they examined the binding of object identity (color) and loca-
tion in visual STWM. The major difference from the procedure used in the study
by Ross-Sheehy et al. (2003) was that the same three colors and three locations
were used in each cycle, but the colors switched locations from cycle to cycle. It
could be shown that 12.5-month-old infants significantly preferred the changing
stream, indicating that they could rapidly bind location and color in visual STWM.
Interestingly, whereas 6.5-month-old infants could not remember simple color-
location combinations across a 300-ms delay, 7.5-month-olds could bind color and
location as effectively as the 12.5-month-old infants. The authors concluded that
a rapid development in visual STWM binding occurs during a rather narrow time
window, and this development parallels rather dramatic neuroanatomical changes in
the parietal cortex. Apparently, the ability to represent multiple items and the ability
to bind features in visual STWM emerge in about the same time period. As shown
in a more recent study by Ross-Sheehy et al. (2011), even 5-month-olds can bind
an object’s color to its location and encode information about individual items in
multiple-object arrays when provided with specific attention-directing cues.
In another recent study, Oakes et al. (2011) investigated developmental changes
in infants’ visual STWM representations for location, using a variant of their change-
detection procedure. In a first experiment, Oakes et al. (2011) assessed 6-, 8-,
and 12-month-old infants’ memory for the locations of one, two, or three objects.
Findings resembled those obtained for STWM for color in that 6-month-olds sig-
nificantly preferred the changing stream only at set size 1, indicating that their
capacity is limited to one location, whereas 8- and 12-month-olds significantly pre-
ferred the changing stream even at set size 3. Further experiments confirmed that
infants’ STWM for location information emerges by 6 months. By this age, infants
can detect changes in the spatial configuration of three items. Major developmental
Memory Tasks Used with Preverbal Infants 53

changes seem to occur during a rather short time period, given that 8-month-olds
already show memory for multiple locations. In general, these results indicate that
substantial developmental changes in STWM for color and spatial STWM occur
during the first year of life, with the developmental trajectories showing impressive
similarities (for more details, see Oakes et al. 2007). This work also suggests that
infants, like adults, can store representations of three to four items at a time.
Feigenson and colleagues (Feigenson and Carey 2005; Feigenson et al. 2002)
used a more naturalistic setting to validate the latter assumption. In their cracker-
choice task, 10- and 12-month-old infants saw two quantities of desirable objects
(graham crackers) sequentially placed into a pair of opaque buckets and were then
allowed to choose between them. Feigenson and colleagues assumed that this pro-
cedure would serve as a naturalistic test of preverbal children’s STWM given that
the task required the children to maintain and compare representations of the hidden
objects in order to determine which bucket contained more crackers. In one of the
experiments (Feigenson et al. 2002), infants were given a choice between 1 versus
2, 2 versus 3, or 3 versus 4 crackers. For instance, in the 1-versus-2-cracker choice,
the experimenter first showed the infants that the buckets were empty and then
placed one cracker in one bucket and two crackers in the other bucket. The depend-
ent measure was simply which bucket infants chose to walk or crawl to. Results
showed that infants in both age groups successfully chose the bucket containing the
greater quantity. They failed, however, with a choice of 3 versus 4 crackers.
In a next step, new groups of infants were tested on this task with quantities in a
ratio, giving infants a choice between 2 versus 4, 3 versus 6, and 1 versus 4 crack-
ers. This time, infants in both age groups failed to choose systematically between
any of these quantities. According to the authors’ interpretation, this dramatic break-
down in performance illustrates that infants’ ability to remember the hidden objects
was determined by the total number of objects seen and not by the ratio between the
two quantities that were presented. That is, infants succeeded only when one, two,
or three crackers were placed in either bucket, and chose entirely at chance when
required to remember larger numbers. Using a different (manual search) task that
also measured the number of objects infants could remember, Feigenson and Carey
(2005) obtained similar results for 12- and 14-month-old infants. They succeeded
with 1 versus 2 and 2 versus 3 comparisons, but failed with 2 versus 4 and 1 ver-
sus 4 comparisons. The authors concluded that results from studies exploring infants’
STWM capacity using different experimental paradigms have yielded identical pat-
terns of results, indicating that infants’ memories are limited to representing approxi-
mately three to four objects at a time (for a detailed discussion, see Feigenson 2007).

Summary

Overall, there is little doubt that recent research on infants’ STWM using inno-
vative and creative experimental procedures has yielded important new find-
ings. Together, research using different paradigms suggests that infants have
a limited-capacity system that they use to temporarily maintain information.
54 4  Memory Development During the Infant and Toddler Years

Multiple studies using very different paradigms also suggest that young infants
(e.g., 6 months) can retain only one item in STWM, whereas older infants (e.g.,
12 months) can retain three or four objects (Oakes and Luck 2014). These studies
all demonstrated that dramatic developmental changes in STWM occur during the
first year of life, indicating that the development of neural systems, and in particu-
lar, the development and maturation of anterior (prefrontal) and posterior (tempo-
ral and parietal) brain systems, is very important for driving memory development
in infancy. STM seems rather limited in infants younger than 6 months of age but
improves remarkably during the second half of the first year of life. Most research-
ers agree that WM becomes more robust during the second half of the first year.
There is also (limited) evidence that STWM capacity continues to improve in tod-
dlers. For instance, Kagan (1981) administered a memory for location task to large
samples of children in the USA and other countries and found a steep increase in
STWM across the second year. Comparable results were reported by Reznick et al.
(1997) for a large sample of identical and fraternal twins.
More research is needed to further clarify relations between brain develop-
ment and cognitive development. The research summarized by Bell and Morasch
(2007) indicates that a focus on the longitudinal assessment of individual differ-
ences is helpful for understanding brain–behavior relations in the development of
working memory. For instance, the EEG studies conducted by Bell and colleagues
nicely demonstrated that frontal lobe maturation contributes significantly to indi-
vidual differences in the development of WM around the end of the first year (e.g.,
Bell and Fox 1992). Moreover, there is a need for research to validate these new
methodological accomplishments and to provide better controls for possible influ-
ences on performance including attention, interest, vulnerability to distraction, and
aspects of temperament (see Leslie and Káldy 2007; Reznick 2014). There is opti-
mism that new methods can be devised to link together WM research in infants
and young children.
Finally, as noted by Cowan (2007), future research on WM in infants may also
involve an expanded array of topics. Such research may cover not only the develop-
ment of capacity and memory resilience but may also focus on the development of
grouping and recoding processes that are so important for later developments in work-
ing memory. As infants develop, sensory processes allow them to have not only a more
precise perceptual representation with which to work but also a richer set of abstract
codes with which to maintain and manipulate experiences in working memory.

The Development of Long-Term Memory in Infants

and Toddlers

Conditioning Paradigms
As noted by Rovee-Collier and Gerhardstein (1997), both classical and operant
conditioning paradigms provide direct measures of infants’ long-term retention. In
an early case study of classical conditioning, Jones (1930) exposed a 7-month-old
The Development of Long-Term Memory in Infants and Toddlers 55

infant to repeated pairings of a tapping sound [the conditioned stimulus (CS)] and
an electrotactual stimulus [the unconditioned stimulus (UCS)] for five sessions
across several days. An anticipatory galvanic skin reflex served as the conditioned
response (CR) and was established during the first session. As a main result,
Jones found that the infant still exhibited the CR after 6 weeks despite the fact
that no additional conditioning trials had occurred in the meantime. The infant also
showed a (weaker) CR 7 weeks after completing the training sessions.
About 40 years later, Little (1970) used the eyeblink response, a protective
reflex, as a CR in a study that investigated early memory in newborns. When the
UCS is a corneal air puff, the CR that immediately anticipates the UCS is a func-
tional avoidance response. The optimal interstimulus interval (ISI) to establish a
conditioned eyelid reflex in humans and nonhumans is about 500 ms. As pointed
out by Rovee-Collier et al. (2001), the early failures to establish a conditioned
eyelid reflex in infants resulted from using the 500-ms ISI, which was suboptimal
for young children. Little (1970) found that the optimal ISI for eyelid condition-
ing in newborns was about three to four times longer, potentially reflecting an
age-related change in synaptic efficiency. In a subsequent study, Little, Lipsitt, and
Rovee-Collier (1984) demonstrated robust acquisition and retention of the condi-
tioned eyeblink response in very young infants, using the optimal 1,500-ms ISI as
the experimental condition and the nonoptimal 500-ms ISI as the control condi-
tion. Infants received 50 presentations of a CS (CS-tone) that overlapped and ter-
minated with an air puff UCS that caused infants to blink. Infants were trained at
either 10, 20, or 30 days of age and received a retention session 10 days later. As
an example of such a design, infants who were trained for the first time at 20 days
of age served as maturational controls for infants who were first trained at 10 days
of age and tested at 20 days. Only infants in the 1,500-ms ISI condition learned the
association, regardless of age, and infants who were first trained at 20 or 30 days
showed significant savings. The percentage of CRs increased with age, and the old-
est group of infants reached a higher level of conditioning than the youngest group.
Moreover, infants who were first trained at 20 days of age and tested 10 days later
performed significantly better than the 30-day-old infants when tested for the first
time (for more details, see Rovee-Collier and Gerhardstein 1997).
Most of what we currently know about infants’ LTM was obtained in studies
using operant conditioning procedures. Adapting techniques from this para-
digm, Carolyn Rovee-Collier and her colleagues developed creative procedures
that assess the memories of young infants from the infants’ motor actions (espe-
cially foot kicks) across delays of up to 6 weeks (see reviews by Rovee-Collier
and Cuevas 2009; Rovee-Collier and Gerhardstein 1997; Rovee-Collier et al.
2001). According to the authors, the underlying logic of this approach is straight-
forward. Given that young infants cannot say what they recognize, they learn a
motor response to use instead.
In their operant conditioning paradigm of mobile conjugate reinforcement,
an attractive mobile is suspended above an infant’s crib or playpen. The typical
experiment begins with a 3-min baseline period during which a ribbon connects
the infant’s ankle to an empty mobile stand, whereas the mobile is suspended from
56 4  Memory Development During the Infant and Toddler Years

another stand on the opposite crib rail. Thus, the infant can view the mobile but
cannot move it. Next comes a 9-min reinforcement phase (acquisition) during
which the ankle ribbon is connected to the mobile. Most infants quickly learn the
contingency between their kicking and the movement of the mobile. They figure
out that their kicks are linked to the movement of the mobile. After the conditional
response is acquired, indicated by a learning criterion of kick responses that are
clearly above the baseline rate, the memory phase of the experiment begins. The
3-min long-term retention test occurs after delays ranging from a few minutes to
several weeks. Here, the infant is again hooked up to the ribbon with a mobile
overhead. If he or she resumes kicking at rates greater than baseline when the rib-
bon and mobile are not connected, it implies memory of the earlier reinforcement
phase. If the kicking rate is low and comparable to baseline, the implication is that
the infant has forgotten the earlier contingency. Because responding is not rein-
forced during the long-term retention test, infants’ performance reflects only what
they bring to the test session and does not reflect new learning at the time of test-
ing (Rovee-Collier and Cuevas 2009).
Using variants of this procedure, Rovee-Collier and her colleagues demon-
strated LTM in very young infants. The findings are generally impressive, illus-
trating that young infants are able to encode, store, and retrieve memories after
extended periods. For example, it was found that infants at 2 months of age can
remember the mobile for up to 3 days, infants at 3 months of age show retention
for up to 5 or 6 days, and by 6 months, they remember for as long as 2 weeks (Hill
et al. 1988). The length of time across which behavior toward the mobile can be
retained can be extended by providing “reminders” of the mobile during the delay
and is also affected by the amount of training the infants receive, the distribution
of training, and the affect that infants display during training (for a review, see
Rovee-Collier and Cuevas 2009).
Because the mobile task is inappropriate for use with infants older than
6 months, Rovee-Collier and colleagues developed a second task for infants
between 6 and 18 months of age. For this train task, children learn to press a lever
to move a miniature train around a circular track. During the long-term retention
test, the lever is deactivated. Because 6-month-olds show identical memory behav-
ior on both tasks (Hartshorn and Rovee-Collier 1997), the operant train task can
be viewed as an upward extension of the mobile task. Subsequent studies using
both tasks have shown that retention increases linearly between 2 and 18 months
of age (Hartshorn et al. 1998; see Fig. 4.5). As noted by Rovee-Collier and Cuevas
(2009), the retention function provided no hint that LTM abruptly improves either
at the end of the first year when a qualitatively different memory system is thought
to mature (Nelson 1997) or during the second year when infants develop language.
One focus of the research conducted by Rovee-Collier and her colleagues was
the contribution of context to infants’ long-term retention. In these experiments,
aspects of the acquisition and testing environments were changed; for instance, the
types of cribs the infants were tested in or the colors or patterns of the crib lin-
ers. In one study, 6-month-old infants were tested with standard procedures, but
the testing situation was made very distinctive (Rovee-Collier et al. 1992). The
The Development of Long-Term Memory in Infants and Toddlers 57

Fig. 4.5  Standardized 14
reference functions for
the maximum duration of 12
retention (in weeks) of infants

Maximum Retention (weeks)

who were trained and tested
in the operant mobile and 10
train tasks and the deferred Train Task
imitation (puppet) task using 8
a standardized procedure with
age-calibrated parameters.
Differences in the slopes of 6
the two functions are due
solely to different training 4
parameters. (Reprinted from
Hartshorn et al. 1998, p. 74)
2 Mobile Task

0 Deferred Imitation
0 3 6 9 12 15 18
Age (months)

infant sat in an infant seat placed in a playpen, which was draped with a distinctive
cloth (for instance, a yellow liner with green squares). Some infants were tested
24 h later in the same situation (the “no change” condition), whereas others were
placed in a playpen draped with different colors (the “context change” condition).
As a main result, infants in the “no change” condition showed significantly better
long-term retention than infants in the “context change” condition, illustrating the
importance of context for infant memory. Findings from these and other related
studies indicate that children do not respond to the context “as a whole,” but
rather, they process specific components of the context. For instance, they do not
seem to associate the training cue and the context. As noted by Rovee-Collier and
Cuevas (2009), however, there is evidence that infants’ memories become less con-
text-dependent with age. Whereas 6-month-olds cannot recognize the cue in a dif-
ferent context 1 day later, 9-month-olds can. Although the original cue presented
in a different context cannot reactivate the memories of either 6- or 9-month-olds,
it can do so in 12-month-olds. According to Bauer et al. (2011), the context speci-
ficity of memory in the mobile conjugate reinforcement task suggests that the type
of memory measured by this task may reflect implicit learning (but see Rovee-
Collier et al. 2001, for a different view).

Deferred Imitation Tasks

This paradigm was originally suggested by Piaget (1952) who assumed that it is
not until late in the second year of life that infants possess symbolic abilities and
thought. In deferred imitation tasks, children observe a model and imitate the
58 4  Memory Development During the Infant and Toddler Years

modeled behavior some time thereafter, retrieving from memory some representa-
tion of the observed behavior. Beginning in the mid-1980s, the technique was devel-
oped as a memory test for infants and young children (e.g., Abravanel and Gingold
1985; Bauer and Shore 1987; Meltzoff 1985). Most experiments follow a procedure
in which infants are shown a series of novel actions performed by an adult model,
often involving a new toy. Infants do not play with the toy at this time, but only
observe the model. Some time later, infants are again shown the toy and encouraged
to interact with it. In a series of classic studies, Meltzoff and colleagues were able
to challenge Piaget’s position that deferred imitation should not appear before the
ages of 18–24 months. When deferred imitation was examined under highly con-
trolled experimental conditions, even 9-month-olds could imitate an experimenter’s
actions after a 24-h delay, and 14-month-olds could exhibit deferred imitation after a
4-month delay (see Meltzoff 1988, 1995). As the novel activities were modeled only
briefly, were modeled without instructions, and were not performed by the infants
themselves prior to the retention test in Meltzoff’s studies, it is likely that the infants’
imitations were based on the stored representations of their previous observations,
thus indicating recall rather than simple recognition (Courage and Howe 2004).
For more than a decade, 9 months were considered to be the youngest age at
which infants would exhibit deferred imitation. However, building on Meltzoff’s
work, Hayne and colleagues (e.g., Barr et al. 1996; Collie and Hayne 1999; Hayne
and Herbert 2004; for a review, see Hayne 2007) were able to show that even
6-month-olds can master such tasks. Hayne and colleagues investigated develop-
mental changes in deferred imitation by 6- to 30-month-old infants and toddlers.
In one of the tasks typically used with 6-month-olds (Barr et al. 1996), infants
watched the experimenter remove a mitten from a hand puppet, shake it to ring
a bell inside, and replace the mitten. Barr et al. (1996) found that 6-month-old
infants could imitate these actions 24 h later if the demonstration lasted 60 s but
not if it lasted only 30 s. An infant’s imitation score is the number of actions (0–3)
performed within 90–120 s of touching the puppet. Although younger infants were
less accurate in their imitations (i.e., produced fewer actions) than older infants,
even 6-month-olds performed most or all of the actions. Moreover, 12-month-olds
did not generalize the modeled action to a new object as readily as older (18- and
21-month-old) infants did. These data thus provide evidence of short-term recall
as early as 6 months. There was also compelling evidence of ordered recall after
the delay: Whereas 75 % of the 6-month-olds produced one action after 24 h, only
25 % of them memorized more than one step of the sequence.
In a variation of deferred imitation, Bauer and her colleagues used elicited imi-
tation in which the experimenter provides verbal prompts during modeling, and
infants imitate immediately and occasionally during the retention interval as well
(e.g., Bauer et al. 1994; Bauer and Shore 1987). Taken together, the data obtained
by using this paradigm demonstrate impressive LTM in very young children. For
instance, Bauer and Shore (1987) found that 1- to 2-year-olds exhibited retention
after 6 weeks. Their results also indicated that the structure of the target event influ-
enced the outcome. That is, infants’ imitation of events that can only be exhibited
in a special temporal order (causal or enabling events) is superior to their imitation
The Development of Long-Term Memory in Infants and Toddlers 59

of events that can be performed in any order (arbitrary events). By 9 months of age,
developments in ordered recall ability are already apparent (Carver and Bauer 1999).
When the task was to imitate novel two-step event sequences after a 5-week delay,
almost half of the 9-month-olds showed recall of these sequences. Older infants (11-
to 24-month-olds) represented order information in their recall of longer sequences.
Moreover, their recall was facilitated and prolonged if the components of the events
were familiar and contained enabling relations (Bauer 1995; Bauer et al. 2002).
Subsequent retention tests after increasing delays of 3 to 12 months showed an age-
related increase in the robustness of long-term retention (Bauer 2002).
Although procedural differences between deferred imitation and elicited imi-
tation have been considered to be negligible, a systematic comparison between
the two approaches revealed that this is not true. For instance, Hayne, Barr, and
Herbert (2003) tested the effects of practice on 18-month-old infants’ memory
performance. All infants were tested using an imitation paradigm. Whereas some
of the infants practiced the actions before retention (elicited imitation), others
did not (deferred imitation). As a main result, Hayne et al. (2003) found that the
opportunity to practice the actions made a difference when long-term retention
was assessed 6 weeks later by using a brief reminder treatment. The reminder pre-
vented forgetting by infants who had practiced the actions before but did not have
similar effects for those infants who did not practice. The opportunity to prac-
tice not only affected long-term accessibility of the memory but also had positive
effects on generalization trials.

Does the Age of 9 Months Indicate an Important Transition

Period?

Findings from several studies using elicited imitation procedures suggest that the age
of about 9 months indicates a transition period in infants’ abilities to store and/or
retrieve information across extended periods of time. Whereas about half of the chil-
dren showed evidence for temporally ordered recall memory, the other half did not
seem able to master such tasks (Carver and Bauer 1999; Bauer et al. 2001). In fact,
Carver et al. (2000) reported an association between infant brain activity as meas-
ured by event-related potentials (ERPs) and behavioral evidence of ordered recall.
Those infants showing earlier evidence of ordered recall had shown evidence of dif-
ferential ERP responses to pictures of event sequences that were new to them.
Another important change occurring when infants are about 9 months of age
concerns the emergence of independent locomotion. According to Rovee-Collier
(1996), infants who begin to crawl are faced with the ongoing challenge of recog-
nizing old objects in new places. As a consequence, they may become more resil-
ient to changes in context. On the basis of her research, she was able to show that
infants begin to generalize across changes in context at about the age of 9 months.
More direct evidence supporting this assumption comes from a study by Herbert
et al. (2007). Herbert and colleagues compared deferred imitation by infants
60 4  Memory Development During the Infant and Toddler Years

who were crawling with deferred imitation of infants of about the same age (all
9 months of age old) who were not yet crawling. The experimenter demonstrated
a single action with a novel object, and all infants were tested for imitation about
24 h later. Within each crawling condition, some infants were tested with the dem-
onstration object in the familiar context, and the remaining infants were tested
with a novel object in a different context. Whereas all infants imitated the action
with the demonstration object in the familiar context, only the infants who were
already crawling exhibited imitation when tested with a novel object in a differ-
ent context. Herbert et al. (2007) concluded that those infants who were already
crawling showed greater flexibility in memory retrieval than did infants of the
same age who were not yet crawling (for further confirming evidence, see Hayne
and Simcock 2009). These findings are in accord with the position that the onset
of independent locomotion (i.e., crawling) represents not only a major motor mile-
stone, but a more general cognitive milestone as well (see Campos et al. 2000).
There is also evidence that the ability to encode experiences and events in rela-
tional representations (thus allowing a more flexible use of memories) emerges in
early infancy at about the age of 9 months. It is widely assumed that the devel-
opment of relational memory is related to maturational processes in the hip-
pocampus and supporting medial temporal lobe (MTL) structures (e.g., Olson
and Newcombe 2014). One core assumption of the relational processing theory
is that the MTL is important for encoding into memory what has been described
as binding, contextual, or relational information (Eichenbaum 1999). Numerous
behavioral studies suggest that the ability to represent memories in relational net-
works develops gradually during infancy and early childhood (e.g., Sluzenski et al.
2004). In a recent study, Richmond and Nelson (2009) used an eye-tracking meas-
ure to explore the emergence of relational memory. They found that 9-month-olds
can form memories that represent relations among items (scenes and faces) and
maintain them over short delays. According to Richmond and Nelson, their study
highlights the idea that eye movement measures may have the potential to add to
our understanding of the development of relational memory. It should be noted,
however, that Koski et al. (2013) reported discrepant findings for 4-year-old chil-
dren based on the same experimental procedure, suggesting that preschoolers are
capable of forming only weak relational memories. This finding challenges the
view that eye movements provide an accurate measure of memory independent of
conscious remembering. Clearly, more research is needed to validate Richmond
and Nelson’s findings and also the assumption that the age of 9 months represents
an important transition period in infants’ memory development. According to a
recent review of the literature by Olson and Newcombe (2014), the development
of relational memory is characterized by an initial discontinuity around the end of
the second year of life, followed by a period of at least 4 years during which this
kind of memory slowly strengthens. Olson and Newcombe linked this process to
the development of the hippocampus, which matures rapidly from ages 0–2 years,
followed by slower growth in subsequent years.
On the other hand, however, research based on the elicited imitation paradigm
clearly indicates that the long-term retention of ordered actions emerges near the
The Development of Long-Term Memory in Infants and Toddlers 61

end of the first year of life. As shown in a large-scale study of remembering and
forgetting over the course of the second year (Bauer et al. 2000), the ability to
remember undergoes significant development and consolidation during the second
year of life. In this study, Bauer et al. (2000) tested 360 infants at the ages of 13,
16, and 20 months on an elicited imitation task, exposing them to the same six
multistep event sequences at each of the three sessions, spaced 1 week apart. The
infants returned for delayed recall assessments after delay intervals of 3, 6, 9, or
12 months, with the delay condition serving as the between-subjects manipulation.
Here, infants were tested for recall of the six-event sequences to which they had
previously been exposed as well as on three novel event sequences as a within-
subjects control. In all cases, infants first experienced a delayed recall period
during which they were prompted by the event-related props alone, followed
by verbal reminders of the event sequences (see Bauer et al. 2000, for details of
the procedure). As a main result, 78 % of the infants in the youngest age group
(i.e., the 13-month-olds) showed temporally ordered recall after a delay interval
of 1 month, a score reliably greater than chance. Beyond that month, the num-
ber of 13-month-olds who performed at higher levels for previously experienced
events than for novel events did not exceed the chance level. About 72 % of the
16-month-olds recalled the six-event sequences after 6 months, and 67 % of the
20-month-olds showed comparable performance after 12 months. Bauer et al.
(2000) concluded that the reliability for long-term ordered recall increases signifi-
cantly during the second year of life (for more recent validations, see Lukowski
and Bauer 2014). Furthermore, they found that memory became increasingly
robust across the second year of life and that age differences were particularly
apparent under conditions of greater cognitive demand. However, Bauer et al.
(2000) also noticed that accurate recall of sequences that were arbitrarily ordered
(and thus did not contain enabling relations) was a later development. It was not
until about 20 months of age that infants performed above chance levels on such
tasks after short delays (i.e., after 2 weeks). By the age of 28 months, infants
recalled such arbitrarily ordered sequences well after a delay (Bauer et al. 1998).
Overall, it appears that the range of behaviors that infants can imitate after a
delay increases with age beginning with facial and body movements, then spread-
ing to specific actions applied first to a specific object and then to similar objects,
and finally to intended actions and social goals (Rovee-Collier et al. 2001). This
steady developmental progress seems to reflect developmental increases in infants’
motor and cognitive abilities as well as age changes in their social competencies.

Effects of Memory Reinstatement and Reactivation

As shown above, there is overwhelming evidence that young infants can remem-
ber training experiences for rather long time intervals. However, there is also no
doubt that these experiences will be forgotten after some delay without further
reminders. Reminder paradigms such as memory reinstatement and memory
62 4  Memory Development During the Infant and Toddler Years

reactivation have been used to explore the effectiveness of additional memory

cues. It seems important to distinguish between the two procedures. In the rein-
statement procedure, memory is maintained over time by periodically exposing
infants to abbreviated versions of the original event (e.g., the contingency in the
mobile task). In the reactivation procedure, the forgotten memory was recovered
by a single exposure to some of the original training conditions at the end of the
retention interval. As Rovee-Collier and Gerhardstein (1997) put it, reinstatement
prevents forgetting, whereas reactivation alleviates it.
According to findings by Campbell and Jaynes (1966) based on a study with
infant rats, the installment of periodic reinstatement reminders helps to keep the
memory for the original learning situation alive and prolongs retention intervals
considerably. Similar findings were also reported by Cornell (1979), who tested
5- to 6-month-olds for recognition memory of briefly presented visual stimuli.
In one of his experiments, substantial delayed recognition was observed when
a short exposure of materials was introduced immediately prior to the recogni-
tion test. A subsequent experiment in this study revealed that the effects found
in the first experiment were not due solely to familiarity with the experimental
procedure.
In general, reinstatement has been used with older children (e.g., Howe et al.
1993), whereas reactivation has been used with infants. Several studies carried out
by Rovee-Collier and colleagues with young human infants based on the mobile
task confirmed the general validity of the findings presented by Campbell and
Jaynes (1966) and Cornell (1979). For instance, in a reactivation study conducted
by Rovee-Collier et al. (1980), 3-month-old infants were trained on the mobile
task, allowed to forget it, and then briefly exposed to a memory prime 2–4 weeks
after training. As a main result, priming restored learned responses to their origi-
nal levels. Interestingly, the reactivated memory was forgotten at the same rate as
the original one. In subsequent studies, both results were confirmed for infants of
different ages with single primes doubling the maintenance of a memory and mul-
tiple primes prolonging it even more (for reviews, see Hayne 2007; Rovee-Collier
and Cuevas 2009).
Findings reported by Hildreth and Hill (2003) indicate that there is an UL
to how much time can elapse after an experienced event before a reactivation
reminder becomes inefficient. As shown in Fig. 4.6, the absolute UL of reactiva-
tion increases logarithmically between 3 and 12 months. However, the relative UL
of reactivation, defined as the ratio of the absolute UL to the maximum amount of
time the initial event is remembered, is constant across age and is about four times
longer than the original retention. As noted by Rovee-Collier and Cuevas (2009),
there is a possibility that the relative UL of reactivation is constant across the life
span.
Using similar procedures such as behavioral re-enactment, researchers have
shown that young children’s memory for events that occurred during their infant
and toddler years are retained and can be remembered for several months even
though memory for these events becomes increasingly fragmentary with the pas-
sage of time. For instance, McDonough and Mandler (1994) reported evidence of
The Development of Long-Term Memory in Infants and Toddlers 63

Absolute UL of Reactivation Relative UL of Reactivation

32 5:1
Maximum Duration (weeks since training)

28
4:1
24

Ratio of Ul : Orig Retention

20
3:1
16

12 2:1

8
1:1
4

0 0
0 3 6 9 12 3 6 9 12
Age (months)

Fig. 4.6  The upper limit (UL) of reactivation (the longest delay after training at which a forgot-
ten memory can be reactivated) at 3, 6, 9, and 12 months of age. Left panel The absolute UL,
expressed as weeks since original training (3-month point from Greco et al. 1986). Right panel
The relative UL, expressed as a ratio of the absolute UL to the maximum duration of original
retention. (Reprinted from Hildreth and Hill 2003)

recall of specific actions in 2-year-olds who had participated in these actions about
1 year earlier. Moreover, as shown by Bauer et al. (2000), repeated retention tests
improve the recall of multistep action sequences in elicited imitation experiments,
showing an age-related increase in the robustness of long-term retention.

The Impact of Language Acquisition on Toddlers’ LTM

One crucial difference between infants and toddlers concerns the emergence of
language. There is no doubt that the acquisition of language skills provides new
ways for young children between 2 and 4 years of age to encode and retrieve expe-
rienced events. In fact, research using imitation paradigms has repeatedly shown
that adults’ verbal cues facilitate memory performance by preverbal and early-
verbal children (e.g., Bauer et al. 2000; Hayne and Herbert 2004). For example,
Hayne and Herbert (2004) presented 18-month-olds with a deferred imitation
task and tested their imitation 4 weeks later. Some infants received “full narra-
tion” at both the demonstration and test trials, meaning that the event goal and
the target actions were verbally described. Other infants received what the authors
called “empty narration,” typically used in studies of deferred imitation in which
only a few general suggestions and comments were made. Although both types
of verbal cues facilitated performance during the long-term test, the infants in the
64 4  Memory Development During the Infant and Toddler Years

“full narration” condition clearly outperformed the infants in the “empty narra-
tion” condition. As emphasized by the authors, the verbal cues were most effective
when presented at the time of the test rather than at the time of demonstration.
Thus, by the age of 18 months, infants can use verbal retrieval cues to facilitate
long-term retrieval. Similar findings were also obtained in elicited imitation stud-
ies in which infants were allowed to practice the target actions prior to the reten-
tion interval (for a review, see Bauer et al. 2000).
An interesting related issue concerns the question of how memory changes
when toddlers begin to describe experienced events in their own words. One
straightforward approach has been to compare young children’s verbal and non-
verbal recall of a past event (e.g., Bauer et al. 1998; Bauer and Wewerka 1995,
1997). For example, Bauer et al. (1998) analyzed the spontaneous verbal expres-
sions of some of the 16- and 20-month-olds who had taken part in Bauer et al.’s
(2000) study. At the time of testing, the children were between 22 and 32 months
of age. Although the researchers did not encourage verbal comments, most chil-
dren spontaneously produced verbalizations in the sessions, some of which were
indicative of memory for the past events. Similarly, Bauer and Wewerka (1995)
noticed that at the time of testing, children made more mnemonic verbal com-
ments that referred to an experienced event with the props present when they were
tested with objects they had seen before than they did when they were tested with
novel objects. Bauer and colleagues concluded that early-verbal children are able
to spontaneously express their memories even after long delays when the context
is supportive. Accordingly, young children seem to have verbal access to their pre-
verbal experiences.
As noted by Hayne and Simcock (2009), such a conclusion may be too opti-
mistic given that none of the children exhibited verbal recall in the absence of
familiar objects. In subsequent studies that assessed children’s verbal comments
prior to when the children saw the props, almost no spontaneous verbal recall
for the past event was recorded (e.g., Bauer et al. 2002). Even if children have
some verbal access to their preverbal memories, that access seems rather incom-
plete and rare. This conclusion was supported by a study that assessed 2- to
4-year-old children’s memories for a unique event (Simcock and Hayne 2003).
In this study, children were shown how to operate a special machine that could
“shrink” several toys (the Magic Shrinking Machine). When tested 24 h later, the
verbal reports provided by the older children in the sample were more compre-
hensive, with the 3- to 4-year-olds recalling more than twice as many items than
the younger children. However, as shown in Fig. 4.7, verbal reports were rather
sparse even for the oldest children in the sample who knew the words required to
describe the events. In comparison, most of the 3- to 4-year-olds could correctly
identify the toys on tests of recognition and were also able to successfully repro-
duce the actions that were required to operate the machine. Simcock and Hayne
(2003) inferred that toddlers continue to rely primarily on their nonverbal repre-
sentations of events and that early-verbal children may find the task of translat-
ing memories originally encoded in some kind of nonverbal format into language
very difficult.
The Development of Long-Term Memory in Infants and Toddlers 65

0.8
Proportion Recalled

Memory
Measure:
0.6

Verbal Recall
0.4
Photo Recognition
Re-enactment
0.2

0
2 Years 3 Years 4 Years
Age Group

Fig. 4.7  Age-related changes in verbal recall, photorecognition, and re-enactment after a 24-h

delay by 2- to 4-year-olds who were tested using the Magic Shrinking Machine (based on data
from Simcock and Hayne 2003)

Determinants of Young Children’s Long-Term Recall

Although the findings summarized above seem to indicate that memory devel-
opment in infancy and the ability to retrieve information after longer time inter-
vals depends mainly on chronological age, this variable may be only one of the
determinants of how long a memory survives and how rapidly forgetting occurs.
As noted by Bauer et al. (2011), it is very unlikely that we will ever be able to
establish “growth-chart”-type functions for the anticipated durations of memory
traces for infants of different ages. Memory performance is multiply determined,
and a number of factors such as the features of the task, the intensity and duration
of training periods, and the delay between learning and testing have been shown
to affect infants’ recall. Moreover, as indicated by Rovee-Collier (1995), the “time
window,” that is, the limited period of time within which successive stimuli are
integrated, may determine infants’ long-term retention. A time window opens
when an event occurs and shuts when it can no longer be retrieved. Predictions
derived from the time-window construct are that (a) successive events are inte-
grated when the second event occurs within the time window of the first, (b) each
retrieval expands the width of the time window, and (c) the later in the time win-
dow an event is retrieved, the longer it will be retained into the future. As reported
by Rovee-Collier and Cuevas (2009) in more detail, recent data have found
support for these predictions across infancy (e.g., Hsu 2007). The conclusion
that Rovee-Collier and colleagues drew from these findings is that even though
younger infants forget more rapidly, their retention intervals are determined
experientially, not maturationally. That is, the width of the time window and the
66 4  Memory Development During the Infant and Toddler Years

number as well as the strength of associative links between the target memory and
other events affects long-term retention. Rovee-Collier and Cuevas (2009) thus
propose that younger infants forget more rapidly because they have fewer associa-
tions to which the memory can be linked.
The question of developmental changes in the long-term retention of informa-
tion is particularly difficult to address because the retrieval of information about a
previously experienced event depends on what was originally encoded about the
event (see Courage and Howe 2004; Howe 2000). Conclusions about developmen-
tal changes in LTM are particularly problematic when age-related effects in initial
learning were not experimentally controlled for. Given that older infants learn faster
than younger infants and that individual differences in learning rates contribute to
individual differences in retention rates (i.e., slower learners forget more rapidly),
younger infants will forget more rapidly because they are slow learners. As a conse-
quence, age differences observed in long-term retention become difficult to interpret.
The studies that routinely controlled for levels of learning at the end of acquisi-
tion (e.g., Hartshorn et al. 1998; Hartshorn et al. 1998; Howe and Courage 1997)
found that there are developmental differences in long-term retention independ-
ent of developmental differences in learning. For instance, Howe and Courage
(1997) investigated infants’ forgetting of novel activities across a 3-month interval.
They observed a developmental decline in forgetting as a function of age such that
12-month-olds forgot more rapidly than 15-month-olds, who in turn forgot more
about the events than 18-month-olds. Similarly, Hartshorn et al. (1998) found that
infants from 2 to 18 months of age who were trained on the same learning crite-
rion showed a linear increase in retention as a function of age. Although it is not
possible to equate levels of initial learning across age in deferred imitation and
elicited imitation studies, the studies that included immediate retention control
groups found that there were reliable increases in both the amount of informa-
tion retained and the duration of retention with increasing age across the ages of
6–24 months (e.g., Barr and Hayne 2000; Bauer et al. 2000; Meltzoff 1995). In
order to eliminate differences in encoding processes as a potential source of devel-
opmental differences in long-term recall, Bauer (2005) reanalyzed the data from
Bauer et al. (2000), matching subsets of 13- to 16-month-olds and subsets of 16-
to 20-month-olds for levels of encoding. The amount of information the infants
forgot over the delays was then examined. For both comparisons, older infants
exhibited less forgetting relative to younger infants. So there is evidence that age
differences account for considerable amounts of variance in the recall data.
However, as emphasized by Courage and Howe (2004), we still do not know
whether individual differences in encoding ability, consolidation, or retrieval
mediate developmental improvements in infant and toddler memory. In a num-
ber of studies, Howe and colleagues (e.g., Howe 2000; Howe and Brainerd 1989;
Howe et al. 1993) used the trace integrity framework and its associated math-
ematical model to disentangle the effects of storage and retrieval processes in
infants’ and children’s long-term retention. The trace integrity framework assumes
that storage and retrieval are processes that lie on a single continuum and that
traces consist of collections of primitive elements such as features or nodes. The
The Development of Long-Term Memory in Infants and Toddlers 67

key to initial learning success is integrating features into a single cohesive struc-
ture in memory. Traces that are not well integrated tend to disintegrate more
quickly, and the original memory trace loses its cohesion and distinctiveness.
Howe and colleagues used a variety of memory tasks, materials, and experimental
manipulations to test the assumptions of this theoretical framework (for a review,
see Howe and O’Sullivan 1997). As a main result, they found that with increasing
age, infants and toddlers (as well as older children) were better able to maintain
information in storage. Thus, forgetting is dominated by storage failure rather than
retrieval failure, and storage failure also decreases with age in childhood. This
finding is also in line with observations by Bauer et al. (2000), who concluded that
their data strongly implicate storage and consolidation as opposed to retrieval pro-
cesses as the major sources of developmental change.

Kinds of Memory: Implicit and Explicit Memory in Infancy

As already noted above (cf. Chap. 3), most memory researchers agree that human
memory does not consist of a unitary system. It is widely believed that memory is
comprised of different systems or processes that are supported by different neural
substrates and serve different functions. In the adult memory literature, a popu-
lar distinction is the nondeclarative (implicit) and declarative (explicit) memory
dichotomy (e.g., Squire 1992). The nondeclarative system represents a variety of
nonconscious abilities such as learning habits and skills, priming, and some forms
of conditioning, all of which are not consciously accessible. The declarative mem-
ory system refers to the recognition or recall of objects or events that are available
to consciousness. Within this system, a further distinction has been made between
episodic and semantic memory (Tulving 1972, 2002), broadly characterized as the
difference between “knowing” and “remembering.” Whereas semantic memory
refers to knowledge of general facts such as “Paris is the capital of France,” epi-
sodic memory is about events in particular places at particular times, emphasiz-
ing the what, where, and when (e.g., “I twisted my ankle at our school’s soccer
field last Friday”). Although the semantic and episodic components of declara-
tive memory were originally thought to be parallel systems, Tulving (2002) more
recently suggested that the relation may be more hierarchical than parallel, with
the assumption that episodic memory emerges out of semantic memory.
The distinction between different types of memory was originally derived from
studies exploring adult cognition but also seemed very important for developmen-
tal scientists who assumed that declarative and nondeclarative memory rely on
different neural substrates that show different courses of development. As noted
by Bauer and colleagues (e.g., Bauer 2007; Bauer et al. 2011), a variety of brain
regions are involved in implicit memory processes, including the neocortex (prim-
ing), striatum (skill learning), and cerebellum (conditioning). These regions are
assumed to develop very early and to support the early emergence of nondeclara-
tive memory (see Lukowski and Bauer 2014; Nelson 1997; Richmond and Nelson
68 4  Memory Development During the Infant and Toddler Years

2009, for reviews). By contrast, explicit declarative memory depends on a com-

plex neural network that includes medial temporal lobe structures (including the
hippocampus and perirhinal cortex), the medial diencephalon, and higher cortical
association areas such as the prefrontal cortex (for reviews, see Bachevalier and
Mishkin 1994; Zola and Squire 2000). Initially, researchers believed that human
infants could exhibit only nondeclarative memory skills, at least during their first
year of life, and that declarative memory emerged around the end of the first year
(e.g., Schacter and Moscovitch 1984). However, findings from recent studies on
infant memory indicate that such an assumption is probably not correct, and that
the situation is rather complex.
It seems obvious that the distinction between explicit and implicit memory is
harder to work with when investigating infant memory than when studying adult
memory, mainly because it is difficult if not impossible to determine whether
infants show conscious memory behavior. Given that infants are preverbal and
cannot speak, procedures analogous to those used to test declarative memory in
adults but that do not require verbal instructions or verbal expression had to be
developed. As shown in the previous chapter, assessments of memory dissocia-
tions in amnesic adults (i.e., illustrations of impaired performance on tasks tapping
declarative memory and normal performance on tasks tapping implicit memory)
have proven to be useful in this regard. Further evidence supporting the distinc-
tion between explicit and implicit memory in the adult memory literature comes
from various sources. For instance, experimental memory studies with normal
adults have shown dissociations between the two types of memory such that the
same experimental manipulations had different effects on tasks tapping explicit
memory (i.e., recognition and recall) and tasks of implicit memory (e.g., priming).
Similarly, brain imaging studies assessing recognition memory in normal adults
have shown robust and selective activation in the hippocampal region, and stud-
ies using animals with lesions to specific areas in the medial temporal lobe have
indicated that the lesions impair recognition and recall but do not affect perfor-
mance on tasks that tap implicit memory (see Bauer 2009; Nelson 1997, 2000;
Richmond and Nelson 2009; Zola and Squire 2000). As emphasized by Richmond
and Nelson (2008), neuropsychological research has shown that the hippocampus
is crucially involved in declarative memory performance, and there are also behav-
ioral studies indicating that rudimentary “pre-explicit” declarative memory may be
evident early in infancy, which begins to give way to adult-like explicit memory
abilities around 8 months of age (see Nelson 1997).
Do infants indeed show signs of declarative memory, and is it possible to
classify the major experimental paradigms used to assess long-term retention in
infants described above according to whether they assess explicit versus implicit
memory? As a developmental memory researcher not specializing in infancy, I
was impressed to see how much time and energy the experts have invested in the
discussion of the issue of multiple memory systems (e.g., see Bauer 2007; Bauer
et al. 2007, 2011; Hayne 2007; Rose et al. 2007; Rovee-Collier 1997; Rovee-
Collier et al. 2001). Particularly interesting in this regard is the second part of
the book edited by Oakes and Bauer (2007) addressing LTM in infancy and early
The Development of Long-Term Memory in Infants and Toddlers 69

childhood. This book not only very nicely presents discrepant views on this topic
but also provides sophisticated comments by distinguished researchers on the vari-
ous chapters. In the following, an attempt is made to briefly summarize the most
important aspects of this discussion. For more details, I refer the reader to Chaps.
7–12 of Oakes and Bauer’s inspirational book.
What kind of memory is assessed by the major experimental paradigms used
in research on infant LTM (i.e., the VPC task, the mobile conjugate reinforcement
task, and the deferred/elicited imitation task)? There is disagreement regarding the
VPC task. Whereas Richmond and Nelson (2008), Rose et al. (2007), and Hayne
(2007) are convinced that the task assesses declarative visual recognition mem-
ory, Snyder (2007) and Mandler (2007) assume that it measures implicit memory.
To support their view that the VPC task taps explicit memory, Rose et al. (2007)
cite evidence from neuropsychological work with nonhuman primates, showing
impaired performance of monkeys with lesions in the medial temporal lobe on the
VPC task. They refer to similar outcomes in studies with amnesic adults, who also
exhibit impaired performance on VPC tasks. Finally, evidence from experimental
studies on adult versions of the VPC tasks indicates that adults show preferential
looking for novel stimuli just as infants do and even more importantly that their
preferences in VPC tasks were correlated with their confidence judgments regard-
ing traditional recognition responses after longer delays. Richmond and Nelson
(2008) note that the VPC task passes both the amnesia filter and the parameter
filter as a measure of declarative memory: That is, patients with damage to the
medial temporal lobe exhibit no preference when tested on the VPC task after
some delay, and performance on the VPC task is affected by study time, retention
interval, and context change in the same way that these variables affect adults’ per-
formance on other measures of declarative memory.
As pointed out by Snyder (2007), however, there are problems with this appar-
ently supporting evidence. For instance, the neurological damage in amnesic patients
and monkeys with damage to the hippocampus participating in the VPC studies was
not confined to the hippocampus. Moreover, the VPC tasks used in studies with nor-
mal and amnesic adults varied significantly from those used with infants and thus
do not represent a suitable basis for comparison. In light of the available evidence,
Snyder (2007) concluded that novelty preferences in VPC tasks do not constitute a
measure of explicit memory. She suggested that greater attention to novel stimuli
may result from a property of the visual system rather than a recognition response
per se, assuming that repetition suppression, that is, a reduction of neuronal
responses in the occipital temporal visual processing pathway is responsible for the
shift in attention from an old to a new stimulus. Snyder’s argument is based on evi-
dence that the visual system is biased toward the processing of new stimuli with an
associated reduction in neuronal responses to the visual processing pathway.
A better understanding of the kind of memory assessed with VPC tasks is
provided by studies that have used ERPs to investigate the neurophysiologi-
cal processes underlying infants’ novelty preferences. In particular, the use of
ERPs provides evidence for which brain areas are involved in infants’ preferen-
tial looking. More precisely, if novelty preferences are associated with increased
70 4  Memory Development During the Infant and Toddler Years

amplitudes of ERP components in anterior frontal temporal regions, this would

support an explicit memory interpretation. However, if novelty preferences are
associated with decreased amplitudes of ERP components in the posterior occip-
itotemporal regions, this would support the suppression repetition argument and
favor the implicit memory interpretation. In fact, the only study investigating this
issue (Snyder et al. 2006) supported the latter assumption, showing that preference
for novel faces and objects was accompanied by decreases in electrical activity in
the temporal regions. So there is preliminary evidence for the implicit memory
hypothesis but clearly not enough. I agree with Newcombe and Crawley’s (2007)
claim that the debate about which form of memory is indexed by the VPC task is
not resolved. Future studies focusing on this issue are clearly needed.
An important point made by Snyder (2007) is that a one-to-one mapping
between performance on the VPC task and a single underlying competency seems
unlikely. Thus, her argument is that performance on the VPC reflects the interac-
tion between visual attention and memory and that memory is one process that
guides visual attention on such tasks. This assumption should be followed up
more precisely in future studies. Furthermore, as noted by Mandler (2007) and
Newcombe and Crawley (2007), future research on this issue should also consider
the dual-process nature of recognition memory in adults (i.e., the assumption that
adult recognition judgments are jointly determined by feelings of familiarity and
by specific recollection; see Ghetti and Lee 2014; Ornstein and Light 2010). Given
that unlike adult recognition memory tests, the VPC task does not include deliber-
ate requirements, the responses shown on the VPC may be called “primitive recog-
nition” (Mandler 2007). Although they reflect access to the familiarity component,
they are probably nondeliberate and nonconscious in nature.
Does performance on deferred/elicited imitation tasks index explicit mem-
ory? Most infant memory researchers think so (including those discussing issues
of infant LTM in the volume edited by Oakes and Bauer 2007). Bauer and col-
leagues (e.g., Bauer 2002, 2007; Bauer et al. 2007, 2011) list several reasons
for arguing that imitation-based tasks measure explicit memory and can be con-
ceived of as nonverbal tests of deliberate recall. One reason is that, once young
children acquire language, they are able to talk about events experienced on pre-
vious imitation tasks (e.g., see Bauer et al. 2002). This suggests that children’s
event representations are explicit and declarative. Moreover, imitation proce-
dures assess ordered recall, not recognition. The actions and their order must be
encoded during presentation and must later be retrieved from a representation of
the original event in the absence of ongoing support, a process that closely mir-
rors the demands of verbal recall paradigms (Lukowski and Bauer 2014). Finally,
imitation-based tasks pass the “amnesia test.” That is, adults with medial tempo-
ral lobe amnesia, in whom explicit memory processes are compromised, were also
impaired on an imitation-based task using multistep sequences, and this finding
did not hold for a control group of adults (McDonough et al. 1995). Overall, this
evidence suggests that imitation-based tasks are measures of explicit memory.
Is this also true for the conjugate mobile reinforcement paradigm? Rovee-
Collier and colleagues have discussed at length that infants’ behavior on this task
The Development of Long-Term Memory in Infants and Toddlers 71

indexes explicit memory, emphasizing the fact that the mobile and deferred imi-
tation tasks are more similar than they are different (e.g., Hayne 2007; Rovee-
Collier 1997; Rovee-Collier et al. 2001). Bauer et al. (2007) also acknowledge
that there are apparent similarities in the contents and functions of memory as
well as and in the rules by which memory seems to operate in conjugate reinforce-
ment and imitation-based tasks. On both tasks, infants learn about and remember
objects, and there are also similarities across tasks in the rates of learning, indi-
cating that learning is rapid on both tasks and that older infants learn at a faster
rate than younger ones. Finally, on both tasks, changes between encoding and
retrieval contexts are tolerated with increases in the ages of the infants, indicating
that memory becomes less context-dependent as a function of age. Nonetheless,
Bauer et al. (2007) believe that the memory functions involved in the mobile con-
jugate reinforcement tasks represent implicit memory. These memory functions
are assumed to develop at an earlier age and most likely depend on the cerebellum
and certain deep nuclei of the brain stem.
Given the many apparent similarities between the two tasks, the question arises
as to why the two tasks should be indicative of different memory systems. Bauer
et al. (2007) offer two answers to this question. First, they assume that some of the
similarities are elusive, using the demonstration of memory for temporal order as an
example. In reinforcement paradigms, the test of infants’ memory for information
about order is whether they change their rate of responding when temporal order
is violated. On imitation-based tasks, on the other hand, temporal order must be
encoded at the time of presentation, and temporal order has to be reproduced from
memory. Whereas sensitivity to temporal patterns on the mobile task may occur even
in the absence of conscious awareness, deliberate memory of the temporal sequence
of actions is required on deferred/elicited imitation tasks. Second, Bauer et al. (2007)
pointed out similarities in mnemonic behavior even as it “crosses” different memory
systems. Given that the same neural structures participate in multiple memory net-
works, there is reason to expect similarities in mnemonic behavior.
As noted by Newcombe and Crawley (2007), one problem in evaluating what
kind of memory is tapped by operant conditioning tasks is that there do not appear
to be any such studies with amnesic adults, imaging studies, or studies with non-
human animals. Thus, the question about whether the mobile conjugate rein-
forcement task paradigm assesses explicit memory still remains open. Given that
most infants showing explicit memory in deferred imitation studies are 6 months
of age or older, we still lack definite evidence regarding explicit memory in the
first 6 months of life. On the other hand, findings also contradict the original view
mentioned above that there are no signs of declarative memory before the end of
the first year of life. The findings basically seem to be in accord with the posi-
tion that the development of implicit and explicit memory is not strictly hierarchi-
cal, and they seem compatible with the view that the memory systems that support
implicit and explicit memory are both present from early in infancy (Courage and
Howe 2004; Rovee-Collier et al. 2001).
Given that declarative memory emerges very early in human development,
another obvious question is whether infants and toddlers also exhibit episodic
72 4  Memory Development During the Infant and Toddler Years

memory as well. Although it seems premature to make any claims about this
issue, it is possible that memory in infants and toddlers is primarily semantic
(Newcombe and Crawley 2007). Such a developmental pattern would be quite
adaptive, as it is clearly more relevant for very young children to learn about the
world in a general way than to know when, where, and from whom information
was obtained. Such a sequential pattern would also be in accord with Tulving’s
(2002) speculation that episodic memory grows out of semantic memory. This
assumption would also help to explain the fact that children younger than 2 years
of age can form explicit memories, but later in their lives will not retain memo-
ries for autobiographical episodes (Bruce et al. 2000). As noted by Newcombe and
Crawley (2007), semantic memory ability may strengthen during infancy, as indi-
cated by the increasing generalization of memory across contexts.
On the other hand, however, there is now also evidence that the minimum age
for the development of episodic memory is likely to be much earlier than previ-
ously assumed. Although it is difficult to say with certainty whether preverbal
infants’ recall of event sequences is episodic, research on the elicited imitation
paradigm shows that infants remember specific events after substantial delays,
with memory for special features associated with higher levels of recall (see
reviews by Bauer et al. 2011; Lukowski and Bauer 2014). Thus, several infant
researchers argue that episodic memory is evident at least by the end of the second
year of life. Prospective data on hide-and-seek tasks also suggest that the mini-
mum is at least 3 years (DeLoache 1984; Hayne 2007; Hayne and Simcock 2009).
The available evidence suggests that the earliest roots of episodic memory can
be found around the second birthday, at about the time that long-lasting autobio-
graphical memories begin to appear. There is also reason to assume that the first
episodic memories are fragile and that the years between ages 2 and 6 are full of
rapid developments in episodic skills (Newcombe and Crawley 2007). Supporting
evidence for this assumption comes from research on autobiographical memory
(see Chap. 5), indicating, for example, that the ability to determine whether an
event really occurred or was only imagined improves markedly in this age range
(Sluzenski et al. 2004).

Summary

The review of research on infants’ LTM in this chapter provides plenty of evidence
of remarkable progress in the field. The development of sophisticated research
tools and techniques during the last three decades has yielded important informa-
tion about young infants’ memory abilities and has thus changed former beliefs.
Although infants and toddlers cannot tell us what they remember, the use of innova-
tive techniques has shown that young infants can readily form new and relatively
permanent associations between objects and events that are physically present, acti-
vate memory representations of objects and events and a physically present object
or event, and also activate memory representations of absent objects and events.
The Development of Long-Term Memory in Infants and Toddlers 73

Research based on operant conditioning techniques such as the mobile conjugate

reinforcement task indicates that the duration of LTM in young infants increases
linearly from about 1–2 days for infants trained at 2 months to 13–14 weeks for
infants trained at 18 months (see Rovee-Collier and Cuevas 2009). Research based
on deferred and elicited imitation techniques also shows remarkable strides in recall
across the first year of infancy and beyond. Whereas the temporal extent of memory
seems limited in the first 6 months of life, 6-month-olds already exhibit evidence
of recall for 24 h, and 9-month-olds recall multistep sequences after delays of
1 month. This research has also demonstrated the important roles of encoding and
memory consolidation processes for the long-term retention of objects and events,
related to developmental trends in neural “hardware.” There is ample evidence that
re-exposure to an event can facilitate long-term recall in infants and young children.
Systematic studies of reinstatement effects in infants have also shown that multiple
reinstatements can extend memories for very long time intervals, with the timing of
a reminder exposure also affecting recall. Reminders that are introduced later, when
a memory is no longer active, are more effective at enhancing the recall of older
children (see Hudson and Grysman 2014).
Recent research in cognitive neuroscience indicates that there is correspond-
ence between the time course of changes in memory behavior and performance
during the first year of life and what is known about the development of aspects
of the temporal cortical network that supports declarative memory (Nelson 1997;
Bauer 2007, 2009; Richmond and Nelson 2008). Several experiments have shown
that when initial levels of learning are equated across age, infant age, and the
length of the retention interval are positively correlated. That is, older infants
retain information about objects, people, and events across longer intervals than do
younger infants, and older toddlers integrate more information across longer time
periods than younger toddlers, allowing them to use what they learn after increas-
ingly longer delays (Howe et al. 2009; Hayne and Simcock 2009). As the neural
structures that subserve memory mature, memories become more durable and less
vulnerable to decay. As shown in the last section of this chapter, infant memory
experts disagree about whether the classic experimental tasks used to assess long-
term retention actually tap declarative memory. Overall, the experimental find-
ings related to this issue seem to indicate that although brain structures subserving
implicit procedural memory may be developed very early in life, explicit memory
systems may be available much earlier than generally assumed. The prolonged
development of encoding, consolidation of memory traces, retention, and retrieval
processes significantly contributes to age-related changes in declarative memory
performance throughout infancy and early childhood.
There is no doubt that the field has made considerable progress in understand-
ing what develops when and how these developments contribute to behavioral
changes in performance. In particular, neuropsychological research investigat-
ing brain–behavior links in very young children has provided many new insights.
However, it seems important to note that identifying a correlation between
the relative maturation of brain systems underlying memory and age-related
changes in behavior is only the first step toward improving our understanding of
74 4  Memory Development During the Infant and Toddler Years

the mechanisms of memory development (Richmond and Nelson 2008). Future

research on the development of recall memory in infants and toddlers may ben-
efit from improvements in neuroimaging tools that allow us to look inside the
developing brain but also from exploring infants’ and young children’s social
environments, which can have strong facilitating effects on further development
(Lukowski and Bauer 2014). Finally, as already shown in research focusing on
infants’ STM, future studies may benefit from a closer look at individual differ-
ences. For instance, research by Bauer, C.A. Nelson, and colleagues (e.g., Bauer
et al. 2006), using both behavioral data and measures of brain activity, has nicely
demonstrated that individual differences in consolidation and storage processes
are related to variability in long-term recall memory. Although the progress in
research on infant memory relating brain development and behavioral develop-
ment conducted during the past three decades has been remarkable, much more
has to be learned about how advances in brain development can contribute to
developmental changes in memory behavior and performance.
Chapter 5
The Development of Event
and Autobiographical Memory
in Childhood and Adolescence

Unlike implicit memory, episodic memory involves conscious awareness. This

memory system usually refers to memory for episodes and events in a person’s
life. Most of what people remember during their everyday lives is about things
that happened to them or about events. Although event memory is explicit in that
a person is aware that he or she is remembering, the encoding of most events is
unintentional, meaning that the person does not specifically try to remember the
event while it is being experienced.
The term “event” is rather broadly defined. According to Nelson (1986), events
“involve people in purposeful activities, and acting on objects and interacting
with each other to achieve some result” (p. 11). Event memory can be conceived
of as an amalgam of incidental and deliberate memory, with information usually
encoded without the intent to remember but with deliberate forces operating at
the level of reconstructing the experience (Ornstein et al. 2006). When such event
memory is about personal experiences, it is usually described as autobiographi-
cal memory. Although the mechanisms underlying event memory and autobio-
graphical memory appear to be rather similar, autobiographical memories differ
from most other event memories in that they are infused with a sense of personal
involvement in the event, implying that one has emotions, special thoughts, and
reactions that are associated with such events (Bauer 2007). Obviously, the term
event memory represents the broader category, and autobiographical memories
tend to deal with unique events that happened at a specific place and at a specific
time. However, the difference is not always easy to grasp. There is evidence that
event memories take on more and more autobiographical features over the course
of the preschool years, and this trend is reflected in children’s narratives. For
instance, children increasingly enrich their narratives with subjective experiences
and perspectives (e.g., “I fell down and was so embarrassed”), and this enrichment
illustrates the significance of the event for the child (Bauer 2006; Fivush 2001).

© Springer International Publishing Switzerland 2015 75

W. Schneider, Memory Development from Early Childhood Through
Emerging Adulthood, DOI 10.1007/978-3-319-09611-7_5
76 5  The Development of Event and Autobiographical Memory …

There is a widespread view in the adult literature that event knowledge is hier-
archically organized in a general autobiographical memory system (Belli 1998;
Conway 2005). According to this view, personally experienced events are struc-
tured in autobiographical memory by hierarchically ordered types of memo-
ries (i.e., extended, summarized, and specific) that vary in their scope. Whereas
memories for extended events reveal the temporal nature of autobiographical
memory and nest all other types of memories, memories for summarized events
emphasize the thematic aspect of autobiographical memory, that is, the common
themes that underlie events of the same type. These memories are abstractions
of the typical aspects of similar events. The most detailed type of memory at the
bottom of the hierarchy is for specific events that include perceptual and episodic
information. This type of memory thus involves details that can be used to recon-
struct a single event.
Although we know a lot about the characteristics and functions of auto-
biographical memory in adolescence and adulthood, considerably less is known
about its genesis and early development. To date, the social cultural developmen-
tal theory presented by Katherine Nelson and Fivush (2004) can be conceived of
as the most ambitious and comprehensive theoretical account of the emergence
of autobiographical memory and its further development as their theory assumes
that autobiographical memory incorporates many different concepts and skills as
illustrated in Fig. 5.1. These include language ability, narrative understanding,

Self-representation
Self-representation
Cognitive self
Theory
Theoryof mind
of mind

Complex language Rep

Beginning Representational
resentational
Language

Core Self
Basic Memory (implicit/explicit) Semantic Episodic AUTOBIOGRAPHICAL
Social Interaction/Communication/ Memory Memory MEMORY
Intentionality

Conversations
Conversations about
about the
the
past
past and
and future
future
Mental concepts

Temporal concepts

Narrative
Narrativestructure
structure
and
and
content
content

1yr. 2yr. 3yr. 4yr. 5yr.

Fig. 5.1  Hypothetical relations in developments from 1 to 5 years of age leading to the emer-

gence of autobiographical memory. Larger arrows indicate more direct influences; double-headed
arrows indicate reciprocal influences. Years (yr.) in the bottom scale indicate approximate ages
when influences come into play on average in normal development. Areas above the center are
presumed to be more endogenous and those below more exogenous as sources of development.
(Adapted from Nelson and Fivush 2004, p. 490)
5  The Development of Event and Autobiographical Memory … 77

temporal concepts, self-concept, consciousness, and social processes. As indicated

in Fig. 5.1, various sources contribute to the development of autobiographical
memory across the period from infancy to the late preschool years. All compo-
nents shown in the figure are subject to developmental processes in their own right
but also influence the emergence of autobiographical memory at particular points
in time. The theoretical model illustrates the complexity and diversity of develop-
mental trajectories, indicating that several variables contribute to the emergence
of autobiographical memory. In the following review of the literature, the roles of
these variables will be examined in more detail.
Given the similarities in the conceptualizations of event memory and autobio-
graphical memory and their treatment in the relevant literature, the present over-
view of the literature will not provide separate accounts of the development of
these two aspects of episodic memory but will use the terms “event memory” and
“autobiographical memory” more or less synonymously.
As already described in detail in the preceding chapter, even preverbal chil-
dren can show event memory for ordered events across long periods of time as
indicated by their ability to imitate sequences of actions they have seen before.
For instance, Bauer and her colleagues (e.g., Bauer et al. 2000) tested memory
for action sequences in 360 children between the ages of 1 and 3 years. Whereas
almost 80 % of the 1-year-olds could retain temporally ordered memories for
about 1 month, most older children in the sample were able to retain these mem-
ories for more than 6 months (for confirming evidence in a different paradigm,
see also Meltzoff 1985; Rovier-Collier 1997). As noted by Bauer (2006, 2009),
changes in basic mnemonic processes such as encoding, consolidation, storage,
and retrieval are associated with this accomplishment, with consolidation and stor-
age contributing the most to the observed developmental changes.
Although some of these children were able to translate those early events into
language later on, this translation seemed to depend on the age and language level
of the child at the time of the experience (see also reviews by Peterson 2002;
Reese 2014). Several researchers have proposed that it is children’s increasing
sophistication with language occurring between their second and fourth years of
life that is primarily responsible for the development of autobiographical memory
(Fivush and Hamond 1990; Leichtman and Ceci 1993; Nelson 1993). Beginning in
the third year of life when most children are able to talk about the past quite flu-
ently, verbal assessment procedures become a viable alternative, and similar meth-
ods of interviewing young children about the same or different events at different
ages can be easily employed. This option provides new possibilities and has initi-
ated a very active research paradigm, yielding a wealth of data on preschool and
schoolchildren’s event memory.
The first empirical studies on the development of event memory in preschool
and schoolchildren were conducted in the early 1980s, and interest in this aspect of
memory has increased steadily since then. Although Schneider and Pressley (1989)
did not deal with this issue in the first edition of their book, they felt the need to
add a separate chapter on children’s autobiographical memory in the second edi-
tion (Schneider and Pressley 1997). Also, whereas the topic of event memory was
78 5  The Development of Event and Autobiographical Memory …

only briefly addressed in Mandler’s (1983) chapter in the fourth edition of the
Handbook of Child Psychology, it was devoted a whole chapter in the handbook’s
sixth edition (Bauer 2006; for recent reviews, see also Fivush 2014; Hudson and
Mayhew 2009; Reese 2014). In the following, an attempt will be made to summa-
rize the major outcomes of research on children’s event memory.

Memory for Routine Events (Script-Based Memory)

How do young children remember the experiences of their everyday lives? How
are these memories organized? Research from the last three decades indicates that
children and adults organize memory for recurring events in the form of “scripts”
or general event representations (Nelson 1986; Nelson and Gruendel 1981).
Scripts are general knowledge structures that specify the sequence of expected
actions in everyday events and are used to guide the comprehension of events or
text information (Hudson and Mayhew 2009). For instance, a restaurant script
might involve driving to the restaurant, entering the restaurant, getting seated at
a table, reading the menu, ordering food, eating the food, paying the cashier, and
exiting the restaurant. As noted by Price and Goodman (1990), such an event epi-
sode can be conceived of as a hierarchy of event subunits. The hierarchical struc-
ture of a restaurant-attending event is given in Fig. 5.2. It shows the activities
described above at the highest level, together with subunits such as more detailed
actions that are carried out on objects in the environment. For instance, the actions
and objects that constitute the activity of paying the cashier in an American restau-
rant include receiving the check, calculating the tip, and giving the correct sum of
money to the cashier (Hudson et al. 1992).
Substantial research has demonstrated that even very young children organ-
ize information temporally and causally in a script-like fashion (Bauer 2007;
Nelson 1996). To study young children’s memory for recurring events, a suitable
method is to ask young children questions about familiar routines such as attend-
ing birthday parties, making cookies, and going grocery shopping or to a restau-
rant. This is exactly what Nelson and her colleagues (e.g., Nelson and Gruendel
1979, 1981) did in their pioneering research on scripts. Children were asked to
tell the experimenters what happens during such events and were then prompted
when necessary. Nelson and colleagues found that children as young as 3 years
had general and temporally organized knowledge of recurring events. The results
of these and later studies (e.g., Nelson et al. 1983; Fivush and Slackman 1986)
demonstrate that preschoolers are able to give consistent and veridical reports of
what typically happens during such events, although older children’s scripts are
usually more detailed than those of younger children. This ground-breaking dis-
covery illustrated the importance of familiarity with a domain or event for memory
performance, challenging the prevailing assumption of that time that young chil-
dren’s general event memories are unorganized and idiosyncratic (e.g., Piaget and
Inhelder 1973). Instead, this research showed that children’s episodic memory is
Memory for Routine Events (Script-Based Memory) 79

Event: Going to a Restaurant

Activities: Read menu Order food Eat meal Pay

Actions: open menu server arrives server serves receive check

read menu order appetizer pick up utensils calculate tip

entres

choose items appetizer eat appetizer give money

entres

(talk with companions)

server clears plates

serves entree

eat entree

server clears plates

Fig. 5.2  Hierarchical structure of going to a restaurant event (Hudson et al. 1992)

temporally organized and that from a very early age, their reports include the com-
ponent actions of events in the correct temporal sequences. Thus, young children’s
event knowledge is general and well organized. Apparently, these children are able
to form a general representation of an event based on only a few experiences.

Developmental Differences in Event Recall

Although there are surprising developmental continuities in event reports, there are
also developmental differences (Fivush 1997; Hudson and Nelson 1986). Younger
children’s verbal reports are always more skeletal than those of older children and
adults. One typical aspect of young children’s reactions when asked “what hap-
pens” in familiar events is that they report general actions using the impersonal
pronoun “you” and the timeless present tense, for instance, “you buy things” when
80 5  The Development of Event and Autobiographical Memory …

you go to the supermarket (Hudson and Mayhew 2009). Whereas children of all
ages report component actions of events in the appropriate temporal order, older
children’s reports are more complex and elaborate than younger children’s reports.
One explanation for this age difference could be that older children’s script reports
are more complex and elaborate because, in general, they have had more expe-
rience with most events than younger children. Indeed, many of the age-related
changes noted above took place as children accumulated experience with events.
However, although experience contributes to developmental differences, it does
not seem to completely account for these differences. For instance, in laboratory
research in which children of different ages were given comparable amounts of
experience with an event, age differences remained, with older children produc-
ing more elaborate reports than younger children (e.g., Farrar and Goodman 1990;
Fivush et al. 1992; Price and Goodman 1990).
One problem with interpreting the obtained developmental differences is that the
use of verbal measures may result in underestimations of young children’s event
knowledge. One needs to make a distinction between the verbal report of the event
and the underlying representation of that event (Nelson 1986). Older children’s
verbal reports may be more complex and elaborate because of their superior lan-
guage skills. Younger children’s event representations may be as complex as those
of older children, but they are just not able to express this complexity in language.
To test this assumption, several studies used nonverbal assessment procedures such
as picture-sequencing tasks or behavioral re-enactments (e.g., Fivush and Mandler
1985; Hudson 1988; Ratner et al. 1986). For instance, studies using nonverbal reen-
actment procedures such as deferred imitation have shown that young children can
remember events well even after long delays (McDonough and Mandler 1994).
Overall, findings from these studies indicate that verbal reports indeed underesti-
mate the exhaustiveness of preschool children’s event knowledge. However, they
also suggest that young children’s event knowledge is not as flexibly organized as
the event knowledge of older children (see also Fivush and Slackman 1986). In par-
ticular, younger children’s event representations are not organized around goals to
the same extent as the event representations of older children and adults.

Developmental Differences in Event Recall Format

Do young children represent events in a hierarchical format, that is, in terms of

superordinate units (e.g., main goals) and subordinate units (e.g., action and activ-
ity levels) from the very beginning? The available research suggests that there
may be developmental differences in this regard as well. For instance, research by
Ratner and Smith (Ratner et al. 1986; Smith et al. 1987) focused on the hierar-
chical structure of young children’s emerging event representations after a single
experience with a novel event. Smith et al. (1987, Experiment 2) asked 5-year-
olds and adults to remember a hierarchically organized event, making fun dough,
immediately after the experience. Each participant was randomly assigned to a
Memory for Routine Events (Script-Based Memory) 81

no-cue and several cue conditions that focused on goals and actions. In their verbal
recall of the event, children tended to mention the more important actions more
frequently than the less important actions (as determined by adult importance
ratings), thus indicating that young children are more sensitive to the core con-
tent components than to the hierarchical structure after a single experience with a
novel event. Overall, children relied on verbal cues provided by the interviewer to
a greater extent than adults who reported more superordinate information in both
the no-cue and cue conditions. This suggests that, with increasing age, internal
cues are more often generated and used to direct retrieval. Although young chil-
dren do not seem to represent event knowledge in a hierarchical structure during
the early phases of script acquisition, they are able to temporally organize actions
and recognize the most important actions (Ratner et al. 1986).
One of the most robust findings in this research paradigm is that children report
the component actions of events in the temporal order in which the events were
experienced. Yet some everyday events are more temporally constrained than oth-
ers, and children seem to be sensitive to these differences. Most studies have found
that events that follow a logical order such as going to a birthday party are almost
always reported in that sequence, whereas events that follow a more flexible order
are reported in more variable sequences. Regardless of variations in temporal con-
straints, however, even young children report events in their correct temporal order
after only a few experiences with the event (Hudson and Nelson 1986; Slackman
and Nelson 1984; Ratner et al. 1986). Apparently, children are able to form a general
set of expectations based on the first encounter. They seem to form scripts after one
experience with a new event, trying to recount it in a generalized temporally ordered
framework. Fivush (1997) concluded from these findings that scripts and episodic
memories do not fall along a continuum in the memory system but rather seem to
reflect different ways of representing events that can be present simultaneously.
There are only a few studies that have directly compared young children’s
memory for script-based versus unfamiliar novel information. Children in the
Munich Longitudinal Study were presented with several narratives that dealt with
familiar and less familiar events (e.g., a birthday party, playing in the afternoon,
moving to another town). Whereas the birthday party story followed a familiar
script and a well-known temporal order, the playground story included famil-
iar elements but was not based on a familiar sequence of events. Finally, the
“removal” story referred to an event that preschoolers were not particularly famil-
iar with. The three stories were similar with regard to the number of propositions,
the number of sentences, and the number of events covered. As a main result, sub-
stantial increases in text recall were observed between the ages of 4 and 6, with
4-year-olds freely recalling about 20 % and 8-year-olds recalling about 50 % of
the information provided in the texts. Although text recall was better for script-
like information than for the unfamiliar story, the difference was not pronounced.
Individual differences in text recall were rather stable across time from the very
beginning of the longitudinal study, regardless of the text. The fact that the mean
correlation between the familiar and the unfamiliar stories (birthday party and
removal) was substantial (r  = 0.65) indicates that differences in familiarity and
82 5  The Development of Event and Autobiographical Memory …

script structure did not significantly affect memory–performance differences

among children (see Knopf 1999; Schneider et al. 2009). The authors concluded
that the ability to learn, store, and recall text information develops rather early in
childhood and progresses at a similar pace for the majority of children.
One basic advantage of script knowledge is that it gives us an awareness of
what usually happens during any given occurrence of an event, and this allows us
to predict and anticipate actions in our world. Young children’s tendency to organ-
ize information according to familiar scripts seems to result in their tendency to
fail to remember specific information about routine events (Fivush and Hamond
1990). Rather than recalling novel aspects of familiar events, young children are
more focused on routine information. Why is this so? According to Nelson (1986),
young children’s memory is so closely tied to recurring events because this mem-
ory has the adaptive value of permitting children to predict the likelihood of events
in the future. Although there is not such a payoff for novel events and variations in
routine events, children eventually also manage to remember specific event infor-
mation rather than exhibiting just generalized event memory.

Memories for Single Instances of Routine Events

To understand the process of script acquisition more completely, it seems impor-

tant to investigate how children deal with event variability, that is, variations from
one experience to the next that are also true for routine events. For instance, the
items bought during a visit to the supermarket may vary from time to time even
though the main action components remain the same. How do children come to
represent this kind of event variability, and how does event variability affect recall
for children of different ages? In a study by Fivush et al. (1992), the effects of
temporal order and event variability on young children’s developing representa-
tions of events were systematically explored. Three- and 5-year-old children par-
ticipated in three different events: making edible “fun dough” (a logical-invariant
event), making a shape collage (a logical-variable event), and sand play (an arbi-
trary-invariant event). For each age group, half of the children experienced events
only once prior to recalling them verbally and behaviorally, whereas the other half
experienced the events four times and recalled them verbally and behaviorally
after each occasion. The objects and actions in the “making edible fun dough”
event remained invariant across all experiences of the event. This was not true for
the “shape collage” event, in which the actual shape as well as some of the other
materials changed each time. Unlike the other two events, the “sand play” event
did not follow any logical order. However, every time the child experienced this
event, the component activities were performed in the same prespecified order,
and the actions and objects were invariant across experiences. See Table 5.1 for a
description of the component units of each event.
Overall, the results indicated that the structure of the event had an important
effect. Children recalled more and sequenced their recall better for logically con-
nected than for arbitrarily connected events after the first experience and also
Memories for Single Instances of Routine Events 83

Table 5.1  Component units of each event (adapted from Fivush et al. 1992, p. 192)
I. Logical-invariant event: Making fun dough
   1. Put ingredients in bowl
   2. Mix dough
   3. Roll out dough with rolling pin on cookie sheet
   4. Cut out cookies with cookie cutter
   5. Put cookies on plate with spatula
II. Logical-variable event: Making a shape collage
   1. Trace shape (bunny, tree, duck, flower, watermelon) with crayon
   2. Cut out shape with scissors
   3. Squeeze glue onto shape
   4. Glue material (cotton, leaves, feathers, petals, seeds)
   5. Color (ears, trunk, beak, center, rind) with marker
III. Arbitrary-invariant event: Sand play
   1. Put sand in hand mold with shovel
   2. Squeeze sand through sifter
   3. Draw in sand with rake
   4. Pour sand through funnel
   5. Dig in sand with sand drill

recalled more from logically connected events after several experiences with these
events. Whereas the effects of temporal order were qualitatively similar across age,
the effects of event variability were not. A comparison of the 3-year-olds and 5-year-
olds showed that the older children had no difficulty remembering the changes in
the shape collage event. When asked to re-enact the event with a new set of varia-
tions, they were able to do so, regardless of whether they had one previous experi-
ence or four. By contrast, the 3-year-olds had more difficulty when asked to reenact
the event with a new set of items after only one experience with the task. They
needed more experiences with the task, each with a different set of items, to reenact
the event with a fifth set of items (for similar results, see Bauer and Fivush 1992;
Kuebli and Fivush 1994). Obviously, younger children experience difficulties repre-
senting minor variations within an event and need more experience with how events
can vary from time to time in order to build this variability into their representation
of the general event. An interesting next question is how they eventually manage to
differentiate and represent single occurrences of recurring events.

Memory for Fixed and Variable Items

A common method used to explore this issue is to examine children’s memory

for fixed and variable items. In such experiments, children are provided with a
series of events in which some elements remain constant and others vary. Recall
and recognition for fixed and variable events are then compared, and the effect of
experience (i.e., different numbers of repetition) on memory is also assessed. For
84 5  The Development of Event and Autobiographical Memory …

instance, Farrar and Goodman (1990) asked 4- and 7-year-olds to participate in

a novel event consisting of four “animal game” activities in four different loca-
tions. During standard visits, children enacted the same activities in every session.
During deviating visits, new activities replaced the standard activity. Children in
the experimental group experienced the standard visit either once or three times
before experiencing the deviating visit. Children in the two control groups experi-
enced only one type of visit, that is, either a single standard visit or a single devi-
ating visit. As a main result, children’s recall was not affected by the particular
kind of variation experienced. However, the amount of experience with the event
proved to be important. Children in the two control groups who had only one
experience, either a standard or a deviating visit, recalled the event rather well.
The children in the experimental group who experienced one standard and one
deviating visit recalled significantly less information about both visits than chil-
dren in the two control conditions. With increasing experience with the standard
visit, this event was well recalled, but the deviating event was still poorly recalled.
Repeated experience with the standard visit seemed to enhance older but not
younger children’s ability to keep the two types of visits separate in memory.
A follow-up study using the same age groups (Farrar and Goodman 1992) con-
firmed these results, once again showing that the two visits were confused more
by younger children, and deviations were better recalled by the older children.
However, younger children in the experimental group recalled more information
from the deviating visit when that visit came last instead of first. Taken together,
these studies show that increasing experience with the same event improves recall
of that event, but at the same time, it diminishes recall for one-time variations.
However, multiple experiences with the same event also lead to less confusion
between standard and deviating events. Overall, results indicate that young chil-
dren reliably and accurately report memory for fixed events but have more difficulty
recalling variable items in recurring events. Thus, general event knowledge enhances
memory for fixed items but also produces distortions in the recall of variable items
(for similar findings, also see Kuebli and Fivush 1994; Pearse et al. 2003; Roberts
and Powell 2006). Although the general-event-knowledge schema can provide a per-
son with knowledge of the main event categories that are filled with specific details
when the person recalls a specific event that is related to the general routine, scripts
sometimes distort children’s recall of their experiences. In particular, young children
experience difficulty distinguishing between what they actually experienced and the
events specified in the script (Adams and Worden 1986; Hudson 1988).

The Schema-Confirmation-Deployment Model

as an Explanatory Approach

The finding that younger children organize their memory for general and specific
event episodes differently than older children confirms the predictions based on
the schema-confirmation-deployment model originally developed by Goodman
Memories for Single Instances of Routine Events 85

(Farrar and Goodman 1990). According to this model, when children encounter
a new event, they try to use an available event schema to comprehend this new
event. If the event is novel and no schema is available, all potential useful event
information from this new event is retained as information about a new event, and
no distinction between typical and atypical components is made. However, if an
appropriate schema is available and can function as a coherent mental unit, children
tend to focus on schema-typical information, a process that is called the schema-
confirmation phase. After a schema is confirmed, the schema-deployment phase
begins. In this phase, attention is directed toward processing new schema-atypical
information. One implication of the model is that the schema-confirmation phase
may take longer for younger children because of their information-processing and
knowledge constraints. Thus, younger children require more experience than older
children to establish both general event representation and a separate representation
of atypical event information. As a consequence, younger children may be more
likely than older children to merge separate event episodes in memory, particularly
during the acquisition of the representation of a general event. As noted by Hudson
and Mayhew (2009), this model is useful for understanding how and why children
of different ages attend to and recall different information from their first encoun-
ters with novel variable events (for confirming evidence, see Farrar and Goodman
1992; Kuebli and Fivush 1994). However, the model does not predict the differen-
tial processing of event information once schema deployment is achieved (Hudson
et al. 1992), and it cannot explain variations in memory for details of recurring
events after the representation of a general event has been constructed.
An interesting question is whether findings for immediate recall also gener-
alize to delayed recall. In a study by Hudson (1990), 3- and 5-year-old children
were assigned to either a repeated or an episodic experience of a creative move-
ment workshop. In the repeated condition, four workshops were attended in which
the specific actions involved in two of the five activities varied each time, and the
remaining activities were performed in the same way each time. In the episodic
condition, the children participated in only one workshop, either the first or the last
of the repeated condition. Half of the children in each group were asked to recall
the events immediately and also 4 weeks later. The other half of the children were
asked to recall the workshops 4 weeks later. Hudson found that children’s imme-
diate recall of the workshops was generally quite accurate regardless of whether
they had experienced one or four workshops. However, the pattern of results was
different for delayed recall. Although children in the repeated condition recalled
more information about the first workshop than the children in the episodic condi-
tion did, the children in the repeated-condition group also reported more intrusions
than the other children. Thus, repeated experience increased the amount but not
the accuracy of recall as illustrated by the finding that the children reported expe-
riences from the other workshops and had not been able to maintain the separa-
tion of information across workshops. There was also an effect of recall practice
on the results in that the 5-year-olds who recalled the events twice reported fewer
intrusions in delayed recall than the 5-year-olds who did not immediately recall the
events. Interestingly, practice did not affect the younger children’s intrusion rate.
86 5  The Development of Event and Autobiographical Memory …

The Importance of Memory Cues

Research with young children generally indicates that they have great difficulty
recalling specific episodes of recurring events without further prompting. For
instance, Pillemer et al. (1994) asked 3- to 5-year-olds about last night’s dinner at
home. Although children did not report much specific information initially, they
did so after they were provided with additional cues. Providing cues appears to
be particularly helpful when the specific episode deviates considerably from the
general script. Thus, a class trip to an archeology museum may seem different
from visits to museums in general because children dug for artifacts in a sand-
box or made clay models of what they had found. Fivush et al. (1984) found that
children were able to recall specific details of their trip to the archeology museum
both immediately after the visit and also 6 weeks later. Similar findings were also
reported in experimental studies that asked children to recall script-based stories in
which atypical actions were embedded. Both children and adults tended to recall
atypical events quite well immediately after the story presentation. In delayed
recall, however, children and adults tended to report script information that was
not present in the original story, and they had difficulty recalling the atypical
elements (Adams and Worden 1986; Smith and Graesser 1981). On the basis of
assumptions of popular script models such as the one developed by Schank and
Abelson (1977), Smith and Graesser concluded that atypical actions are tagged in
memory, yielding better immediate recall of such actions than of typical actions
that are more easily confused with script knowledge. However, atypical actions
are more easily forgotten over time, whereas typical actions are more likely to be
recalled or to even intrude into recall when they did not take place but are consist-
ent with expectations (Ornstein et al. 2006). Such “typicality effects” have been
found repeatedly, particularly in younger children (Adams and Worden 1986).
This suggests that the memories of younger children rely more heavily on scripts
than the memories of older children. Yet, this cannot explain the persistence of
many atypical episodes in children’s long-term memory (see Fivush et al. 1984;
Hudson and Fivush 1991). It may be the case that different kinds of deviations are
remembered differently and that more significant deviations from a script can be
retained in memory for longer periods of time than smaller deviations.
In summary, children’s reports of specific episodes of recurring events show a
great deal of continuity across development. Younger and older children as well as
adults have difficulty recalling a specific instance of an event that does not deviate
much from the general script. However, even young children can recall episodes
in great detail when given specific cues. Episodes that differ considerably from the
script are generally well recalled even without specific cues. Atypical actions that do
not have a great deal of core relevance to the goal are well recalled initially but seem
to fade over time as the episode is reconstructed to conform to the general script.
Young children’s recall is more negatively affected by variations across instantia-
tions of the event than older children’s recall, particularly in the initial phases of
forming and consolidating a script (see Hudson et al. 1992, for more details).
Memories of Novel and Salient Events 87

Memories of Novel and Salient Events

Memory of Novel Events

Several studies with young children have demonstrated that novel and unu-
sual events can be remembered for longer periods of time, indicating that epi-
sodic memory is not constrained to routines and recurring events. These studies
have also shown that even toddlers are able to recall specific novel information in
some detail and can retain these memories for remarkably long time periods (for
reviews, see Fivush 1997; Reese 2014). For instance, Fivush et al. (1987) inter-
viewed 2- to 3-year-old children about events that had occurred in the recent past,
either within the past 3 months or more than 3 months ago. As a main result, they
found that all children in the sample recalled at least one event that had happened
6 months or more in the past and were able to recall as much information about
distantly experienced events as they could about more recently experienced events.
Some limitations of these early studies were that all the children were asked about
different events that had occurred at different times and that there was only a little
information available about how frequently these events had been discussed since
their occurrence.
Subsequent longitudinal studies on the development of autobiographical mem-
ory have addressed these concerns. For instance, Hamond and Fivush (1991) asked
4- to 6-year-old children to recount a family trip to Disney World. Not only were
all the children asked about the same event, but they were also all asked the same
set of structured questions. Furthermore, parents were asked to estimate how often
the events had been discussed in the family. Surprisingly, children who had been
to Disney World 18 months previously recalled as much as children who had been
there 6 months previously. All children recalled a considerable amount of informa-
tion about their experience, and the older and younger children did not differ in the
number of information units they provided. The age groups did differ, however,
in how elaborate the reports were, with older children providing more detailed
responses. With age, children focused more on the unique and novel aspects of
these events and less on the scripted and routine aspects. Older children also
tended to provide more orienting (who, where, and when) as well as evaluative
(how and why) information in their accounts (Fivush et al. 1995).
Based on findings on long-term recall in preverbal children (see Chap. 4), we
might expect to find increases in the length of time across which preschool children
can remember event information as they get older. The available data are somewhat
mixed. There is little evidence that nonverbal memories cross the language barrier to
become verbal memories. For instance, Peterson and colleagues (e.g., Peterson and
Bell 1996; for a review, see Peterson 2002) collected information on children’s inju-
ries and emergency room treatments. Their findings showed that most children who
were 1 year of age old at the time of the injury recalled nothing of the injury or hos-
pital treatment several years later. However, children who were 2 years of age old at
the time of the injury recalled many core features and details of the hospital treatment
5 years later (see Peterson and Parsons 2005). This finding corresponds with those
88 5  The Development of Event and Autobiographical Memory …

reported in Chap. 4 that showed that it is difficult for most children to express early
events with language later on. Whereas only a few nonverbal children could cross the
language barrier and talk about a novel event several months after the experience,
most children who were about 2 years of age and able to express their memories
through language at the time of the experience were able to maintain these memories
over time (Bauer et al. 2004; Jack et al. 2012).
Overall, the findings from several longitudinal studies on the development of
autobiographical memory have suggested that memories stabilize around the age
of 3 once most children are able to talk about the past fairly fluently. For instance,
the study by Hamond and Fivush (1991) suggested only minor age differences in
the number of units of information reported by children who had been 3 versus
4 years of age at the time of the critical event. Other studies have indicated that
the memories of older children are more robust than those of younger children.
For example, Quas and her colleagues (Quas et al. 1999) assessed the memory of
children 3–13 years of age for a painful medical procedure that they had experi-
enced between the ages of 2 and 6 years. Most of those children who were older
at the time of the medical procedure gave detailed reports of the event. In com-
parison, the younger children’s reports were generally vague. In fact, none of the
children who had been 2 years of age at the time of the procedure provided clear
memory for it later, whereas most of the children who were at least 4 years of age
at the time of the critical event remembered it. See Peterson and Whalen (2001)
and Pillemer et al. (1994) for similar findings.

Memory for Salient Events

An alternative to creating novel events for children to experience either repeatedly

or on one occasion is to explore children’s memory for a specific instance of natu-
rally occurring events—either recurring or unique. In recent years, a considerable
amount of work on young children’s memory for unique personally experienced
research has been published. In some studies, children were exposed to a range of
specially crafted stimulus events, such as visiting a pirate ship (Murachver et al.
1996) or a pretend zoo (McGuigan and Salmon 2004).
This is the approach taken by Peter Ornstein and his colleagues who investigated
children’s memory for specific pediatric examinations (for reviews, see Ornstein
et al. 2006, 2008). In one of the first studies of this type, Baker-Ward, Gordon,
Ornstein, Larus, and Clubb (1993) assessed 3-, 5-, and 7-year-olds’ retention of the
details of a routine visit to the doctor. Most children were interviewed twice: first,
immediately after the checkup and then after a delay of 1, 3, or 6 weeks. The inter-
views began with open-ended questions (e.g., “Tell me what happened during your
checkup?”) and were followed by more specific questions. The children were also
asked potentially misleading questions about activities not included in the checkups.
The main findings of this study are presented in Figure 5.3. As illustrated in the top
panel (A) of this figure, even the 3-year-olds were able to report most of the features
Memories of Novel and Salient Events 89

Fig. 5.3  Percent of features Open-ended Specific

correctly reported in response
to open-ended and specific (a) Test Condition: 3-year-olds
probes by test condition at 100
ages 3 (panel A), 5 (panel B), 90
and 7 (panel C) years. Note
80
that the data presented for
the initial test are averaged 70
across the three delay groups. 60
(Adapted from Baker-Ward 50
et al. 1993)
40
30
20
10
0
Initial 1 week 3 weeks 6 weeks

(b) Test Condition: 5-year-olds

100
90
80
70
60
50
40
30
20
10
0
Initial 1 week 3 weeks 6 weeks

(c) Test Condition: 7-year-olds

100
90
80
70
60
50
40
30
20
10
0
Initial 1 week 3 weeks 6 weeks
90 5  The Development of Event and Autobiographical Memory …

of the event (about 75 %). Nonetheless, Figure 5.3 also demonstrates clear age-
related differences in performance. In contrast to the 5- and 7-year-olds, the 3-year-
olds showed lower levels of overall recall and produced less information than the
older children in response to open-ended questions and thus had to be asked more
specific questions before they provided information about the experience. Finally, a
comparison of the bars across the three panels reveals that the younger children evi-
denced more forgetting than the older children over the 6-week time interval. This is
also reflected by their reduced ability to distinguish between activities that had and
had not been included in their medical checkups.
Although the ability to remember specific events for a longer period of time is
already evident at this early age, it undergoes further changes with development,
affecting both the number of events remembered and the robustness of memories
for specific events (for a review of relevant studies, see Bauer 2006).

Effects of Experience and Prior Knowledge

The event memory literature clearly demonstrates that the earlier views of chil-
dren’s memory in which young children’s recall was considered to be quite limited
(e.g., Myers and Perlmutter 1978) have to be corrected. Nonetheless, the find-
ings by Baker-Ward et al. (1993) indicate that there are considerable age-related
changes in the ability to remember salient events that cannot be fully explained
by lower levels of language proficiency and an incomplete mastery of narrative
conventions (e.g., Greenhoot et al. 1999). These findings raise several questions
regarding age-related changes in the encoding, storage, and retrieval of memo-
ries. There is evidence that the type of experience and the amount of prior knowl-
edge contribute significantly to the findings. For instance, Murachver et al. (1996)
examined effects of type of experience on memory in a study of children’s event
memories acquired through either direct experience, observation, or stories.
Children at 5 and 6 years of age received either one or three exposures to a novel
event: visiting a pirate room. Regardless of age, those children who directly expe-
rienced the events produced more complete and accurate verbal recall a few days
later than children who either observed the event or heard stories about it.
Several other studies have pointed to the impact of prior knowledge on event
memory. When previously acquired knowledge is lacking, as in a situation in which
a child experiences a novel event, long-term recall can be facilitated when knowl-
edge is gained while the experience takes place. For instance, Ornstein et al. (1998)
used data from a study of 3- to 7-year-olds’ memories of an invasive and novel
radiological procedure (Merritt et al. 1994) to look at how providing information
to children during the event about the stressful and unfamiliar medical procedure
affected their subsequent memories of the event. The procedure involved urinary
bladder catheterization. Some children were provided with a verbal description of
the catheter and its insertion, whereas other children did not receive this procedural
narrative. As a main result, children in the procedural narrative group recalled more
Memories of Novel and Salient Events 91

information both immediately after the experience and 6 weeks later than children
who were not given such a description. Importantly, these differences could not be
attributed to age differences or to different levels of stress during the procedure.
Thus, these findings suggest that information that is obtained during an unfamiliar
and stressful event enhances remembering.
The impact of prior knowledge on event memory was also demonstrated in a
reanalysis of the 5-year-olds’ recall data from the Baker-Ward et al. (1993) study.
Clubb et al. (1993) rescored the protocols from the Baker-Ward et al. study to cre-
ate memory scores that indicated recall of the various components of the office
visits. These memory scores were then compared with normative knowledge
data obtained in a different study. Although there was considerable variability in
the memorability of the components of the physical examination and children’s
knowledge about the individual features, the knowledge and memory scores were
highly correlated. These findings strongly suggest that what a child knows about
an event can significantly affect the extents to which information about the event is
stored in and retrieved from memory.
However, this does not imply that rich prior knowledge always has posi-
tive effects on event recall. As shown by Ornstein et al. (1998), children’s prior
knowledge can also interfere with subsequent recall as the details of an experience
fade over time. In this study, 4- and 6-year-old children underwent a specifically
constructed physical examination that was both consistent and inconsistent with
knowledge-driven expectations. The examination included some typical medical
procedures such as listening to the heart with a stethoscope but omitted others that
usually occur (e.g., checking the mouth). The omitted features were replaced by
atypical ones, such as measuring head circumference. Ornstein et al. (1998) were
able to show that prior knowledge had both positive and negative effects. The
expected features of the medical procedure were initially better remembered than
the atypical ones. However, about 12 weeks after the checkup, the children made
spontaneous commission errors, reporting certain medical treatments that were not
included in the original procedure. These findings are consistent with the view that
memory representations change over a 12-week time interval. As the children’s
memory for the checkup faded over the course of the 12-week interval, more typi-
cal and generic event representations of visits to the doctor were incorporated into
their reports (for a more detailed discussion, see Ornstein et al. 2006, 2008).

Social–Cultural Contexts of Event Memories:

Effects of Maternal Reminiscing Style

Conversations About Past Events

Although the previous sections demonstrate that much has been learned about
children’s event memory during the past three decades, one important aspect has
not been included in the discussion of developmental issues. Although events
in the real world occur in a social and cultural context, much of the research on
92 5  The Development of Event and Autobiographical Memory …

developing event memories has focused on children’s individual reports of events.

It seems clear that adults play a large role in facilitating children’s understand-
ing and hence their encoding of information in memory. Meanwhile, a grow-
ing number of studies in autobiographical memory development have addressed
social interaction theories on the ways in which autobiographical memory changes
as a function of children’s interactions with their parents. Children begin to talk
about past events almost as soon as they produce their first words, and children’s
skills for recalling experienced events develop rapidly between 2 and 4 years of
age (e.g., Hudson 1990; Nelson 1986, 2014). As discussed by Nelson and Fivush
(2004), during the preschool years, multiple developments in language, self-con-
cept, temporal understanding, as well as memory skills converge to allow for the
emergence of autobiographical memory. From a sociocultural developmental per-
spective, the adult–child conversation serves as an important mediator of devel-
opmental changes in remembering (Farrant and Reese 2000; Fivush 2014; Fivush
et al. 2006). From such a perspective, it is assumed that social interactions in
which children and adults share their experiences together shape autobiographi-
cal memory. In particular, adults structure experiences in ways that are develop-
mentally appropriate, allowing children to develop their autobiographical memory
skills and to focus their attention on salient aspects of an event, thus increasing
their understanding and memory. One of the first of these studies (Fivush et al.
1987) showed that 3-year-old children can provide correct information about
novel, one-time past experiences in response to parents’ “open-ended” prompts
(e.g., “Tell me about Disneyland”) and also recall events in a somewhat coherent
form after delays of several months. However, given that young children’s early
reports of their past experiences are limited both in content and structure, these
conversations are usually initiated and scaffolded by an adult, and in most cases,
the mother. Thus, it is the adult partner who provides most of the content and the
structure, as illustrated in the following example of a mother and her 24-month-old
(slightly modified from Ornstein et al. 2006):

Mother: What else happened at Peter’s house?

Child: (no response).
Mother: We had dinner. What did you eat?
Child: (goes off task).
Mother: What did you do with Peter?
Child: Sesame Street.
Mother: Yeah, you watched Sesame Street on TV. What else did you do with
Peter? Did you guys fight about something?
Child: (shakes head no).
Mother: No? When you were watching Sesame Street?
Child: (nods head yes).
Mother: Yeah. You guys got hungry and tired. Then what happened?
Child: Uh oh.
Mother: Yeah. What happened? Did you bite Peter’s finger?
Social–Cultural Contexts of Event Memories … 93

There is no doubt that the nature of the mother–child conversation about past
events has an impact on how children remember these events. For example, Tessler
and Nelson (1994) compared what mothers and their 4-year-old children talked
about during an event and children’s subsequent memory for the event. They
found that children recalled more information that was mutually discussed by the
mother and the child than information talked about by only the mother or only
the child. Moreover, numerous studies have now established substantial individ-
ual differences in the ways in which parents structure how they reminisce with
their children and that these differences have long-lasting implications for chil-
dren’s developing autobiographical memory skills. The majority of these studies
have focused on mothers and the preschool years. Although there is a consider-
able amount of variability in the ways in which mothers structure past event con-
versations with their preschool children, it seems possible to distinguish between
two different styles, labeled high elaborative versus low elaborative (see Fivush
et al. 2006, for a review). High-elaborative mothers such as the one in the above
illustration ask questions frequently and are more likely to ask detailed questions
about the same event. They continuously provide new information to cue memory
even when their children do not seem to recall spontaneously. In comparison, low-
elaborative mothers do not provide much of a narrative structure for their children.
They have shorter event conversations in which they frequently repeat their own
questions.
There is now plenty of evidence that different parental reminiscing styles are
associated with individual differences in children’s event memory. In the first lon-
gitudinal study of parent–child reminiscing, McCabe and Peterson (1991) found
that parents who focused on a single past event during reminiscing and asked
several questions about that topic had children who told longer stories about past
events as preschoolers and who included information about past events in line with
their parents’ specific questions. This work was extended by Reese et al. (1993)
who studied a sample of mothers, fathers, and preschoolers in a longitudinal panel
design to investigate long-term consistency and change in maternal styles in talk-
ing about past events. Reese et al. (1993) noted that some parents were highly
elaborative in the ways in which they discussed past events with their children and
that later in preschool, these children exhibited richer memories about a variety of
personally experienced events. As illustrated by the lagged correlations in Fig. 5.4,
Reese et al. (1993) demonstrated that the synchronous associations between moth-
ers’ elaborations and children’s memory responses were strong and significant
from the first measurement point on. Moreover, mothers’ elaborations during early
conversations with their 40-month-olds were positively correlated with children’s
contributions of memory information in conversations at 58 and 70 months of age.
Mothers’ reminiscing styles seemed to be consistent over time, at least from the
second time point on, with the direction of effects flowing more from mother to
child than from child to mother. The ways in which mothers reminisced with their
young preschool children continued to affect children’s memory skills more than
2 years later. At the last two time points, however, a highly intercorrelated and
94 5  The Development of Event and Autobiographical Memory …

40-month 46-month 58-month 70-month

time point time point time point time point

.47*
.57**

Mother .22 Mother .46* Mother .70* Mother

elaborations elaborations elaborations elaborations

.47*
.85**
.57** .59**
.50*

.60** .66**
.69***
Child Child Child Child
memory respon memory respon memory respon memory respon
ses ses ses ses

Fig.  5.4  Cross-lagged correlations between maternal elaborations and children’s memory

responses. Reprinted with permission from Reese et al. (1993). Note * p < 0.05, ** p < 0.01,
*** p < 0.001

reciprocal pattern was found, indicating bidirectionality both concurrently and lon-
gitudinally. Thus, by the end of the preschool period, not only did mothers’ elabo-
rations affect children’s memory responses, but children’s memory responding was
also shaping mothers’ reminiscing styles.
Haden et al. (1997) clarified that in this same sample, it was mothers’ and not
fathers’ reminiscing style that predicted children’s subsequent memory skills.
This longitudinal finding has been replicated and extended in multiple studies
(for reviews, see Fivush 2014; Reese 2014). For instance, in a study over a decade
later, Haden et al. (2009) conducted a longitudinal investigation of mother–child
reminiscing based on a relatively large sample of over 50 US children and their
mothers, beginning with 18-month-old children and including children’s language
skills as an independent predictor. As a main result, mothers’ style of reminisc-
ing was shaped by children’s language and conversational participation as early as
the toddler years, but mothers’ open-ended elaborative questions were most criti-
cal for children’s memory development in that they fostered children’s active par-
ticipation in reminiscing. Overall, findings suggest that a child’s verbal memory in
the preschool years is linked to the mother’s reminiscing style and the child’s own
active participation in event recall from younger ages.
Social–Cultural Contexts of Event Memories … 95

Conversations During Events

Just as adult–child conversations about the past can influence the reporting of
information, verbal interactions that occur during an ongoing event may also have
positive effects on memory. The few studies to date that have examined mother–
child talk during an event suggest that children of high-elaborative mothers pro-
duce longer and more detailed reports of these experiences.
In an initial short-term longitudinal study, Haden et al. (2001) followed a sam-
ple of preschool children from 2.5 to 3.5 years of age, observing mothers and their
children as they engaged in three specially constructed activities: at 30 months, a
pretend camping trip; at 36 months, a bird-watching adventure; and at 42 months,
the opening of an ice-cream shop. Mother–child interactions in the family’s liv-
ing room were videotaped during the events, providing a precise record of how
each dyad interacted when dealing with component features of the events (e.g.,
the backpack and sleeping bag in camping). Haden et al. (2001) explored whether
recall of the various components varied as a function of the type of talk (e.g., joint
verbal, mother-only verbal, child-only verbal, no verbal). An analysis of the pro-
portion of features of the three events remembered by the children in response to
open-ended questions revealed a striking effect of joint talk as the events unfolded.
For each of the activities, the features that were handled and discussed by both the
mother and the child were better recalled than those that were jointly handled but
talked about only by the mother. Not surprisingly, the proportion of features cor-
rectly remembered increased with age. Overall, the findings from this longitudinal
study suggest that the nature of mother–child interaction as an event unfolds influ-
ences encoding and subsequent remembering. Children as young as 2.5 years of
age show memory benefits from conversations about events in the here and now
(see also Tessler and Nelson 1994).
Given the primarily correlational nature of this finding, Boland et al. (2003)
designed an experiment in which mothers were trained to use specific conversational
techniques such as Wh-questions, follow-ins that encourage discussion of aspects of
events the child is talking about, and positive evaluations to enhance their children’s
understanding of unfolding events. The intervention turned out to be successful such
that trained mothers produced significantly more of the targeted conversational tech-
niques than untrained mothers did. The effects of the training did not vary as a func-
tion of the children’s language skills. Of even greater interest, the children’s recall of
the target event was affected by the training that their mothers received such that the
children of trained mothers recalled more details of the event.
Several other intervention studies focusing on past events have yielded similar
outcomes (e.g., Peterson et al. 1999; Reese and Newcombe 2007). For instance,
Peterson et al. (1999) taught mothers from low-income families to use a highly
elaborative style when reminiscing about a past event with their 3-year-olds.
Trained mothers managed to increase the degree to which they used an elabora-
tive style, and their children showed higher levels of recall 1 and 2 years later when
interviewed by an unfamiliar adult than children of mothers who had not been
96 5  The Development of Event and Autobiographical Memory …

instructed. Similarly, Reese and Newcombe (2007) trained mothers in elaborate

reminiscing when their children were about 18–30 months of age. After a baseline
assessment of children’s language and self-awareness and mothers’ reminiscing
style, half of the mothers were assigned to receive training in elaborative remi-
niscing every 2 months over the ensuing year. The control-group dyads were also
visited an equal number of times and were administered assessments of the chil-
dren’s language and memory. The results of both short-term and long-term posttests
showed positive effects of the intervention procedure. Two months later, trained
mothers asked more open-ended questions than did untrained mothers. About
15 months later, the children of trained mothers provided more memory elabora-
tions during shared recall than did children of untrained mothers but only if they
had begun the study with a stronger self-concept. This finding is in accordance with
the longitudinal correlational results of an independent study (Harley and Reese
1999). These findings were also basically confirmed in a more recent study by van
Bergen et al. (2009) who conducted a 6-month longitudinal study of Australian
mothers and their preschoolers.
Although these parent intervention studies all showed some positive training
effects, they also indicated that these effects were less pronounced when children
were very young and did not have a firm sense of self and when the intervention
periods were rather short. However, there were no effects of maternal education
levels on the success of the intervention.
The consistency of findings across delay intervals, events, and different stud-
ies supports the view that mother–child interaction events contribute substantially
to children’s understanding and memory. As noted by Fivush (2014), the narrative
retelling of our experiences is the key in the development of autobiographical mem-
ory as it provides a coherent structure of our past. Although the research by Ornstein
and colleagues has convincingly shown that mother–child interactions during an
unfolding event are helpful for building and improving memories of this event, it
is still an open issue whether differences in the timing of interviews and talking
about events make a difference. For instance, McGuigan and Salmon (2004, 2005)
engaged 3- and 5-year-old preschoolers in a series of unusual entertaining activities
in the laboratory and questioned them about their experiences under different con-
ditions, using children’s final free recall as the dependent variable. Whereas some
children were interviewed before the event occurred to find out about the impact of
their expectations, others were interviewed during or after the event, using either
a high- or a low-elaborative reminiscing style. Finally, another group of children
talked with an adult about the event as it was unfolding. The adult talked with some
of these children in a highly elaborative way and with others in a low-elaborative
way. All these conditions were crossed and counterbalanced. As a main result, it
was found that elaborate questioning after the event best predicted the accuracy and
amount of children’s recall, although talking during the event was also facilitative for
the 5-year-olds. Children’s recall was enhanced to a greater extent by the degree to
which they contributed to the talk relative to the adult.
In a more recent study by Hedrick et al. (2009), an experimental design
was used to examine the effects of elaborative talk during or after an event on
Social–Cultural Contexts of Event Memories … 97

children’s event memory reports. Sixty preschoolers were randomly assigned to

one of four experimental conditions that varied according to a researcher’s use of
high- or low-elaborative talk during or after a camping event. In a memory conver-
sation 1 day after the event, children who engaged in high-elaborative talk during
or after the event reported more information than children who engaged in low-
elaborative talk during or after the event. Moreover, 3 weeks later, when a stand-
ard memory interview was conducted with all of the children, high-elaborative talk
during the event influenced the children’s memory reports. Thus, the outcomes of
these studies seemed to depend—at least to some extent—on the delay interval
between conversation and recall.

Relevance of Social–Cultural Context: Reminiscing

and Socioemotional Development

Please note that reminiscing style has been broadly measured in terms of elabo-
ration in most studies. The construct of elaboration is a complex one and can be
conceptualized along multiple dimensions (see Fivush et al. 2006). It not only
focuses on more cognitive aspects such as memory skills but also includes emo-
tional aspects of reminiscing such as sharing and evaluating the conversation part-
ner’s contributions to the evolving story. Some research suggests that elaborative
content and emotional affirmations might play separate roles in children’s devel-
oping autobiographical memory skills. Given that autobiographical memory is
about constructing coherent narratives about the self, mothers’ reminiscing style
is related not only to children’s memory development but also to several aspects
of their social and emotional development (see Fivush 2009; Fivush et al. 2006;
Nelson 2014). The assumption here is that mothers who provide and structure
more coherent and elaborated narratives about past experiences facilitate their
children’s understanding of the personal past, helping them to create a coherent
(i.e., temporarily extended) sense of self and personal relationships (Fivush and
Zaman 2014). Moreover, the argument is that creating a coherent narrative about
emotional experiences helps children to understand the relevance of emotions and,
for instance, to better cope with negative emotions.
Meanwhile, several studies have shown that mothers with a more elaborative
reminiscing style facilitate the development of a more coherent and differenti-
ated self-concept in their young children (e.g., Bird and Reese 2006; Welch-Ross
et al. 1999). In addition, mothers who use more internal state language when remi-
niscing with their preschoolers have children who demonstrate a more coherent
self-concept (Wang et al. 2010). There is also evidence that maternal reminiscing
style is closely related to children’s developing emotion regulation. Mothers who
use more internal language and are highly elaborative have children who display
more sophisticated emotional skills and who are better able to label emotions and
to understand the relations between emotions and behavior (Fivush 1997). More
elaborative mothers have children who display more emotional stability and also
98 5  The Development of Event and Autobiographical Memory …

show more adaptive coping strategies when dealing with traumatic and stressful
events (Bauer et al. 2007; Laible and Thompson 2000; Sales and Fivush 2005).
This relationship has also been demonstrated in longitudinal work. Reese and
colleagues extended their longitudinal studies of social interaction to include the
role of children’s self-awareness and their sociocognitive and socioemotional
development in their growing autobiographical memories in a longitudinal study
of more than 50 children (for details, see Reese 2002, 2009). As a main result,
children with advanced self-recognition at the age of 18 months showed steeper
increases in their independent remembering of past events than children with less
advanced self-recognition. By early adolescence, participants with a stronger
sense of past self-assessed in early childhood had more accurate memories from
early childhood (Reese et al. 2010). Other longitudinal research with this same
sample demonstrated that young children’s secure attachment to their mothers
played a role in how effectively they internalized their evaluations of past events
(Newcombe and Reese 2004). Taken together, these findings indicate that chil-
dren’s autobiographical memory is amplified when their mothers adopt an elabora-
tive, autonomy supportive, and emotionally open style of reminiscing with them
during early childhood (Reese 2014).
There is evidence that the positive role of maternal elaborative reminisc-
ing style holds across different cultures, even though systematic differences can
also be observed. According to several theorists (e.g., Triandis 1989), differences
in cultural expectations may be influential in that European American cultures
support a more independent notion of self (autonomy), whereas Asian cultures
support a more communal concept of self (relatedness). For autobiographi-
cal memory, mothers from European American cultures thus may foster a more
detailed sense of the self in the past, whereas mothers from Asian cultures may
want to downplay the independent self, supporting the reconstruction of a more
communal past. In fact, several cross-cultural studies have demonstrated that
mothers from Western middle-class cultures are on the whole more elaborative
during past event conversations than mothers from non-Western cultures (e.g.,
Leichtman et al. 2003; Mullen and Yi 1995; Schröder et al. 2013; Wang 2001,
2014). For instance, Schröder et al. (2013) examined the conversations of 164
mothers from seven different cultural contexts (Western and non-Western). The
children’s provision of memory elaborations was high in Western middle-class
families, low in non-Western rural farming families, and moderate in non-Western
middle-class families. Across contexts, maternal evaluations prompted children
to contribute memory elaborations. Interestingly, maternal elaborations predicted
children’s memory in non-Western rural families but not in other contexts.
The studies by Wang (2001) and Wang et al. (2000) support the idea that Asian
mothers are less elaborative than American mothers and that they do not focus on
the independent self as an agent of past experiences. Whereas European American
mothers tend to emphasize children’s personal experiences during reminiscing,
Chinese mothers tend to focus on reminiscing as a way to teach moral values.
According to Miller and colleagues (e.g., Miller et al. 1997), European American
mothers, regardless of social class, share a goal of autonomous reminiscing for
Social–Cultural Contexts of Event Memories … 99

their children, and they achieve this goal by asking many questions about past
events. Thus, European American mothers are more elaborative overall during
reminiscing than mothers from Asian cultures. Less elaborative reminiscing pre-
dicts a later age of earliest memory, as has been found in Asian populations (for
reviews, see Fivush et al. 2006; Nelson and Fivush 2004).
There are also cultural differences in the ways in which emotions are incorpo-
rated in mother–child conversations. American parents seem to focus on emotions
in reminiscing more than Asian parents do (Wang 2001), and American children
include more emotions in their independent narratives than Asian children do (Han
et al. 1998). This pattern of findings was also confirmed in more recent longitudi-
nal studies. For instance, Wang (2007) conducted a cross-cultural longitudinal
study on the development of autobiographical memory with European American,
Chinese American, and mainland Chinese families; this study showed that European
American mothers were more elaborative reminiscers as a group than either Chinese
American or Chinese mothers. In the same sample, Wang (2008) also noted the
unique role of children’s understanding of emotions at age 3 for their ability to
remember specific events later in the preschool period regardless of culture. Overall,
the findings suggest that cultural differences in maternal elaborative style seem to
have implications for the contents of autobiographical memories across the life span.
On the other hand, however, it should not be overlooked that within-culture
variations that occur between subcultural groups and individuals are common. For
instance, Asian American parents have been found to emphasize independence and
personal sufficiency to their children in addition to the family interdependence that
is also equally valued in their socialization practices (Wang 2014). So it seems dif-
ficult to characterize mother–child conversations as predominantly autonomy ori-
ented versus relatedness oriented as has been attempted in several cross-cultural
comparisons (e.g., Schröder et al. 2013). Growing up in modern society, children
constantly encounter new technologies and new ways of thinking, and this makes
the construction of the cultural self and autobiographical memory an extremely
complex process.

How Long Do Autobiographical Memories Last?

This question can be answered only through longitudinal research. Whereas many
areas of memory development research suffer from the problem that longitudinal
evidence is scarce, this is not true for the domain of autobiographical memory. The
fact that numerous studies have explored the emergence of autobiographical mem-
ory is particularly impressive given that long-term longitudinal studies are not only
expensive but are also plagued by numerous methodological and practical problems
(see Reese 2014; Schneider 1989). The field of autobiographical memory defini-
tively benefits from the fact that a range of longitudinal studies using very differ-
ent methods have accumulated evidence informing us of the conditions under which
young children are able to maintain memories over longer periods of time.
100 5  The Development of Event and Autobiographical Memory …

Although there are large individual differences in young children’s language

skills, their ability to talk about the past, and their interest in talking about the
past, a number of findings concerning long-term recall have been cross-validated
in several studies. For instance, it was repeatedly found that from around age 3
on, children can memorize novel and salient events across the preschool years.
A consistent finding is that the salience and emotional valence of the event influ-
ences long-term recall. Overall, negative salient events such as a personal injury or
a visit to the emergency room are generally better remembered than more positive
experiences, even though memories for the latter are also impressive and can be
maintained over long time intervals. There is multiple evidence that young chil-
dren maintain their memories for negative salient events over long delays of 1 to
5 years (Burgwyn-Bailes et al. 2001; Peterson and Parsons 2005). For instance,
the fact that negative experiences are retained particularly well in the long term
was demonstrated by Fivush et al. (2004). The children in their study who had
experienced Hurricane Andrew at the ages of 3 or 4 years remembered the event
6 years later. Similarly, almost 80 % of the 7-year-olds in the study by Peterson
and Parsons (2005) recalled a medical emergency that had occurred at age 2. By
contrast, only 20 % of the children who had experienced the Magic Shrinking
Machine (supposed to be a positive event) in early childhood remembered it
6 years later (Jack et al. 2012). Interestingly, rapid forgetting of both positive and
negative events has been observed in middle childhood. Repeated interviews using
open-ended questions may strengthen children’s memory in the long-term (e.g.,
Goodman and Quas 2008; Tizzard-Drover and Peterson 2004). In Tizzard-Drover
and Peterson’s (2004) study, however, additional interviews strengthened long-
term memory only for preschoolers but not for older children (see Ornstein et al.
2006, for similar findings).
Given the generally rapid rise in the forgetting of early experiences beginning
by middle childhood, it seems surprising that when older children do recall an
event from early childhood, they sometimes present more details about the event
than they provided in earlier interviews regardless of whether the event was posi-
tive or negative (Fivush et al. 2004; Morris et al. 2010). This reminiscing phe-
nomenon could be related to positive effects of repeated questioning but may also
be due to the children’s growing experience with narratives and their advanced
perspective taking (Reese et al. 2011). Although this means that more information
is provided by older children, recall has been shown to be less accurate over time.
A critical variable in long-term autobiographical memory is the organization
of memory at the time of initial recall. Given that young preschool children are
less able to present a coherent narrative, their memories of salient events do not
fare as well in the long-term as those of older preschoolers (Pillemer et al. 1994;
Morris et al. 2010). Most long-term studies of autobiographical memory in child-
hood have found that the thematic coherence of individual memories near the time
of the critical event was the best predictor of their retention some time later (see
Reese et al. 2011, for a review).
Overall, the findings suggest that individual memories are retained at fairly
impressive levels across the preschool period, particularly if the to-be-remembered
Social–Cultural Contexts of Event Memories … 101

event was negative, novel, and salient to the child. The forgetting of early child-
hood experiences appears to increase exponentially in middle childhood around
the age of 8, after which it begins to stabilize. By the time children enter ele-
mentary school, their long-term memory of personal events is stable and strong
regardless of the nature of the event (see Bauer 2007; Reese 2009, for reviews).
Forgetting an entire event is rare, and memories of salient events can last for
5 years or more (Hudson and Fivush 1991; Jones and Pipe 2002). During middle
childhood, children are able to provide elaborative detail and coherent reports of
early experiences, and also tend to evaluate their experiences by including more
internal state language and reflections on the outcome (see Reese et al. 2011, for
a review). Developing cognitive and social skills in adolescence provides the basis
for new forms of autobiographical narratives, the life narrative, including the abil-
ity to engage in greater levels of hypothetical reasoning, to understand causality as
well as conventional time, and also the abilities to take the perspectives of others
and to integrate emotions (Fivush et al. 2010; Habermas 2007).
If memory during early childhood and beyond is relatively good, why is it that
as adults we cannot remember our earliest experiences? This problem of nearly
completely forgetting our early memories by adulthood has been labeled infantile
amnesia. Although the phenomenon as such has been well known for more than
50 years, it was only during the past two decades that several theoretical explana-
tions have been provided; these will be summarized below.

Infantile Amnesia

Although the neural “hardware” necessary to encode, store, and retrieve informa-
tion is present at birth (and before), very young children’s memory is rather frag-
ile. This frailty of early memory has been assumed to cause “infantile amnesia,”
that is, the phenomenon that adults seem unable to recall events that happened to
them in their infancy and toddler years, a finding repeatedly documented in the
literature (e.g., Bauer 2006, 2014; Courage and Howe 2004; Howe and Courage
1993, 1997; Nelson and Fivush 2004). The general absence of autobiographical
memory before the age of about 3 years has stimulated numerous explanatory
attempts and empirical studies (for reviews, see Howe et al. 2009; Nelson 1993;
Schneider and Pressley 1997).
The term infantile or childhood amnesia was first coined by Freud (1905/1953)
who pointed to the fact that memory of one’s past life is not present from birth.
Numerous studies exploring adults’ earliest memories carried out over the course of
more than 100 years produced one of the most consistent findings in the psycho-
logical literature (see reviews by Dudycha and Dudycha 1941; Pillemer and White
1989; Rubin 1986; White and Pillemer 1979): The earliest childhood memory of
the populations studied (primarily European American males) was dated on aver-
age at 3.5 years of age. The same average age of earliest memory was found regard-
less of whether the source of the data was a survey, a narrative, or a response to
102 5  The Development of Event and Autobiographical Memory …

a memory cue (Bauer 2014). At the same time, wide individual differences were
reported, with some adults claiming memories from about 2 years, and some consid-
erably later, that is, not before 7 or 8 years of age (Usher and Neisser 1993; West and
Bauer 1999). Several studies also reported differences with regard to gender and cul-
ture, such that women had memories from earlier in life than men, and individuals
from Western cultures reported memories from earlier in life than individuals from
Eastern cultures (for reviews, see Fivush and Zaman 2014; Wang 2014).

Empirical Findings

Several methods have been used to assess the existence of early memories in chil-
dren and adults. The most common procedure asks participants to report their ear-
liest memories. This method was already used in the first empirical study on this
issue conducted by Henri and Henri (1898). In their study, 123 adolescents and
adults (age range: 16–65 years) answered a series of questions about their earliest
recollections. Although there were large individual differences in the age at which
these early memories occurred, most early memories fell between the ages of 2
and 4 years.
One of the obvious problems related to this method is that many adults may
have difficulty providing a precise date for their earliest memory unless it is tied
to some datable event, such as the birth of a sibling (Nelson and Fivush 2004).
Nonetheless, it is impressive that the average age of the earliest memory has been
consistent across most studies that have used this approach. Tustin and Hayne
(2010) calculated the grand mean of the average age of earliest memory found in
these studies and reported a finding of 3.48 years, with modes that ranged from 3
to 4 years.
To overcome some of the problems related to this approach, Usher and Neisser
(1993) asked college students to recall events that were known to have occurred
when participants were 2, 3, 4, or 5 years of age. The critical events included the
birth of a sibling, a family move, and the death of a relative. Interestingly, the age
of earliest memory depended on the event in question. On average, participants
recalled a sibling birth that occurred as early as age 2, whereas the death of a fam-
ily member was not recalled before the ages of 3 or 4. As noted by Nelson and
Fivush (2004), this finding indicates that infantile amnesia is not an all-or-none
phenomenon but illustrates that salient personal experiences that vary along
dimensions of emotionality and distinctiveness may be differentially retained into
adulthood. Overall, however, the average age of the earliest personal memory
(3–4 years of age) was very similar to that found in previous studies. The robust-
ness of this finding is indicated by a reanalysis of four recent studies on this topic
conducted by Bauer (2014). This reanalysis yielded an average age of earliest
memory at 38 months, with a range of 28–45 months.
Another methodology that has been used to study early memories asks adults
to recall as many memories as possible from specific time periods across their life
Infantile Amnesia 103

span (Rubin 1986). Using this procedure, researchers (e.g., Wetzler and Sweeney
1986) were able to estimate a “forgetting curve” for personal memories. As a main
result, it was found that there were significantly fewer memories before the age
of 7 than expected by using a linear function of normal forgetting, and that there
was a dramatic drop-off of memories before the age of 3, implying an accelerated
loss of memories from early childhood and suggesting that the average age of the
earliest memory was about 3 years (for a similar finding, see Weigle and Bauer
2000). According to Bruce et al. (2000), autobiographical memories become more
continuous at the age of about 5 years, substantially later than the age of earliest
memories.
In sum, the large body of research conducted with adults concurs that there is
little recall of events from before the age of 3–4 years. This finding seems para-
doxical in the light of the numerous prospective studies on autobiographical mem-
ory summarized above, all of which showed that even 2- and 3-year-olds have a
well-functioning memory system and can readily talk about past events, including
those that happened several months or even more than a year earlier (e.g., Peterson
and Whalen 2001). Thus, young children have demonstrated verbally accessible
memories of events that occurred when they were about 3 years of age, and yet
adults seem to have no recall of any of those events (Fivush and Hamond 1990).
To solve this puzzle, it seems worthwhile to investigate whether the phenomenon
of infantile amnesia is restricted to adults or can also be found in children. Several
more recent studies have focused on this issue, interviewing children about their
earliest memories, and also comparing childhood amnesia in children, adolescents,
and adults (see Bauer 2014, for a review).

Explanations of Infantile Amnesia and Their Evaluation

How does the empirical evidence summarized above relate to the numerous theo-
retical explanations of infantile amnesia? Historically, a variety of hypotheses have
been developed to explain the phenomenon. For instance, Freud (1905/1953) pro-
posed that memories that are too arousing for the ego, particularly traumatic ones,
will be repressed and screened from consciousness. Overall, there is not much
empirical support for this assumption. First, adults typically have more memo-
ries of early childhood than would be expected from Freud’s model of repression.
Second, several longitudinal studies on the development of autobiographical mem-
ory have shown that traumatic events from early childhood are particularly well
remembered in the long term (Peterson 2002).
Another category of theoretical accounts has suggested that very early personal
experiences were never registered properly initially or were rapidly lost from long-
term memory. For example, findings from early neuropsychological studies on
memory in infancy have suggested that infantile amnesia results from the inability
to form and retain explicit memories in infancy and early childhood. As shown
above, evidence for long-term recall during the first 2 years of life speaks against
104 5  The Development of Event and Autobiographical Memory …

this hypothesis because infantile amnesia generally extends to 3 years and later
(Nelson 1994). A related hypothesis claims that infantile amnesia phenomena may
result from memory loss because of problems identifying the origins of the mem-
ories or weak representations of and mismatches between contexts (e.g., Rovee-
Collier and Hayne 2000). For instance, because infants lack language, they encode
their experiences visually or acoustically but not symbolically. Theorists favoring
a retrieval failure hypothesis thus assume that memories formed without language
become inaccessible later on when nonverbal encoding gives way to verbal encod-
ing (e.g., Neisser 1962). However, as was shown above, the empirical support for
this assumption is not very convincing given that preverbal experiences under
some circumstances can later be described once language is acquired (e.g., Bauer
et al. 1998). Meanwhile, numerous studies have demonstrated that the assumption
that infants and very young children lack the symbolic capacity to form long-term
memories is no longer tenable (see Chap. 4).
Another explanation of infantile amnesia suggests that adults do not have many
memories of early childhood because no such memories were formed during the
first years of life due to a general cognitive deficit. For instance, Piaget (1962) pro-
posed that children younger than 2 years of age do not have the capacity for cogni-
tive representations. On the basis of the assumption that very young children are
unable to represent objects and events in their absence, these children should not
be able to recall past events. Both the research on infants’ long-term memory abili-
ties described in the preceding chapter and the prospective longitudinal studies on
the development of autobiographical memory summarized above do not support
this classic assumption.
Given that so many explanatory hypotheses are not supported by recent empiri-
cal data, are there any hypotheses that can be reconciled with the existing data-
base? The answer is yes. Several hypotheses can be linked to empirical findings.
For instance, Leichtman and Ceci (1993) assumed that infantile amnesia is a con-
sequence of the child’s developing information-processing system. In accordance
with basic assumptions of fuzzy-trace theory (Brainerd and Reyna 2002, 2014,
see discussion later in this chapter), Leichtman and Ceci proposed that there is
a developmental shift in how events are represented, such that young children
encode events primarily in terms of verbatim (precise) memory traces, whereas
older children and adolescents rely more on gist (less precise) traces. Given that
verbatim traces are more susceptible to forgetting than gist traces, heavy reliance
on verbatim traces makes memories from infancy and early childhood unavailable.
Gist traces become increasingly available by the early school years, about the time
when more memories can be retrieved.
Another explanation was provided by Howe and Courage (1993, 1997b; Howe
et al. 2009) who proposed that important specific conceptual changes play a role
in the explanation of infantile amnesia. Here, the assumption is that the capstone
event in the demise of infantile amnesia does not occur in the memory mecha-
nisms themselves but in the onset of the cognitive sense of self, as indicated by the
toddler’s recognition of self in a mirror, which is typically found between 16 and
24 months of age (Lewis and Brooks-Gunn 1979). As emphasized by Howe et al.
Infantile Amnesia 105

(2009), there is reason to assume that the factor most important for the emergence
of autobiographical memory is the emergence of the cognitive self, late in the sec-
ond year of life. This cognitive self enables a new knowledge structure whereby
new experiences can be organized as personal. This position seems consistent with
Tulving’s (2002) notion of “autonoetic consciousness,” or an awareness that a cer-
tain event happened to “me” in the past. In a similar vein, Perner and Ruffman
(1995) assumed that episodic memory requires experiential awareness—that is,
the consciousness of having experienced remembered events—which is not pos-
sible before the age of about 4 or 5 years. Thus, the development of autobiographi-
cal memory is accompanied by an understanding of the perceptual origins of one’s
own knowledge (“theory of mind”).
In fact, prospective longitudinal studies on the development of autobiographical
memory have confirmed the importance of the emerging self. In line with Howe
and Courage’s (1993) theory of the role of self-awareness in offsetting infantile
amnesia, children who recognized themselves in the mirror were found to have
more robust event memories and to make faster progress in independent autobio-
graphical reports relative to children who took longer to develop self-recognition
(Harley and Reese 1999; Reese 2014). Children with advanced self-recognition
showed steeper increases in their independent remembering than children with less
advanced self-recognition. By early adolescence, children with a stronger sense of
a past self in early childhood had denser memories from early childhood in which
their autobiographical memories were more closely spaced (Reese et al. 2010).
On the other hand, the longitudinal data analyzed by Reese et al. (2010) did not
support Perner and Ruffman’s (1995) proposal that children’s autobiographical
memory is uniquely linked to changes in children’s theory of mind even though
maternal reminiscing style supports both children’s autobiographical memory and
their understanding of mind.
Whereas it is clear that the notion of self in memory is crucial for autobio-
graphical memory formation and to the issue of infantile amnesia, the cognitive
self is only one important factor that helps in overcoming the amnesia barrier. For
instance, although Leichtman and Ceci (1993) mainly referred to the tenets of
fuzzy-trace theory to explain the infantile amnesia phenomenon, they also noted
that children become increasingly facile with language during their early elemen-
tary school years, using it not just for communication but also as a representa-
tional system. In fact, several theorists have proposed changes in language and
how language is used as an explanation for the phenomenon of infantile amnesia
(Fivush and Hamond 1990; Nelson 1993, 1996). These alternative theories about
the delayed onset of autobiographical memory focus on the roles of emerging lan-
guage and social interaction.
Empirical support for the importance of language can be found in Simcock
and Hayne’s (2002) research, which showed that 2- and 3-year-old children were
not able to translate a preverbal memory into language 6 months later. But lan-
guage is not fully developed by 3 years of age, and other language-related changes
may further influence children’s ability to remember events from the past. Social
interaction, in particular, the sharing of experiences with others linguistically,
106 5  The Development of Event and Autobiographical Memory …

helps children to understand that language can be used to share memories with
others (for reviews, see Fivush 2009; Nelson and Fivush 2004; Reese 2002). As
was shown above, there is growing evidence that talking about past experiences
with one’s parents and family facilitates the construction of a personal history. In
particular, the prospective longitudinal research conducted by Reese and her col-
leagues has demonstrated that other factors such as early mother–child interaction
also play a major role. For instance, Reese et al. (2010) found that maternal remi-
niscing style in the form of elaborative questions not only remained the dominant
influence on autobiographical memory in early childhood but also affected auto-
biographical memory in adolescence. Adolescents had earlier and more volumi-
nous memories from childhood and more insightful memories from adolescence
if their mothers used an elaborative reminiscing style. This finding is compatible
with the view that the cognitive self in the toddler years is but one factor that is
mediated by others such as maternal discourse style as well as cognitive and emo-
tional processing of event information in predicting later memories (Fivush and
Zaman 2014; Nelson and Fivush 2004; Wang 2008).
Although recent prospective longitudinal studies have been helpful in identi-
fying myths concerning the phenomenon of infantile amnesia and have also pro-
vided meaningful explanations for its occurrence, the findings have indicated
that some unresolved issues remain. For instance, recent empirical findings have
challenged the view that episodic memory does not initially emerge until the age
of 4. For instance, the children in the Tustin and Hayne (2010) study accurately
reported information of events that occurred when they were much younger than
4 years of age. Thus, the boundary of childhood amnesia (3.5 years) cannot be
explained by a lack of episodic memory during early childhood. However, a recent
study by Scarf et al. (2013) showed that even though 3-year-olds form episodic
memories, they fail to retain those memories following a delay of 30 min. By con-
trast, 4-year-olds retained episodic memories over much longer delays (up to a
week). The findings of this study support an alternative explanation for childhood
amnesia. That is, although 3-year-olds can form episodic memories, they fail to
retain them over long delays. These results as well as those by Tustin and Hayne
(2010) provide an explanation that is based on forgetting, indicating that forget-
ting makes at least some contribution to childhood amnesia during adulthood. A
comparison of preschoolers’ and adults’ reports of their earliest memories illus-
trates the case. When Tustin and Hayne interviewed children and adults about their
earliest childhood memories, 5-year-olds recalled events that took place when they
were about 1 year old, whereas adults showed the traditional lack of memories
prior to 3.5 years of age. These findings support the assumption that young chil-
dren form episodic memories beginning very early in childhood but that they for-
get these memories over longer periods of time.
Also, there is little doubt that theories highlighting the roles of children’s lan-
guage development, their self-concepts, as well as the role of social interaction
(or more likely, an integration of all these approaches) provide important expla-
nations for the occurrence of the infantile amnesia phenomenon as well as for
individual variations in children’s or adults’ access to early memories. However,
Infantile Amnesia 107

they are less successful in accounting for the longitudinal shift in the age of chil-
dren’s earliest memories found by Peterson et al. (2011). In the Peterson et al.
study, 4- to 13-year-old children were asked for their three earliest memories in
an initial interview and again 2 years later. The findings of that study indicated
that children’s age at the time of their first memory increased as their current age
increased, and their age at the time of their first memory shifted upward by several
months between their initial and follow-up interviews. Also, older children were
more likely to provide the same memories in both interviews, and cues were more
helpful for the older than for the younger children. In a reanalysis of these data,
Peterson et al. (2014) examined influences on the survival of children’s earliest
memories. The data from the Peterson et al. (2011) study were coded for inclusion
of emotional terms and thematic, chronological, and contextual narrative coher-
ence. Results of logistic multilevel modeling indicated that the prediction of the
memory’s survivability over the 2-year period was enhanced by emotion and each
dimension of coherence. On the other hand, the uniqueness of the event and mem-
ory cues were not associated with differences in retention.
Although the Peterson et al. (2014) study provides interesting information
on predictors of memory survivability, we still do not know why the bounda-
ries of infantile amnesia are flexible or why the consistency of earliest memories
increases as children get older. Also, why is it that early memories are so rapidly
forgotten in middle childhood and even more rapidly forgotten by younger than
older children (Reese 2014)? Research focusing on the neural substrates of auto-
biographical memory points to the importance of the protracted development of
cortical structures that affect the encoding, consolidation, and retrieval of early
memory traces (for reviews, see Bauer 2007, 2014). Thus, is it likely that access to
early memories is gradually lost because the information is too poorly organized
to be retrieved over very long delays. This interpretation is supported by the find-
ing that the distribution of early memories generated by young children best fits an
exponential function rather than a power function and that early in the preschool
years, the rate at which memories are forgotten outstrips the rate at which they
are retained (Bauer et al. 2007). Later, during the preschool years, remembering
begins to outstrip forgetting, resulting in a more robust and reliable memory.
Recent brain research also indicates that changes in hippocampal neurogenesis
may play an important role (Akers et al. 2014). Throughout life, new neurons are
continuously added to the dentate gyrus, remodeling hippocampal circuits and
leading to forgetting of established memories. Rates of hippocampal neurogen-
esis are high early in life but decline substantially with age. During infancy, when
neurogenesis levels are elevated, high rates of decay may render hippocampus-
dependent declarative memories inaccessible at later time points. In fact, infantile
forgetting or amnesia is observed across a wide range of species. As demonstrated
by Akers et al. (2014) through a series of experimental studies with mice, reduc-
ing neurogenesis at this developmental stage can increase the persistence of
hippocampus-dependent memories. Moreover, increasing neurogenesis levels at
a later stage of development promoted forgetting in adult mice. These findings
indicate that there is a trade-off between plasticity—the ability to incorporate
108 5  The Development of Event and Autobiographical Memory …

new information—and stability—ensuring that the process of incorporating new

information does not degrade information already stored in episodic memory.
Although these new and interesting findings contribute to our understanding of
infantile amnesia, they do not seem sufficient to explain the large loss of memory
observed during the transition from infancy to young childhood. As noted above,
the emergence of the cognitive self, developmental shifts in the representation of
events, and increases in language abilities provide additional explanations for the
infantile amnesia phenomenon.
Overall, numerous developmental studies have indicated that the primary func-
tion of autobiographical memory is to develop a life history in time and to do so
by telling others about events of the past (this position is also supported by evolu-
tionary developmental psychology; see Bjorklund and Sellers 2014). Representing
events from one’s life is clearly a social process that begins early in development
(Fivush 2009). The capacity to organize information more coherently in memory
is critical to most memory advances in childhood and seems closely related to the
emergence of the cognitive self (see Howe et al. 2009). Subsequent advances in
language and social cognition seem very important for developmental changes
in autobiographical memory. An interesting aspect of autobiographical memories
is that they simultaneously reflect elements of the original experience as well as
interpretations of those events that occur when the events are discussed with sig-
nificant others.
One thing that is certain about autobiographical memory is that our memories
of ourselves are never perfect. Given that autobiographical memory is reconstruc-
tive memory and that social interactions and long-term knowledge affect the origi-
nal encodings, there is always a risk of memory errors. The reliability of children’s
memory reports has been a core issue in applied psychology. The accuracy of
autobiographical memory is crucial in the context of forensic investigations, and
numerous studies have been conducted to investigate developmental trends in chil-
dren’s eyewitness memory.

Eyewitness Memory

Eyewitness memory represents one specific form of event memory or autobio-

graphical memory that emphasizes the accuracy and not so much the amount of
information recalled about an experienced event. As children are increasingly
called upon to provide testimony in legal cases, issues about how much and how
accurately children remember and the degree to which they are influenced by sug-
gestion and misinformation have become a high-priority research interest, particu-
larly in the USA (see Bruck and Ceci 1999; Ceci and Bruck 1993, 1998; Goodman
et al. 2014; Pipe and Salmon 2009; Qin et al. 1997, for reviews). According to
Ceci and Bruck’s (1998) rough and conservative estimate concerning the situa-
tion in the United States, about 100,000 children testify annually in criminal and
civil cases. In view of this large number of children providing statements of legal
Eyewitness Memory 109

importance, it seems important to know more about the reliability of children’s

statements. Core issues of interest to both developmental psychologists and the
legal profession include age differences in remembering self-related events, the
long-term properties of such memories, and their resistance to misleading post-
event information.
Two different research paradigms have been used to explore children’s and
adults’ eyewitness memory. In laboratory-style experimental event memory stud-
ies, researchers assess the quantity and quality of the information participants
report about staged or naturally occurring events. Because the event is known,
one specific advantage of this approach is that not only the amount of detail but
also the accuracy of memory reports can be evaluated. The latter is not true for
field or naturalistic studies because here the child is typically the only source
of information about what happened, with rare exceptions, and thus the accuracy
of children’s testimony cannot be directly assessed. However, such studies have
the advantage of ecological validity, and it is possible to explore how laboratory
findings generalize to the real-world setting of forensic interviews. Most research-
ers agree that the convergence of findings from these two approaches provides the
most robust basis for recommendations from the laboratory to the field (see Pipe
et al. 2004; Pipe and Salmon 2009; Paz-Alonso et al. 2009). Although the follow-
ing review draws heavily from experimental research, findings from naturalistic
field studies will also be considered.
In most experimental studies, children witness events that are either staged at
their schools or videotaped or are asked to recollect some potentially traumatic
experiences, such as visits to the doctor or dentist. Usually, children are not told
that they will be asked later to report what they can remember. During the inter-
view period following the event, children are typically first questioned using gen-
eral prompts, usually followed by more specific questions. There are, of course,
many variations depending on the purpose of the study. However, in most cases,
children’s memories for specific events are probed, often with the purpose of see-
ing how likely children are to change their answers when leading questions are
asked by an interviewer.
How reliable is children’s testimony? When a child is the primary source of
information in a forensic investigation, how do we know that we can trust his or
her testimony? When children do falsely remember information to cues, does
this actually change their memory representation? In most respects, eyewit-
ness memory is no different from any other type of event memory. Consistent
with other research on children’s event memory, substantial age differences
have been found when examining immediate free recall, that is, recall at the end
of the experimental session. Levels of recall in responses to general questions
are typically low in preschool and early school-age children and increase with
age (e.g., Baker-Ward et al. 1993; Cassel and Bjorklund 1995; Ornstein et al.
1992; Poole and Lindsay 1995). Despite low levels of recall, what preschoolers
recall is usually accurate and central to the event. Thus, most young children
possess the cognitive capacity necessary for accurate testimony. Age differ-
ences may diminish or even disappear under certain conditions, for instance, if
110 5  The Development of Event and Autobiographical Memory …

an event is particularly salient or personally meaningful to young children. In

most studies, levels of correct recall increased when more specific cues were
provided. Unfortunately, in most cases, such increases were also accompanied
by increases in the number of inaccurate responses, thus reducing the level of
overall accuracy, particularly in younger children. See the parallel findings in
research on aging (e.g., Shing et al. 2008).
When examining the accuracy of reports, that is, the ratio of incorrect to correct
information remembered, most studies have reported no age differences (Howe
et al. 1994; Rudy and Goodman 1991). What young children recall is highly accu-
rate and central to the witnessed event (Goodman et al. 1987; Baker-Ward et al.
1993; Poole and White 1995). For example, in a video involving a boy and a girl
in a park in which the boy steals a bike, young children typically recalled the bike
theft but were less likely than older children or adults to mention aspects of the
setting or characteristics of the bike (Cassel and Bjorklund 1995; Cassel et al.
1996). Thus, young children’s free recall has been found to be typically low, accu-
rate, and about the core aspects of the event.
As is the case for memory in general, levels of correct recall of an eyewit-
nessed event increase for people of all ages when more specific cues are provided
(Cassel et al. 1996; Ornstein et al. 1992; Poole and Lindsay 1995). However,
unbiased specific cues can also increase the number of inaccurate answers,
although not always (Howe et al. 1994; Poole and Lindsay 1995; Lindberg et al.
2000). Increased error rates in answers to unbiased cued questions were found
in a study by Cassel et al. (1996) who showed kindergarten, second-grade, and
fourth-grade children and adults a video, followed by free recall and then a set of
unbiased questions. Cassel and colleagues reported that the unbiased cues yielded
age-related increases in levels of correct recall for all subjects but also produced
increased levels of inaccurate recall, reducing the overall accuracy of recall, par-
ticularly for the kindergarten children (see also Bjorklund et al. 1998; Roebers and
Schneider 2000). Although eyewitness memory improves with age, differences
seem most pronounced for the period between 4 and 7 years of age. Most studies
that included samples of adults illustrated that older elementary schoolchildren’s
free recall of witnessed events was comparable to that of adults (e.g., Bjorklund
et al. 1998; Cassel et al. 1996; Roebers and Schneider 2005).
Despite age differences, most young children possess the cognitive capacity
necessary for accurate testimony. For instance, as already noted above, research
assessing children’s memory for traumatic and stressful events (e.g., memory for
a physical checkup) indicates that even 3-year-olds can provide accurate accounts
of personally experienced events (Baker-Ward et al. 1993; Goodman et al. 1991).
Age differences may diminish or even disappear under certain conditions. That is,
if an event is particularly salient or personally meaningful to young children, this
event becomes highly memorable. For instance, Saywitz et al. (1991) examined
children’s memory for a physical checkup in which half of the 5- and 7-year-old
children experienced genital touching. Saywitz and colleagues found age effects
in free recall for the children in the nongenital condition but not for the subgroup
that had experienced genital touching. Thus, under at least some conditions, even
Eyewitness Memory 111

preschoolers can provide relevant and reliable information about their experiences,
although eyewitness memory performance in such situations improves with age.

How Long Do Eyewitness Memories Last?

We know from cross-sectional experimental research on autobiographical memory

with adults (e.g., recall and recognition of familiar names and faces) that long-
term memory can be fairly accurate. For instance, Bahrick et al. (1975) asked
392 participants to recall and recognize the names and portraits of the members
of their high school graduating class. Participants were divided into nine groups
according to the elapsed time since their high school graduation, which varied
between 3 months and 57 years. As a main result, Bahrick et al. (1975) found that
recognition memory was very good: On average, participants were able to identify
about 90 % of the names and faces of the members of their class, and the vis-
ual information was retained virtually unimpaired for at least 35 years. Although
free recall performance was considerably lower in general, it varied as a function
of the nature of the interpersonal relationship. As may be expected, the names of
close friends and romantic interests were almost perfectly recalled, regardless of
the time interval between graduation and memory test. Thus, social context was
shown to be an important determinant of recall performance, whereas it did not
have much impact on recognition memory.
Regarding long-term memory of events, there is evidence that traumatic child-
hood memories can be maintained into adulthood, but these memories are not
immune from forgetting or distortion. On the one hand, Berntsen and Rubin (2006)
were able to show that Danish Second World War survivors, who were between 59
and 90 years old when they were interviewed about their childhood memories of the
German invasion of 1940 and subsequent surrender 5 years later, still had vivid and
extensive memories of these events. So there is no doubt that traumatic experiences
can be stored in memory for a very long time. On the other hand, however, there is
also evidence that such memories will not last forever. We know from research on
the eyewitness memory of concentration camp survivors (Wagenaar and Groeneweg
1990) that victims can forget the physical appearance and names of people who
abused them badly and almost caused their deaths some 40 years earlier.
How long do children’s memories of witnessed events last? Are there develop-
mental differences in long-term recall and forgetting rates? This seems like an
important question given that in forensic interviews, children are asked to recall
experiences or events they witnessed weeks, months, or even years earlier (Paz-
Alonso et al. 2009; Pipe and Salmon 2009). Although most developmental stud-
ies on the issue have not assessed children’s long-term recollections, several have
investigated children’s memories of specific events for periods ranging from several
weeks to 2 years (e.g., Flin et al. 1997). This is important for the legal system given
that children are sometimes asked in forensic interviews to remember events that
they witnessed a long time ago (Paz-Alonso et al. 2009; Pipe and Salmon 2009).
112 5  The Development of Event and Autobiographical Memory …

Although the results of studies on children’s eyewitness memories are not

totally consistent, they indicate that age differences in the accuracy of recall
increase with increasing delays, at least when delays are longer than a month.
Most research assessing the effects of delays of 1 month or less has reported that
the total amount of information recalled and the overall accuracy of recall (i.e.,
the ratio of inaccurate to accurate information) was comparable to what it was
when recall was tested immediately after witnessing the event regardless of age
(Baker-Ward et al. 1993; Cassel and Bjorklund 1995). However, age differences in
recall accuracy have been found with longer delays. For instance, Poole and White
(1993) reported that, although 6-, 8-, and 10-year-olds recalled about as much
accurate information as adults following a 2-year delay, they recalled significantly
more inaccurate information (about 20 %) than the adults (7 %), but no differences
were observed between the three groups of children (see also Flin et al. 1992).
The conclusion from these studies is that age differences in accuracy have been
found but only when memory is assessed after extended delays. This finding is in
accordance with the experimental research on long-term retention and forgetting
conducted by Brainerd, Howe, Reyna, and colleagues (e.g., Brainerd et al. 1993).
How can this pattern of findings be explained? One explanation has been
offered by fuzzy-trace theory (FTT; Brainerd and Reyna 1998, 2005; Reyna and
Brainerd 1995). According to this influential dual-process theory of cognitive
development, most cognition is intuitive in that it is based on fuzzy memory rep-
resentations. Brainerd and Reyna (2014) propose that memory representations
exist on a continuum from literal, verbatim representations to fuzzy, skeletal, gist-
like traces, and that multiple memory traces exist for any event one experiences.
Memory representations of an event are not clearly either verbatim or fuzzy but
may exist in a variety of forms simultaneously, each of which is independent from
the others (Reyna and Kiernan 1994). The type of trace people use will depend
on their age, the nature of the task, the durability of the trace, and the process-
ing preferences of the individual. An important assumption in fuzzy-trace theory is
that people of all ages prefer to use fuzzy traces when solving problems, although
this preference varies with age. Early in life, children are biased toward storing
and retrieving verbatim traces, although they also extract gist. A verbatim–gist
shift occurs sometime during the early elementary school years, with children then
showing a gist bias. According to FTT, age differences in the accuracy of recall are
due to the greater rate of decay in verbatim (exact) relative to gist traces. Verbatim
traces, favored by young children, deteriorate more rapidly than the gist or fuzzy
traces preferred by older children, resulting in a greater loss of information over
delays and corresponding increases in proportions of incorrect recall. The verba-
tim traces favored by young children are especially susceptible to deterioration,
making it highly unlikely that these memories would be available years after their
original encoding (Leichtman and Ceci 1993).
A host of factors other than age and length of delay have been found to affect
the amount and accuracy of children’s eyewitness memories (Reyna et al. 2002).
For instance, children with high IQs were found to provide higher levels of eyewit-
ness recall than their peers with lower IQs did, and children rated as shy by their
Eyewitness Memory 113

mothers did more poorly when answering specific questions about an observed
event than children of the same age who were not shy (Roebers and Schneider
2001). Also, beneficial effects of prior knowledge on memory were reported in
several studies (e.g., Clubb et al. 1993; Schneider and Bjorklund 1992). Moreover,
characteristics of the interview can greatly affect what children remember and
the accuracy of their recollections (for reviews, see Bruck et al. 2006; Ceci and
Bruck 1998). Overall, children recall more when given neutral cues than when
asked open-ended questions, but the accuracy of their recall declines because they
also recall more false information. There is also evidence that children’s personal
involvement in an experience affects memory accuracy. In general, directly par-
ticipating in an event is likely to result in stronger and more accurate memories for
both children and adults, compared with merely watching the event’s occurrence
(e.g., Murachver et al. 1996; Roebers et al. 2004; Rudy and Goodman 1991). Also,
interviewer characteristics, such as whether the interviewer is friendly and sup-
portive, have been found to influence the level of recall, particularly in younger
children (Roebers and Fernandez 2002; Schwarz and Roebers 2006). Another rele-
vant factor discovered in recent years and discussed in some detail below concerns
children’s metacognitive and background knowledge, in particular their monitor-
ing skills and their familiarity with the event.

The Role of Metacognitive and Background Knowledge

The importance of knowledge for the quality of eyewitness memory has been con-
firmed in numerous studies. As shown above, even young children can increase the
accuracy of their testimony when provided with adequate incentives. One inter-
esting finding concerns young children’s problems with spontaneously using the
“I don’t know” option, even when they are definitively unsure about the correct
answer (Roebers and Schneider 2000). This may be the case because they feel
obliged to answer all questions in order to meet the interviewer’s expectations
(Roebers et al. 2007). However, even young schoolchildren can be instructed to
enhance the accuracy of their testimony by screening out wrong answers and do
so spontaneously when given explicit incentives for accuracy (Koriat et al. 2001;
Roebers and Schneider 2005). Accordingly, young children can regulate how they
report their memories to produce a more accurate record of past events when they
are allowed and encouraged to screen out wrong answers (e.g., by saying “I don’t
know”) and when they are explicitly motivated to do so. Children given incentives
to be accurate in their recall are, indeed, more accurate than children not given
incentives (Roebers et al. 2001; Zajac and Karageorge 2009).
Meanwhile, numerous studies have demonstrated the relevance of metacogni-
tive knowledge, that is, children’s ability to monitor ongoing cognitive activities
(e.g., Keast et al. 2007; Roebers 2002; Roebers et al. 2007; Roebers and Schneider
2005). Roebers and colleagues assessed children and adults’ metacognitive moni-
toring in event recall and recognition tasks. For instance, Roebers et al. (2007) had
114 5  The Development of Event and Autobiographical Memory …

8- and 10-year-old children as well as adults watch a brief video. Participants were
asked immediately after the session to indicate how sure they were about whether
they would remember specific aspects of the video in 2 weeks (judgment-of-
learning interview). Two weeks later, participants were seen again and questioned
individually about the contents of the video. As a main result, children 8 years of
age and older already showed appropriate metacognitive monitoring. Participants’
judgments of learning and confidence judgments were higher for answerable than
for nonanswerable questions regardless of age, and the intraindividual consistency
in metacognitive knowledge across these two indicators was high. These findings
are in accordance with those of previous studies that also documented relatively
good metacognitive monitoring skills in young elementary schoolchildren and
only minimal developmental changes across the late elementary school years and
beyond (Roebers 2002; Roebers and Howie 2003). Young children’s metacogni-
tive monitoring appears to be insufficient when they answer misleading questions.
Those studies that included younger children (kindergarteners or first graders)
found that this age group tended to be overly optimistic as indicated by their con-
fidence judgments with regard to questions that they did not answer correctly.
Although the number of false alarms usually decreases with increasing age, post-
event misinformation generally has negative effects on recall performance regard-
less of age.
An identification lineup task is a special kind of eyewitness report in a foren-
sic context in which participants’ facial recognition memory is tested. There is
evidence that children perform more poorly on such tasks than adults and that
developmental improvement can be found between the ages of 4 and 10 but not
thereafter (Davies 1993). Apparently, young children seem to attend more to what
people do than what they look like. Interestingly, age-related differences are more
obvious when the test pictures do not include the target actor than when the target
is present in the lineup (Parker and Carranza 1989; Roebers 2003). In such situa-
tions, children’s confidence judgments are clearly not predictive of identification
accuracy.
This was demonstrated by Keast et al. (2007) whose study focused on the con-
fidence–accuracy (CA) relation, that is, the importance of identification confidence
for the discrimination between accurate and inaccurate identifications in a lineup.
Participants in this study were 12-year-old children and young adults. They viewed
a brief video of a credit card theft in a restaurant and were then presented with thief
and waiter lineups. When watching the video, all participants were informed that
they would be asked some questions about the video later. After viewing the video,
they completed a 15-min filler task and then worked on the lineups, making an
identification decision and indicating how confident they were that their decision
was correct. The target was present in about 50 % of the lineups. Although children
displayed higher confidence when making correct target-present identifications, a
result that is also typically found for adults, the children’s confidence was lower
when they made lineup rejections than that of adults, both correct and incorrect.
Consistent with previous research, the confidence–accuracy relations were gener-
ally modest, not exceeding a correlation of .33. Interestingly, a meaningful relation
Eyewitness Memory 115

between participants’ confidence and accuracy for all sets of stimulus materials and
instructional conditions was found for adults but not for the children. However,
additional prompting helped to improve the children’s monitoring skills. When
children were explicitly provided with the option to respond with “I am not sure,”
their false identification rates in target-absent lineups was substantially reduced.
This finding indicates that deficient metacognitive monitoring processes may be at
last partly responsible for children’s typically low performance on such face-recog-
nition tasks (for a more detailed discussion, see Roebers 2014).
Taken together, the findings by Keast et al. (2007) and Roebers et al. (2007)
indicate that children’s metacognitive monitoring skills in eyewitness contexts
seem to differ across settings: Whereas the diagnostic value of children’s identifi-
cation confidence in lineup situations (face-recognition tasks) seems questionable,
their confidence judgments in recall situations can assist police in determining
which of the items recalled are more likely to be accurate than others as long as
the questioning is unbiased (see Koriat et al. 2001, for supporting evidence).
As was shown above, individual differences in children’s background
knowledge for a given event also play an important role in eyewitness memory
(Elischberger 2005; Ornstein et al. 2006). Children’s recollections of stressful
and invasive medical procedures are related to their knowledge of the procedures.
Ornstein et al. (1998) included both typical and atypical features in a mock physi-
cal examination in a study with 4- and 6-year-old children. Children were inter-
viewed about the exam both immediately and 12 weeks later. When long-term
memory was tested, children of both ages recalled more typical features correctly
than atypical features, a finding that reflects the positive effects of the knowledge
base. However, there was also a negative side to knowledge in this study; that is,
children from both age groups were more likely to erroneously say that a typical
event happened when it did not. Thus, their false alarms were higher for typical
than for atypical events. Overall, however, research assessing the impact of back-
ground knowledge on children’s autobiographical memory has demonstrated that
young children are less susceptible to misleading information and better at recall-
ing experienced events when they possess a clearly developed script of the relevant
situation. Regardless of age, domain knowledge has a positive impact on eyewit-
ness memory (Ornstein et al. 1998).

Importance of the Salience of Events

Stressful events such as medical procedures are not only experienced by most
children, they are also particularly salient because of their threatening contents.
Developmental studies on the recall of stressful and traumatic events have typ-
ically found developmental differences such that older children tend to be more
accurate and complete than younger ones in their reports of nonstressful and
stressful events (for a review, see Paz-Alonso et al. 2009). Young children’s memo-
ries of stressful experiences such as a visit to the doctor for a physical examination
116 5  The Development of Event and Autobiographical Memory …

seem quite accurate (see Baker-Ward et al. 1993; Goodman and Quas 1997;
Ornstein et al. 1997, 2006). However, the relevant literature is not consistent with
regard to the assumption that higher levels of distress are associated with better
memory even though some findings are supportive. Direct comparisons of reports
of traumatic and nontraumatic events have revealed that narratives about traumatic
events such as a devastating tornado were about twice as long as those referring to
nontraumatic events experienced at about the same time, regardless of age (Ackil
et al. 2003; Fivush et al. 2004). Although the range in the ages of the children par-
ticipating in the study was large, age differences in recall were not pronounced.
The results of this research and other studies (e.g., Fivush et al. 2003) suggest that
children’s memory reports of traumatic and nontraumatic experiences differ quali-
tatively such that narratives about traumatic events are more complete and better
integrated (Bauer 2006).
However, as noted by Pipe and Salmon (2009), generalizing from laboratory
studies to the forensic context is not straightforward given that the events about
which children are interviewed in forensic contexts (e.g., sexual abuse) are not
always salient or significant for them at the time. Moreover, it seems likely that
in many studies in which children have shown good retention (e.g., dealing with
a hurricane or injuries), the children were probably reminded of their experiences
by family members or friends, and this is unlikely to occur in the case of sexual
abuse, which in most cases is a very private experience that is rarely discussed
with others (Fivush 1997). Such differences may affect the memorability of stress-
ful and traumatic events.

Age Differences in Suggestibility

Perhaps the single most investigated area of eyewitness testimony concerns age
differences in suggestibility (for reviews, see Brainerd and Reyna 2005; Bruck
and Ceci 1999; Ceci and Bruck 1998; Reyna et al. 2002). The primary concern of
developmental researchers has been whether children are more susceptible to sug-
gestion than adults. Needless to say, studies of children’s and adults’ suggestibil-
ity have important forensic implications as well. In most suggestibility paradigms,
participants witness an event and are later presented with some post-event infor-
mation that contradicts events observed earlier (misinformation) or are asked sets
of misleading questions that suggest an inaccurate “fact” (Ceci et al. 1987; Quas
et al. 2007). There is little doubt that people of all ages are susceptible to both mis-
information and misleading questions. As noted by Roebers and Schneider (2002),
this finding is consistent across situations, as indicated by considerable group sta-
bility across witnessed events. Roebers and Schneider (2002) examined the con-
sistency of 6-, 8-, and 10-year-olds’ eyewitness memories of two different videos.
Considerable group consistency in response patterns was found across the two
events. This was also true for the issue of suggestibility to misleading questions.
In particular, children from the youngest age group (i.e., the 6-year-olds) in most
Eyewitness Memory 117

cases complied with the interviewer and accepted most of the false suggestion in
both interview situations, thus showing high levels of suggestibility across the two
situations.
Although this has been a controversial area of research, most studies that have
looked for age differences in suggestibility have found them, with preschool chil-
dren being particularly susceptible to suggestion (see reviews by Bruck et al.
2006; Pipe and Salmon 2009). About 90 % of the studies that included develop-
mental comparisons reported significant age differences such that preschoolers
were found to be the most suggestible age group (Bruck and Ceci 1999; Ceci and
Bruck 1993). For instance, individual differences in preschoolers’ behavioral prob-
lems and inhibitory control have been shown to influence children’s suggestibil-
ity and their provision of inaccurate information (Schaaf et al. 2008). Moreover,
it has been shown in several studies that interviewer characteristics such as
interviewer bias substantially affect the accuracy of young children’s reports. For
instance, Thompson et al. (1997) conducted a study in which 5- and 7-year-olds
viewed a staged event that could be construed as either abusive or innocent. When
questioned by a neutral interviewer or by an interviewer whose interpretation was
consistent with the activity viewed by the child, children’s accounts were factu-
ally correct. However, when questioned by interviewers who had been given false
information about the event, children often made incorrect accounts consistent
with the biased interviewers’ scripts.
Age differences in suggestibility have also been confirmed in other studies that
used realistic and externally valid designs by simulating repeated questioning in a
legal context.
For instance, Bjorklund, Cassel, and their colleagues (Bjorklund et al. 1998;
Cassel and Bjorklund 1995; Cassel et al. 1996; Roebers and Schneider 2005)
simulated a prototypical eyewitness experience that spanned a 1-month period
beginning with a participant observing a videotaped event and ending with the
participant being asked to recall the matter in a trial. Children of different ages
and adults went through several questioning sessions, involving free recall and
leading questions suggesting either correct or incorrect “facts” in the same man-
ner in which attorneys might question someone who had witnessed a petty theft.
Finally, subjects were questioned about 4 weeks after they had observed the event
using both correctly leading and misleading questions in the same interview. As a
major result, Bjorklund, Cassel, and their colleagues reported age differences in
responses to repeated suggestive questioning, with kindergarten children follow-
ing misleading questions and changing their answers more often than older par-
ticipants (see Fig. 5.5). As shown in this figure, whereas participants of all ages
tended to change their answers when questioned about noncentral (peripheral)
items, only the kindergarten children showed this pattern for the central items.
However, on the final multiple-choice questions, the kindergarten children were
able to provide the correct answer as often as they had to the initial questions,
despite intervening errors.
The latter finding indicates that young children’s erroneous answers do not nec-
essarily reflect an actual change in memory representation. It is likely that some
118 5  The Development of Event and Autobiographical Memory …

100

Correct
80
Percentage of Responses

Incorrect

60

40

20

0
6-Year-Olds 8-Year-Olds Adults

Fig. 5.5  Percentage of correct and incorrect responses by age to misleading questions. (Based

on data from Cassel and Bjorklumd 1995)

of the young children’s compliance with misleading questions is related to the

specific context of the interview, in particular, the social demand characteristics
of the situation (Ceci and Bruck 1998; Reyna et al. 2002). For example, children
may interpret a repeated question as evidence that the (adult) interviewer was not
satisfied with their first answer and that a different answer must be more correct.
This is particularly likely in the studies conducted by Bjorklund and colleagues
discussed earlier in which the follow-up question was asked immediately after a
similar question. In fact, there is evidence that young children are less susceptible
to misinformation when given by a child rather than an adult authority figure and
that repeated questioning does not necessarily lead to “mind changing” and inac-
curate recall in young children (Ceci and Bruck 1993; Poole and White 1991).
However, as noted by Ceci and Bruck (1993), preschool children are dispropor-
tionately more vulnerable to suggestion than either schoolchildren or adults. If sug-
gestions are strong and persistent enough, young children may confuse the source
of the information, thinking that the misinformation was something they actually
experienced and not something they heard someone else say. In this context, an
interesting question is whether children’s memories of events actually change as a
consequence of misleading questions. There is indeed evidence that when sugges-
tions are strong and persistent, young children do not change their newfound inter-
pretations of previous events when interviewed again after a longer delay.
For instance, Leichtman and Ceci (1995) assessed the effects of negative stere-
otyping and suggestions on preschool children’s recollections of an event that hap-
pened at their school. An unfamiliar person, Sam Stone, visited the classroom for
Eyewitness Memory 119

a short time, talking to the teacher and listening to a story. Children in the “stereo-
type condition” were given misleading background information about Sam Stone
before his visit. Children in the “suggestion condition” were interviewed several
times after Sam Stone’s visit and given misinformation about the visit. Children
in the “combined condition” received both the negative stereotype and misinfor-
mation before the visit and were repeatedly interviewed thereafter. Finally, chil-
dren in the control condition received neither the stereotyped information nor the
misinformation about Sam. When children were given an open-ended interview
2 weeks after the visit, Leichtman and Ceci found that the highest levels of false
reports about Sam’s visit came from children who received both the stereotyped
information before the visit and the misinformation thereafter. Impressively, the
percentage of erroneous answers increased to more than 70 % for the younger
preschoolers when the children were asked specific follow-up questions concern-
ing whether Sam had ripped a book or got a teddy bear dirty. Obviously, many
children firmly believed that these events actually happened, even after being told
that the events were just made up. Thus, it seems that false memories of plausible
but extraordinary events are relatively easy to put in young children’s minds (see
Goodman and Clarke-Stewart 1991, for a similar example).
Although most studies reviewed so far have emphasized a particularly strong
impact of suggestive interview techniques on the accuracy of young children’s
reports, it seems important to point out that children older than 6 years of age and
adults are also suggestible across a wide range of events. However, the magnitude,
boundary conditions, and factors involved in suggestibility at these ages have not
been adequately examined so far (Bruck and Ceci 1999). Similarly, it should be
noted that there are pronounced individual differences in susceptibility to mislead-
ing information. Some preschoolers are very resistant to interviewers’ sugges-
tions, whereas some older children immediately follow the slightest suggestion.
Although we previously discussed several factors (e.g., IQ, compliance, knowl-
edge about the event, etc.) that might contribute to these differences, we are still a
long way from understanding the sources of these differences.
Why are younger children so susceptible to the effects of misinformation and
suggestion? Again, fuzzy-trace theory (Brainerd and Reyna 1998, 2005, 2014;
Reyna and Brainerd 1995) offers an explanation. According to fuzzy-trace theory,
young children are more reliant on verbatim (exact) memory traces than older
children. Because verbatim traces deteriorate rapidly, they may not be available
when post-event information is provided or when suggestive questions are asked.
Thus, erroneous information may be incorporated with “real” memories and may
thus become indistinguishable from them. In fact, research has shown that false
memories can be more resistant to forgetting than real memories (e.g., Brainerd
and Mojardin 1999; Brainerd and Reyna 2002). Therefore, it appears that under
certain circumstances, children can come to believe their implanted memories and
treat them as real. The issue of developmental trends in false memory creation is
an interesting one and will be taken up again in more detail below.
120 5  The Development of Event and Autobiographical Memory …

Source Memory Problems

The studies by Leichtman and Ceci (1995) and Goodman and Clarke-Stewart
(1991) clearly indicate that young children’s memories can be changed by expo-
sure to consistent misinformation about events, causing them to confuse the source
of the information, making them believe that the misinformation is actually some-
thing that they experienced and not something they heard someone else say (Ackil
and Zaragoza 1995).
The issue of source monitoring, which refers to the ability to specify con-
textual information surrounding a memory, has received considerable attention
during the past two decades. More often than not, people remember events that
really happened. Moreover, in most cases, they also remember whether the events
happened to them or somebody else. Other times, however, people confuse an
actual event with an imagined event or an event that happened to them with an
event that happened to others. That is, source monitoring is not perfect, even in
adults (Drummey and Newcombe 2002; Lindsay et al. 1991; Johnson et al. 1993).
Research has shown that young children often experience problems monitoring the
source of their memory. They sometimes have difficulty determining whether they
actually performed an act or just imagined it (Ackil and Zaragoza 1995; Foley
et al. 2010). Such findings have caused some researchers to propose that young
children’s increased susceptibility to suggestion may largely be due to problems
with monitoring the source of their knowledge. Some research supports these
speculations, showing that young schoolchildren who are poor at source monitor-
ing are more prone to the effects of suggestion (Ceci and Bruck 1995; Mazzoni
1998). Moreover, children who were given source-monitoring training made fewer
false statements to misleading questions (Thierry and Spence 2002).
There is evidence for developmental trends in source monitoring, even though
the findings are not consistent and the time course of developmental improvements
is not entirely clear. For instance, studies exploring children’s theory of mind
have found considerable source-monitoring difficulties in 3-year-olds, whereas 4-
and especially 5-year-olds are quite good at specifying how they learned some-
thing (Gopnik and Graf 1988; Perner 1991; Perner and Ruffman 1995). Similarly,
Welch-Ross (1995) reported good source memory for older preschoolers, and
Lindsay et al. (1991) found that 4-year-olds were surprisingly good at a source
task, having difficulty recalling the sources only when the sources were highly
similar to one another (see also Foley et al. 1987). Although these findings indicate
that children’s source monitoring seems to function well by the age of 4 years,
other research using different experimental procedures has not confirmed this mes-
sage. For instance, to chart developmental trends in source monitoring, Drummey
and Newcombe (2002) used a more difficult source task that they took from the
literature on source amnesia in the elderly and adapted it for 4-, 6-, and 8-year-old
children. They found that children’s source-monitoring abilities seemed to undergo
a dramatic shift between the ages of 4 and 6 years. Drummey and Newcombe also
found a substantial correlation between changes in source monitoring and a meas-
ure of prefrontal functioning.
Eyewitness Memory 121

Developmental differences in children’s source monitoring were also reported

by Mazzoni (1998), who investigated the impact of source monitoring and self-
efficacy on 6- and 9-year-old children’s suggestibility. The children witnessed an
event in their classroom and were questioned about it 1 week later. Participants’
source-monitoring ability was assessed by two separate tasks. Individual differ-
ences in source-monitoring ability were found to play a role in suggestibility for
the younger but not for the older children, whereas self-efficacy predicted suggest-
ibility in older children only. Further evidence supporting the assumption of devel-
opmental trends in source monitoring after the age of 4 years comes from a study
by de Chastelaine et al. (2007) who investigated improvements in source memory
performance in children, adolescents, and adults by using event-related potentials
(ERPs). Source memory improved with age, and the authors inferred from their
findings that age-related changes were due to the development of attentional con-
trol mechanisms that facilitated the accurate selection of task-relevant memories.
Taken together, the findings from the studies summarized above confirm the
assumption that children’s source-monitoring abilities are not well developed
before the age of 6 but improve steadily thereafter, probably affected by the matu-
ration of the prefrontal cortex and the development of control functions. There is
little doubt that the poor source-monitoring abilities found in preschool children
do contribute to their suggestibility in eyewitness investigations.

Developmental Trends in False Memory Creation

How easy is it to create a false memory in a person, that is, to get children or adults
to believe an event happened to them when it never really did? Research conducted
by Elisabeth Loftus and colleagues has demonstrated that false memories from
childhood can easily be planted in the mind of an adult. For instance, Loftus and
Pickrell (1995) presented college students with four scenarios of specific childhood
events that were based on reports of parents and older siblings. One event, being
lost in mall at age 5, never actually occurred. The participants were asked to write
as much about each event as they could remember and were later interviewed about
the events. They accurately remembered, on average, almost 70 % of the true events.
However, they also “recalled” about 25 % of the false events, sometimes vividly.
False memory creation using similar procedures has also been demonstrated in
preschool children (Ceci et al. 1994). These authors interviewed children throughout
an 11-week period about events that might have happened to them. For example,
children were asked whether they remembered ever getting their fingers caught in
a mousetrap. Figure 5.6 provides the percentage of false reports across the sessions
separately for 3- to 4-year-olds and 5- to 6-year-olds. Although few children admit-
ted to experiencing these false events in the early interview sessions, the percentage
of false assents increased as a function of session. By the end of the study, more than
50 % of the younger preschoolers and about 40 % of the older age group assented
that these events happened, indicating that preschool children are even more suscep-
tible to creating false memories than adults.
122 5  The Development of Event and Autobiographical Memory …

60

3- to 4-year-olds
50
Percentage of False Assents

5- to 6-year-olds

40

30

20

10

0
1 2 3 4 5 6 7 8 9 10
Sessions

Fig. 5.6  Percentage of false reports over session for 3- to 4-year-olds and 5- to 6-year-olds.

(Based on Ceci et al. 1994; adapted from Bjorklund 2012, p. 338)

Such “implanted memories” as well as other memory errors may depend in

part on the salience and plausibility of the suggested events (Ghetti and Alexander
2004; Pezdek et al. 1997). For example, Pezdek and Hodge (1999) found that a
plausible false event such as having been lost in a shopping mall was more likely
to be falsely acknowledged by both children and adults as a real memory than was
an implausible episode about receiving a rectal enema: Whereas about 33 % of the
5- to 7-year-old children “remembered” being lost in a mall, only about 5 % of the
children “remembered” being given the enema.
Another rather new series of studies conducted by Gabrielle Principe and her
colleagues have examined how rumors may influence young children’s memory
of events. This research was stimulated by the idea that suggestive factors outside
the courtroom and the formal interview context could also contaminate children’s
memory. For instance, memories for specific events may be affected by sharing
them with peers or family members. In a first study exploring the issue of mne-
monic effects of memory sharing, Principe and Ceci (2002) let a group of 3- to
5-year-olds participate in a staged event (an archeology dig). Only one-third of the
children in the witness condition observed the archeologist ruin the two target arti-
facts (“target activities”), whereas another third of the children in the classroom
condition did not witness the target activities but were the classmates of the chil-
dren in the witness condition. The last third of the children in the control condition
did not see the target activities and also had no chance to interact with the other
Eyewitness Memory 123

children. Principe and Ceci (2002) asked children who recalled events seen only
by their classmates whether they themselves had seen these activities take place
or had merely heard about them from someone else. Half of the children received
neutral interviews, whereas the remaining half were questioned in a suggestive
manner. Interestingly, children in the classmate condition who did not witness the
target activities also reported these activities. In particular, those children in the
classmate condition who were exposed to the suggestive interviews were more
likely than the children in the control condition to report having actually seen
these activities. This finding suggests that natural conversations with cowitnesses
not only may induce false reports but may also lead to source confusion.
Subsequent research (e.g., Principe et al. 2006, 2010) using a slightly different
experimental paradigm has confirmed these findings, again showing that rumors
can leak into memory. Principe et al. (2006) had some preschoolers in the same
classroom overhear an adult allege a fictitious rumor that a certain event occurred
in their school. A second group of children did not overhear the adult, but instead
interacted freely with their classmates who were exposed to the rumor. A third
group who had no contact with the two other groups actually experienced the
event. When questioned later, the children exposed to the rumor, either directly
from the adult or secondhand from their classmates, were as likely to report expe-
riencing the rumored event as the children who actually experienced it. Children
in the rumor conditions who said they saw the event provided more vivid descrip-
tions of the (non)event than children who said they did not see the event, suggest-
ing that the rumor produced a true false memory, complete with perceptual details.
In a subsequent study, when some children were warned that a false rumor was
going around, this reduced the number of false reports of rumored events in the
5- to 6-year-olds, but did not affect the 3- to 4-year-old children’s reports. The
research conducted by Principe and colleagues also indicates that children are
especially likely to embellish information picked up by peers, showing the impor-
tant role of peers as a source of errors and that memory-sharing situations can be
a potent source of children’s report contamination (for more details, see Principe
and Schindewolf 2012).
One final interesting aspect of this research concerns the impact of misinformed
parents on children’s memory. Principe et al. (2012) had children watch a specific
event (a magic show) and then gave the children’s mothers a letter that asked them
to discuss the show naturally with their children. Mothers in the suggestion let-
ter condition received misinformation, whereas those in the neutral letter condition
did not receive the false suggestion. Results indicated that the letter was a potent
form of suggestion. The suggestive letter children were more likely to wrongly
report the suggested event, with about 60 % of the children claiming that they saw
the rumored event. Interestingly, the influence of the misinformation manipula-
tion varied as a function of maternal reminiscing style. In line with the prediction
from the studies on autobiographical memory described above, children of high-
elaborative mothers provided more detailed narratives of the actually experienced
portions of the magic show than children of low-elaborative mothers. However,
children with high-elaborative mothers were two times more likely than those with
124 5  The Development of Event and Autobiographical Memory …

low-elaborative mothers to wrongly report the rumored event. Children with high-
elaborative mothers also provided more narrative detail than those with low-elab-
orative mothers. These findings demonstrate that the highly elaborate style usually
associated with more skilled autobiographical remembering is also linked with
increases in errors in children’s memory when mothers are exposed to misinfor-
mation about their children’s experiences (Principe and Schindewolf 2012).
Overall, the picture that emerged from developmental research on children’s
false memories during the past two decades has been quite consistent. That is,
false memories have been found to decrease with age throughout childhood and
adolescence (e.g., Bruck and Ceci 1999; Ceci and Bruck 1993). Age declines have
been reported in numerous memory suggestion experiments that were intended
to evaluate the effects of manipulative interviewing practices (Ceci et al. 2007;
Marche 1999). During the past decade, however, several published experiments
have tested fuzzy-trace theory’s prediction that some of the most powerful forms
of false memory in adults are considerably attenuated in children. Brainerd and
Reyna (2007, 2014; Brainerd et al. 2008) have developed an extensive theory-
based research program directed at questions about the origins and development
of false memories, for both memories that are spontaneously produced and those
that were implanted by suggestive questioning. On the basis of FTT, together with
the assumption that the knowledge base becomes increasingly rich and intercon-
nected in later childhood and adolescence, they predicted that the occurrence of
false memories would increase during the course of childhood and adolescence
for materials that make it easy to form semantic connections among learning mate-
rials. In fact, in reviewing many experiments, they found that in some common
domains of experience in which false memories were rooted in semantic/meaning
connections among word lists or events, age increases in false memory were the
rule and not the exception.
For instance, Brainerd et al. (2008) reviewed 17 articles that contained 26
experiments in which false recall or false recognition was measured using seman-
tically related materials (Deese–Roediger–McDermott = DRM lists) in children
and adults who ranged in age from 3 years to their early 20s. Almost all of these
experiments provided impressive evidence of age increases in intrusions and
false alarm rates. Similar evidence of developmental reversals has been found
in research on memory suggestion. Brainerd et al. (2008) concluded from their
review that such developmental reversals occur when memory errors are rooted
in the formation of meaning connections between events, and when it is difficult
to use verbatim memory for actual events to suppress those errors. Moreover, this
effect seems particularly pronounced for verbal materials but may not generalize
to pictorial stimuli. Howe (2006, 2008a) confirmed the developmental reversal
effect for semantically related word lists by finding that the false recall of unpre-
sented category exemplars more than doubled between the ages of 5 and 11. When
lists were presented as pictures, however, this age trend disappeared: False recall
levels for the pictures did not differ as a function of age. The question of why pic-
torial presentation has such a powerful suppressive effect on age increases in false
memory awaits further investigation.
Eyewitness Memory 125

Overall, this relatively new research program on false memory development

leaves no doubt that younger children are less susceptible to false memory illu-
sions than older children and adults, and in particular, when experimental manipu-
lations are designed to draw attention to semantic relations in the list structure.
Fuzzy-trace theory (FTT) anticipates many of these findings and also provides
a good fit to the available data. False memory rates are reduced by the presence
of verbatim traces and increased by the presence of gist traces. Although young
children are capable of constructing both types of traces, the ability to construct
traces that connect meaning (gist) across list items improves with age, which also
increases the susceptibility to false memories.
According to Howe (2008b), what is lacking in FTT is a precise description
of the gist-connecting mechanism. One basic assumption of FTT is that connect-
ing meaning (gist) relations across multiple exemplars of the same category during
encoding is a key to false (and probably true) memory development. This notion
is hard to tie down conceptually and runs counter to findings that even young
schoolchildren encode semantic relations, even though they may not use them dur-
ing retrieval. More research is needed to evaluate the claims of FTT against the
alternative account offered by Howe (2008b); that is, associative-activation theory
(AAT). This theory relies on a direct activation mechanism rather than an explicit
gist-connecting one. The activation mechanism is assumed to change with corre-
sponding changes in children’s experience, practice, and knowledge base, changes
that lead to faster and more efficient activation of concepts and their semantic rela-
tions in memory. Many of the same false memory phenomena can be explained
by AAT and FTT. As noted by Brainerd et al. (2008), Ghetti (2008), and Ornstein
and Light (2010), there is a need for research designed to provide empirical evi-
dence that directly links independent measures of gist or meaning extraction to the
occurrence and magnitude of false memory.

Familiarity and Recollection

In recent years, there has been increased interest in dual-process retrieval mod-
els (see Ghetti and Lee 2014; Jacoby 1991; Malmberg 2008; Yonelinas 2002, for
reviews) that propose distinct components in recognition memory. The discrimi-
nation of true from false memories is a topic central to most contemporary dual-
process models of memory. These models propose two processes—recollection
and familiarity—that underlie both recall and recognition. In most dual-process
models, recollection is defined as the consciously controlled intentional use of
memory that allows for the retrieval of qualitative details of a past event. The pro-
cess of recollection is frequently associated with the subjective experience of vivid
remembering and involves remembering particular aspects of a prior episode such
as perceptual details or the source of information. Familiarity, by contrast, usually
refers to experiences of prior events that may arise from activated semantic rep-
resentations (e.g., as present in the DMR tasks described above). The familiarity
126 5  The Development of Event and Autobiographical Memory …

process is considered to be relatively automatic, less effortful, and less depend-

ent on intention. Although dual-process models differ in many respects, they typi-
cally characterize recollection as a deliberate, attention-demanding, and rather
slow process, whereas familiarity is assumed to be a relatively automatic pro-
cess that is recruited more rapidly. For instance, fuzzy-trace theory distinguishes
between identity and similarity processes in memory judgment, thus reflecting the
recollection–familiarity difference (Reyna and Brainerd 1995). Research in adult
populations has provided convincing behavioral and neurological support for this
dual-process distinction (e.g., Yonelinas 2002).
The relative contribution of recollection and familiarity to memory perfor-
mance differs from task to task. For instance, in recognition tasks, familiarity
plays a central role even though recollection may contribute to some extent. By
contrast, memory tasks that require participants to recall details about the con-
text in which the event originally occurred rely heavily on recollection. Again,
however, certain encoding conditions that encourage the processing of item and
context information may result in an increase in the contribution of familiarity to
source decisions (Ghetti and Lee 2014). One of the advantages of dual-process
models in developmental studies is that they allow for testing the possibility that
the processes underlying suggestibility may change even under conditions in
which the overall magnitude of false memory effects undergoes little age-related
change (Reyna et al. 2002).
For instance, in a series of experimental studies, Holliday and Hayes (2001,
2002) presented 5- to 9-year-old children with picture stories followed by mis-
leading post-event details that were either read to the children or generated by the
children in response to misleading cues provided by the experimenter. There was
only minimal evidence of age-related changes in children’s suggestible responses
across a wide range of recognition tests. However, subsequent process dissociation
analyses revealed that automatic processing (familiarity) contributed to the accept-
ance of misinformation in children regardless of whether the misinformation was
read or generated. However, there was a particularly strong misinformation effect
in the youngest children when the misinformation was generated, indicating that
younger children are more likely than older children to accept suggested items
automatically—according to their familiarity.
As a direction for future research, Brainerd et al. (2008) suggested that
researchers should examine changes in the relative degrees to which recollec-
tion and familiarity contribute to false memories as children develop. So far, the
number of studies that have used estimation methods to examine the development
of recollection and familiarity in childhood is still relatively small (Brainerd and
Reyna 2014; Ghetti and Lee 2014; Ornstein and Light 2010). Using a variety of
materials and techniques, these studies have presented a fairly consistent view
despite their methodological differences. There is plenty of evidence that both
recollection and familiarity develop during early childhood. However, whereas
all studies report reliable increments with increasing age for recollection in child-
hood, the pattern for familiarity is not as consistent. Overall, there is evidence that
familiarity seems to stabilize by middle childhood, whereas recollection continues
Eyewitness Memory 127

to improve into adolescence (Brainerd et al. 2004; Ghetti and Angelini 2008; Ofen
et al. 2007). These general trends are observed not only in accurate memories but
also in false memories (Holliday 2003). Growth in recollection seems to outpace
changes in familiarity, but increases in the degree to which gist is relied on may be
substantial under some conditions (see Brainerd et al. 2004).
An interesting issue raised by Ghetti (2008) is whether the observed age-related
increases in false memory are driven primarily by increases in the selected age
range rather than by a gradual linear increase across all age groups. For instance,
the literature based on the DRM paradigm reviewed by Ghetti seems to indi-
cate that the age effects observed in false memory formation were driven by the
youngest children in the various samples (mostly preschoolers), whereas no dif-
ferences were observed between groups of younger and older schoolchildren. In
other research, only adults exhibited significantly higher rates of false memories
than children did, but no age differences were found within the sample of chil-
dren. According to Ghetti, it is possible that the development of gist processing is
steeper between early and middle childhood or between late childhood and adult-
hood than it is during middle childhood. Clearly, there is a need for more research
on this issue and the development of an overarching model that represents all pro-
cesses that underlie false memory formation.
In sum, as noted by Ghetti and Lee (2014), a relatively small amount of lit-
erature on the development of recollection and familiarity during childhood has
converged to show distinct developmental trajectories. More research is needed to
provide a full picture of factors that might affect the development of these pro-
cesses. The review of developmental cognitive neuroscience studies provided
by Ghetti and Lee indicates that this just emerging field has already yielded new
interesting insights. For instance, this research has begun to show that there is
considerable improvement during middle childhood in not only the prefrontal-
cortex-mediated strategic mechanisms but also the medial-temporal-lobe-mediated
mechanisms that support the binding and contextual processing that are critical for
recollection. There is reason to assume—and future research may show—that the
same might be true for mechanisms supporting familiarity.

Facilitating Young Children’s Performance in Eyewitness

Situations

As shown above, different interviewing procedures may induce false memories.

Interviewing techniques that encourage witnesses to make sense of their experi-
ences or to adopt a more meaningful or less verbatim criterion are apt to increase
gist-based false recognition and recall (Brainerd and Reyna 1998). Repeatedly
cuing meaning or inferences strengthen these gist representations. The research
on false memory creation summarized above has convincingly shown that external
suggestions (e.g., as in misinformation paradigms) can effectively insinuate them-
selves into the autobiographical memories of individuals.
128 5  The Development of Event and Autobiographical Memory …

Given the various memory problems of young children in eyewitness situations,

researchers are still seeking to uncover the optimal ways by which to elicit accurate
eyewitness memory from these children. Not surprisingly, how children are inter-
viewed can greatly affect their recollections (Bruck et al. 2006; Ceci et al. 2000).
The type of questions asked influences what children remember. As noted above,
young children tend to recall relatively little information when asked open-ended
free-recall questions, but what they recall tends to be highly accurate. Meanwhile,
promising interview procedures such as the cognitive interview (CI; McCauley and
Fisher 1995; Saywitz et al. 1992) have been developed. Such procedures enhance
accurate memory reports and simultaneously decrease inaccurate reports. As shown
by Saywitz and colleagues (1992), the CI significantly increased the number of cor-
rect facts that 7- to 11-year-old children reported compared with a standard inter-
view condition. Also, simply asking children to mention everything they have seen
and heard seems sufficient to increase levels of accurate recall in young children
(Poole and Lindsay 1995; Elischberger and Roebers 2001). Moreover, an interper-
sonally warm interview atmosphere increases children’s memory and the accuracy
of their memories (Goodman et al. 1991). The more open-ended the questions, the
less likely young children are to report incidents that did not occur. Also, accu-
rate responses are most likely when the interviewer is neutral in tone and does not
attempt to mislead the child with suggestive questioning.
In addition to verbal prompts, a number of other techniques and training proce-
dures have been explored to help children remember and report their experiences. For
instance, various nonverbal techniques have been used in forensic contexts (Saywitz
and Snyder 1993). Positive effects on memory have been found for techniques that
include revisiting the environmental context in which the event occurred, and provid-
ing children with items or representations of items associated with the to-be-remem-
bered event. Finally, “mental context reinstatement,” a component of the CI described
above, has been effective with both children and adults (for more information, see
Pipe and Salmon 2009). Taken together, the numerous studies on children’s eyewit-
ness memory have not only led to important theoretical advances in our understand-
ing of children’s autobiographical memory but have also demonstrated their important
practical potential for guiding interview procedures in the relevant applied settings.

Summary

This chapter has shown that our understanding of the development of autobiographi-
cal memory has improved considerably during the past three decades. Event mem-
ory in general, and more specifically eyewitness memory, has been one of the most
investigated areas in cognitive development lately. When Schneider and Pressley
(1989) summarized the state of the art, research on children’s event memory was
just beginning, with studies conducted by Jean Mandler, Katherine Nelson, and their
colleagues providing the first insights into the importance of script knowledge for
young children’s general representation of events. This early research stimulated
Eyewitness Memory 129

hundreds of studies on the development of autobiographical memory. These stud-

ies have not only illustrated that even young children can memorize salient events
for a long time, but they have also identified important sources of development and
individual differences. For instance, language development and maternal reminisc-
ing style as well as indicators of socioemotional development have been found
to be highly relevant in this respect. The hypothesis that children learn to narrate
through dialogue experiences has been confirmed in numerous studies (see reviews
by Fivush 2014; Reese 2014). This impressive research program has also helped in
evaluating traditional perspectives and in producing new hypotheses concerning the
infantile amnesia phenomenon, which has been discussed in the literature for more
than 100 years. Although we are still far from completely understanding this phe-
nomenon, there is reason to assume that several factors have significantly contrib-
uted to offsetting it, such as the development of the cognitive self and the beginning
of verbal mother–child dialogues about children’s experiences.
Recent research on the development of autobiographical memory has also
enriched our understanding of young children’s problems with testifying in legal
contexts. Overall, the findings of recent eyewitness memory research have con-
firmed several findings from the early period (e.g., Whipple 1911) such as age
differences in the recall of events and the particularly large effects of misleading
questions and repeated questioning on young children’s memory reports. On the
other hand, however, recent studies have not supported the assumption held by the
early eyewitness memory researchers that young children are generally unreliable
witnesses. Although the available literature indicates that young children’s eye-
witness memory is typically less accurate than that of older children and adults,
research findings have also shown that, in general, young children are capable of
providing accurate eyewitness memory about personally significant events in their
lives. One merit of contemporary research is that it has identified the conditions
that increase the probability of obtaining accurate memory reports from young
children. For instance, new and promising interview techniques that avoid the
problems of previous interrogations and provide the conditions for optimal recall,
particularly in young children, have been developed. Adults interviewing young
children need to establish good rapports, use a neutral tone, and avoid social
demand characteristics as well as leading questioning (Pipe and Salmon 2009).
An interesting aspect of recent research on eyewitness memory is that false mem-
ories actually increase as a function of age under specific circumstances (i.e., the
use of semantically related materials). Dual-process theories such as FTT provide
a convincing explanation for this developmental reversal, pointing to the relevance
of recollection and familiarity processes and their developmental changes. As noted
by Ghetti and Lee (2014), the developmental cognitive neuroscience of recollec-
tion and familiarity is just emerging, indicating that developmental changes in these
processes might occur across multiple cortical and subcortical regions. There is rea-
son to assume that future behavioral and neuropsychological research will further
improve our knowledge about the emotional, social, and cognitive factors that influ-
ence children’s reports and event memories and our understanding about which fac-
tors affect the accuracy of memory in childhood and adolescence.
Chapter 6
The Role of Basic Memory Capacities
and Working Memory

How does modern research on memory development differ from the historical
approaches outlined in Chap. 2? As has been shown in the preceding chapters, one
of the crucial differences between classic and modern approaches concerns a shift
from an emphasis on describing developmental differences in memory perfor-
mance to an emphasis on identifying the underlying mechanisms of developmental
differences and developmental change. Another difference concerns the theoretical
framework used to describe and explain memory development, that is, a shift from
very general theoretical assumptions to testable models.
Since the mid-1960s, research on memory development has been strongly
influenced by theoretical models derived from information processing and neu-
roscience approaches (see the reviews by Bauer 2006; Kail 1990; Schneider and
Bjorklund 1998, 2003; Schneider and Pressley 1997). Developmental psycholo-
gists began looking at changes in children’s thinking in terms of a computer meta-
phor. From this perspective, memory development can be seen as reflecting either
hardware (the capacity of memory systems and the speed at which information
can be processed) or software (e.g., use of strategies). As already shown above,
developmental research on memory has been strongly influenced by multistore
memory models developed in cognitive psychology that distinguished between a
sensory register (SR), a short-term store (STS), and a long-term store (LTS; see
Atkinson and Shiffrin 1968). Please note that multistore conceptions of memory
have been supported by research carried out in biology, psychopathology, and neu-
roscience (Bauer 2007; Squire 1999). For instance, research exploring short-term
memory (STM) and long-term memory (LTM) processes in infancy has provided
strong evidence for such a view (see Chap. 3).
Although research on infancy (see Chap. 3) and research on autobiographi-
cal memory (see Chap. 5) have accumulated over the past few decades, the vast
majority of experimental studies on memory development between the mid-1960s
and the late 1980s have been carried out on elementary schoolchildren, adoles-
cents, and adults and have mainly dealt with explicit memory (i.e., the conscious

© Springer International Publishing Switzerland 2015 131

W. Schneider, Memory Development from Early Childhood Through
Emerging Adulthood, DOI 10.1007/978-3-319-09611-7_6
132 6  The Role of Basic Memory Capacities and Working Memory

remembering of facts and events). It was repeatedly found that substantial

improvements in declarative memory could be observed for the age range between
6 and 12 years, which roughly corresponds to the elementary school period in
most countries. In order to explain these rapid increases in memory performance
over time, different sources or determinants of memory development have been
identified. According to most contemporary memory researchers, changes in basic
memory capacities, memory strategies, metacognitive knowledge, and domain
knowledge all contribute to developmental changes in memory performance.
There is also broad agreement that some of these sources of development contrib-
ute more than others and that some play an important role in certain periods of
childhood but not in others (see Schneider and Pressley 1997; Siegler 1998).
This chapter deals with the development of basic memory capacity and its
contribution to the development of memory skills. I will use this term to refer to an
individual’s limited resources or capacity for holding a small amount of informa-
tion in mind in an active available state and to temporarily manipulate this infor-
mation if necessary. The term thus comprises the concepts of short-term memory
capacity and working memory capacity. The role of memory capacity in the devel-
opment of memory behavior and performance has been a controversial point of
discussion in the literature. As shown in Chap. 2, the development of short-term
memory has been a major issue since the very beginning of scientific memory
research, and its special role in the storage and consolidation of incoming informa-
tion has been highlighted by several theoretical models. Although there is no sin-
gle information processing theory of cognitive development (for a recent review,
see Halford and Andrews 2011), most approaches are built on the assumption
that incoming information moves serially through a system of connected memory
stores when individuals try to store, consolidate, and eventually retrieve this input.
The idea of a limited capacity is central to information processing approaches;
this means that we can deal with only a few pieces of information at any single
point in time. In particular, the STS is assumed to have a small capacity for storing
information for a short while, that is, for current attention. If nothing is done to the
information in the STS, it will be lost.
From a developmental perspective, one of the most interesting issues debated
in the literature concerns the question of whether STS capacity increases with
age. Since the 1970s, numerous studies have addressed this topic. In the present
chapter, evidence concerning age-related changes in short-term memory capacity
and working memory capacity will be examined, and the possible implications of
findings from theories of memory development will be discussed in more detail.
Please note that the term “short-term memory” will not be used interchangeably
with the term “working memory.” Although working memory (WM) is not com-
pletely distinct from short-term memory, it differs from the latter in that it refers
not only to the temporary maintenance of information in a memory store but also
to additional activities related to this information. Whereas short-term memory
involves only the storage and reproduction of information, working memory
also includes the capacity to transform the information that is being held in the
STS. For instance, one relies on working memory when the task is to memorize
The Role of Basic Memory Capacities and Working Memory 133

the information kept in the STS in a different order (as in the backward memory
span task) or when the task is to retain partial results while solving an arithmetic
problem without paper. There is empirical evidence that short-term and working
memories reflect separate but highly related constructs (e.g., Engle et al. 1999).
The two constructs are also linked to different neural subsystems in that WM
relies more on the dorsolateral prefrontal cortex (PFC), whereas maintaining infor-
mation in mind and not manipulating it (STS) does not need the involvement of
the dorsolateral cortex. Imaging studies have shown frontal activation in the vent-
rolateral PFC only when memory maintenance was the goal of the task (Diamond
2013). Since the mid-1970s, the term “working memory” has been more dominant
in the field after Baddeley and Hitch (1974) came up with a general model that
demonstrated that a single module could not account for all kinds of temporary
memory. Thus, the findings concerning “short-term memory” and “working mem-
ory” will be discussed separately below.

The Development of Short-Term Memory Capacity

One of the oldest and most controversial issues concerning children’s information
processing is whether the amount of information they can actively process at one
time changes with age. Age differences in the capacity of the STS were typically
found in developmental studies that used memory span tasks. Memory span tasks
have simple structures and provide reliable data. Such tasks require participants
to repeat, in the exact order, a series of rapidly presented items such as digits or
words. Participants are usually presented sequences of stimuli at a constant speed
(e.g., one item per second). Short sequences (e.g., three items) are used first, with
subsequent presentations increased by one item at a time until the participant can
no longer reproduce the entire sequence. A child’s memory span is defined as the
longest sequence recalled correctly and is considered to reflect the size of the STS.
From the very beginning, developmental studies on children’s memory span have
shown that developmental improvement on these tasks was determined in part by
the type of material being memorized, with the slopes of developmental memory
span curves varying as a function of item type (e.g., digits, letters, or words).
Figure  6.1 illustrates developmental differences as a function of type of learning
material. According to Dempster (1978), the differences in slopes may reflect dif-
ferences in item encoding facility. As shown in Fig. 6.1, memory span increases
steadily with age, with the comparably largest memory spans reported for the digit
span measure. Obviously, the digit span of 5-year-olds (about four items) dou-
bles that of 2-year-olds (about two items). Although the average memory span of
9-year-olds (six items) is considerably higher than that of the 5-year-olds, there
is not much increase in digit span until adulthood, with the digit span of adults
averaging around seven items, which is in accordance with the “magic” number
7 (+ and −2) reported by Miller (1956) as the maximum number of slots that peo-
ple can hold in STS.
134 6  The Role of Basic Memory Capacities and Working Memory

6
Number of Items
5

3
Word Span
2
Letter Span
1 Digit Span
0

Age

Fig. 6.1  Developmental differences in digit span, letter, span, and word span. (Adapted from
Dempster 1981)

The pattern of developmental changes observed for more complex memory

span tasks indicates that increases can be observed until late adolescence. For
instance, longitudinal work carried out in the Munich Longitudinal Study on
the Genesis of Individual Competencies (LOGIC; Weinert and Schneider 1999;
Schneider and Bullock 2009) using the same listening span (sentence span)
measure for the same participants from age 4 to age 23 showed continuous span
increases until the age of 17 but no increases thereafter (Schneider et al. 2009). In
this STS version of the listening span task, participants first had to listen to a set
of sentences read to them by the experimenter and then had to recall each sentence
verbatim. The test was constructed with 75 sentences of three to seven words in
length. There were 25 sets of sentences in all, including five sets of 1 to 5 sen-
tences. The prerecorded sentences in each set were presented to the participants as
a playback from tape. For sets containing more than one sentence, the sentences
were read in quick succession. At the end of each set, participants heard a tone that
was a cue to recall each sentence in the set verbatim. Participants began with the
five 1-sentence sets and were then presented with the five 2-sentence sets and so
forth. Testing was terminated when the participants failed all five sentences in a set
at a particular level. Participants were given credit for correctly recalling a set as
long as they repeated each sentence verbatim, regardless of the order in which they
repeated the sentences. The total number of correct sentences and the total number
of words reproduced in the correct order were used as dependent variables.
Over the course of the longitudinal study, sentence span sum scores varied
between two and 56 (i.e., number of sentences correctly reproduced across trials).
Sentence span did not improve between the ages of 4 and 5 but increased signifi-
cantly between adjacent measurement points until adolescence (i.e., age 17). There
was no significant difference in sentence span found for the last two measurement
The Development of Short-Term Memory Capacity 135

points (ages 17 and 23), indicating that a peak in verbal memory capacity was
reached by late adolescence. As shown in Fig. 6.2, particularly pronounced
increases in sentence span were observed between the ages of 7 and 8 and the ages
of 12 and 17, respectively. The number of sentences remembered correctly was
about 15 during the preschool years and about 25 around the end of the elementary
school years and went up to more than 35 during the adolescence years.
Do these findings indeed indicate that basic memory capacity increases stead-
ily across childhood and adolescence? In a seminal paper, Chi (1976) evaluated
this assertion. The results of her analyses did not provide much support for such
an assumption. The available evidence suggests that the parameters characterizing
short-term memory such as capacity and rate of information loss are comparable
across different age groups. On the other hand, however, substantial age-related
differences were observed in the use and acquisition of control processes such as
rehearsal and grouping. Chi concluded from her analysis that what appears to be a
short-term memory capacity limitation in children is actually a deficit in process-
ing strategies as well as in processing speed.
Although age differences in memory span have been found to be stable in
numerous studies, we still need to ask whether this is mainly due to developmen-
tal differences in strategy use and information processing speed as Chi concluded.
In an extensive review of the literature, Dempster (1981, 1985) explored whether
strategic variables such as rehearsal, grouping, and chunking as well as nonstra-
tegic variables such as item identification speed and item ordering could explain
developmental differences in memory span. His critical evaluation of the available
evidence lets him to conclude that strategic variables are not a major source of
memory span differences even though effects of rehearsal, grouping, and chunking

45
40
35
30

25
M

20
Younger
15
Older
10
5
0
Total nr. of Total nr. of Total nr. of Total nr. of Total nr. of Total nr. of
Correct Correct correct Correct Correct Correct
Sentences Sentences Sentences Sentences Sentences Sentences
W3 W4 W5 W7 W10 WE11

Fig. 6.2  Development of sentence span, as a function of age group (Schneider et al. 2009, p. 74,
Fig. 2)
136 6  The Role of Basic Memory Capacities and Working Memory

on performance in memory span tasks could not be ruled out. On the other hand,
he inferred from the findings that one nonstrategic variable, namely item identifi-
cation speed, is a major and the only really relevant source of span differences. A
closer inspection of this review article reveals that Dempster was a very cautious
and conservative evaluator of research outcomes with a tendency to play down the
relevance of strategic and nonstrategic variables whenever the database was incon-
sistent. In the following, the impact of these variables on developmental differ-
ences in memory span will be briefly reconsidered by using a more comprehensive
database. The role of information processing speed will be given special attention,
given that it was the only variable that was shown to explain developmental differ-
ences in short-term memory capacity in Dempster’s analysis.

The Impact of Strategic Variables on Memory

Span Development

Age differences in digit span, for which marked developmental improvement

occurs (see above), have been hypothesized to be due to developmental increases
in the spontaneous use of rehearsal. Early developmental studies provided some
evidence for such a conclusion but could not rule out alternative interpretations.
For instance, Frank and Rabinovitch (1974) presented 8-, 10-, and 12-year-old
children with both a digit span task and a running memory span task. The lat-
ter differs from the conventional memory span task in that sequences of variable
and unspecified supra-span length are presented and that participants are told to
recall as many items from the end of the list as possible. Given that participants
are unable to predict the end of the list, rehearsal does not seem to be a reasonable
strategy. In fact, Frank and Rabinovitch found that the age differences obtained on
the running memory span task were about two-thirds smaller than those obtained
on the conventional memory span task, supporting the assumption that active
rehearsal may be a source of developmental differences. However, the reduced
age differences on this task can just as readily be attributed to the use of alter-
native strategies such as chunking in conventional memory span tasks (Dempster
1978, 1981).
Subsequent analyses have been slightly more supportive of the conclusion that
rehearsal while learning lists of digits is more common at the end of the grade-
school years than among younger children. For instance, when Hitch et al. (1991)
had 5-year-olds label pictures in a memory span task, their performance improved,
suggesting that they did not verbally label to-be-remembered items as they were
presented. In contrast, when 11-year-olds labeled the items, their short-term mem-
ory performance worsened, suggesting that the labeling interfered with covert
multiple repetition rehearsal processes. In summary, there is some evidence that
increases in rehearsal with increasing age can account at least in part for devel-
opmental differences in memory span (but see Jarrold and Hall 2013, for a more
skeptical view).
The Development of Short-Term Memory Capacity 137

Another strategy hypothesis was that active grouping and regrouping of the to-
be-learned materials would mediate developmental differences in memory span.
This hypothesis followed from the observation that adults spontaneously group
items given in a memory span task and that adults generally remember more items
on lists that contain temporal or rhythmic groupings (Dempster 1981). The effect
of grouping on memory span was studied frequently in the 1970s and 1980s (for
reviews, see Dempster 1981, 1985; Schneider and Pressley 1997). Unfortunately,
the studies that investigated the effects of experimenter-imposed grouping proce-
dures on memory span did not provide clear-cut results. Although in some studies,
the predicted interactions between age and grouping variables were found, indicat-
ing spontaneous grouping in older but not in younger children, these studies were
plagued by methodological problems such as ceiling effects for older participants
and floor effects for younger children. As the data were consistently mid-range,
there was little evidence of a developmental interaction (e.g., Engle and Marshall
1983; Halford et al. 1985; Samuel 1978). More recent studies (e.g., Bailey et al.
2009; Dunlosky and Kane 2007) have replicated earlier findings, demonstrating
that age difference in effective strategy use (i.e., grouping) could not explain the
age-related variance in span performance and thus confirming the view that there
is little support for the grouping hypothesis.
A more promising hypothesis concerns the impact of chunking on the develop-
ment of memory span. Chunking refers to the recoding of two or more nominally
independent items of information into a single familiar unit. This process depends
on knowledge of the items and is thus a knowledge-specific strategy (Chi 1978).
From this perspective, age differences in memory span are presumed to be due
to larger information sequences being encoded as chunks with increasing age,
permitted by the expansion and increasing conceptualization of knowledge with
development. In one of the first studies to test the chunking hypothesis, Dempster
(1978) compared the short-term memory performances of 7-, 9-, and 12-year-old
children who were presented with various materials such as numbers, consonants,
words, and nonsense syllables. Developmental differences were obtained when
the materials were chunkable but not when chunking was not possible given the
material. Unfortunately, Burtis (1982) failed to obtain supportive evidence. Burtis
presented 10- to 14-year-olds with easy-to-chunk, somewhat difficult to chunk,
and difficult-to-chunk consonant sequences. Regardless of age, the participants in
this study seemed to chunk all of these materials, thus undermining the conclu-
sion that chunking alone could account for age-correlated increases in the recall of
consonants. Although the findings by Burtis (1982) are based on supra-span lists
and thus cannot be generalized safely to the memory span task, the discrepancy
between the outcomes of the two studies is puzzling.
One obvious problem with the studies by Burtis and Dempster is that in nei-
ther piece of research were students’ prior knowledge and familiarity with the con-
sonant sequences determined directly. A classic study carried out by Chi (1978)
controlled this factor with more complex materials. She compared the abilities of
chess experts and novices in memorizing chessboard positions. Most fascinating
from a developmental perspective, the 10-year-old children were the experts in this
138 6  The Role of Basic Memory Capacities and Working Memory

study and the adults the novices. Although the children performed more poorly
on a digit span task, the results were reversed for recall of the chess positions.
Chunking was implicated in the post-experiment interviews, with child experts,
indicating that they viewed the whole chessboard as the unit of learning. The chil-
dren, who knew a lot about chess, were in a better position to create chunks.
These findings clearly support the chunking hypothesis. However, a caveat seems
in order here. Although there is little doubt that the experts and novices differed sys-
tematically in the way they reconstructed the chess positions, it appears to be more
difficult to identify chunks in performance than was assumed by Chi (1978). In a
careful replication study based on a much larger sample of child and adult chess
experts and novices, Schneider et al. (1993) found no support for the assumption
proposed by Chase and Simon (1973) and adopted by Chi that all chess pieces
reconstructed within a 2-s interval belonged to the same chunk and that chunk size
assessed for a 2-s inter-response latency time (IRT) differed between experts and
novices. Whereas the IRT measure did not lead to clear-cut results in the Schneider
et al.’s (1993) study, a “collective reconstruction” measure turned out to be more
informative. Using this index, Schneider and colleagues found qualitative differ-
ences between experts and novices in that the majority of the experts followed a
similar sequence when reconstructing the pieces, whereas the novices’ initial
reconstruction patterns were much more variable and unpredictable. Despite these
methodological problems, the studies by Chi (1978) and Schneider et al. (1993)
confirmed the importance of chunking for performance on memory span tasks.
The robustness of age-related span differences increases the attractiveness of
the interpretation that the actual capacity of the STS increases with age. As appeal-
ing as this is, however, it is too simple. The research described above has made
it clear that memory span is not a domain-general phenomenon that is essen-
tially identical regardless of what type of information is being remembered (see
Dempster 1978). Rather, how much a person knows about the stimuli he or she is
remembering definitively affects memory span such that knowledge presumably
has an effect by influencing processing speed.

The Impact of Item-Ordering Ability on Memory

Span Development

Some researchers (e.g., Chi 1977; Huttenlocher and Burke 1976) have suggested
that age differences in memory span may be due in part to the requirement that
materials on memory span lists be reproduced in order. Thus, one implication of
the item-ordering hypothesis is that the disadvantage of young children relative
to older children and adults should be greater when an ordered scoring method is
used than when recall is scored without regard to position errors (Dempster 1981).
In support of this hypothesis, Chi (1977) reported 100 % improvement in
recall performance by 5-year-olds when the order of the items was not consid-
ered. Similarly, Case et al. (1982) demonstrated that higher memory spans were
The Development of Short-Term Memory Capacity 139

obtained when the ordering of information was ignored (unconstrained word

span) than when correct serial ordering was required (serial word span). Findings
from the Munich Longitudinal Study on the Genesis of Individual Competencies
(LOGIC; Weinert and Schneider 1999; Schneider and Bullock 2009) confirmed
this finding. The mean differences between unconstrained and serial span were
particularly pronounced for the early measurement points when children were 4
and 5 years of age, respectively. The difference between scores decreased with
increasing age, and the format of recall (unconstrained vs. serial) no longer mat-
tered from the age of 6 years on. The longitudinal findings also indicated that
there was a substantial increase in serial word span from age 4 to age 6 but only
moderate increases during the elementary school period and beyond. Correlations
between unconstrained and serial word spans varied between 0.52 for the 4-year-
olds and 0.80 for the 10-year-olds, indicating increasing correspondence between
the two span measures over time (Schneider et al. 2009). Thus, although the data-
base is not extensive at this point, there is evidence that age-related changes in the
ability to order items are responsible for some of the considerable growth in mem-
ory span during early childhood, supporting the hypothesis that sequential order-
ing skill may be part of the cause of age differences in memory span.

Information Processing Speed and Short-Term

Memory Capacity

Developmental Changes in Information Processing Speed

Several researchers exploring the development of short-term memory capacity
in the 1970s and 1980s proposed that differences in the item identification pro-
cess may be a major source of span differences (e.g., Case et al. 1982; Chi 1977).
There is no doubt that processing speed generally increases with age during child-
hood and adolescence. For instance, the rate of item identification (i.e., the time
necessary to recognize or name to-be-remembered items when they are first pre-
sented as pictures or words) decreases with age. Chi (1977) investigated recogni-
tion time for faces in 5-year-old children and adults and discovered that the time
needed by the children was much greater than the time needed by the adults.
The research carried out by Robert Kail on developmental increases in pro-
cessing speed provided further evidence for the importance of this variable for the
development of short-term memory capacity. In a series of analyses, Kail (1986,
1988, 1991, 1992) observed that there are consistent increases in processing speed
with development. The increases in speed were so consistent that essentially the
same mathematical function related speed of processing to developmental level
for a variety of tasks (e.g., speed of tapping, mental addition, picture matching).
Kail argued that this finding suggests that a common pool of cognitive resources is
used to perform the various tasks that were studied and that this pool of resources
increases with development. His basic result has demonstrated good replicability:
Processing speed increases with development such that the curve of this relation
140 6  The Role of Basic Memory Capacities and Working Memory

is virtually identical for task after task whether the task is novel when executed
or has been practiced before (see also Hale et al. 1993). Also, it was shown in
longitudinal growth curve modeling studies that the speed with which cognitive
processes are executed increases substantially in childhood and less rapidly in ado-
lescence: Whereas linear models yielded relatively poor fit to the data, the fit of
exponential and quadratic models was much better (Kail and Ferrer 2007).
This developmental change in processing and mental speed has been assumed
to reflect, at least in part, the growth of a global mechanism. That is, the mecha-
nism is not specific to particular tasks but instead is a fundamental property of
the emerging information processing system. The assumption of a global mecha-
nism that accounts for developmental changes in processing speed is backed up by
two lines of evidence. First, age-related variance in speeded measures is virtually
eliminated through statistical control of other measures of processing speed. Also,
when age differences in task-relevant knowledge are experimentally controlled
for, age differences in memory span no longer occur (cf. Dempster 1985). Second,
children’s reaction times are often a simple multiple of adult reaction times
across a wide variety of tasks and conditions (Kail and Salthouse 1994). It seems
important to note, however, that the nature of the global mechanism still remains
unclear. For instance, findings from dual-task experiments rule out the possibility
that this hypothetical mechanism can be equated with increases in the total amount
of processing resources. That is, when children of different ages are required to tap
a key as quickly as possible while working on a cumulative rehearsal task (e.g.,
Guttentag 1984), developmental differences in rehearsal efficiency are typically
observed, a finding that can be explained by older children’s greater amount of
practice and processing efficiency and not by a larger total processing capacity per
se. Older children simply need fewer resources than younger children to perform
the task successfully (Kail 1991).
Is the underlying mechanism really domain-general? There is evidence that it
may not be as global as previously assumed. For instance, Kail and Miller (2006)
carried out a longitudinal study with children who were tested on a large number
of speeded tasks when they were 9 years of age and retested 5 years later using the
same test battery. The primary aim of the study was to determine whether process-
ing speed in the language domain develops at the same rate as global processing
speed. As a main result, developmental change in processing speed was greater
on nonlanguage tasks (sampling global processing speed) than on language tasks.
Also, processing speed was faster on language than on nonlanguage tasks at age
9 but not at age 14. Thus, there was only modest support for the assumption that
developmental changes in processing speed are due to a global domain-general
mechanism. As noted by Kail (2004), however, evidence for domain-specific
change is not necessarily inconsistent with global change. So the issue should not
be framed in “either–or” terms. It is perfectly plausible that speeded performance
can be a product of global processes as well as some domain-specific ones.
The role of knowledge also cannot be ignored. A study by Rabinowitz et al.
(1994) explored age constraints on speed of processing on a lexical decision task,
trying to unconfound age and experience by using two lexical decision tasks and
Information Processing Speed and Short-Term Memory Capacity 141

two age groups of seventh and ninth graders who attended German gymnasi-
ums. There were two groups of ninth graders, one that had received instruction in
English (their second language) for 5 years and one that had received only 1 year
of instruction in English. The seventh graders had all received 3 years of instruc-
tion in English. One of the two lexical decision tasks was provided in German,
and the other was in English. When the lexical decision task involved German
words and nonwords, the older participants responded more quickly than the
younger ones. However, when the stimulus items were English words and non-
words, this age-related progression was disrupted, and response speed was related
to experience with English as a second language. According to the authors, the
results of this study are consistent with the view that age exerts a weak constraint
on children’s speed of processing in a lexical decision task. In addition, experience
seems to play an important role in determining the time required to make a lexical
decision.
Cultural differences may also serve as a mediating factor. In a recent study, Kail
et al. (2013) compared developmental changes in processing speed in children
ranging in age from 4.5 years to just under 11 years of age. The children came
from China, Korea, and the USA. As a main result, it was found that although pro-
cessing speed was comparable across subsamples at the youngest age (4.5 years),
it developed more rapidly in Chinese children than in US children. There was
also some evidence that speed of processing developed more rapidly in Korean
children, but this finding was not consistent across measures. Kail and colleagues
concluded that their results did not provide direct evidence regarding possible
underlying mechanisms but that the findings were compatible with both genetic
and cultural accounts.
Confirming evidence comes from a study by Demetriou et al. (2005) who
compared processing speed in Greek and Chinese children from 8 to 14 years of
age. The Chinese children outperformed the Greek children on all speed-related
tasks and tasks addressing visuospatial processing but not in general intelligence
or in processes in which the two groups had equivalent experience. Regarding the
cultural account, Demetriou and colleagues concluded that the advantage in the
Chinese group was associated with the massive practice in visuospatial processing
that is required to learn the Chinese logographic writing system. Overall, however,
they demonstrated that the architecture and the developmental patterning of the
various processes were basically the same in the two ethnic groups.

Relations Between Information Processing Speed

and Memory Span

Since the mid-1970s, numerous studies have shown that the rate of item identifica-
tion is correlated with memory span performance. Chi (1977) found that the differ-
ences in memory span between kindergarten children and adults were drastically
reduced when the time of picture presentation was reduced by half for the adults,
142 6  The Role of Basic Memory Capacities and Working Memory

and thus, the effects of processing speed differences were reduced. Similarly,
Case et al. (1982) discovered a monotone and almost linear negative relation
(r = −0.74) for children from 3 to 6 years of age between reaction time rate and
memory span, and this relation remained significant when the effect of age was
removed statistically. Case et al. (1982) also found that children and adults had
comparable memory spans when tested on materials that were named at the same
rate by children and adults. Nicolson (1981) reached a similar conclusion with a
sample of 8- to 12-year-old children, reporting no age-related memory span differ-
ences when reading speed was equated across age levels.
Overall, the relation between rate of processing and memory span appears to
be very pronounced both between and within age groups. Individuals who process
information rapidly have relatively large memory spans. Such a causal relation
between speed of activation and memory span performance has been repeatedly
shown for verbal articulation speed. For instance, Hulme and colleagues (e.g.,
Hulme et al. 1984; Hulme and Tordoff 1989; Roodenrys et al. 1993) demonstrated
reliable relations between increasing speech rate and memory span with develop-
ment. The close connection between speech rate and memory span was also shown
in studies by Hitch et al. (1989, 1993) who carried out experiments with 5-, 8-,
and 11-year-old children. Hitch and colleagues compared the respective roles of
perceptual identification times and articulation rates in predicting auditory span.
Their findings indicated that articulation rate turned out to be a better predictor of
memory span than perceptual time in both studies regardless of age. The correla-
tion between articulation rate and memory span was almost perfect: The faster the
children could articulate words, the higher their memory spans for words.
Other studies have confirmed this finding. For instance, the results of a path
analysis carried out by Kail (1992) indicate that age and processing speed inde-
pendently contribute to articulation rate, which determines memory span. In
a similar vein, Kail and Park (1994) demonstrated relations between increas-
ing speech rate and memory span increases with development for samples of
American and Korean children and adults. For example, in these studies, when
participants’ speech rates were slowed when longer words were presented on the
to-be-remembered lists, memory span declined as would be expected if speech
rate was important for determining capacity. Causal modeling procedures used
by Kail and Park (1994) indicate that age-related change in processing time is
associated with a decrease in the time required to articulate numbers and letters,
with articulation rate determining memory span (see Fig. 6.3). In both samples,
processing time decreased with age, which led to an increased articulation rate.
In turn, individual differences in articulation rate directly influenced memory
span. Overall, the results were quite similar for the American and Korean sam-
ples, with the only exception being that chronological age directly influenced
speech rate in the Korean but not in the American sample. This path means that
age-related change in articulation time was not entirely mediated by age-related
changes in rate of processing and that other age-related changes were involved in
Information Processing Speed and Short-Term Memory Capacity 143

Korean Sample .85 Identical Pictures

Process
Time Number Comparisons
.82

-.63
.36

.81 Digit Span

.36
Age Memory

.88 Letter Span

-.33

.96 Naming Digits

Speech
Time
Naming Letters
.91

U. S. Sample
.86 Identical Pictures
Process
Time
.83 Number Comparisons

-.63
.46

.79 Digit Span

.17
Age Memory

.64 Letter Span

-.42

.88 Naming Digits

Speech
Time
Naming Letters
.98

Fig. 6.3  Path diagrams for the Korean and USA samples. Latent constructs are depicted by cir-
cles and measured constructs by squares and rectangles. (Kail and Park 1994, p. 289)
144 6  The Role of Basic Memory Capacities and Working Memory

the Korean sample. However, this difference in findings may be more apparent
than real, given that this path was significant for the English-speaking children
in the study by Kail (1992). Other variables associated with speech rate, such as
the duration of the words and the lengths of a person’s pauses between words, are
also related to memory span as expected (i.e., the slower, the smaller the span)
and with development (i.e., shorter durations and pause times with increasing age
were associated with greater working memory span; Cowan 1994, 2005; Cowan
et al. 1998).
Analyses of language differences also support the conclusion that speech rate
is an important determinant of short-term memory capacity. For instance, Chen
and Stevenson (1988) analyzed digit span differences between American and
Chinese children in terms of naming speed. Chinese digits can be pronounced
more quickly than English digits, and the digit spans of Chinese children are
longer than the spans of American children. Chen and Stevenson’s analysis
ruled out alternative explanations of this cross-cultural difference (for confirm-
ing evidence, see also Geary et al. 1993). Another impressive illustration of
the importance of naming speed for digit span comes from a study by Ellis and
Hennelly (1980) who recruited a sample of children whose first language was
Welsh and whose second language was English. Digit spans were assessed in
both languages. Intraindividual comparisons revealed that significantly larger
digit spans were obtained in English, the children’s second language, than in
Welsh. Although this finding may be surprising at first glance, it is in accord-
ance with the Chinese–English comparisons described earlier in that larger digit
spans were found for the language in which the names of numbers can be more
quickly articulated. Thus, it seems likely that the speed of item identification
(e.g., recognizing a picture or a word) and the speed with which the name of an
item can be spoken (i.e., speech rate) or both in combination (Henry and Millar
1991; Kail and Park 1994; Roodenrys et al. 1993) are important determinants of
memory span.

Capacity, Working Memory, and Cognitive Development:

Theoretical Approaches

The findings by Dempster, Kail, and others, who reported age-related changes in
processing speed concomitant with age-related changes in cognitive task perfor-
mance, are consistent with theories attributing changes in cognitive development
to changes in general information processing capacity (e.g., Baddeley and Hitch
1974; Case 1985; Halford 1982, 1993; Pascual-Leone 1970). These theories also
deal with the question of whether there are really structural increases in capacity
(i.e., more slots) with development, or just increases in slots due to increases in
chunking and processing speed. In the following, the assumptions of these models
will be briefly summarized, and differences as well as commonalities will be dis-
cussed in more detail.
Capacity, Working Memory, and Cognitive Development: Theoretical Approaches 145

Neo-Piagetian Models

The first widely regarded neo-Piagetian theories were those that viewed stage-
related developmental changes as caused by differences in information processing
capacity. Although the number of neo-Piagetian models of cognitive development
has increased rapidly over the past 35 years (cf. Case 1985; Demetriou et al. 1993;
Demetriou et al. 2002; Fischer 1980; Halford 1982, 1993; Pascual-Leone 1970,
1987), the different models show a number of common postulates (for reviews, see
Case 1995; De Ribaupierre and Bailleux 1994, 1995). One of these postulates is
that general stages should be defined in terms of an upper limit at which children
of a given age or cognitive level can function. A second postulate is that working
memory or attentional capacity plays a strong, if not causal role, in determining
the upper limit.
The first very influential neo-Piagetian model of working memory growth was
presented by Pascual-Leone (1970, 2000). Pascual-Leone’s theory of construc-
tive operators has prompted a number of studies on the development of mem-
ory capacity. According to Pascual-Leone, the central concept behind “limited
capacity” theories is that stages reflect “the endogenous growth of maturationally
driven mental attention mechanisms” (2000, p. 843). He believes that a quantita-
tively specifiable parameter can be used to explain many of the qualitative phe-
nomena described by Piaget. This is “central computing space” or “mental space,”
also called M-space. A child’s cognitive capacity is represented by M, which is
the maximum number of schemata that a person can activate, coordinate, or both
activate and coordinate simultaneously. As noted by Case (1995), the concept of
M-space would be referred to as working memory today. Pascual-Leone hypoth-
esized that there is a linear increase from 3 to 16 years of age in this mental capac-
ity. M is assumed to be a combination of a constant called “a,” which is shared
by all children, and a variable labeled “k,” which changes with age. Thus, for any
given child, mental capacity can be expressed as M = a + k, with the value of k
increasing with age. Increments in k are assumed to be representative of transi-
tions between Piagetian stages (cf. Fig. 6.4).
M-space reflects the number of items one can hold in short-term memory at a
given time. It is assessed with working memory tasks. Designing tasks that permit
assessment of the k-component is not easy. Pretraining of the task is necessary so
that the child can acquire the executive skills to carry out the task. The task must
involve some mental transformations so that it taps many aspects of short-term
capacity that are presumably represented by k. In addition, it is important to sup-
press strategies such as chunking. That is, the size of each chunk must be known
by the experimenter in order to determine k, and this is very difficult to do if par-
ticipants are recoding materials into larger chunks.
Pascual-Leone (1970) relied heavily on one particular task to provide a meas-
ure of M-space, the compound visual stimulus information task (CVSI). This
task requires children between 5 and 11 years of age to react motorically by clap-
ping or raising their hands in response to various visual stimuli (e.g., a square,
146 6  The Role of Basic Memory Capacities and Working Memory

Central
executive

Visuospatial Episodic Phonological

sketchpad buffer loop

Visual Episodic Language

semantics LTM

Fluid Crystallised
systems systems

Fig. 6.4  The revised model of working memory (Baddeley 2000). (Reproduced with permission
from Baddeley 2000)

the color red). After participants master stimulus–action pairings (e.g., a square
calls for clapping, red calls for raising hands), the critical measurements are
taken. Combination stimuli are presented (e.g., a red square), and participants are
required to make every response that would be appropriate for such a stimulus
(e.g., both clap and raise hands). The compound stimuli vary in complexity. The
results obtained by Pascual-Leone were perfectly consistent with his model in that
the complexity of stimuli that children could respond to increased with age. For
instance, 5-year-olds could perform correctly when compound stimuli were made
up of two stimuli–action sequences, whereas 11-year-olds could respond correctly
to compounds requiring five stimulus–action sequences. A number of other stud-
ies have also provided data consistent with Pascual-Leone’s assumption that there
are age-correlated increases in M-space (e.g., Burtis 1982; Case and Serlin 1979;
Globerson 1983).
Despite the empirical support for the model, there are a number of objec-
tions that can be raised against Pascual-Leone’s conclusion that mental capacity
increases with age. First, from a theoretical perspective, it is somewhat disturb-
ing that the amount of space allocated to executive action is considered to be con-
stant across childhood (i.e., equal to the invariant parameter “a” at all age levels).
Just the fact that mental processes are executed more rapidly with increasing age
suggests that the functional capacity used by executive schemata may change.
In particular, with increasing age and increasing speed, it might be expected that
the amount of space required for the executive actions decreases. In addition, the
assumption that stimulus–action pairings are equal to one same-sized chunk at all
Capacity, Working Memory, and Cognitive Development: Theoretical Approaches 147

ages poses difficulties. The speed of recognition and execution of these pairings
should also increase with age and thus reduce the capacity required to attend to
and execute a single pairing. In short, speed of information processing was not
taken into account by Pascual-Leone (1970).
Other problems with the empirical tasks used to test the model concern con-
foundings of procedure and age. For instance, the number of reactions required
in Pascual-Leone’s M-space tests was systematically confounded with age. That
is, younger children were given fewer stimulus–action pairings to learn and pro-
duce in reaction to compound stimuli. See Trabasso and Foellinger (1978) for a
thorough review of these problems and Pascual-Leone’s (1978) responses. A final
problem with the data reported in support of Pascual-Leone’s original model is
that sometimes different measurement procedures produced different age norms
(e.g., compare Pascual-Leone 1970, with Globerson 1983), and sometimes the
same methods produced different age norms (e.g., compare Case 1972, with
Globerson 1983).
Despite the critiques of Pascual-Leone’s original model, the revised version
is still influential and well received among neo-Piagetians (see Case 1985; De
Ribaupierre 1999, 2002; De Ribaupierre and Bailleux 1994). According to Case
(1995), the conceptual progression has had a dialectic character. That is, critiques
of Pascual-Leone’s original model led to new studies, which led to refinements of
the model, which in turn led to further critiques and so on.
In the most updated Pascual-Leone’s model (Pascual-Leone and Baillargeon
1994; Pascual-Leone and Johnson 1999, 2011), working memory is not con-
ceived of as a unitary system. Instead, it corresponds to a set of highly activated
schemes that concur with the performance on a given task. When new informa-
tion is processed, a larger set of schemes is activated, most of them automati-
cally. These schemes represent the “activation field.” Depending on the situation,
certain schemes in this field are inhibited, whereas others are more strongly acti-
vated, defining the “field of mental attention.” Pascual-Leone posited that three
mechanisms contribute to the selection and activation of schemes in this field:
(a) M-space; (b) an inhibition mechanism lowering the weight of task-irrelevant
but usually highly activated schemes, and (c) executive schemes responsible for
performance control (i.e., for planning and monitoring and for the distribution of
resources). As I will show later, the role of these three mechanisms is comparable
to the “central executive” in the working memory model developed by Baddeley
(1986; Baddeley and Hitch 1974). However, whereas Baddeley did not decompose
the “central executive” any further, Pascual-Leone assumed that several processes
are at play.
The model exemplified by Pascual-Leone and his colleagues essentially pro-
posed that children’s performance on reasoning tasks such as conservation can be
accounted for by how much information they can hold in mind at a given time.
This model proposes a domain-general set of cognitive resources that children
allocate to various cognitive operations that increase with age. Unmodified, such
domain-general models are not well suited for explaining results such as those
reported by Chi (1978) in which child chess experts had a greater memory span
148 6  The Role of Basic Memory Capacities and Working Memory

for chess positions than adult novices did but not when digits served as the stimuli.
Some cognitive feats seem to be domain-specific, with skills in one domain that do
not generalize to other domains.
How can a theory that postulates a domain-general pool of resources account
for such findings? The model of capacity development presented by Case (1985,
1995) proposed that age differences do not exist so much in children’s absolute
mental capacity but rather in the efficiency with which children use their mental
capacity (Case 1995; Kee 1994). Case argued that the amount of memory space
required to operate on stimuli functionally decreases with increasing age, given
that executive actions operate more efficiently. This in turn frees up more space for
the storage of materials and, to a large extent, accounts for the increase in mem-
ory span. Some of these improvements can be attributed to maturation. However,
children also become increasingly adept at acquiring information and using strat-
egies. This ability fosters greater efficiency and results in a heightened speed of
processing.
Case distinguished between storage space and operating space when concep-
tualizing his model of working memory. Storage space is the mental space that a
person has available for storing information, whereas operating space is the men-
tal space that can be allocated to the execution of cognitive operations. These
two components make up the total processing space. It is thus assumed that the
relation between memory span and processing speed reflects a trade-off between
operation and storage within a central total processing space that does not change
during development. More specifically, as each new developmental skill is prac-
ticed and mastered, the increase in processing efficiency frees attention (or process-
ing space) that can then be applied to coordinate new strategies. Support for this
assumption came from an experimental study (Case et al. 1982) with young chil-
dren ranging in age from 3 to 6 years. On the basis of their theory, Case and col-
leagues predicted that there would be a relation between operational efficiency (as
reflected by the speed of identification) and storage space (as reflected by children’s
memory span). The data supported this prediction, showing a substantial relation
between storage space and operational efficiency that was also related to age.
The conceptualization of working memory growth presented by Case has
received much attention in the field. However, its validity has been challenged not
only by researchers in the neo-Piagetian tradition (e.g., Halford 1993), but also by
scientists using a different theoretical approach derived from information process-
ing theories.

Information Processing and Working Memory Models

Only a few scientists who currently derive their memory capacity models from
information processing theories argue that there is a single set of resources
that influences all aspects of cognition. However, as already noted above,
there is research that supports the idea that, in combination with more specific
Capacity, Working Memory, and Cognitive Development: Theoretical Approaches 149

mechanisms, a general processing mechanism plays a substantial role in cogni-

tive development. For instance, the models developed by Kail and colleagues are
based on such an assumption. The (life span) developmental theory proposed
by Kail and Salthouse (1994) assumes that a global processing speed mechanism
influences cognitive performance in different domains either directly (faster global
processing results in enhanced performance) or indirectly by influencing the speed
with which more domain-specific processing mechanisms (such as rehearsal) are
executed. Each of these factors (global processing speed, domain-specific process-
ing speed, and cognitive performance) can be influenced by other factors that are
associated with age, such as maturationally based change involving the myelina-
tion of nerves in the associative “thinking” area of the brain or knowledge.
On the basis of evidence of regular age-related changes in processing speed
(discussed earlier), Kail and Salthouse (1994) proposed that processing speed is a
cognitive primitive or fundamental parameter of humans’ cognitive architecture.
Differences in this parameter change with age in regular ways and influence cogni-
tive functioning by constraining how much information can be handled at any one
time. The speed of processing model presented by these authors does not claim
that all aspects of cognitive performance are determined exclusively by capacity.
Instead, capacity is seen as but one of several forces in human performance and
change. However, there are usually strong relations between age and processing
speed throughout the life span (which may vary somewhat according to specific
experiences), and variations in speed appear to contribute to age-related variations
in cognition. According to Kail and Salthouse, although experience can play a role
in age differences in speed, a general mechanism limits speeded performance.
They suggest that further research should explore the nature of processing speed
because it is an unusual construct that may be linked to explicit changes in a per-
son’s neural structure and functioning on the one hand and to higher-order men-
tal processes such as reasoning and abstraction on the other. Please note that De
Ribaupierre et al. (2011) recently confirmed the importance of processing speed
(and inhibition) for working memory development from a life span perspec-
tive using an experimental design that included children, young adults, and older
adults. The same structural equation model accounted for working memory per-
formance in all three age groups, thus indicating that age differences in working
memory are driven by age differences in processing speed and inhibition, a find-
ing that appears to be in line with both neo-Piagetian and information processing
accounts of working memory development.
As indicated by the causal modeling procedures used by Kail and Park (1994,
see Fig. 6.3), speech rate (an indicator not assessed in the study by De Ribaupierre
et al. 2011) is also an important predictor of short-term memory capacity. This
component was explicitly considered in Baddeley’s (1986, 2000, 2007; Baddeley
and Hitch 1974) working memory model, which defined working memory as a
multicomponent system. The original working memory (WM) model (Baddeley
and Hitch 1974) consisted of three components (in contrast to a unitary view of
short-term memory) composed of the central executive (CE), conceptualized
as a higher-order control mechanism or attentional system with limited capacity,
150 6  The Role of Basic Memory Capacities and Working Memory

and two “slave” systems: the phonological (articulatory) loop (PL) and the visu-
ospatial sketch pad (VSSP). The CE is the most important component in the
system as it ensures that working memory resources are directed and used appro-
priately to achieve the person’s memory goals. The CE represents an attentional
control system, coordinating the various working memory activities. It has been
studied to a relatively small degree in the past.
The phonological loop is a temporary phonological store that maintains and
processes verbal and acoustic information (speech sounds) and lasts about 1–2 s.
As decay in this store is rapid, verbal information needs to be rehearsed by subvo-
cal articulation. The PL is the most studied part of the system. It consists of two
components: a phonological store, which is relatively passive and to which verbal
material (presented auditorily) has automatic access, and a rehearsal mechanism,
which helps to maintain stored items as well as to recode verbal material presented
visually. The PL is temporarily limited and contains as many items as can be
rehearsed in about 2 s. Thus, it is closely linked to the articulation rate.
The VSSP is responsible for holding and manipulating visual and/or spatial
information and is also assumed to hold any verbal information stored as an image
for short periods of time. Experimental studies exploring age-related changes in
this component are less numerous although their number has increased in recent
years. There is experimental evidence indicating that the VSSP is very different
from phonological short-term memory and that it consists of two different sub-
components, that is, a visual and a spatial one (Logie and Baddeley 1990).
Although Baddeley’s original WM model was developed to describe memory
processes in adults, there is developmental evidence that the basic modular struc-
ture of this model is present from 6 years of age and possible earlier (Gathercole
et al. 2004). Overall, the model has successfully accounted for a large body of
experimental research (see below). However, various criticisms of the model
have led to some significant modifications. For instance, it became clear that
there was a need to account for the effects of LTM on working memory capacity,
and these effects had not been considered in the original model. Thus, Baddeley
(2000) added a fourth component to the WM model; he labeled it the “episodic
buffer.” This new component represents the greatest change to the original model.
It is described as a “multimodal” temporary store, which means that it is not lim-
ited to storing information in just one modality (e.g., verbal or visual), but that it
binds together information from different sources of the WM system. For instance,
information about a scene may comprise visual information, speech sounds, and
movements. All of this information is integrated in the episodic buffer (which
is linked to LTM), and this helps a person to form a coherent memory episode.
Although the capacity of the episodic buffer is not clearly specified, it is believed
to be limited to the number of chunks or episodes that can be maintained simulta-
neously. The revised model of working memory is shown in Fig. 6.5.
It should be noted that the current view of the CE described in the revised WM
model has changed somewhat (Baddeley 2000, 2007; Henry 2012). It is no longer
regarded as having an unlimited capacity for storage. Rather, it is responsible only
for the control and allocation of attention. Thus, its core role is in allocating atten-
tion, and it does so by focusing, dividing, and switching attention. Another new
Capacity, Working Memory, and Cognitive Development: Theoretical Approaches 151

Central executive processes exert some control over

attention (effortful control)

Activated portion of long-term memory

includes features limited by decay and
Stimulus
feature-specific interference
events can
recruit
Focus of attention
attention
includes chunks limited
(orienting
to a few at any one time
responses)

Fig. 6.5  A depiction of the theoretical model of Cowan (1988, 1999) as it pertains to short-
term and working memory. Short-term memory refers here to information in the activated portion
of long-term memory. Working memory here includes short-term memory plus central executive
control processes that allow the focus of attention (a subset of the activated portion of long-term
memory) to maintain some information especially well (Cowan 2014; Henry 2012, p. 5)

aspect is that the link between various working memory components and LTM
now comes via the new component (i.e., the episodic buffer). The new role of the
CE in the revised model is that of a broad attentional control space, and it is proba-
bly similar to the type of system that many researchers refer to as “executive func-
tions.” See Baddeley (2007) and Henry (2012) for a more detailed account of the
revised WM system.
Another model that is quite attractive for developmentalists was introduced
by Cowan (1988, 1999; for more detailed reviews of this model, see Cowan 2014;
Cowan and Alloway 2009; Towse and Cowan 2005). This model was suggested as
a general processing model, taking into account the role of attention processes to
a greater degree than Baddeley’s (1986) original working memory model. Cowan’s
(1988) model postulates that all incoming stimulation has some influence on the
information processing system, activating parts of LTM. This model focuses on the
span of apprehension, that is, the amount of information people can attend to at a
single time. For unattended stimuli, the activated part of LTM may be limited to the
gross physical features of the stimuli. However, if these features change abruptly, the
stimuli not only activate LTM but also recruit attention. This will result in a deeper
and more complete analysis of the changed stimuli (Cowan and Alloway 2009).
It should be noted that Cowan’s model is very similar to Baddeley’s (2000,
2007) revised working memory model. One difference is that Cowan’s model is
less modular. It does not distinguish between two slave systems that process either
verbal or nonverbal information mainly because Cowan was not sure how to divide
memory up in a meaningful way. Although he did not deny a difference between
152 6  The Role of Basic Memory Capacities and Working Memory

phonological and visuospatial representations, he did not exclude the possibility

that, for instance, the phonological store might also work for nonverbal sounds.
Thus, it was simply stated that the amount of interference between one stimulus
and another depends on the similarity of the features of the two stimuli.
Cowan (1988) proposed that a subset of the activated LTM is in the focus of
attention (see Fig. 6.6). Later research suggested that the focus (i.e., the span of

(a) 5

4
ITEMS RECALLED

Spatial Supression
1 Verbal Supression
Control

0
5 6 8 9

AGE

(b) 5

4
ITEMS RECALLED

Spatial Supression
1 Verbal Supression
Control

0
5 6 8 9

AGE

Fig. 6.6  a Mr Peanut task: mean number of correct items for each age group and suppression
condition. b Corsi blocks task: mean number of correct items for each group and suppression
condition. (Kemps et al. 2000)
Capacity, Working Memory, and Cognitive Development: Theoretical Approaches 153

apprehension) typically includes three to five meaningful chunks of information in

normal adults (Cowan 2001, 2005). The focus of attention actually probably func-
tions in a manner similar to Baddeley’s (2000, 2007) episodic buffer except that
Baddeley was not very specific about the attentional requirements of that buffer.
As noted above, there is a lot of similarity between the revised Baddeley’s model
and the Cowan’s model. As for the terminology, activated memory can be called
short-term memory, and the focus of attention can be called working memory
(Engle et al. 1999).
In recent years, variations in working memory models have concerned develop-
mental aspects that were not described by the Baddeley’s and Cowan’s models, in par-
ticular developmental changes concerning the trade-off between processing speed and
storage space as first discussed by Case (1985). Please recall that Case’s model was
based on a resource-sharing assumption, indicating that processing and storage share
a common and limited resource, and explaining age-related increases in recall by the
fact that the processing part of the memory span task becomes less and less resource-
demanding as processing efficiency increases. For instance, the study by Case et al.
(1982) using the counting span task appeared to support this assumption. For this task,
children are asked to count dots on cards out loud and then to recall the number of
dots presented on each card. The number of cards to be counted varies, and the maxi-
mum number of cards (and their values) that the children are able to remember consti-
tutes their counting span. Case et al. (1982) explained the age-related increase in the
counting span with an improvement in the efficiency of the counting operation.
Towse and Hitch (1995), Hitch et al. (2001), Hitch (2006) challenged this
resource-sharing assumption and suggested a task-switching model instead,
assuming that children switch between the two functions of working memory (i.e.,
storage and processing) according to the structure of the task and that short-term
memory traces suffer from time-related decay during the processing phases. As
a consequence, Towse and Hitch provided a different explanation for age-related
increases in counting span. They assumed that because the traces of memory items
decline during processing episodes (i.e., counting), recall performance should
depend on the duration of the delay between item encoding and recall. Because
older children are faster at counting arrays of dots, they benefit from shorter delays
between encoding and retrieval at the end of the series, and this explains their bet-
ter recall. Thus, this model considers the developmental increase in working mem-
ory span to be a mechanistic by-product of the increase in processing speed and
the resulting reduction in the maintenance delay.
Barrouillet and Camos (2007; Barrouillet et al. 2004) developed an alterna-
tive developmental model that they labeled the time-based resource-sharing
(TBRS) model. According to this approach, maintenance of information in work-
ing memory and recall performance on complex span tasks rely on the same lim-
ited resource (i.e., attention) and depend on a specific strategy called attentional
refreshing. This strategy is the ability to switch attention during the processing
episodes to briefly reactivate (refresh) decaying memory traces. This active strat-
egy is most likely related to the executive function component in Baddeley’s work-
ing memory model. One interesting aspect of the TBRS model is that it leads to a
154 6  The Role of Basic Memory Capacities and Working Memory

new conception of cognitive load. That is, the model assumes that those complex
tasks that almost always occupy attention prevent the switching and refreshing
of memory traces. Therefore, they involve a high cognitive load and lead to poor
recall. In this model, the cognitive load that a given task requires is a function of
the proportion of time during which this task occupies attention. From a develop-
mental perspective, the emergence of executive functions (in the sense of self-reg-
ulation skills) around the ages of 6 or 7 should facilitate the use of the attentional
refreshing strategy, with older children making better use of attention control and
thus being better able to refresh decaying memory traces (Barrouillet et al. 2004;
Barrouillet and Camos 2007).

Empirical Tests of the Various Theoretical Working

Memory Models

Although the most influential models were just described above, it should be noted
that there are still more theoretical conceptualizations of working memory (e.g.,
see the review by Cowan and Alloway 2009). From a developmental perspective,
however, the models described so far appear to offer the most potential. Overall,
these models seem helpful because they attempt to describe the working memory
system as a whole. They also provide testable assumptions concerning the follow-
ing questions: Is there an increase in basic memory capacity over time? Is there
decay? Is there a capacity limit? Does attention span increase with age? The fol-
lowing sections will provide an overview of empirical studies that have tested the
core assumptions of the various working memory models.

Empirical Tests of Pascual-Leone’s Model

It was already mentioned above that the original working memory model devel-
oped by Pascual-Leone (1970) was initially a controversial topic of discussion
(e.g., by Trabasso and Foellinger 1978). However, subsequent evaluations of
the revised model have been more supportive. For instance, De Ribaupierre and
Bailleux (1994, 1995) carried out a comprehensive longitudinal study with 120
children to test several of the developmental predictions of the Pascual-Leone’s
model. There were four cohorts of children who were 5, 6, 8, and 10 years of age
at the first measurement point; they were tested annually across a time period of
5 years using several spatial working memory measures. As a main result, differ-
ent outcomes for the various working memory tasks could be explained by model
predictions such as limits in the amount of information that could be processed
within a given time period. Moreover, the substantial stability in working mem-
ory scores over time was in line with the hypothesis that limits in M-space power
exert constraints on cognitive performance. Contrary to expectations, however,
Empirical Tests of the Various Theoretical Working Memory Models 155

the increase in M-space was relatively linear over the first 3 years of the longi-
tudinal study regardless of the age cohort under study. This linear increase was
not congruent with the neo-Piagetian postulate of developmental stages lasting
for 2 years. The authors argued that their findings were not sufficient to refute
Pascual-Leone’s claim that there are developmental stages with duration of 2 years
because they could be explained by a combination of developmental and learning
effects. That is, changes with age may be linear when they result from a combina-
tion of M-power and effects of learning, as indicated by retest effects in the data.
One of the goals of the study carried out by De Ribaupierre and Bailleux
(1994, 1995) was to illustrate the similarities between the Pascual-Leone’s and
Baddeley’s models of working memory in their references to similar mecha-
nisms of mental attention and executive schemes. According to the authors, the
longitudinal findings of the two studies illustrated that the two different models
might be complementary rather than contradictory. In fact, De Ribaupierre and
Bailleux realized during the course of their study that some of their findings,
which were in accordance with Pascual-Leone’s predictions, could just as well be
interpreted using Baddeley’s model.
This issue was further explored in a developmental study by Kemps et al.
(2000) who selected two working memory tasks, one typically used by Pascual-
Leone (the Mr. Peanut task) and one more often used in the Baddeley working
memory tradition (the Corsi blocks task). Whereas the Mr. Peanut task was con-
structed as a measure of M-capacity, the Corsi blocks task was designed as a neu-
ropsychological test for assessing spatial working memory and associated with the
VSSP in Baddeley’s model (e.g., Hitch et al. 1989). The two tasks are both clearly
spatial but differ in several respects. The original unicolored variation of the Mr.
Peanut task adapted from Case (1985) presents a two-dimensional outline of a
clown figure to which a number of purple dots are attached to an exposure time
of about one dot per second. Children are then presented with a blank outline of
the clown figure and asked to point to those parts of its body that previously had a
dot on them. The apparatus for the Corsi blocks task used in this study consisted
of a set of nine black blocks arrayed in a quasi-random pattern on a black wooden
board. The experimenter tapped a particular sequence of blocks at a rate of one per
second, and the children were then required to tap out the same pattern immedi-
ately afterward.
Kemps and colleagues chose these two working memory tasks to demonstrate
that the models developed by Pascual-Leone and Baddeley could indeed comple-
ment each other, making their own contribution to the interpretation of empirical
results. For instance, typical of Pascual-Leone’s model is the assumption of a step-
wise increase in M-space, with age-related increments of one unit every 2 years.
On the other hand, typical of developmental studies generated from Baddeley’s
perspective is the hypothesis of a progressive complementation of visual coding
by phonological coding with age (e.g., Hitch et al. 1989). The authors included
articulatory and spatial suppression conditions (i.e., continuously repeating the
word “the” in the first condition and continuously tapping four plates in the latter
condition when working on the two working memory tasks) in their experimental
156 6  The Role of Basic Memory Capacities and Working Memory

design because phonological recoding seemed to be relevant in the Mr. Peanut

task, particularly for older children, whereas spatial chunking seemed possible in
the Corsi blocks task.
The results of their study, which assessed working memory in children ranging
from 5 to 9 years of age (5-, 6-, 8-, and 9-year-olds), confirmed most of their hypoth-
eses. That is, both tasks yielded a developmental trend in performance, and the age-
related data regarding Mr. Peanut were broadly consistent with Pascual-Leone’s
assumption of a stepwise improvement in working memory span. More specifically,
the scores of children who were supposed to belong to the same developmental stage
(i.e., the 5- and 6-year-olds) were not significantly different, whereas the scores of
children who were supposed to belong to different stages (i.e., the 8- and 9-year-
olds) were significantly different. As shown in Fig. 6.6, span scores on the Corsi
blocks task also showed a stepwise increase, at least for the younger age-groups,
with 5- and 6-year-olds not differing from each other and 8-year-olds scoring sig-
nificantly higher than the 6-year-olds. However, the findings that the groups of 8-
and 9-year-olds did not differ from each other on the Corsi blocks tasks and that the
scores on the Corsi blocks tasks were reliably higher than those for the Dr. Peanut
task across age groups were not consistent with the hypotheses. Given that no group
of 7-year-olds was included in the sample, it seems difficult to rule out the assump-
tion of continuous developmental change. The developmental patterns shown in
Fig. 6.7 indicate a rather steep performance increase between the ages of 6 and 8 for
both working memory tasks, and it seems likely that 7-year-olds would have differed
reliably from both the 6- and the 8-year-olds. This is at least what the longitudinal
findings reported by De Ribaupierre and Bailleux (1994) clearly confirmed for the
Mr. Peanut task when using considerably larger subsamples.
In accordance with the authors’ hypotheses, concurrent verbal suppression did
not affect performance on the Corsi blocks task. However, it caused a substantial

(a) (b)
2.5 8
Mean memory span

2 7
Mean recall rate

1.5
6
1
5
0.5

0 4
Phase 1 Phase 2 Phase 3 Phase 1 Phase 2 Phase 3

Control Fast Control Fast

Fig. 6.7  a Mean recall rates for 8-year-old children. The “control” group were asked to recall
digits at whatever speed they liked in Phases 1, 2, and 3: The “fast” group were asked to recall
the items as quickly as they could without making errors in Phase 2. b Mean memory spans for
these two groups are displayed (Henry 2012, p. 105)
Empirical Tests of the Various Theoretical Working Memory Models 157

impairment in the performance of the older children on the Mr. Peanut task,
whereas it did not interfere with the recall of the younger children. This finding of
a differential effect of articulatory suppression according to age groups on the Mr.
Peanut task is difficult to reconcile with Pascual-Leone’s model because no age-
related difference is expected in allocating M-capacity to schemes that represent
verbal or spatial information. However, Baddeley’s model can provide a plausible
explanation for this developmental effect, given that it assumes a verbal recoding
of visually presented information in older but not in younger children (Halliday
et al. 1990; Hitch et al. 1989).
Although not all of the findings were compatible with the two working mem-
ory models, Kemps et al. (2000) concluded that each model was found to make a
specific contribution to the interpretation of the data. Pascual-Leone’s model pro-
vided a coherent account of age-related increases in M-space, whereas Baddeley’s
model explained the developmental contributions of the phonological and visu-
ospatial components, in particular the effects of articulatory and spatial suppres-
sion. The complementarity assumption between the two theoretical models was
also confirmed. Whereas Baddeley’s model could offer an explanation of the dif-
ferential age-related effects of articulatory suppression in the Mr. Peanut task,
Pascual-Leone’s model could—at least to a certain extent—provide a develop-
mental account of the age-related increases in performance on the Corsi blocks
task. Although the two models converge with regard to the notion of age-related
growth in working memory capacity, they differ in their underlying assumptions.
According to Pascual-Leone, M-space increases in accordance with Piaget’s stages
of cognitive development, but Baddeley attributes the development of (verbal)
working memory to an age-related increase in articulation.
It is interesting to note that Pascual-Leone (2000) applauded Kemps et al.’s
(2000) efforts but did not agree with their conclusions. In Pascual-Leone’s view,
his model of constructive operators explains all the data analyzed by Kemps and
colleagues without the need for working memory theory. However, Pascual-Leone
referred to the original Baddeley (1986) model and not to the revised version,
which included the episodic buffer (Baddeley 2000, 2007).

Tests of Case’s Resource-Sharing Model or Working Memory

and its Competitors

Another series of empirical tests assessed the validity of Case’s (1985) resource-
sharing assumption by comparing it with predictions of models that also con-
sidered trace decay and its reactivation as additional components such as the
Baddeley’s model and the time-based resource-sharing (TBRS) model devel-
oped by Barrouillet and Camos (Barrouillet et al. 2004; Barrouillet and Camos
2007; Camos and Barrouillet 2011a). As already noted above, Towse and Hitch
(1995) had challenged the trade-off hypothesis forwarded by Case et al. (1982),
which suggests that the age-related span increase is due to an improvement in the
158 6  The Role of Basic Memory Capacities and Working Memory

processing component, which demands a smaller proportion of total processing

space as the age of the individual increases. Case thus assumed that a trade-off
between two activities (processing and storage) competes for a single limited total
processing space.
Towse and Hitch (1995; Towse et al. 1998) proposed an alternative to Case’s
limited cognitive space hypothesis. They assumed that the strength of the memory
trace in the short-term storage space weakens as the interval between storage and
recall increases. The higher counting span of older children might be due to the
fact that they count more quickly than younger children and that this speed advan-
tage reduces the time period during which the information has to be retained. The
“memory decay hypothesis” (also labeled the “task-switching hypothesis”) pro-
posed by Hitch and colleagues does not require the recourse to notions of “operat-
ing space” or the sharing of a “total processing space” for processing and storage
in order to account for the difficulty of the counting span task. Their model con-
siders the developmental increase in working memory span to be a mechanistic
by-product of the increase in processing speed and the resulting reduction in the
delay of maintenance.
To decide between the trade-off hypothesis and the memory decay (task-
switching) hypothesis, Towse and Hitch (1995) manipulated the difficulty of
the counting span task while keeping its executive time constant. The traditional
counting span task is a dual-task procedure that consists of counting target objects
that are mixed up with other objects. The difficulty of the counting task increases,
for instance, when the similarity between target objects and other objects
increases. Towse and Hitch (1995) as well as Barrouillet and Camos (2001) were
able to confirm the basic assumptions of the memory decay hypothesis when using
the counting span task. Accordingly, the developmental increases observed in the
counting span task appeared to result from shorter retention periods in older chil-
dren. Given that children count more quickly, this advantage in speed reduces the
period during which the relevant information has to be retained and thus reduces
decay.
However, Barrouillet and Camos (2001) argued that the resource hypothe-
sis is not incompatible with the memory decay hypothesis. For instance, Cowan
(1999) proposed that working memory functioning is constrained by the lim-
ited number of items that can be maintained in the focus of attention due to
a resource limitation but also by a temporal decrease in the activation of items,
the second of which is a memory decay phenomenon. Although the findings by
Towse et al. (1998), who manipulated the retention interval and held the cognitive
cost of the task constant, clearly supported the memory decay hypothesis, these
authors did not exclude the possibility that the limited resources hypothesis was
also valid. As emphasized by Barrouillet and Camos (2001), a working memory
task involving highly demanding processing is needed to more accurately evalu-
ate the resource hypothesis. Thus, they used a more complex operation span task
that required participants to perform arithmetic operations while remembering let-
ters for recall. Their findings no longer supported the memory decay hypothesis
because participants in the operation span condition showed poorer memory spans
Empirical Tests of the Various Theoretical Working Memory Models 159

than participants in an easier “baba” span condition, which required participants

to say “baba” continuously while carrying out the memory task. Please note that
the retention period was held constant across conditions. According to the strong
version of the memory decay hypothesis, the span should depend only on the
retention period no matter what the competing task is. Overall, the results were
in accordance with the limited resources’ hypothesis, which claims that the differ-
ence between the operation span and the “baba” span can be accounted for by the
increased cognitive load required to solve arithmetic problems.
Subsequent tests of the two hypotheses proposed by Case and Towse and Hitch
provided by Gavens and Barrouillet (2004) and Bayliss et al. (2005) controlling
for both the duration of the retention period and processing efficiency revealed that
the increase in processing efficiency was not the sole factor underpinning devel-
opmental changes in working memory. Individual differences in storage ability
were found to explain additional variance in children’s complex span performance.
Bayliss et al. (2005) suggested that this storage ability mainly reflects differ-
ences in the process of the refreshment of memory traces, thus supporting one of
the basic assumptions of the TBRS model (Barrouillet et al. 2004). As outlined
above, this model identifies three main sources of working memory limitations:
the amount of available attention, the phenomenon of decay, and the efficiency of
the mechanism by which memory traces are reactivated through rapid attentional
switching. Regarding the last item, the model suggests that when individuals are
not engaged in processing activities during complex span tasks, they take advan-
tage of free pauses to reactivate or rehearse the memory items. The model fur-
ther assumes that the cognitive load that a given task involves is a function of the
proportion of time during which this task occupies attention, thus preventing the
refreshment of memory traces (Barrouillet et al. 2009).
Barrouillet et al. (2009) carried out a series of experiments involving chil-
dren from 5 to 14 years of age to investigate the role of this reactivation process
in developmental differences in working memory span. As a main result, they
observed that the reactivation process is efficient from age 7 on and increases
in efficiency until late adolescence. This refreshment, which, according to the
TBRS model, involves the covert retrieval process through attentional focus-
ing described by Cowan et al. (1994), is faster in older than in younger children
such that preschoolers are unable to take advantage of the free pauses that can
occur during processing. According to the authors, this study provided basic sup-
port for the TBRS model and its main assumption of resource sharing achieved
through a rapid switching of attention. The latter process permits participants to
reactivate memory traces while performing concurrent activities, at least for chil-
dren from early school age and older. Given that rapid switching of attention was
not observed in the younger age group, Barrouillet et al. (2009) further concluded
from their findings that the Towse and Hitch task-switching model appeared to
be more appropriate for accounting for working memory in preschoolers than the
TBRS model.
In a subsequent study, Camos and Barrouillet (2011b) tested the hypothesis of
a developmental shift from a macrolevel switching with a passive maintenance
160 6  The Role of Basic Memory Capacities and Working Memory

of the to-be-remembered items (task-switching model) to a microlevel switch-

ing combined with an active strategy of refreshing memory traces (TBRS model)
in working memory. In a complex span task, 6-year-old children’s recall perfor-
mance seemed to depend only on the duration of the span task but not on the cog-
nitive load the task involved. On the contrary, the cognitive load of the complex
span task, which constrained the amount of time that was available for refresh-
ing memory traces, determined the amount of recall exhibited by the 7-year-olds
in the sample, whereas the duration of the task did not have any impact. Camos
and Barrouillet (2011b) concluded from their findings that the task-switching
model described by Towse and Hitch (1995; see also Hitch 2006) was suitable for
explaining the working memory functioning of the 6-year-olds and that the pattern
of performance found for the 7-year-olds reflected the goal-directed microlevel
switching described by the TBRS model.
The data obtained for schoolchildren and adults are basically in accordance
with the TBRS model’s core assumption that two main factors account for devel-
opmental differences in working memory spans. One factor is the time needed to
perform the task’s processing component, which determines the time during which
memory traces decay (see also Case 1985; Towse and Hitch 1995). The second
more novel factor concerns the maintenance mechanisms’ efficiency, which deter-
mines how much information can be reactivated during a given period of time.
In a series of experiments, Gaillard et al. (2011) were able to show that the sig-
nificant working span differences between third and sixth graders observed under
standard conditions are reduced but not eliminated when processing efficiency is
equated across age groups, indicating—contrary to Case’s (1985) basic assump-
tion—that working memory spans do not depend solely on processing efficiency.
The importance of the second factor is illustrated by the fact that age differences
in the memory spans of the participants in the Gaillard et al.’s (2011) study eventu-
ally disappeared when reactivation times were tailored to processing speed, that
is, when the younger children were given more time to reactivate their memory
traces (for a similar conclusion, see Cowan et al. 2010). Evidently, age-related dif-
ferences in processing speed and in the efficiency of the maintenance mechanism
both account for developmental differences in working memory span.

Empirical Tests of the Baddeley’s Model

So far, Baddeley’s model of working memory has received a lot of attention, and
numerous studies have been carried out to test its specific assumptions. The major-
ity of these studies have addressed aspects of the phonological loop, in particu-
lar relations between speech rate and memory span. As noted above, Baddeley’s
working memory model emphasizes the fact that information processing is
restricted by capacity limits and time limits. Accordingly, it places substantial
emphasis on the speed of verbal processes. Several studies on the development of
working memory have reported age differences in the speed with which words can
Empirical Tests of the Various Theoretical Working Memory Models 161

be articulated and the corresponding differences in memory span (e.g., Gathercole

and Hitch 1993; Hitch et al. 1989; Hulme et al. 1984; Roodenrys et al. 1993). For
instance, Hulme et al. (1984) demonstrated that the articulation rate increased with
age and was highest for the words with the fewest syllables. When articulation
rates were plotted against memory span scores, one linear function described the
data regardless of age. Thus, individuals from the age of 4 to adulthood recalled as
much as they could articulate in about 1.5 s. Given that the average memory rate is
so closely related to the articulation rate at the group level, a basic assumption—
representing a strong version of the working memory hypothesis and derived from
Baddeley’s model—was that developmental improvements in articulation rate are
responsible for developmental increases in verbal working memory (see also Hitch
et al. 1989, 1993).
However, subsequent studies have indicated that such a strong version of the
working memory hypothesis may not be adequate. For instance, such a version
would also imply that each individual child’s memory span should be closely
related to his or her articulation rate. Although Henry (1994) still found substantial
correlations between articulation rate and memory span when the scores of 5- to
10-year-old children were calculated at the individual level (r = 0.60), the relation
was no longer significant when the effects of age were partialled out (r = 0.15).
These findings were broadly replicated in a study by Ferguson et al. (2002) using
considerably larger samples (n = 39–58) per age group. The correlations between
verbal memory span and articulation rate within each age group ranged between
−0.03 and −0.41, indicating that the relation between the two components was
relatively moderate within age groups. Overall, these findings suggest that the
strong version of the working memory model, which assumes that articulation rate
is the only driving force behind developmental increases in verbal memory span, is
not supported even though articulation rate appears to play some role in the devel-
opment of memory span (Henry 2012). Partial support for the articulation rate
hypothesis comes from a study by Henry and Millar (1991) who asked children
aged 5 and 7 years to recall words for which the two groups had shown the same
articulation rate. As a main result, the usual age difference in memory span was
reduced to nonsignificance at least for high-frequency words. This finding is com-
patible with the articulation rate hypothesis. However, the fact that memory spans
for low-frequency words, even when matched for articulation speed, still differed
between the two age groups is not in accordance with the general hypothesis that
articulation rate is the only relevant factor.
Kail (1997) proposed a slightly revised articulation rate hypothesis, arguing
that phonological skill is important in addition to articulation rate when it comes
to predicting verbal memory span in young schoolchildren (6- to 10-year-olds).
The logic behind this assumption was that being fast at accessing phonological
representations of items in LTM is helpful during recall if the memory trace has
begun to decay. The findings of his study confirmed the basic assumption, indi-
cating that both the rate of articulation in the phonological loop and the use of
phonological information in LTM influence the development of verbal memory
span. Ferguson and Bowey (2005) replicated this finding using structural equation
162 6  The Role of Basic Memory Capacities and Working Memory

modeling in large samples of children between 5 and 13 years of age. These

authors found that measures of processing speed directly contributed to the avail-
ability of long-term phonological representations and articulation rates. In turn,
the latter two components both contributed independently to verbal memory span.
Given that the simple correlation between articulation rate and memory span in
this study was only moderate (r  = 0.44), Ferguson and Bowey argued that their
findings did not offer strong support for the working memory position.
Several other studies have referred to the importance of long-term knowledge
in the development of verbal memory span. The phonological store is assumed
to hold information about the phonological forms of each word but has a limited
capacity and decays within 2 s. Thus, most participants, when working on span
tasks, are most likely faced with partially decayed traces of phonological forms for
at least some items in the list. In such a situation, the participants may try to guess
the correct word from the partially decayed memory trace, reconstructing the cor-
rect word based on their knowledge of known word forms (i.e., neighbors that
sound similar to the actual word). This process, known as “redintegration,” was
found in experiments with adults but could also be observed in studies with chil-
dren. For instance, Turner et al. (2000) examined the influence of item familiarity
on memory span. They asked adults and children of the ages of 5, 7, and 10 years
to recall either words (familiar items) or nonwords (unfamiliar items). As a main
result, the effects of familiarity on item recall were found from 7 years on, indi-
cating that the redintegration process develops with age, emerging between 5 and
7 years. In a subsequent study, Turner et al. (2004) found that the redintegration
effect depends on the nature of the memory task. That is, the use of redintegration
increased with age for tasks requiring spoken item recall but decreased with age
when position information but not naming was required. Overall, there is evidence
that older children and adults use long-term (phonological or semantic) memory
knowledge to support their recall on verbal memory span tasks.
Another interesting aspect of the articulation rate hypothesis concerns the word
length effect, which means that fewer long words than short words can be remem-
bered when performing memory span tasks. The importance of subvocal rehearsal
has been demonstrated in several studies that used articulatory suppression and
found reduced effects of word length on memory span when rehearsal was pre-
vented by a secondary task (see Baddeley 1986). However, one problem with the
basic assumption that speed of rehearsal is the key factor that accounts for devel-
opmental increases in verbal memory span is that preventing children of different
ages from using rehearsal by applying articulatory suppression procedures does
not equalize their memory spans. For instance, Hitch et al. (1989) compared the
memory spans of 8- and 11-year-old children under normal experimental condi-
tions with spans under articulatory suppression conditions, a procedure whereby
the children had to continuously repeat an irrelevant word when conducting the
word span task. Although suppression reduced memory spans in both age groups,
it did not eliminate age differences. The findings reported by Hitch et al. (1989)
suggest that although the 8- and 11-year-olds in their sample seemed to use ver-
bal rehearsal, this factor could not be the only one responsible for developmental
Empirical Tests of the Various Theoretical Working Memory Models 163

increases in memory span (for similar findings, see Henry and Millar 1991; Hitch
et al. 1993).
Another problem with the phonological loop interpretation of verbal memory
span concerns the assumption that all children must be using verbal rehearsal
when working on memory span tasks. Such a view implies that even preschool
children as young as 4 years of age use verbal rehearsal strategies on memory span
tasks, an idea that contradicts traditional findings on memory strategy develop-
ment (see Chap. 7). The question then is how to explain the extensive amount of
data collected by researchers on word length effects in young children when the
data clearly indicate robust effects of word length on verbal memory span tasks
with children as young as 4 years of age (e.g., Henry and Millar 1991; Hulme and
Tordoff 1989; Hulme et al. 1984; Nicolson 1981). Answering this question is not
an easy task. Cowan and Alloway (2009) suggested that a slight modification of
the verbal rehearsal hypothesis could still work. That is, whether the phonologi-
cal memory trace lasts long enough to result in successful recall should depend
not only on the speed of covert rehearsal but also on the speed of overt spoken
responses. Evidence for the importance of overt response speed came from a study
by Henry and Millar (1991) who found that young children showed effects of
speech rate on memory only when there was a spoken response, whereas older
children showed this response even when the response was nonverbal. In a similar
vein, several other studies have demonstrated that the speed of spoken responses
on memory span tasks increases substantially with age in childhood (e.g., Cowan
et al. 1994, 1998).
In order to clarify the issue, Cowan et al. (2006) tested the following hypoth-
esis: If speed of recall is an important factor in the development of memory span,
then increasing children’s speed of recall should produce a corresponding increase
in memory span. In their study, a sample of 8-year-olds were administered a num-
ber of digit span tasks that required verbal recall. In the first phase of the experi-
ment, children were asked to repeat the lists at whatever speed they liked. Phase
2 included the key manipulation: Half of the children were now asked to speak
their responses as quickly as possible. Phase 3 was a replication of Phase 1. As
shown in Fig. 6.7, there were clear increases in speed of recall in Phase 2 for those
children asked to recall quickly, and this increase was also basically maintained
in Phase 3. However, there were absolutely no increases in memory span for
these children. Cowan et al. (2006) concluded from this outcome that correlations
between speed and memory of any kind must be treated with extreme caution. The
findings did not confirm their earlier assumption that developmental increases in
speed of recall were a major cause of the increase in memory span with age. Some
authors have even questioned whether word length effects have any articulatory
basis at all (e.g., Campoy 2008). The fact that such effects can occur on tasks with
no articulatory requirements presents a challenge for the working memory account
of verbal memory span development.
Taken together, the research evaluating the phonological loop interpretation of
verbal memory span development has provided mixed results. On the one hand,
most studies have confirmed that developmental increases in articulation rates are
164 6  The Role of Basic Memory Capacities and Working Memory

associated with increases in verbal memory span development. There is no doubt

that the phonological loop account of verbal working memory has been very ele-
gant and attractive, generating a great deal of research. On the other hand, several
studies have pointed to problems with the phonological loop account. For instance,
research on developmental trends in word length effects has shown that verbal
rehearsal cannot be the sole factor accounting for developmental improvements
in verbal memory span, mainly because age differences remain when rehearsal is
prevented and because rehearsal is generally not present in children under the age
of 7. Many of the problematic issues with regard to articulation rates and word
length effects are by no means resolved, and more research is needed to explore
the impact of other potentially important factors such as phonological long-term
knowledge and speed of recall (see Henry 2012, for a more detailed discussion of
this issue).
As already noted above, the majority of studies evaluating assumptions of the
Baddeley’s model have focused on the phonological loop component. In compari-
son, developmental research on the visuospatial sketch pad is relatively scarce.
The few studies dealing with this issue have either explored whether there were
different developmental trends for verbal and visuospatial working memory or
tested the assumption that the visual and spatial components of the sketch pad can
be further separated.
Regarding the issue of whether different developmental trends in verbal and
visuospatial working memory can be found, Isaacs and Vargha-Khadem (1989)
compared children’s performance on verbal memory span and visuospatial mem-
ory span, using digit span and Corsi blocks tasks. According to the authors, such
a comparison could be important in the context of memory theories that postu-
late distinct verbal and visuospatial short-term memory systems. Children rang-
ing from 7 to 15 years of age were given both forward and backward versions of
these tasks. Developmental increases in performance were found on both types of
memory span tasks. However, whereas digit span performance was influenced by
order of recall, this was not true for the Corsi blocks task on which increases of up
to an average of five and a half items were observed at age 15. This finding was
interpreted as providing support for those theories that assume the existence of a
separate visuospatial short-term memory module. Similarly, a study by Pickering
et al. (1998) showed that visuospatial working memory is distinct from phonologi-
cal working memory. The children in this study (5- and 8-year-olds) were given
memory span tasks that assessed phonological/verbal memory span (digit span
and letter span) as well as visuospatial working memory (Corsi or spatial span).
Although performance on both span tasks increased as a function of age, there
were no significant associations between these two types of tasks, suggesting that
the two “slave” systems in Baddeley’s model indeed measure different memory
systems.
Several developmental studies on visuospatial working memory have attempted
to establish whether there are separable visual versus spatial subcomponents
within the VSSP as has been found for adults (e.g., Baddeley and Logie 1999;
Darling et al. 2007, 2009). For instance, Logie and Pearson (1997) assessed the
Empirical Tests of the Various Theoretical Working Memory Models 165

development of visual and spatial working memory in children of 5–6, 8–9, and
11–12 years of age, using a visual patterns task and a spatial Corsi blocks task.
On the visual patterns task, children had to remember which squares in a matrix
had been colored in. On some trials, they had to recognize one changed square in
the matrix, and on some other trials, they had to recall the positions of all colored
squares. On the standard Corsi blocks tasks, children had to remember sequences
of blocks arranged randomly on a wooden board and pointed to sequentially by
the experimenter. The rationale for this study was that if visual working memory
and spatial working memory were found to develop at comparable rates, the view
that the VSSP can be conceived of as a unitary system would be confirmed. On
the other hand, differences in the rate of development of these two types of tasks
would indicate that there are two systems, one visual and one spatial. Logie and
Pearson (1997) reported that children’s performance on the two working memory
tasks increased with age but that there was a much steeper age-related increase for
the visual patterns task, suggesting that the visual subcomponent of the VSSP is
distinct from the spatial subcomponent and develops faster in children.
This finding was replicated in subsequent studies. Hamilton et al. (2003) exam-
ined the nature of visuospatial working memory development using conventional
visual span and spatial span measures with children between the ages of 6 and
13 years and adults between the ages of 18 and 38 years. Hamilton et al. (2003)
replicated the finding that distinct developmental rates were found for the two
span tasks: Whereas the visual component of working memory developed quite
quickly between early childhood and adulthood, the spatial working memory com-
ponent showed a slower steadier developmental improvement.
In a similar vein, Pickering et al. (2001) explored the workings of the VSSP, in
particular the unitary system assumption. In their study, memory for visual versus
spatial information was examined in three experiments with children of 5, 8, and
10 years of age. Pickering et al. (2001) followed up on Logie’s (1995) suggestion
that the VSSP consists of two subcomponents that can be thought of as a “vis-
ual cache” and an “inner scribe.” Whereas the visual cache is thought to deal with
visual information such as form and color, the inner scribe, by contrast, is pro-
posed to handle information about movement sequences. Pickering et al. (2001)
wondered whether the difference between visual and spatial memory tasks could
be better captured by contrasting what they referred to as “static” memory for vis-
ual details (no movement information) versus “dynamic” memory for movement
(space-related) sequences. To test this assumption, they designed the “matrices”
task, which used the matrix pattern of the visual patterns test described above with
a facility for presenting information that was to be recalled as either static pat-
terns or dynamic paths. This task therefore captured the key aspects of both the
visual patterns task and the Corsi blocks task. Pickering et al. (2001) also included
a “mazes” task in their first experiment. This task required children to recall two-
dimensional routes through printed mazes as either static or dynamic paths. The
inclusion of this task allowed the authors to investigate whether differing devel-
opmental increases in the static and dynamic versions of the matrices task were
generalizable to other visuospatial memory tasks.
166 6  The Role of Basic Memory Capacities and Working Memory

30

25

Total trials correct 20

Matrices static
15 Matrices dynamic
Mazes static
Mazes dynamic
10

0
5 8 10
Age group

Fig. 6.8  Mean performance on the matrices and mazes tasks as a function of age group and task
presentation format (Pickering et al. 2001, p. 404)

The major findings of Experiment 1 are given in Fig. 6.8. As shown in this fig-
ure, memory for static information was better than memory for dynamic infor-
mation for children at all age levels regardless of task. The significant interaction
between age and task format (static vs. dynamic) provided further evidence of a
developmental fractionation of visuospatial working memory, replicating the find-
ings by Logie and Pearson (1997). The other experiments in this study tested the
impact of verbal recoding on the findings but could not find any evidence for any
influence of this variable. Overall, these findings as well as those by Logie and
Pearson (1997) and Hamilton et al. (2003) provide evidence for the assumption
that visual and spatial information may be maintained by separable subsystems of
the VSSP and that visual working memory and spatial working memory develop at
different rates.
As noted by Henry (2012), the development of “central executive-loaded”
higher-order processing skills has received considerable attention in the literature.
However, not of all this work takes a “working memory” perspective as conceived
of by Baddeley and colleagues. The central executive (CE), despite being the most
important of the three components of working memory, is certainly the least well
defined. Although most researchers consider the CE to be multifunctional and
complex, there remains a considerable debate. Many authors believe that executive
skills should be subdivided into separate but loosely related areas. For instance,
Miyake et al. (2000) used a latent variable approach to explore the structure of
executive functioning in young adults. They found good evidence that set shifting,
working memory, and inhibition constitute separate factors and also reported that
performance on complex working memory tasks was predicted best by the ability
to shift between strategies or rules rather than by updates in working memory or
inhibition (for confirming evidence, see Fisk and Sharp 2004).
Although many studies have adopted the multifactorial approach of Miyake
et al. (2000), only a few have provided an empirical or theoretical justification for
Empirical Tests of the Various Theoretical Working Memory Models 167

the suitability of the model for child populations (Lee et al. 2013). Evidence from
neuroscience suggests that the prefrontal and parietal structures, which support
EFs, have long-lasting maturation schedules that continue into early adulthood
(e.g., Blakemore and Choudhury 2006). Thus, it is perhaps premature to assume
that an EF model based on adults’ data can be easily generalized to children and
adolescents. In fact, the results are somewhat less straightforward with respect to
children (see Henry 2012). Numerous studies have addressed the development of
executive function (EF; for reviews, see Best and Miller 2010; Garon et al. 2008;
Zelazo et al. 2003), but there are not that many developmental studies that have
explored the structure of EF in young children. One exception is the study by
Wiebe et al. (2008) who recruited a large sample of 243 preschool children rang-
ing in age from a little more than 2 to 6 years who completed a battery of age-
appropriate executive control tasks. Confirmatory factor analysis results indicated
that the simplest model including a single executive control factor was supported
over multifactor models in which the tasks were parsed in terms of working mem-
ory and inhibitory control demands. The authors concluded from their findings
that tasks conceptualized as indexes of working memory and inhibitory control in
fact measure a single cognitive ability in young children. In a subsequent study,
Wiebe et al. (2011) replicated this pattern of results with a sample of 3-year-olds,
thus confirming the assumption that EF in preschool children is not yet differenti-
ated into the components observed in adults. Further confirmatory evidence comes
from a study by Welsh et al. (2010; but see Hughes 1998).
Overall, findings from recent studies are equivocal, with a number of studies
failing to find evidence for differentiation into the three subtypes (e.g., Hughes
et al. 2010; Willoughby et al. 2010). Shing et al. (2010) found evidence in accord-
ance with a differentiation hypothesis: Whereas memory maintenance and inhibi-
tion were found to be undifferentiated up to 9.5 years of age, the factors became
separable in 9.5- to 14.5-year-olds. Lee et al. (2013) carried out a comprehensive
study on a large sample of 668 children and a cohort-sequential design with four
cohorts consisting of kindergarten children as well as second, fourth, and sixth
graders. The children were tested once per year across a time span of 4 years. As
a main result, the study showed that the structure of the EF varied over a 10-year
period from early childhood to the mid-adolescent years. Results of various con-
firmatory analyses indicated a process of differentiation from a two-factor struc-
ture in early childhood to a three-factor structure in the teenage years. Findings
for the 11- and 14-year-olds suggested a period of transition during which the
two- and three-factor solutions provided almost equivalent model fits. Overall,
the authors found the process of EF differentiation across 6- to 15-year-olds to be
slow, with the change from a two- to three-factor solution occurring over a pro-
tracted period.
This study and related research have thus confirmed the assumption that the
structure of EFs and their contents change with age. We know from numerous
studies that EF components relevant for adult cognitive performance such as set
shifting, updating information in memory, and inhibition undergo rapid develop-
ment during early childhood and the preschool period (e.g., Diamond 2006; Espy
168 6  The Role of Basic Memory Capacities and Working Memory

1997; Jacques and Zelazo 2001). However, it is not clear whether these are the
only EF components that determine working memory performance and whether
the structure and contents of executive functioning in children resemble those of
adults. Several studies have found evidence for differing executive skill areas,
thus making interpretation difficult. For instance, Welsh et al. (1991) tested a large
sample of children (3–12 years) on a range of EF measures and identified three
factors that were interpreted as “speeded responding,” “set maintenance,” and
“planning.” Using an even larger sample of 400 Finnish children (7–12 years),
Klenberg et al. (2001) found evidence for four EF subcomponents, which they
labeled “fluency,” “visual attention,” “auditory attention,” and “inhibition.” These
factors are somewhat difficult to interpret given the variations in the terminologies
used. Furthermore, as noted by Henry (2012), the EF tasks used in these studies
were rather complex and could have involved more than one executive function.
More recent studies have attempted to examine executive functioning in chil-
dren with respect to the tripartite structure (working memory, inhibition, and
set shifting) proposed by Miyake et al. (2000). For instance, Lehto et al. (2003)
investigated a dimension of executive functioning in a sample of 8- to 13-year-
old Finnish children using a range of EF tasks. Both exploratory and confirmatory
factor analyses yielded three interrelated factors that resembled those identified
by Miyake et al. (2000). Lehto et al. (2003) interpreted these factors (with some
reservations) as reflecting updates in working memory, inhibition, and set shift-
ing. Subsequently, Huizinga et al. (2006) also adopted the conceptual framework
of Miyake et al. (2000) to investigate developmental differences in EF, includ-
ing large samples of 7-, 11-, 15-, and 21-year-olds in their study. Huizinga et al.
(2006) found that the three EF components set shifting, working memory, and
inhibition reached adult levels between the ages of 11 and 15 years. Please note
that this finding is in accordance with the outcome of Lee et al.’s (2013) study (see
above). Structural equation modeling analyses revealed that the three components
were clearly separable as their correlations were quite modest. One problem with
the inhibition component was that the three measures assessing inhibition were not
well related to each other and did not represent a common latent variable (for a
similar problem, see van der Sluis et al. 2007).
Overall, the findings summarized above indicate that some caution must be
used in following the interpretation that executive functioning is structured simi-
larly in children and adults, given that the EF tasks used in the various develop-
mental studies described above differed in several respects from those used by
Miyake et al. (2000). However, the fact that the findings from these developmental
studies are roughly comparable supports the view that EF may be similarly struc-
tured throughout development, at least from middle school age on. Moreover, the
outcome of several developmental studies indicates that different aspects of execu-
tive skill seem to improve at different rates and that adult levels of performance
are reached earlier on measures of set shifting and inhibition than on measures of
planning and fluency (e.g., Davidson et al. 2006). Although the findings have not
been consistent, it appears that inhibition shows particularly striking improvement
during the preschool years and less change later on, whereas working memory and
Empirical Tests of the Various Theoretical Working Memory Models 169

set shifting show more gradual linear improvements with age. Apparently, some
EF components facilitate the development of other EF components. For example,
as WM and inhibition seem to develop ahead of shifting (Davidson et al. 2006),
perhaps a certain level of WM and inhibition has to be developed before children
can use them toward the development of shifting behaviors (Best and Miller 2010;
Garon et al. 2008). However, some developmental differences may depend on the
exact nature of the tasks, which vary between studies. Given that task-related fac-
tors such as the degree of complexity influence the observed developmental trajec-
tory of a latent EF component, future research needs to carefully consider how
differences in these task factors may shape the developmental trajectory of the
various EF components.
The research situation concerning the developmental relations between the
three EF components seems inconsistent. For instance, St Clair-Thompson and
Gathercole (2006) found evidence that inhibition and working memory were
separate components of EF but that switching was not. In a more recent study,
McAuley and White (2011) used a latent variable approach to assess the struc-
ture of EF in a sample of 153 individuals ranging from 6 to 24 years of age. They
confirmed St Clair-Thompson and Gathercole’s (2006) finding that inhibition and
working memory were separable factors and also found that processing speed was
a third separable factor by 6 years of age (measures assessing set shifting were not
included).
The fact that inhibition and working memory are not closely related is surpris-
ing at first glance. It seems that many inhibitory tasks, particularly complex ones,
also place demands on working memory. Inhibition is an old concept that is con-
sidered foundational for EF and working memory (e.g., Best and Miller 2010;
Miyake et al. 2000). It refers to an active suppression process such as the removal
of task-irrelevant information from working memory (Harnishfeger 1995).
According to the model of inefficient inhibition in working memory proposed by
Bjorklund and Harnishfeger (e.g., Bjorklund and Harnishfeger 1990; Harnishfeger
and Bjorklund 1994), differences in the ability to keep task-inappropriate thoughts
out of working memory influence task performance. The central idea in this model
is that young children not only have great difficulty ignoring task-irrelevant stimuli
in their environments, but that they also have a difficult time keeping task-irrele-
vant thoughts out of working memory. This problem effectively reduces functional
working memory space (for confirming evidence, see Lorsbach and Reimer 1997;
Lorsbach et al. 1998). Given the proposed theoretical relation between inhibition
and working memory, the finding that empirical correlations between (complex)
inhibition and working memory measures tend to be low is unexpected and dif-
ficult to explain.
As noted by Diamond (2013), there is disagreement between EF researchers
over whether inhibition is separate from WM or whether inhibition is a behavioral
product of exercising WM rather than a separate cognitive skill. Another view is
that WM and inhibition depend on the same limited capacity system. As a con-
sequence, increasing the demand on either affects one’s ability to do the other
(e.g., Engle and Kane 2004). Furthermore, there is disagreement regarding the
170 6  The Role of Basic Memory Capacities and Working Memory

basic structural EF–WM model. Although most EF researchers refer to WM as a

subcomponent of EFs, several researchers use the term WM more broadly so that
it becomes roughly synonymous with EFs. Finally, functions of the central execu-
tive in Baddeley’s (2000) working memory model include inhibitory control and
cognitive flexibility (e.g., multitasking, shifting between tasks).
The different theoretical views outlined above certainly complicate the situ-
ation. I agree with Diamond (2013) that reserving the term WM to mean only
holding information in mind and working with it should facilitate the discussion.
WM as defined in this sense is clearly associated with inhibition. It seems obvi-
ous that WM can support inhibitory control, for instance, in situations in which
one concentrates especially hard on the information one is holding in mind, thus
increasing the likelihood that this information will guide one’s behavior and also
decreasing the likelihood of an inhibitory error. On the other hand, inhibitory con-
trol also supports WM. In order to keep your mind focused on what you want to
focus on, you must inhibit internal and external distractions. Thus, it seems obvi-
ous that the two concepts are theoretically related but also dissociable. As noted by
Henry (2012), the working memory model proposed by Baddeley and colleagues
has rather little to say about the development of the central executive in children.
This part of the model has been rather unexplored until quite recently. The model
simply assumes that executive WM requires that either the verbal storage system
(the phonological loop) or the visuospatial storage system (the VSSP) works in
concert with a coordinating central executive (Gathercole et al. 2004). From a the-
oretical point of view, it seems to make sense to distinguish between what Henry
(2012) labels “executive-loaded working memory (ELWM),” that is, the type of
working memory that requires processing plus storage, and the other measures of
executive functioning discussed above, including fluency, inhibition, set shifting,
and planning. In the following, developmental trends in ELWM will be explored.
As already noted above, working memory involves the ability to maintain and
manipulate information over brief periods of time without relying on external cues
or aids (Alloway et al. 2006; Huizinga et al. 2006). Neuroimaging research sug-
gests that working memory tasks vary by how much the task elicits PFC activity
(Luciana et al. 2005). It seems reasonable to distinguish between more simple and
complex WM tasks when estimating the amount of executive control involved in
task solution. For instance, the digit span and word span backward tasks described
above seem to require less executive load than more complex tasks such as the
sentence (listening) span and operation span tasks. In fact, findings by Siegel and
Ryan (1989) confirm the assumption that developmental trends differ for WM
tasks that differ in cognitive load. These authors compared the performance of
7- to 8-year-olds, 9- to 10-year-olds, and 11- to 13-year-olds on a relatively easy
counting span task and a more demanding sentence span task. There were clear
improvements with age on both tasks, with higher levels of recall and a steeper
increase in performance found for the easier counting span task.
According to Gathercole et al. (2004), the executive component of WM is suf-
ficiently developed by age 6 to be used in complex span tasks. Gathercole et al.
(2004) also found that simple and complex tasks had similar developmental
Empirical Tests of the Various Theoretical Working Memory Models 171

trajectories, indicating a linear increase between the ages of 4 and 14 and a leve-
ling off between ages 14 and 15 across nearly all tasks that were examined. Please
note, however, that their definitions of “simple” and “complex” WM tasks differed
from the ones provided above in that “complex” tasks always required an interac-
tion between the storage and central executive functions (i.e., working memory),
whereas the “simple” tasks focused on the storage component (short-term mem-
ory). For instance, the digit span forward task (a STS task) was considered to
be simple, whereas the digit span backward task (classified as simple by Siegel
and Ryan) was defined as complex. Thus, the finding that similar developmental
improvements were found regardless of task complexity may be explained by the
specific definition of complexity in this study.
Results of other studies seem to confirm Siegel and Ryan’s finding that devel-
opmental trends differ for simple and complex working memory tasks. For
instance, Luciana et al. (2005) used a battery of nonverbal (visuospatial) WM
tasks and found that the developmental course of WM depended on the complexity
and thus the executive demands of the task, with the less demanding tasks mas-
tered earlier in development. Conklin et al. (2007) extended the work by Luciana
et al. (2005) by using a battery of verbal and spatial WM tasks across the ages
of 9–17 years. Interestingly, children and adolescents tended to perform better on
the spatial than on the verbal WM tasks even though the tasks were thought to
tap similar cognitive processes. Overall, however, the developmental trajectories
were similar between the two types of WM tasks and differed according to task
complexity.

Neuropsychological Evidence

Although the vast majority of research on working memory has been carried out
using solely behavioral experiments, some studies have explored working mem-
ory and its development at both behavioral and neural levels. Such research has
used neuroimaging techniques, such as positron emission tomography (PET)
and functional magnetic resonance imaging (fMRI), to determine which brain
areas are active when people perform working memory tasks and to determine
which cognitive functions are mediated by these areas. For instance, Smith and
Jonides (1997) used PET techniques to explore the architecture of working mem-
ory, in particular the distinction of multiple memory systems as featured in the
Baddeley’s model. In a series of sophisticated neuroimaging studies with adults,
they found that different neural circuits mediated verbal, spatial, and object work-
ing memory, with spatial memory being right lateralized and verbal as well as
object memory being left lateralized. In addition to lateralization, some of the
cortical regions involved in the various kinds of working memory seemed to
be distinctive. That is, there was activation in an occipital region when the task
tapped spatial memory, and there was activation in Broca’s area when verbal
working memory was assessed. Moreover, within the circuitry for verbal working
172 6  The Role of Basic Memory Capacities and Working Memory

memory, there was a separation between the components underlying storage (the
posterior parietal cortex) and those mediating rehearsal (frontal regions such as
Broca’s area and the premotor area). In contrast to storage and rehearsal func-
tions, the processes that operated on the executive function seemed to be medi-
ated by the dorsolateral PFC. Smith and Jonides (1997) concluded that their
results provide a neural basis for cognitive models of working memory and also
support the assumption of distinctive features emphasized in Baddeley’s model of
working memory.
What about the neural basis of developmental trends? The bulk of the child
neuropsychological literature clearly supports the view that the neural structures
that support working memory, particularly the PFC, undergo protracted devel-
opment (e.g., Diamond and Goldman-Rakic 1989; Luciana and Nelson 1998).
Developmental studies that have explored structural and functional changes in the
brain related to working memory development are still rare, mainly due to the fact
that many of the imaging procedures used in adults cannot be easily applied to
studies of normally developing children. For instance, it is difficult to use fMRI
in children younger than 5–6 years of age, primarily because of their inability to
sit still for a long enough period of time. Thus, most of the developmental stud-
ies that have assessed the functional neuroanatomy of working memory were car-
ried out with schoolchildren and adolescents. Most of them also focused on the
development of nonverbal visuospatial working memory. For instance, Nelson
et al. (2000) used fMRI to examine spatial working memory in 8- to 11-year-
old children, with the goal of examining the functional anatomical organization
of nonverbal working memory during middle childhood. Findings on a spatial
working memory task that has previously been used with adults demonstrated that
children showed a pattern of brain activation remarkably similar to that of adults.
Activation was not restricted to the posterior parietal region but also included the
dorsal aspects of the frontal cortex (middle and superior frontal gyri). The find-
ings differed from those obtained in an earlier study (Casey et al. 1995) in that
less activation was seen in the ventral areas of the PFC. Overall, however, the evi-
dence from the Casey et al.’s (1995) and Nelson et al.’s (2000) studies paralleled
the results obtained from studies of spatial working memory in adults. Apparently,
the neural substrate thought to underlie spatial working memory may be adultlike
prior to the onset of puberty.
There are only a few neuropsychological studies that have focused on a wider
age range, including both children and adolescents. Klingberg et al. (2002) used a
visuospatial working memory task that was sensitive to measuring developmental
changes in WM capacity. A total of 13 children and adolescents between 9 and
18 years of age performed the task, while their brain activity was measured with
fMRI. In order to minimize interindividual differences in behavior during scan-
ning, the authors intentionally created a ceiling effect by keeping the WM load
low. Differences in WM capacity were assessed using a similar task outside the
scanner, and this task was entered as a covariate (together with age) in the vari-
ous analyses. The main result was that the adolescents showed higher cortex
activation than the younger children did. A positive correlation with age was
Empirical Tests of the Various Theoretical Working Memory Models 173

found bilaterally in the superior frontal sulcus, in the intraparietal and superior

parietal cortex, and in the left occipital cortex. Furthermore, a correlation between
WM capacity and WM activity was found in the left superior frontal sulcus and
in the left intraparietal cortex. Klingberg et al. (2002) inferred from their findings
that the correlation between age and WM-related activity was due to the age trends
in WM capacity. They also noted that there are several maturational processes
that coincide in time with the increase in WM capacity. The parietal area is one of
the last in the brain to myelinate and is also involved in a wide range of activities
requiring attention and spatial WM. According to the authors, the development of
functionality in these areas may be closely linked to increases in WM capacity and
the development of cognitive abilities during childhood and adolescence.
In a related study, Scherf et al. (2006) explored whether similar kinds of func-
tional reorganization in the brain support visuospatial working memory develop-
ment from childhood to adulthood. Thus, the goal of their study was to investigate
developmental changes in functional brain circuitry that support WM improve-
ments. Scherf et al. (2006) proposed that the early acquisition of WM requires
functional specialization within brain regions (e.g., parietal regions), maintenance
of representations (e.g., dorsolateral prefrontal cortex = DLPFC), and the ability
to act on those representations (e.g., premotor regions). On the other hand, adult-
level performance on WM tasks should involve both specialization within these
necessary regions and efficient functional integration between these regions.
A total of thirty 8- to 47-year-olds (three age groups: children, adolescents, and
adults) were tested on a visuospatial WM task using fMRI. Although participants
from all three age groups performed well, the children tended to be less accurate
than the adolescents and adults. Moreover, quantitative and qualitative changes in
functional brain circuitry were observed across the three age groups. Quantitative
changes reflected an age-dependent refinement of local connectivity and compu-
tations within essential WM regions such as the DLPFC, anterior cingulate, and
caudate nuclei. Qualitative changes reflected the refinement of functional connec-
tivity among more distal regions, such as the incorporation of premotor and infe-
rior parietal regions. The authors concluded from their findings that in the course
of these developmental transitions, both children and adolescents appear to com-
pensate for less integrated functional connectivity within a region by relying on
regions that play a much smaller role in the adult network. Apparently, the matu-
ration of adult-level cognition involves a combination of increasing localization
within the necessary regions and their integration with performance-enhancing
regions.
As noted by Klingberg et al. (2002), one obvious problem with the use of dif-
ferent tasks to manipulate the cognitive load or the executive demands on WM
is that it is difficult to ensure that non-EF processes are equivalent across tasks
and that they do not influence age differences in performance (see also Best and
Miller 2010). Luciana and Nelson (1998) addressed this important issue by using
only one task that varied across trials in its cognitive demands. They showed that
the development of ELWM occurred gradually from childhood on with contin-
ued refinement through adolescence, particularly for complex task versions that
174 6  The Role of Basic Memory Capacities and Working Memory

required the maintenance and manipulation of multiple items. In accordance with

the behavioral results, evidence from neuroscience (fMRI) supports the view that
ELWM shows a protracted developmental course that leads to localized activity
within the PFC (Kwon et al. 2002).
Overall, the findings from neuropsychological and behavioral studies on
ELWM indicate that performance on complex WM tasks, that is, those requir-
ing a great deal of processing while maintaining information in either the
phonological store or the VSSP, improves steadily and almost linearly from
childhood to adolescence. As noted above, neuropsychological research indi-
cates that there are both qualitative and quantitative changes in the brain acti-
vation patterns underlying visuospatial WM from late childhood to adulthood
(e.g., Scherf et al. 2006), with a shift in activity from the ventromedial regions
to the frontal regions (including the right dorsolateral PFC) found in adoles-
cence and a shift to more localized activities from the right to the left dorsolat-
eral PFC observed during the transition from adolescence to young adulthood.
Empirical evidence regarding the issue of whether such changes in the neural
system are also accompanied by behavioral changes is both scarce and incon-
sistent: Although Luciana et al. (2005) and Alloway and Alloway (2013) found
that improvements in visuospatial working memory continued until adult-
hood, Schneider et al. (2009) did not observe any developmental change in a
complex verbal span measure between late adolescence and young adulthood.
Alloway and Alloway (2013) reported that performance peaked by the age of
30 and that there was relatively little change across the life span such that 70-
to 80-year-olds performed at levels that were comparable to 14- to 19-year-old
teenagers.
One interesting aspect of combined neuropsychological and behavioral
research on working memory development concerns the question of whether brain
activity data can play a unique role in predicting children’s WM capacity. Findings
from a recent longitudinal study by Ullman et al. (2014) indicate that this may be
the case. These authors recruited 62 participants ranging in age from 6 to 20 years
for their experiment. Cognitive tests included several WM tasks and a reasoning
task and were provided at two measurement occasions within a 2-year interval.
Imaging data (fMRI) were collected at Time Point 1. Using a rather sophisticated
methodological procedure, the authors were able to show that the neuroimaging
data could explain variance in the second working memory assessments in addi-
tion to that explained by the cognitive tests at baseline (t1). Accordingly, the neu-
roimaging prediction indicated that there was unique fMRI-derived information
about future capacity that could not be derived from the working memory tasks
themselves. These findings thus suggest that future cognitive development can be
predicted from anatomical and functional fMRI signals above and beyond what
can be explained by cognitive tests. The specific predictive ability of neuroimaging
may be due to the fact that it is possible to distinguish a subcortical system pre-
dicting future cognitive development from a cortical system associated with cur-
rent cognitive capacity. Clearly, more research is needed to explore this interesting
hypothesis.
Empirical Tests of the Various Theoretical Working Memory Models 175

Evidence of the Episodic Buffer

Given that the episodic buffer is the most recent component of Baddeley’s work-
ing memory model, it does not come as a surprise that there has been little empiri-
cal research that has attempted to test the assumption that there is a separate
episodic buffer component in typically developing children. As noted above, this
component of working memory is supposed to be a limited capacity system that
depends heavily on executive processing but that differs from the central executive
in that it is primarily concerned with the storage of information rather than with
attentional control. It is capable of binding information from a number of differ-
ent sources into chunks or episodes and hence the term “episodic.” It was con-
ceptualized as a buffer in the sense of providing a way of combining information
from different modalities into a single multifaceted code (Baddeley 2000, 2003).
It also provides access to LTM, which can be used to improve working memory
performance. According to Baddeley (2003), developing highly constrained prose
recall tasks could be a suitable strategy for testing this component. Instead of the
usual narrative structure of prose recall tests, such more constrained tasks employ
a simple and standardized matrix structure involving a few episodes describ-
ing events that can be combined into a coherent episode of a specified range of
features. Given that all of these features can be addressed by specific probe-type
recall questions, this constrained prose recall procedure has considerable potential
for the study of long-term forgetting in children and adults (Baddeley et al. 2013).
To my knowledge, the first study attempting to investigate the episodic buffer
in children was conducted by Alloway et al. (2004). These authors argued that the
recall of spoken sentences should be a suitable paradigm because repeating sen-
tences involves the integration of information from temporary memory (to sup-
port the verbatim recall of individual words and their order) with the products
of semantic and syntactic analysis by the language processing system. Thus, this
recall measure was viewed as a way of capturing how a child might integrate or
“bind” information from the phonological loop with knowledge and analyses car-
ried out on the sentences by the language processing system. Alloway et al. (2004)
assessed a large sample of more than 600 children ranging between 4 and 6 years
of age. All of the children were given several memory span tasks, nonverbal abil-
ity and phonological awareness measures, as well as two sentence repetition tasks.
Confirmatory factor analyses testing the assumptions of the Baddeley (2000)
working memory model indicated that a modular structure including an episodic
buffer distinct from both the central executive and the phonological loop fits the
data well, even though sentence repetition was highly associated with both the
phonological loop and the central executive factors. Alloway et al. (2004) inter-
preted their findings as being compatible with the view that sentence repetition
taps the episodic buffer and that the buffer integrates information from working
memory, LTM, and language processing systems.
The view that the episodic buffer is in place before children enter school
was also confirmed by Sluzenski et al. (2006) who examined feature binding in
176 6  The Role of Basic Memory Capacities and Working Memory

4- and 6-year-olds. Sluzenski et al. found that the ability to bind information about
different aspects of pictures (animals and backgrounds) increased with age, an
outcome that was also confirmed for a verbal binding task. They also noticed that
there were modest correlations between the two binding tasks, and this could be
interpreted as evidence in favor of a general episodic buffer component that binds
information from different sources together (Henry 2012).
Age differences in feature binding were also investigated by Cowan et al.
(2006) who looked at older children (8–10 years of age) and compared them with
younger and older adults on a task that required the binding of visual object infor-
mation with spatial information. Cowan et al. (2006) found that children were less
likely to notice a binding change than adults, although the poorest performance
was shown by older adults. Binding appeared to improve with age into young
adulthood and then showed a decline in older adults. In this respect, it seems
important to note that Shing et al. (2008) argued for a two-component model of
episodic memory, including an associative component that appears similar to the
concept of binding discussed above as well as a strategic component. Using a
paired-associate learning paradigm, Shing et al. (2008) found that the associative
component seemed to develop earlier in children ranging between 8 and 15 years
of age than the strategic component. In their view, children possess a “readily
functional” associative ability that can be used very effectively by schoolchil-
dren. By contrast, strategic binding draws on executive skills, which are slower to
develop, with the potential for improvements until young adulthood. Subsequent
research has supported this two-component framework of episodic memory devel-
opment from a life span perspective, indicating that the associative component is
relatively mature by middle childhood and declines in old age and that young chil-
dren’s difficulties with episodic memory tasks are primarily due to their low levels
of strategic operations (Shing and Lindenberger 2011; Shing et al. 2010).
In sum, the numerous attempts to evaluate the components of Baddeley’s work-
ing memory model were moderately successful, indicating that it makes sense
to distinguish between the various parts of the theoretical framework. However,
findings also showed that several questions remain regarding the developmental
aspects of the model and the explanation of developmental changes, an issue that
should be tackled in future research.

Evaluation of Cowan’s Working Memory Model

Like Baddeley’s model, the model of working memory developed by Cowan is

inherently hierarchical in its structure as it distinguishes between long-term rep-
resentations, activated LTM representations, and the focus of attention, a subset
of activated representations. The focus of attention is assumed to function in a
manner that may be similar to Baddeley’s episodic buffer except that Baddeley
was not very precise about the attentional requirements of that buffer (Cowan and
Alloway 2009). According to Cowan and colleagues (e.g., Cowan et al. 2002),
Empirical Tests of the Various Theoretical Working Memory Models 177

working memory is not just the activated portion of LTM. Rather, it also requires
feature binding processes that are necessary, for instance, to recover the tempo-
ral sequence in which events take place and to mark other episodic information
(e.g., determining which items were activated after others). In comparison with
Baddeley’s model, Cowan’s model is less modular and emphasizes memory pro-
cesses more than memory domains, proposing that the activation of features is
fundamental for the memorability of stimuli. Although the two models are mov-
ing closer to each other in the postulation of an episodic buffer for working mem-
ory, several points of divergence remain (Towse and Cowan 2005). For instance,
whereas Baddeley’s (2000) model emphasizes that working memory systems per
se may not undergo major structural changes with development, Cowan suggested
that acoustic information may be lost more rapidly in younger children and that
the capacity of the focus of attention and the rate of transfer of information into
that focus of attention probably also change with age (Cowan et al. 2000, 2002).
Cowan has questioned the idea that a central global processing rate explains work-
ing memory development. For instance, Cowan (1999) suggested that there may
be separate processing rate parameters involved in memory search activities and
in phonological processing operations, explaining why these predict independent
sources of variance in children’s memory performance.
Although a far greater body of research has been built on Baddeley’s frame-
work, several studies have tested major assumptions of Cowan’s model. One
important aspect of Cowan’s (1999) model of working memory concerns the idea
that even in adults, the focus of attention is quite limited in task settings that pre-
vent participants from using rehearsal and from combining items to form larger
chunks of information. Cowan (2001) demonstrated that in such cases, the aver-
age capacity of the focus of attention for normal adults is about four unconnected
chunks in both verbal and visuospatial domains.
As noted by Cowan (2014), we still do not have a clear understanding of
why pure memory capacity (without the involvement of strategic processes such
as rehearsal or chunking) is limited to three to five items in adults regardless of
whether items stem from verbal or visuospatial domains. According to the model
shown in Fig. 6.6, all of the capacity supposedly comes from what can be held
in the focus of attention. From a developmental perspective, an important ques-
tion concerns the issue of whether the span of apprehension represents an absolute
capacity of short-term memory and whether it increases with age.
Support for such ideas comes from a developmental study carried out by
Cowan et al. (1999). In the critical condition of the experiment, first and fourth
graders as well as adults played a computer game. Over earphones, they also heard
series of digits that they were asked to ignore. Occasionally and unexpectedly,
they were asked to recall the most recently presented set of digits they had heard
in the exact order in which the digits were presented. Given that participants were
not explicitly attending to the digits, it was unlikely that they used strategies to
remember them. Thus, performance on this task seemed to be a fair test of the
span of apprehension. The average span of apprehension was about 3.5 digits for
adults, about 3 digits for the fourth graders, and about 2.5 digits for the first-grade
178 6  The Role of Basic Memory Capacities and Working Memory

children. Cowan and colleagues interpreted these significant age differences as

reflecting a true developmental difference in the capacity of STS.
These findings were confirmed in a subsequent study (Cowan et al. 2000) on a
sample of second graders, fifth graders, and adults using a different experimen-
tal design that prevented rehearsal and also using varying retention intervals until
the memory tests were given. This study tested the hypothesis that developmental
improvement in working memory occurs in part because there is a faster decay of
information in younger children. Cowan et al. (2000) again examined memory for
lists of spoken digits that were to be ignored during their presentation and thereafter
so that spoken stimuli could not be rehearsed. The retention interval was 1, 5, or 10 s
after the end of the list until a memory test was given. The participants (second grad-
ers, fifth graders, and adults) played a silent picture-name-rhyming computer game
that occupied their attention and speech sound phonological encoding processes dur-
ing the presentation of the digit lists. Although there was no test of most spoken lists,
the rhyming game was occasionally interrupted, and participants were asked to recall
the digits in the most recent list. The difficulty of the lists was equated across partici-
pants by presenting each participant with list lengths that equaled his or her memory
span. As shown in Fig. 6.9, there was substantial forgetting of the ignored lists with
increasing retention intervals regardless of age group (solid squares). Almost no for-
getting was observed for attended lists (open circles), thus showing that there is a
substantial benefit of attention for memory. Given that there were no age differences
in forgetting rates and decay of ignored lists, the authors concluded from their find-
ings that developmental differences in decay cannot explain developmental improve-
ments in memory span or working memory performance. The results showed that
working memory capacity increased with development, from just over two items in
the early elementary school years to an adult capacity of about four items on average.
Another interesting issue that was addressed in some of Cowan’s work con-
cerns the question of whether the number of slots in working memory stays the
same across age groups but with more efficient use of the slots in older individu-
als, or whether older individuals have more slots in working memory (i.e., more
basic capacity) than younger ones. As is well known from research on expertise
development (see Chap. 8), increasing the efficiency of encoding can produce
larger chunks and thus yields better recall performance. In a study by Gilchrist
et al. (2009), first graders, sixth graders, and college students were presented sets
of simple unrelated sentences for verbatim recall. Gilchrist et al. (2009) judged the
efficiency of chunking by assessing how well the words in a remembered clause
hung together. Individual differences in efficiency were assessed by comparing the
age groups on how much of a clause was recalled after it had been accessed. As a
main result, no age difference in this efficiency was found. Nonetheless, there was
still a substantial age difference in the number of clauses accessed. That is, young
children accessed (recalled words from) fewer clauses than did older children or
adults, but no age differences were found in the proportion of words recalled from
accessed clauses. Gilchrist and colleagues took this finding to indicate develop-
mental growth in the number of slots in working memory (for confirming evi-
dence, see Cowan et al. 2010).
Empirical Tests of the Various Theoretical Working Memory Models 179

0.9
Proportion Correct Recall

0.8

0.7

0.6
Ignored
0.5 Speech
0.4 Attended
Speech
0.3

0.2

0.1

0
1s 5s 10s 1s 5s 10s 1s 5s 10s
Grade 2 Grade 5 Adult

Ignored Lists
Final Serial Position
1
0.9
Proportion correct

0.8
0.7 Grade 2
0.6 Grade 5
0.5 Adult
0.4
0.3
0.2
0.1
0
1s 5 s 10 s
Test Delay (seconds)

Fig. 6.9  Results of Cowan et al. (2000) on developmental change in the loss over time of unat-
tended digit lists. Top panel (Cowan et al., Fig. 2, p. 163), all serial positions; bottom panel.
(Cowan 2014, Fig. 4), last serial position only

All in all, the empirical findings presented by Cowan and colleagues are con-
sistent with Baddeley’s and Cowan’s types of working memory models, which
allow for a general cross-domain type of storage, more domain-specific interference
mechanisms of storage, decay over time, and attention-demanding executive pro-
cesses as possible sources of developmental change. Concepts such as chunk capac-
ity, speed, knowledge, strategies, control of attention, and differences in the decay
of memory traces all contribute to the development of working memory throughout
the life span (Cowan 2014; Cowan and Alloway 2009). It remains to be determined
in future research how the various components are interrelated in different stages of
development and whether the number of causal concepts can be eventually reduced.
180 6  The Role of Basic Memory Capacities and Working Memory

Summary

This overview of developmental research on basic memory capacity and working

memory shows that the area has been very active since the early 1980s. On the
one hand, this is certainly related to the fact that theoretical models such as those
developed by Baddeley, Case, Cowan, and Pascual-Leone have attracted a lot of
attention and have stimulated empirical research to test basic assumptions. On the
other hand, this research has also demonstrated that the concept of working mem-
ory has been important to other domains. Working memory reflects the ability to
hold small amounts of information in a temporary store and to act on that informa-
tion for the purpose of solving problems. Working memory is thus foundational
to the execution of many other cognitive tasks, and the capacity of a person’s
working memory would seem to put limits on the type of higher-order cognitive
operations one can perform (Bjorklund and Sellers 2014). Its relevance has been
demonstrated in domains such as learning disabilities as well as reading and math-
ematics (e.g., see Bull and Espy 2006; Cowan and Alloway 2009; Gathercole et al.
2006; Henry 2012; Schuchardt et al. 2008; Welsh et al. 2010), supporting the view
that working memory is central to many areas of cognition.
Although Schneider and Pressley (1997) stated that there is a lack of consensus
in the domain, more recent findings indicate that the situation has improved mean-
while. Most researchers agree that the “total processing space” is not invariant but
rather increases steadily with age (e.g., Cowan 2014; Halford and Andrews 2011;
Pascual-Leone 2000). Several factors such as knowledge and strategy use have
been found to influence performance on memory span tasks. If precautions are
taken to prevent the use of strategies such as rehearsal and chunking, the result-
ing memory spans vary between 2 and 4 items and thus are much lower than the
magic number 7 suggested by Miller (1956).
Numerous studies have demonstrated that developmental changes in informa-
tion processing speed affect developmental changes in memory capacity, suggest-
ing a rather global mechanism or a “cognitive primitive,” which has been assumed
to be domain-general by several researchers. Overall, however, the numerous tests
of working memory models do not confirm the view that a single set of resources
influences all aspects of working memory performance. Different developmental
patterns for simple and complex span tasks were found, and there is also plenty
of evidence from behavioral and neuropsychological studies that verbal working
memory and visuospatial working memory show different developmental trends.
Regarding the latter component, it appears that visual working memory develops
faster than spatial working memory, indicating that the processing of visual and
spatial information may be maintained by separable subsystems of the VSSP in
Baddeley’s model and suggesting a fractionation of these two subcomponents.
Recent advances in working memory modeling have also shown that individual
differences in both processing speed and memory trace decay account for devel-
opmental differences in complex span tasks (Barrouillet et al. 2009). Although
there has not been much research on the development of executive control and
Summary 181

the episodic buffer in the context of the revised Baddeley’s model, the available
evidence supports the assumption that developmental improvements in executive
control and feature binding abilities contribute to increases in both verbal work-
ing memory and nonverbal working memory. Similarly, there is also plenty of evi-
dence supporting Cowan’s assumption that the capacity of the focus of attention
changes with age.
Although it is still unknown why developmental growth in working memory
capacity occurs, there is reason to assume that different factors such as brain
growth, knowledge increases and strategy acquisition, developmental changes in
information processing speed, and rates of memory trace decay all contribute to
this developmental process. Overall, then, age differences in basic memory capac-
ity and working memory are influenced by maturational and experiential factors.
Maturational factors place biological limits on how quickly children can process
information and retain items in their STS. However, speed of processing is also
influenced by experiential factors (e.g., knowledge base), making it clear that
the development of basic memory abilities is a result of the dynamic interaction
between biological and experiential factors that vary over time.
Chapter 7
The Development of Encoding
and Retrieval Strategies

As already noted in Chap. 2, the “cognitive turn” in North American experimen-

tal research stimulated developmental psychologists’ interest in children’s infor-
mation processing skills. Since the mid-1960s, numerous studies have investigated
the role of strategies in memory development. Research interest in strategies such
as rehearsal and semantic organization was partially motivated by the crucial role
that such control processes play in general multistore memory models.
The modern era of memory development began with the pioneering work
by Flavell and his associates on the study of strategies (Flavell et al. 1966) and
dominated the field well into the 1980s (see the example in Fig. 7.1). Historically,
examinations of children’s memory strategies can be traced to mediation theories
developed by neo-behaviorists. In framing their research questions concerning
deliberate memory in children, John Flavell and colleagues were influenced by
mediational approaches to the study of learning in which the central issues con-
cerned children’s use of words as mediators of performance on various discrimina-
tion learning tasks (see Harnishfeger and Bjorklund 1990; Howe and O’Sullivan
1990, for a historical sketch). Their early research was concerned with children’s
use of verbal mediators in the storage and recall of verbal information: “In the pre-
sent context, ‘mnemonic mediation’ will be somewhat arbitrarily restricted to refer
to those cognitive activities which could be deliberately undertaken for the pur-
pose of storing and retrieving input … Developmental transitions from nonproduc-
tion to in our own studies” (Flavell 1970, p. 193f.).
Flavell and colleagues focused in particular on conscious and intentional mem-
ory activities, although they always recognized the importance of incidental learn-
ing for the development of children’s knowledge.
John Flavell’s pioneering work established rehearsal and organization as strat-
egies that develop between 5 and 10 years of age (e.g., Flavell 1971). Memory
strategies were broadly defined as mental or behavioral activities that achieve cog-
nitive purposes and are effort-consuming, potentially conscious, and controllable

© Springer International Publishing Switzerland 2015 183

W. Schneider, Memory Development from Early Childhood Through
Emerging Adulthood, DOI 10.1007/978-3-319-09611-7_7
184 7  The Development of Encoding and Retrieval Strategies

Fig. 7.1  John Flavell
working with Christof
Schneider on a memory
recognition task. (Stanford
1982)

(Flavell et al. 2002; Pressley and Van Meter 1993). During the 1970s and 1980s,
children’s strategy use was regarded as the major source of developmental dif-
ferences in memory performance. Around that time, strategic memory was at the
center of early investigations in this area. This continues to be the case today,
although the topic no longer dominates the field as it once did (see Bjorklund et al.
2009; Pressley and Hilden 2006).
Strategies can be executed either at the time of learning (encoding) or later
when information is accessed in long-term memory (retrieval). The majority
of developmental studies have dealt with encoding strategies, such as rehearsal,
organization, and elaboration strategies carried out during study. Other investiga-
tions have focused on retrieval strategies, that is, the processes by which informa-
tion is accessed in long-term memory. The early studies on encoding and retrieval
strategies explored the extent to which children of different ages produce such
strategies and whether these “mediators” seem to “work.”
Identification of deficiencies in strategy use. Flavell’s group studied two
potential difficulties in great detail. One possibility is that young children do not
produce mediators that can effectively promote learning (i.e., they have a produc-
tion deficiency). Alternatively, children might produce mediators while learning,
but the mediators fail to facilitate performance as they do for older children and
The Development of Encoding and Retrieval Strategies 185

adolescents (i.e., the children have a mediation deficiency; see Reese 1962). As
Flavell (1970) and many others (e.g., Brown and DeLoache 1978) have demon-
strated, mediation deficiencies in young children’s memory strategies are quite rare,
and most difficulties are described as production deficiencies. That is, many young
children initially fail to spontaneously generate relevant task-appropriate strategies
but are able to produce such strategies in response to prompts or training.
The usefulness of the distinction between production and mediation deficiencies
was made apparent in the study by Keeney et al. (1967). First, it was determined that
some first graders used a verbal rehearsal strategy during list learning, whereas other
first graders failed to do so. Memory was clearly better for those children who pro-
duced the rehearsal strategy. However, when the production-deficient children were
taught the rehearsal strategy, their memory improved dramatically. Thus, the children
who failed to use the strategy spontaneously had a production deficiency but not a
mediation deficiency. Flavell (1970) concluded from these findings that production
inefficiency may be a fairly frequent precursor to efficient strategy production.
More recently, Miller (1990; Miller and Seier 1994; see also Bjorklund and Coyle
1995; Bjorklund et al. 1997) identified a third type of inefficiency that he labeled
a utilization deficiency. A child showing such a problem can spontaneously gen-
erate a strategy in response to a request for remembering, but the strategy either
does not “work” or it works but is not as effective as the strategies of older children.
According to Miller, Bjorklund, and colleagues, a utilization deficiency occurs dur-
ing the early phases of strategy acquisition. Although many developmental research-
ers have embraced this concept, there has been substantial debate about its frequency
and meaning (Bjorklund et al. 2009; Schneider 2011; Waters 2000). Recent research
has concluded that a utilization deficiency is context-dependent and constitutes only
one of several possible patterns of relations between strategy use and recall. For
instance, Schwenck et al. (2007) administered a series of sort-recall trials to first-
and third-grade children. Some children were taught how to use an organizational
strategy, whereas others were not. A utilization deficiency would be indicated if chil-
dren’s strategy use increased over trials but their levels of recall did not. This pat-
tern was observed for several first and third graders. However, other children showed
increases in both strategy use and recall on adjacent trials as would be expected if
strategy use had a positive impact on memory performance. Interestingly, still other
children showed an increase in recall in the absence of an increase in strategy use.
Although this research showed that utilization deficiencies are real, it also confirmed
the view that such a deficiency constitutes only one of the possible patterns of rela-
tions between strategy use and recall (Bjorklund et al. 2009).
As noted by Ornstein and Light (2010), the use of strategic deficiency con-
structs has played an important role in focusing attention on the processes that are
central to performance on deliberate memory tasks, and these constructs have also
helped to identify the nature of problems faced by young children in comparison
with older children and adolescents. However, in order to understand the nature of
young children’s problems when dealing with sets of to-be-remembered materials,
it seems important to focus on actual behaviors to the extent to which this is pos-
sible (see also Ornstein et al. 1988; Waters 2000).
186 7  The Development of Encoding and Retrieval Strategies

Evidence of Strategic Behavior in Young Children

Early research on strategy development has provided little evidence that inten-
tional strategic memorizing occurs before 6 years of age (Schneider and Pressley
1989). The problem with the early research was that it followed a “modal mem-
ory” study model (Brown and DeLoache 1978; DeLoache 1980). That approach
involved presenting a sample of very young children with a memory task that is
composed of unfamiliar materials that are presented in an unfamiliar context. In
general, researchers have used the nonstrategic behavior produced by preschool-
ers in such situations as a baseline against which the progress of older children
has been demonstrated. The basic problem with such an approach is that little was
learned about preschoolers’ competencies; rather, we learned only about what they
could not or would not do. Brown et al. (1983) noted that two major techniques
need to be used to demonstrate early competence: (a) simplifying the tasks as
much as possible in order to reveal their cognitive requirements in the easiest way,
and (b) situating the experiment in a familiar context. From the early 1980s on,
several studies have followed this advice and have been able to indicate that pre-
schoolers have more strategic competence than was originally assumed.

Evidence of Early Retrieval Strategies

It was already shown in Chap. 3 that very young children are relatively good at
finding hidden objects, indicating that they can use retrieval strategies in search
contexts. Intentional memory for locations was researched intensively during the
1980s. For instance, Sophian and Wellman (1983) focused on young children’s
emerging understanding that objects are not always found where they were hidden,
using the “A-Not-B” paradigm described in Chap. 3. Children who were 30 months
old understood this fact better than children who were about 24 months old, even
though the younger children already showed appropriate search strategies.
Other research showed that even children younger than 2 years of age can use
appropriate retrieval strategies under favorable circumstances. DeLoache et al. (1985)
asked 18-month-olds to remember the location of a familiar toy animal (Big Bird)
that was hidden under a pillow in a room. Although children were then distracted
with attractive toys for several minutes, they frequently interrupted their activities
to check Big Bird’s hiding place, using a number of rudimentary strategies such as
pointing, “peeking,” and naming to be able to retrieve the toy after the delay. These
activities seemed to be intentional attempts to remember the locations because they
did not occur in a control condition in which Big Bird was put on top of the pillow.
Although the deployment of these behaviors was not related unambiguously to suc-
cessful remembering, their use suggests that young children already have a rudimen-
tary understanding of the need to do “something” in response to a memory demand.
With harder tasks, however, intentional preparation for retrieval is not observed
until later. In a study by Wellman et al. (1975), the children’s main goal was to
Evidence of Strategic Behavior in Young Children 187

remember where an object was hidden—under one of three identical cups—from the
time an experimenter left the room until her return. Three-year-olds but not 2-year-
olds showed what appeared to be consciously intentional memory behaviors including
watching and touching the correct cup. According to Ornstein and Light (2010), these
behaviors may be characterized as “proto-strategies,” that is, potential precursors
of effective strategies that develop during enjoyable activities in highly salient and
meaningful situations. Overall, the findings summarized above confirm that young
children’s memory is particularly good for spatial locations. This is true not only for
hide-and-seek tasks but also for reconstruction tasks (Ellis et al. 1987) and the game
“Concentration” (Baker-Ward and Ornstein 1988; Schumann-Hengsteler 1996).
In a more recent study with young children, Haden et al. (2011) used a multitask
battery tapping nonverbal memory and language skills in a sample of 60 children
at 18, 24, and 30 months of age. The unique aspect of this study was that the same
memory tasks (elicited imitation, hide-and-seek task, working memory for loca-
tion) were presented thrice in a longitudinal design. Thus, it was possible to explore
whether the three tasks showed similar interrelations at different measurement
points. Overall, Haden et al. (2011) found moderate interrelations among memory
measures regardless of measurement point. Interestingly, individual differences in
language skills were related to performance on the nonverbal memory tasks.
Several studies have explored developmental trends in young children’s use of
retrieval strategies. This research has well established that visible retrieval cues
that are strongly associated with the hidden object are readily used by preschool-
ers. Young children seem to benefit from retrieval cues when the task is very sim-
ple and retrieval cues are visible to them. For instance, Gordon and Flavell (1977)
demonstrated that finding a picture of a doctor or a fireman that was hidden in
one of four folders was easy for 3-year-olds when a picture of a thermometer or a
fireman’s hat was placed on the outside of the folder containing the doctor or the
fireman. Although the 3-year-olds in this study had no problem using the visible
retrieval cues, Gordon and Flavell (1977) did not find much evidence that these
young children already had a clear understanding of cognitive cueing (but see
Whittaker 1986, for more positive evidence on this issue).
Another impressive study of preschoolers’ use of retrieval cues was conducted
by Ritter et al. (1973). These authors demonstrated that younger preschool chil-
dren’s (i.e., 3-year-olds’) effective use of retrieval cues occurred only when the
task was very simple. The test materials in this study consisted of six pictures
of people (e.g., a soccer player) and six small toys (e.g., a soccer ball), and each
person was functionally related to one of the toys. The 3- to 5-year-old children
watched as each person entered one of six houses and put their toy in a box such
that the toy in the box was visible at all times. By contrast, the people were placed
out of sight in the house. The children were then shown a “twin” of each person
(six pictures identical to the six people pictures that were hidden). The children’s
task was to show each “twin” how to get to his or her partner. After this task, all
materials were removed except for a single set of people pictures that were placed
face down in front of the children. The children were then asked to recall the
names of the toys. They were free to use the picture retrieval cues if they chose
188 7  The Development of Encoding and Retrieval Strategies

to do so. The majority of children at each age level made use of the visible toys
as retrieval cues during the first task, and there were no striking age differences.
However, the second task yielded a clear developmental pattern, indicating that
there are substantial improvements in young children’s retrieval cue use over the
preschool years. Whereas only about 20 % of the 3-year-olds turned over the pic-
tures of the people and used them as retrieval cues, about 75 % of the 5-year-olds
used this procedure (see Geis and Lange 1976, for complementary results).
Using a slightly different experimental design, Schneider and Sodian (1988) were
able to show that 4- and 5-year-old children know that it is easier to relocate hid-
den items if they are hidden according to semantically related rather than unrelated
cues. About 50 % of the 4-year-olds in their study not only judged semantically
related cues to be more effective than unrelated cues but were also able to justify
their use of retrieval cues by referring to the functional relation between cue and tar-
get. Although most of the criteria to make a cue informative are not fully understood
before the elementary school years (see Beal 1985; Fabricius and Wellman 1983),
3- to 5-year-olds are aware of some of the basic requirements for retrieval cues.
It is one thing to use retrieval cues that are provided. It is quite another for
young children to prepare for retrieval themselves. Ritter (1978) conducted a now-
classic study that is relevant to this point. The task was to prepare to find a piece
of candy hidden in one cup on a turntable containing six cups. Paper clips and
gold stars that could potentially be used to mark the relevant cup were placed near
the turntable. The children in this experiment (3-year-olds, 5-year-olds, and third
graders) were aware of the memory requirement and of the fact that they would
have to close their eyes as the turntable was spun. The children were then given an
opportunity to retrieve the candy. After placing the candy in the cup before spin-
ning the turntable, the experimenter asked each child whether he or she could do
something to help find the candy. Using the paper clips or gold stars to mark the
appropriate cup was defined as spontaneous preparation for retrieval. If the child
failed to mark the cup following the nondirective question, graded prompts were
provided to induce preparation for later retrieval. Although third-grade children
applied the retrieval cues spontaneously without being prompted to do so, all
preschoolers required prompting with 3-year-olds needing more prompting than
5-year-olds. Approximately one-third of the younger preschoolers (aged 3–4;6)
failed to prepare retrieval cues given even the most explicit prompting (see Beal
and Fleisig 1987; Whittaker et al. 1985, for confirming evidence).
These findings indicate that there are substantial improvements between the
preschool and grade-school years in children’s robust and flexible use of retrieval
strategies (see also Sodian and Schneider 1990). Even young preschool children
are able to engage in some planning that may be called strategic. Still, the strate-
gies of young preschoolers lack the effectiveness of the procedures used by older
children. There is evidence of effective strategies, no strategies, and “faulty strate-
gies” (i.e., strategic behaviors that do not help remembering; see Wellman 1988) in
young children. Although preschoolers can make use of and benefit from retrieval
cues, it is unclear when children begin to understand the function of cues in the
acquisition and retrieval of information.
Evidence of Strategic Behavior in Young Children 189

Evidence of Early Encoding Strategies

Several early studies on preschoolers’ memory for objects, pictures, and words
were based on the assumption that young children’s usually poor recall (compared
with their recognition abilities) might be improved if they were provided with spe-
cific prompts such as category cues for recalling categorizable lists during encoding
and retrieval (e.g., Ceci et al. 1980; Emmerich and Ackerman 1978). Although cue
manipulations at retrieval seem to produce dramatic and generally consistent effects,
encoding manipulations (e.g., blocked vs. nonblocked presentations of categorizable
items for study) have been more variably successful. Thus, some researchers have
reported facilitation due to the blocking of categorizable items (e.g., Kobasigawa
and Orr 1973; Perlmutter and Myers 1973; Morrison and Lord 1982), whereas oth-
ers have failed to observe benefits due to blocking (Emmerich and Ackerman 1978;
Garrison 1980). However, overall, this research indicates that very salient prompts
during encoding seem to affect memory. For instance, requiring 4- to 5-year-olds to
sort to-be-learned lists into semantic categories enhances memory of the material
(Moely et al. 1969; Lange and Griffith 1977). Although preschoolers’ classification
behaviors are more dominated by perceptual than taxonomic criteria, they can also
benefit from category cues in addition to color cues (Perlmutter et al. 1982).
A difficulty with the Perlmutter et al. (1982) study was that children were not
aware at encoding that they would be required to recall the material later. What
do sorting preferences look like when preschoolers and kindergarten children are
aware of an upcoming test? Sodian et al. (1986) presented 4- and 6-year-olds with
a number of toys that could be classified according to taxonomic and color crite-
ria. The children in the “sort-and-remember” condition were told to code items
into memory by putting them into groups that the children felt belonged together.
Children in a “play-and-remember” condition were given no instructions specific
to sorting but were rather told that they were allowed to play with the items for a
while before they would be given a memory test. All children sorted more accord-
ing to semantic categories than according to color. Not surprisingly, the provision
of category cues had a more positive effect than the provision of color cues. An
especially important finding was that there was significantly more taxonomic sort-
ing during the study period as well as categorical clustering at recall in the sorting
condition than in the play condition.
The design of the study by Sodian et al. (1986) did not permit conclusions about
whether organizational behaviors could be interpreted as intentional memory strate-
gies. This was due to the fact that children in the play condition were also informed
that they would have to recall the items later. A study conducted by Baker-Ward
et al. (1984) was better suited for this purpose, offering especially convincing evi-
dence that preschoolers consciously memorize. In this experiment, 4-, 5-, and 6-year-
olds were instructed to interact with a set of objects and toys for a 2-min activity
period and were placed in one of three experimental conditions: Target Remember,
Target Play, and Free Play. The children in the Target Remember condition were
told that they could play with all of the objects but that they should especially try
190 7  The Development of Encoding and Retrieval Strategies

to remember a subset of the items (i.e., the target objects). By contrast, the children
in the two play conditions were given no indication that they would have to recall
information later. Children in the Target Play condition received instructions that
stressed playing with a subset of the target items, whereas those in the Free Play con-
dition were given general playing instructions.
The use of an observational coding scheme during the activity period revealed
that even at age 4, the children who were told to remember behaved differently
from those in the play conditions (see Fig. 7.2). For instance, the spontaneous
labeling or naming of objects occurred almost exclusively among the children in
the Target Remember condition, and these children were also more likely to visu-
ally inspect the objects. These differences in the frequency of memory-related
behaviors as a function of experimental condition were found at all age levels and
increased with age. Regardless of age, children in the Target Remember condi-
tion played significantly less than children in the two Play conditions. Baker-Ward
et al. (1984) found it remarkable that even 4-year-olds respond to the challenge of
a memory goal with deliberate study-like behaviors. A second interesting finding
was that the strategic efforts of the younger children did not facilitate remember-
ing. That is, memory-related behaviors affected the recall of only the 6-year-olds
in the study. Similar findings were reported by Lange et al. (1989) who found only
weak relations between young preschoolers’ strategy use and subsequent recall.
More recently, Blumberg and Torrenberg (2005) confirmed this outcome using
a slightly different task with 3- to 4-year-old children. Although most children
in this study used an appropriate selection strategy, use of this strategy was not
related to the correct relocation of items. Baker-Ward et al. (1984) speculated
that this might be because the newly learned strategies were not routinized to
the point that they could actually be useful to the younger preschool children. As
noted by Ornstein and Light (2010), the differential effects of younger and older
preschoolers’ strategic efforts may be due to different underlying strategies. This
possibility is consistent with Baker-Ward et al.’s observation that the younger
children seemed to combine the verbal naming/labeling and manipulation of
objects, whereas the older children put naming together with visual examination.
Apparently, intentionality of memory behavior is only one relevant aspect. Two
other components, namely consistency and effectiveness, must also be considered
in any account of the development of memory.

What Develops Earlier: Encoding or Retrieval Strategies?

An interesting comparison of preschool and kindergarten children’s abilities to use

retrieval strategies in free-recall and story-recall tasks was provided by Hudson
and Fivush (1983). More specifically, these authors compared the effects of cat-
egorical and schematic organization on memory organization in these two age
groups. Children were given two trials on both a free-recall task using categori-
cally blocked words and a story-recall task. It was assumed that young children
Evidence of Strategic Behavior in Young Children 191

(a) Target Remember Target Play Free Play

24
Number of blocks

20
16
12
8
4
0
Naming Play

4-year-olds

(b) Target Remember Target Play Free Play

24
Number of blocks

20
16
12
8
4
0
Naming Play

5-year-olds

(c) Target Remember Target Play Free Play

24
Number of blocks

20
16
12
8
4
0
Naming Play
4-year-olds

Fig. 7.2  Mean number of 5-s blocks of the activity period in which each naming and play
occurred for the 4-year-olds (panel a), 5-year-olds (panel b), and 6-year-olds (panel c) in each
instructional condition. (Adapted from Baker-Ward et al. 1984)
192 7  The Development of Encoding and Retrieval Strategies

already use schematic structures when recalling script-based stories (see the evi-
dence in Chap. 5). Thus, children of both age groups should be able to use the
structure to temporally organize their recall from the first trial on. The predic-
tions regarding performance on the free-recall task were more difficult. If young
children are simply unable to appreciate the categorical structure of the list, their
recall should be poorly organized and should not improve over trials. However, if
they can appreciate the categorical structure but do not use it to deliberately organ-
ize recall, poor performance on the first-recall trial may alert them to the need for
strategy use. Improvement in categorical organization over the two recall trials
would thus indicate strategy use.
As a main result, it was found that children generally recalled the story bet-
ter than the list and that organization increased over trials. Story recall was well
organized in both age groups, suggesting that the schematic structure of the story
automatically guided recall in preschoolers and kindergarten children. Whereas
preschool children’s organization of list recall was poor on both trials, the kinder-
garten children organized list recall better on the second recall trial. The authors
concluded from these findings that there is a developmental progression in the
emergence of retrieval strategies in young children. Preschool children appear to
be unable to use categorical structures deliberately. Although kindergarten chil-
dren do not typically employ categorization strategies spontaneously (thereby
showing a production deficiency), they seem able to use categorical structures stra-
tegically after their first experiences with a free-recall task. Therefore, the ability
to use categorical structures as retrieval guides develops with age, thus reflecting
the development of categorical knowledge.
Such a conclusion was also supported by the results of a more recent study by
Schwenck et al. (2009) who explored young children’s strategy use in sort-recall
tasks in a large sample of more than 500 children ranging in age from 4 to 8 years.
After a baseline trial using neutral instructions, some children were prompted to
use a sorting strategy, whereas other children received a clustering prompt or no
prompt (control condition). Consistent with other research (e.g., Schneider 1986;
Schwenck et al. 2007), the sorting prompt was found to be more effective than
the clustering prompt. Thus, instructing children to sort picture cards according to
semantic categories led to greater recall levels and to more strategy use. Overall,
levels of recall and strategy use were higher for older children and for typical
items as compared with atypical category exemplars. Although this research indi-
cated that even 4-year-olds engage in strategic behavior, the 4-year-olds did not
benefit as much from strategy training as the older children (see also Carr and
Schneider 1991).
As this brief overview indicates, it is not appropriate to classify young children
as nonstrategic. The empirical research carried out in the last three decades has
identified competencies in preschoolers more often than deficits. The hypothesis
that there is a “5–7” shift from nonstrategic to strategic is no longer tenable. Jumps
in strategic competence come earlier, demonstrating that preschoolers are able to
show self-directed and goal-oriented behaviors. There is reason to assume that
mother–child-memory talks influence the acquisition of memory strategies. The
Evidence of Strategic Behavior in Young Children 193

findings on the development of autobiographical and event memory described in

Chap. 5 suggest that young children develop expertise in talking about their past
experiences over the preschool years, which may help them develop new skills
in the use of strategies for retrieving and reporting information from memory
(Ornstein et al. 2006, 2008). Strategic behaviors are especially obvious when chil-
dren are studied in familiar surroundings and with familiar tasks. While noting
that preschoolers are not as deficient in strategic abilities as we once believed they
were, we should not overlook the fact that their memory performance is context-
specific. Indeed, the available evidence suggests that young children may initially
evidence strategic competencies in only some highly salient and supportive set-
tings but not in others (Ornstein et al. 1988). As a matter of fact, their mnemonic
abilities are substantially lower than those of older schoolchildren who display
their memory skills in a wider range of situations, including school-like and labo-
ratory tasks. In the following section, I will examine these skills in more detail.

Encoding Strategies During the Elementary

and Middle-School Years

It is not easy to determine that a strategy affects only encoding, and in fact, it may be
impossible to do so (Waters and Andreassen 1983) because the advantages conferred
by encoding strategies are often not realized until retrieval is required. Encoding
strategies refer to procedures that are deployed during the study of material in prepa-
ration for a subsequent test. Although there are many encoding strategies, a few have
received much more attention from developmental psychologists than others have.
Three of the more prominently researched procedures—namely rehearsal, organiza-
tion, and elaboration—will be reviewed in the following sections.

Rehearsal

Research interest in rehearsal was at least partially motivated by the crucial role of
rehearsal (“control processes”) in multistore memory models and by John Flavell’s
early work, which established rehearsal as a strategy that develops between 5 and
10 years of age. The modern era of research in memory development can be traced
to the seminal study by Flavell et al. (1966) who developed a unique method for
studying list learning in 5-, 7-, and 10-year-olds. The children wore space hel-
mets while preparing for the serial recall of picture lists. The helmets covered
the children’s eyes but exposed their lips, allowing the experimenter to determine
whether the children were verbally rehearsing the materials. Children’s spontane-
ous rehearsal was measured during a study period prior to free recall. Flavell et al.
(1966) reported that the amount of rehearsal increased with age, as did the amount
recalled. There were too few 5-year-olds who rehearsed and too few 10-year-olds
194 7  The Development of Encoding and Retrieval Strategies

who did not to evaluate the effects of rehearsal on recall in these children.
However, about half of the 7-year-olds rehearsed during recall and about half did
not. Seven-year-old children who rehearsed recalled more than 7-year-old children
who did not. Thus, this study established the canonical result that memory strategy
use increases with age and is causally related to levels of recall. In particular, it
was assumed that frequency of rehearsal determines memory performance.
Direct measurement of rehearsal was complemented by Flavell et al. (1966)
and other studies (e.g., Belmont and Butterfield 1977; Hagen and Stanovich 1977)
that used less direct indicators of cumulative rehearsal. The most important of
these was an analysis of recall according to the item’s serial position in the list.
When especially good recall occurred for items early in the list (primacy effect),
rehearsal processes were assumed. Primacy effects were generally not found in
preschoolers and young school children but were observed in older schoolchildren.
Other measures of input activity such as the analysis of children’s pauses while
they studied also supported the conclusion that rehearsal develops. For instance,
although 13-year-olds in Belmont and Butterfield’s (1969) study displayed pro-
gressively longer pause times with the presentation of each new item on a serial
list, pause times were more uniform as a function of list position in younger chil-
dren (i.e., 9-year-olds). It was inferred from these results that the older children
used an active cumulative-rehearsal strategy, whereas the younger participants
probably did little more than verbally label individual items as the items were
presented.
Subsequent research indicated that the relation between frequency of rehearsal
and recall was more complex than initially proposed. Data obtained using the
overt rehearsal technique developed by Rundus (1971) permitted more detailed
developmental conclusions. The advantages of the overt rehearsal method, which
requires that participants memorize all items out loud, are that the quantity and
quality of rehearsal activities can be directly measured and evaluated. Research
using this method by Peter Ornstein, Mary Naus, and their associates has been
especially revealing. For example, Ornstein et al. (1975) required 8-, 11-, and
13-year-old children to rehearse each word on a list at least once in the 5- or 10-s
interval between the presentation of successive words. More specifically, children
were told that they had to repeat the most recently presented word at least once,
and if they wished, that they could practice any other words they liked. By using
this overt rehearsal technique, Ornstein and his colleagues made a detailed study
of quantitative and qualitative differences in the children’s rehearsal. The typi-
cal age effects for serial recall were obtained. Older children both recalled more
items and exhibited a primacy effect. However, there were no age differences in
the total amount of rehearsal nor was the correlation between amount of rehearsal
and recall very large. The fact that frequency of rehearsal was comparable for the
youngest and oldest children runs counter to the earlier findings by Flavell and his
colleagues (1966).
The impact of rehearsal style on recall. Interestingly, Ornstein and his col-
leagues noted developmental differences in the style of rehearsal. Qualitative anal-
yses of the rehearsal sets were very informative, with a rehearsal set defined as
Encoding Strategies During the Elementary and Middle-School Years 195

the number of items rehearsed together. The youngest children tended to rehearse
only one or two unique words during each interval, and Ornstein et al. labeled this
tendency passive or single-item rehearsal. By contrast, older children more often
rehearsed the target word plus several previously presented words, a style Ornstein
et al. labeled active or cumulative rehearsal. On the basis of these and other data,
Ornstein and his colleagues asserted that the important developmental changes are
in terms of style rather than the frequency of rehearsal and that the older chil-
dren’s active rehearsal accounted for both their greater recall and the primacy
effect that was obtained in their data (see also Guttentag et al. 1987; Ornstein and
Naus 1985).
Naus et al. (1977) provided clear evidence that the differences in rehearsal
between younger and older elementary schoolchildren reflected a production
deficiency in the younger children. Third graders who did not use the cumula-
tive-rehearsal strategy spontaneously could be easily trained to use three-item
rehearsal sets. They displayed the primacy effect typical of older children, and
their recall was approximately at the level of sixth graders. This training study
as well as others (e.g., Cox et al. 1989; Ornstein et al. 1977) bolstered the causal
relation between style of rehearsal and memory performance, noting improve-
ments in young children’s recall following cumulative-rehearsal training. Yet,
rehearsal training for young children rarely eliminated developmental differences,
and generalization of the trained strategy was typically weak. On the whole, how-
ever, training young children to use a more effective rehearsal strategy typically
benefitted their memory performance. These findings led to the conclusion that a
critical developmental difference concerns children’s inclination to use a particular
strategy rather than their ability to use it, and this difference indicates that young
schoolchildren are typically production deficient (Bjorklund et al. 2009).
When all of the relevant data are considered, it seems that the essential dif-
ference between the memorization processes of younger compared with older
children are qualitative rather than quantitative. Passive one-word memorization
strategies are replaced by cumulative-rehearsal strategies such that the number of
different items in a rehearsal set eventually reaches three or four (Ornstein and
Naus 1978). These cross-sectional results were complemented by longitudinal
analyses. In a study by Kunzinger (1985), an overt rehearsal task was presented to
7-year-olds who were retested 2 years later. There was an increase in the rehearsal
set size with development (from 1.7 to 2.6 items). Rehearsal frequency was never
correlated with recall. The size of the rehearsal set was correlated with recall only
at the second measurement point. An especially interesting finding was the high
stability in performance. Children who had larger rehearsal sets at the initial time
of measurement also had larger sets at the second measurement point. Kunzinger
also assessed children’s individual stability, that is, each individual’s relative
standing within the reference group. This “lability score” (Bayley 1949), defined
as the across-age standard deviation of an individual’s z-scores, indicated mod-
erately strong individual stability over time, with lability scores ranging between
0.26 and 0.36 (the lower the score, the higher the individual stability). This find-
ing shows that the overall high group stability in developmental changes over time
196 7  The Development of Encoding and Retrieval Strategies

corresponds to most individual change patterns. Guttentag et al. (1987) observed

comparable longitudinal stability between 8;6 and 9;6 years of age.
In an even more comprehensive longitudinal study, Lehmann and Hasselhorn
(2007) assessed developmental changes in rehearsal strategies in 76 children who
were tested five times between the ages of 8 and 10 years. Children’s videotaped
behavior during the learning period was closely examined, and several types of
learning behavior such as labeling, single-word rehearsal, cumulative rehearsal,
and association/elaboration were identified. As a main result, it was found that
there was a gradual shift from a passive single-word rehearsal strategy to the
more active cumulative-rehearsal strategy from age 9 onward. Mean rehearsal set
sizes increased from about 1.1 to 2 over the course of the study, with maximum
rehearsal set sizes increasing from about 4 to about 5. These increases in rehearsal
set size were also associated with increases in recall. Although this finding is in
accordance with former research (Guttentag et al. 1987; Ornstein and Naus 1978),
additional fine-grained analyses carried out by Lehmann and Hasselhorn revealed
that cumulative rehearsal primarily occurred in the early parts of the word lists,
whereas labeling dominated in the later parts of the lists. The major developmen-
tal trend found in this study was that children also showed an incremental use of
cumulative rehearsal in the later parts of lists with increasing age. Interestingly,
however, Lehmann and Hasselhorn (2007) observed the coexistence of several
strategies throughout the elementary school years with a gradual shift from the
less advanced strategies such as labeling to the more advanced strategy of cumula-
tive rehearsal.
The findings summarized above suggest that developmental increases in active
cumulative rehearsal play a crucial role in the explanation of age differences in
free-recall and serial-recall tasks (see Jarrold and Hall 2013, for a more skeptical
view regarding the role of rehearsal in serial-recall tasks). However, the question
remains: To what degree does cumulative rehearsal facilitate the encoding of infor-
mation versus the retrieval of it? Perhaps only active rehearsal transfers informa-
tion from short-term memory into long-term memory. Alternatively, perhaps both
forms of rehearsal permit the transfer of information to long-term memory, but
active rehearsal makes it easier to retrieve the information later. An experiment to
test these two possibilities would compare the effects of various types of rehearsal
in different age groups under testing conditions that differ with respect to retrieval
demands. The comparison of the two test formats would provide information on
whether different rehearsal strategies would produce different encodings, which,
as a consequence, would affect the quality and amount of information stored in
long-term memory (LTM). If developmental differences occur on recall tests but
there are none on recognition tests, it can be assumed that the material is stored in
LTM at all age levels but that younger children experience retrieval problems that
are somewhat linked to their failure to use cumulative rehearsal. If, however, there
are also developmental differences in recognition, this could be taken as evidence
that different rehearsal strategies produce different encodings.
Naus et al. (1977) conducted an experiment along these lines. Third and sixth
graders learned a word list and then took either a recognition or a recall test.
Encoding Strategies During the Elementary and Middle-School Years 197

Consistent with other developmental data, there were significant developmen-

tal differences in recall, especially due to primacy effects. No differences were
found in the recognition data, however. Naus et al. (1977) concluded from their
findings that both active and passive rehearsal seemed to produce the transfer of
information from short- to long-term memory. However, it cannot be excluded
that there are different types of representations in LTM following passive versus
active rehearsal. For instance, it seems possible that the overt cumulative-rehearsal
strategies build interitem associations, and these associations make it easier to
retrieve the items during recall. This assumption was confirmed in recent studies
by Lehmann and Hasselhorn (2010, 2012) who investigated the relation between
rehearsal behavior and recall dynamics. Using data from the Lehmann and
Hasselhorn (2007) longitudinal study, Lehmann and Hasselhorn (2010) found that
the rehearsal of adjacent items in the word lists increased the strength of interitem
associations, thus facilitating participants’ recall in future retrieval attempts. There
was evidence of a “contiguity effect,” that is, the phenomenon that successive
recall is more likely to come from nearby serial positions than from remote serial
positions, and this is also reflected in shorter interitem response times. Another
mechanism that may account for the retrieval benefits produced by cumulative
rehearsal is the self-testing part of this rehearsal activity. Thus, cumulative rehears-
ers have practice with recall more than passive rehearsers before the actual recall
test takes place (Naus and Ornstein 1983; Ornstein and Naus 1978).
Lehmann and Hasselhorn (2012) confirmed this finding for the most part by
assessing rehearsal and recall sequences in a cross-sectional sample of children
between 8 and 10 years of age. There was a more pronounced and prolonged con-
tiguity effect in the older children than in the younger participants. Successive
rehearsal and recall of items in their original order was more pronounced in the
older children, indicating that they were using cumulative rehearsal more inten-
sively than the younger children. Lehmann and Hasselhorn (2012) concluded from
their study that a main feature of item rehearsal with regard to facilitating recall is
the strengthening of interitem associations based on their temporal order within a
list and that this characteristic develops with age.

Organizational Strategies

Bousfield’s (1953) research on subjective organization during learning was the

pioneering work on subject-controlled memorizing activities. This study was one
of the first to explicitly allow inferences about processes that mediate learning
and recall. As already noted in Chap. 2, the study by Bousfield et al. (1958) on
conceptual and perceptual clustering stimulated a flood of developmental research
on factors that affect organization. The development of organizational strategies
was studied with picture and word lists, usually ones containing items that could
be categorized. The items were often selected using age-appropriate norms (e.g.,
Bjorklund et al. 1983) with common items from familiar categories (e.g., animals,
198 7  The Development of Encoding and Retrieval Strategies

furniture, vehicles). Subjective organization was most often measured in devel-

opmental studies using the adjusted ratio clustering (ARC) measure developed
by Roenker et al. (1971) and the ratio of repetition (RR) measure developed by
Bousfield and Bousfield (1966). For both of these measures, values close to one
represent almost perfect organization, and zero indicates random responding.
Typically, the development of organizational strategies is explored in sort-recall
tasks that involve organizing pictures or words into semantic categories. Children
are given a randomly ordered list of categorizable items (e.g., animals, furniture,
and the like). They are then told that their task is to remember the items later on
and that they are free to do anything with the materials that may help their recall.
Following a short study period, children are asked to recall as many stimuli as they
can. Children’s organization of items during study (sorting) and recall (clustering)
is often measured using the clustering indices mentioned above. One of the special
features of these clustering indices is that the degree of organization during encod-
ing (sorting) and retrieval (clustering) can be measured using the same scale, and
this allows for direct comparisons between encoding and retrieval processes.
As noted by Flavell et al. (2002), the developmental story for organizational
strategies is much the same as for rehearsal. A study by Moely et al. (1969) illus-
trates the case. Children ranging in age from 5 to 11 years were shown a set of
pictures that consisted of several animals, pieces of furniture, and articles of cloth-
ing. The pictures were arranged in a circle, and no pictures from the same cat-
egory were adjacent to each other. Children were told that they should try to learn
and remember the pictures. The experimenter said that she would leave for a few
minutes to allow the children to study, during which time the children could do
anything with the pictures that would help them remember the items. As a main
result, it was found that the younger children in the sample (5- to 9-year-olds) did
not sort the items according to their semantic categories, thus showing a produc-
tion deficiency. Only the 10- to 11-year-olds spontaneously categorized the items
as a mnemonic to help them remember at a level significantly higher than chance
(ARC score = 0.6).
Young children’s reluctance to use an organizational strategy in sort-recall tasks
was also illustrated in a study by Salatas and Flavell (1976a). First graders were
presented with a set of 16 pictures comprised of four pictures from each of four
different categories. After naming each picture and each of the four categories,
the experimenter placed the items randomly on a table in front of the children.
The children were then told that they would be asked to remember the pictures
later on and that they could do anything with the pictures that would help them
remember. Free recall was tested after a short study period. Despite instructions
that seemed to bias the children to organize (i.e., sort the items), only 27 % of the
children organized the pictures according to their semantic categories to a signifi-
cant degree.
The findings by Moely et al. (1969) fit the general developmental trends
found for rehearsal. Levels of organization in terms of both sorting and clustering
increased with age and were accompanied by increases in recall (see also Corsale
and Ornstein 1980; Hasselhorn 1992). Studies have also provided evidence that
Encoding Strategies During the Elementary and Middle-School Years 199

young children (preschoolers and kindergarteners) can be easily trained to use an

organizational strategy with corresponding improvements in recall (e.g., Black and
Rollins 1982; Guttentag and Lange 1994; Lange and Pierce 1992). Even preschool-
ers have been found to use organizational strategies and to demonstrate enhanced
levels of recall when instructions are modified to emphasize the importance of
grouping according to the meaning of items (e.g., Lange and Jackson 1974; Sodian
et al. 1986) or when the children are explicitly trained to use organizational strate-
gies (e.g., Carr and Schneider 1991; Lange and Pierce 1992). As emphasized by
Bjorklund et al. (2009), the organizational abilities of preschoolers, kindergarteners,
and young schoolchildren resemble their rehearsal abilities in that the children
demonstrate a production deficiency. In other words, young children are capable
of organizing information for recall, but they rarely do so spontaneously. As with
rehearsal, training children to use an organizational strategy rarely eliminates age
differences, and young children also typically fail to generalize the strategy to new
situations or new sets of materials (Cox and Waters 1986). Thus, a salient feature of
young children’s deliberate use of memory strategies is the context specificity that
characterizes many aspects of their performance (Ornstein and Light 2010).
The only inconsistency between the research findings focusing on the devel-
opment of rehearsal versus organizational strategies concerns the age at which
the strategies were first detected, which is about age 7 for rehearsal and 9 or 10
for organization. Obviously, organizational strategies develop somewhat later
than cumulative rehearsal. This makes sense because the discovery or creation of
semantic relations between items seems like a more complex and demanding pro-
cess than rehearsal.
Interesting questions are whether the organization of items at encoding or at
retrieval is more important for recall and whether there are developmental trends
in the relation between input and output activities. Even young children’s recall
contains a lot of organization when lists containing highly associated items are
learned. Also, younger children (6- to 7-year-olds) will sometimes use organiza-
tional strategies when highly associated items serve as category items (e.g., dog,
cat; salt, pepper), but high levels of clustering using less strongly associated cat-
egorical items (e.g., goat, deer; shoes, hat) are not typically observed until between
9 and 11 years of age (e.g., Bjorklund and DeMarchena 1984; Schneider 1986).
Lange (1978) proposed two possible explanations for young children’s failures to
organize their free recall when left to their own devices to learn potentially catego-
rizable material. One possibility was input failure or the failure to group stimulus
materials clearly and reliably during encoding. The other involved failures during
the retrieval phase to make use of the groupings that were constructed at encod-
ing. Lange’s own work (Lange and Griffith 1977) favored the former interpretation
because he obtained a strong covariation between children’s input categorization
and their recall.
The relation between the organization of sorting and recall was first system-
atically explored in Kee and Bell’s (1981) study, which included children from
Grades 2 and 6 as well as university students. Kee and Bell measured the ARC
cluster indices for both encoding and recall. The oldest participants in the study
200 7  The Development of Encoding and Retrieval Strategies

had high ARC cluster scores at both input and output. By contrast, the two subsam-
ples of children had low ARC scores during input but higher ones at output. The
study also included an instructional component. When the child participants were
instructed to sort items into experimenter-specified categories during encoding,
their ARC scores at encoding and retrieval were high. Instructions to use categori-
zation during retrieval had a much less pronounced effect on organization than the
study instructions did. The study instructions generally improved children’s recall
performance, especially for the second graders. Kee and Bell concluded from their
findings that variations in organizational behaviors during encoding and not during
retrieval produced differences in memory performance in the sort-recall task.
A related issue concerns the question of whether output organization can be
generally conceived of as an indicator of deliberate strategy use. There are now
many demonstrations that the degree of output organization systematically
covaries with associations between the items and the typicality of the item lists
(Bjorklund 1985; Frankel and Rollins 1985; Hasselhorn 1992; Schneider 1986).
Several researchers (e.g., Lange 1978) have argued that young children’s cluster-
ing during free recall may be automatically determined by the structure of the task
materials and does not necessarily indicate conscious strategic operations. A high
degree of output organization for highly associated items probably occurs because
the recall of any particular item more or less automatically triggers the recall of
closely associated words. No consciously controlled strategic encoding processes
are necessary for this mechanism to work. In fact, by the later grade-school years,
it is clear that meaningful interconnections between to-be-learned items can affect
later memory even if learners explicitly used encoding strategies that did not
exploit the meaningful interconnections (Rabinowitz and McAuley 1990).
The studies by Frankel and Rollins (1985) and Schneider (1986) explored the
impact of associative versus categorical relations by factorially manipulating high-
and low-categorical relatedness and high- and low-interitem associations. For
instance, in Schneider’s study, a sample of 64 second-grade and 64 fourth-grade
children were given a sort-recall task with categorizable pictures serving as the
stimulus materials. The pictures were movable and could thus be sorted into piles.
Children were permitted to do anything that they wanted to learn these items. Four
different types of lists were used in the study. One list contained items that were
highly related to the category, with high interitem associations between most of
the items on the list (hereafter, the high related–high associated list). The animals
on this list included a dog, cat, mouse, horse, cow, and pig. The second list was
composed of items highly related to the category but with low-interitem associa-
tions (high related–low associated). The animals on this list included a tiger, ele-
phant, cow, pig, bear, and dog. The third list was composed of items that were
weakly related to the category, although there were some high interitem associ-
ations (low related–high associated). The animals on this list were a goat, deer,
buffalo, hippopotamus, monkey, and lamb. The low related–low associated list
included the animals beaver, rat, alligator, camel, squirrel, and giraffe.
The findings are given in Table 7.1. In general, the fourth graders in the study
employed much more categorical sorting during study than the second graders did.
Encoding Strategies During the Elementary and Middle-School Years 201

Table 7.1  Mean recall, category clustering during sorting and during recall as a function of grade,
category relatedness, and interitem association. (Adapted from Schneider 1986, p. 226, Table 2)
Grade and task High associativity Low associativity
High relatedness Low relatedness High relatedness Low
relatedness
Grade 2
Recall 12.37 10.44 10.75 10.88
Clustering/sorting 0.38 0.32 0.35 0.13
Clustering/recall 0.63 0.38 0.18 0.16
Grade 4
Recall 17.31 15.37 14.19 11.50
Clustering/sorting 0.70 0.77 0.40 0.20
Clustering/recall 0.70 0.57 0.56 0.52

Not surprisingly, the fourth graders also clustered more during recall and recalled
more than the second graders. In addition, there was a main effect of interitem
associativity at recall on clustering, with highly associated lists producing more
clustering. More importantly, there was a striking age by list associativity interac-
tion in the clustering data such that low associativity especially penalized younger
children more than older children. In fact, the clustering of high- and low-associ-
ated lists was approximately equal for the older children in the study. Although the
main effect of associativity and the age by associativity interaction were not sig-
nificant in the recall data, there were strong trends in the recall data mirroring the
clustering data. In general, there were significant correlations between clustering
at study, clustering at recall, and recall.
Similar findings were also reported by Hasselhorn (1992) who compared the
same age groups on free-recall and sort-recall tasks. The deliberate use of an
organizational category increased with age and with the items’ category typicality
and was more frequent in the sort-recall than in the free-recall task. Hasselhorn
concluded from his findings that most 10-year-olds are capable of using seman-
tic organization strategies effectively, particularly when items are typical of the
respective semantic categories.
Although a similar pattern of results was also found in the study by Frankel
and Rollins (1985), there were also slight differences. Frankel and Rollins com-
pared kindergarteners, fourth graders, and 10th graders on sort-recall tasks that
varied according to item typicality and interitem associativity. Whereas the 10th
graders in that study showed high levels of organization in recall whenever cat-
egory relatedness or associative strength were high, kindergarteners and fourth
graders displayed little organization at study, and greater category clustering was
obtained only for the list that was compatible with associative organization (i.e.,
taxonomic factors did not affect recall). Although the finding of the studies by
Frankel and Rollins (1985) on the one hand and those by Hasselhorn (1992) and
Schneider (1986) on the other hand are not consistent with regard to the out-
come for fourth-grade children, they suggest important developmental changes
202 7  The Development of Encoding and Retrieval Strategies

in organizational behavior from the kindergarten period to secondary school age.

Obviously, associative relations play a substantial role in mediating the organi-
zation of young children’s memory but not in that of older elementary school-
children (see Hasselhorn 1990, for similar findings). It seems interesting to note
here that the German fourth graders in the studies by Hasselhorn and Schneider
were more strategic than the American fourth graders recruited by Frankel and
Rollins and also more strategic than the American third graders in the study by
Corsale and Ornstein (1980). I will get back to this issue when discussing the
impact of schooling on strategy acquisition.
Longitudinal evidence. How do organizational strategies develop over time, and
at what age do they begin? The impression one gets when looking at the memory
strategy literature is that children show regular and gradual changes in strategy use
over time. It seems important to note, however, that most data looking at age-related
changes in strategy use are cross-sectional in nature with different children of differ-
ent ages being tested to assess developmental differences (see reviews by Bjorklund
et al. 2009; Kail 1990). Thus, the impression of a gradual continuous increase in
strategy use and corresponding increases in recall inferred from the outcomes of
cross-sectional research need to be validated with longitudinal research.
Until the early 1980s, longitudinal research on memory development was almost
nonexistent, and about 99 % of the relevant studies were cross-sectional in nature.
Thus, advances in our understanding of memory processes in children and adoles-
cents were predominantly based on the analysis of developmental differences
among various age groups but not on developmental changes within the same
group of individuals over time. Given that cross-sectional studies do not permit seri-
ous conclusions about the emergence and variability of memory skills as a function
of age and experience (Ornstein and Haden 2009), most researchers emphasized the
fact that longitudinal studies on memory development were badly needed to provide
information about the developmental course of remembering within individual chil-
dren. There was broad agreement that longitudinal memory studies in which chil-
dren are traced over time are crucial for understanding individual differences in the
relation between early skills and subsequent memory development.
Why is it that research on memory strategy development is mostly based on
the indirect evidence provided by cross-sectional data? One reason is pragmatic:
Longitudinal studies are expensive, require the long-term recruitment of staff and
long-term funding, and also provide several methodological problems regarding the
adequate assessment of change and the handling of missing values (for more details,
see Reese 2009; Schneider 1989). Another reason is more conceptual: There is still
a belief that the basic processes governing human behavior in general and memory
development in particular are universal in nature, sequence, and scope, and that var-
iation occurs primarily in the rate of change rather than in the kind of change.
The few available longitudinal studies clearly show that findings from cross-
sectional research on memory strategy development do not tell us the true story
about individual developmental trends. For instance, using data from the Munich
Longitudinal Study on the Genesis of Individual Competencies (LOGIC),
Schneider and Sodian (1997); Sodian and Schneider (1999) explored the issue of
Encoding Strategies During the Elementary and Middle-School Years 203

whether the impression of gradual developmental increases in strategy use and

recall derived from cross-sectional studies could be confirmed with longitudinal
data. Although Schneider and Sodian first confirmed earlier results on the group
level (which suggested gradual increases in strategy use and recall over time), they
were struck by low-individual stabilities in the strategy and recall variables over
time. A closer inspection of individual change data revealed that gradual steady
increases were rarely observed. Only about 8 % of the children showed a gradual
improvement in their use of organizational strategies (sorting and clustering), as
suggested by the group data. In comparison, about 81 % of the children “jumped”
from the chance level to near perfection between subsequent measurement points.
These findings thus confirm that children go from chance levels of sorting to per-
fection, but they do it at different points in time.
Subsequent analyses on data from the Würzburg-Göttingen longitudinal mem-
ory study (Kron-Sperl et al. 2008; Schneider et al. 2004, 2009) basically con-
firmed this pattern of results. The exceptions were that the percentage of children
“jumping” from the chance level to near perfection was somewhat lower, and the
percentage who showed a gradual increase was higher than that observed in the
LOGIC study (see Fig. 7.3). Overall, the data seemed to square well with other
evidence of substantial inter- and intra-individual variability in strategy use, with
children using different strategies and combinations of strategies on any given
memory problem (Siegler 1996).

90

80

70

60
Frequency (%)

50

40

30

20

10

0
perfect from the start jumped from chance gradual increase nonstrategic
to perfect sorting

LOGIC study (age range 4-12, 5 waves)

Würzburg Longitudinal Study (age range 6-10, 9 waves): Sorting score
Würzburg Longitudinal Study (age range 6-10, 9 waves): Clustering

Fig. 7.3  Individual patterns of strategy acquisition (Schneider et al. 2009, Fig. 1)

204 7  The Development of Encoding and Retrieval Strategies

A drawback of longitudinal studies for assessing patterns of strategy change is

that a lot can happen within 6 months (the time interval between adjacent meas-
urement points in the Würzburg-Göttingen Longitudinal Study), and it may be that
developmental change in strategy use is more gradual than abrupt when shorter
time intervals are chosen (Ornstein 1999). This was addressed in a microgenetic
study by Schlagmüller and Schneider (2002). In this study, 8- to 12-year-old
children (nonstrategists at the beginning) performed a series of nine sort-recall
tasks across 11 consecutive weeks. Patterns of change in strategy use and recall
were then assessed. Consistent with the findings of the two longitudinal studies
described above, it was shown that the transition from nonstrategic to strategic
occurred quickly rather than gradually. Children who acquired the organizational
strategy did so at different times in an all-or-none fashion. Once children started to
use an organizational strategy, they also improved their recall performance.
Several studies on preschool children’s development of event memory have
shown that social context influences memory development (see Chap. 5). One
interesting question explored by Güler et al. (2010) addressed the impact of
maternal strategic behaviors during a collaborative sort-recall task on preschool-
ers’ recall and strategic behavior in a sort-recall task that they completed indepen-
dently about 1 year later. Güler et al. (2010) found relations both concurrently and
longitudinally between maternal behavior, child behavior (i.e., strategy use), and
children’s recall. These findings indicate that mother–child interactions during a
strategic memory task may predict children’s strategy acquisition and that differ-
ences in how mothers use memory strategies with their children are related to chil-
dren’s recall and their use of memory strategies over time.
Taken together, the findings from longitudinal and microgenetic research
clearly extend the cross-sectional database in critical ways by providing a truly
developmental account of the acquisition of a memory skill, thus informing us
about the development of memory in the sense in which Ornstein and Haden
(2009) use the term. A major outcome of this research is that for most children,
the transition from nonstrategic to strategic, at least for organizational memory
strategies, occurs relatively quickly and at different ages for different children
(Bjorklund et al. 2009; Schneider 2011).

Elaboration

For the most part, developmental researchers have studied elaboration strategies
using the paired-associate learning paradigm. As already noted in Chap. 2, early
research using the associative learning paradigm followed the verbal learning
tradition and was not directly concerned with age-related changes (Goulet 1968;
Keppel 1964). This changed with the emerging interest in memory strategy devel-
opment initiated by Flavell and colleagues’ pioneering work, which also triggered
programmatic investigations of developmental differences in children’s and ado-
lescents’ memory (see reviews by Kee 1994; Pressley 1982; Pressley and Hilden
Encoding Strategies During the Elementary and Middle-School Years 205

2006; Rohwer 1973; Schneider and Pressley 1997). Studies assessing elaboration
strategies typically present participants with pairs of items such as a shoe and a
balloon, a cat and an apple, or an arrow and glasses, and participants are asked
to remember the pairs. At testing, the participants are presented one of the paired
items and asked to recall the pair mate. This is a very challenging task unless par-
ticipants use an elaboration strategy.
According to Rohwer (1973), elaboration in noun-pair learning refers to the
process of generating an event that can serve as a common referent for the mem-
bers of each pair. He proposed that the learner’s use of an elaboration strategy is
mainly responsible for successful noun-pair learning. More specifically, elabora-
tion involves associating two or more items by creating a visual or verbal repre-
sentation of them. For example, if a learner was presented the pair “cat–apple” to
memorize, an effective verbal elaboration to link these two words might involve
generating a sentence such as “The cat rolled the apple around.” A visual elabora-
tion concerning this noun pair might involve creating an image of a cat rolling an
apple around.
Early research concerning the use of elaboration strategies revealed produc-
tion deficiencies during the preschool and childhood years (Flavell 1970; Rohwer
1973). Unlike rehearsal and organizational strategies, elaboration strategies are
rarely observed before adolescence and can be conceived of as a “latecomer to the
memorizer’s bag of tricks” (Pressley 1982, p. 303). Experimental research aimed
at determining the developmental level at which learners overcome the elaboration
strategy production deficiency has typically included either an elaboration-pro-
vided condition, an elaboration-generated condition, or both of these conditions
as well as a control condition. If participants in the control condition performed
comparably to participants in the elaboration conditions, it was inferred that con-
trol participants engaged in elaborative activity spontaneously. When a repetition-
control condition was included, it was assumed that repetition would preclude
elaborative processing. Thus, if the performance of participants in the control
group exceeded that of participants in a repetition-control condition, this was also
taken to indicate spontaneous elaboration by control participants. The expectation
in most studies on elaboration strategies was that, at a younger age level, perfor-
mance in the control condition would be lower than in the elaboration condition
but that at a more advanced age-level performance in the control condition would
be comparable to performance in the elaboration conditions, indicating spontane-
ous use of an elaboration strategy.
An early study by Rohwer and Bean (1973) provides an illustration of the typi-
cal data and inferences. These authors compared the effects of various prompts
in participants drawn from five grade levels: first, third, sixth, eighth, and 11th.
In a listen (control) condition, participants were told to attend to and learn a total
of 36 word pairs comprised of familiar nouns. In the rehearsal control condition,
instructions were given to repeat each pair aloud as many times as possible during
the interitem intervals. In the presentation condition, following the presentation
of each pair, the two words were presented again in a sentence context, and par-
ticipants were instructed to repeat the sentences aloud as many times as possible
206 7  The Development of Encoding and Retrieval Strategies

before the onset of the next pair. Finally, in the construction (elaboration) condi-
tion, participants were directed to construct and utter aloud a sentence containing
each noun pair that was presented and to repeat the sentence as often as possible in
the interpair interval.
As a main result, the prompt to construct a sentence was able to promote
effective elaboration in all of the age groups in this study. It was equivalent to
the effects obtained by providing a sentence and clearly more effective than the
instructions to listen or rehearse the noun pairs with one exception: Although
the effect of the listen control condition was indistinguishable from that of the
Rehearsal condition at the first-, third-, and sixth-grade levels, it was stronger at
the eighth-grade level and equivalent to the effects of the sentence conditions at
the 11th-grade level. The authors concluded from this finding that a much less
explicit prompt such as being directed to listen was sufficient for evoking elabora-
tion processes only for eighth- and 11th-graders.
The findings by Rohwer and Bean (1973) also indicated the importance of
individual differences within a given age group. In a subsequent study, Rohwer
et al. (1977) followed up on this issue. In their third experiment, groups of 11-
and 17-year-olds were first asked to learn a list of paired associates under con-
trol instructions. Participants were classified as good, medium, or poor associate
learners based on the outcome of this test. In a next step, all of the participants
were presented with another list of pairs to learn and were placed in either a ver-
bal-generated (elaboration) condition, a repetition-control condition, or a control
condition. Among 11-year-olds of all three learning-ability levels, the elaboration
condition produced better performance than the two control conditions. Although
the same pattern of results was found for the 17-year-old poor and medium learn-
ers, the high-ability participants in the elaboration and control conditions outper-
formed those in the repetition-control condition. Rohwer et al. (1977) concluded
from these findings that the high-ability adolescents in particular used elaboration
strategies spontaneously, and this confirms the importance of individual differ-
ences for elaboration propensity.
Other evidence has corroborated these conclusions (see Pressley 1982; Pressley
and Hilden 2006). For instance, interview studies have confirmed individual dif-
ferences in participants’ reported use of elaboration strategies with their use of
elaboration strategies increasing with age from childhood to adolescence. Also,
elaborative use has been consistently associated with higher learning perfor-
mance (Beuhring and Kee 1987; Pressley and Levin 1977). Taken together, both
experimental and self-report data indicate that improvements in performance on
associative memory tasks are due primarily to an increase in the tendency to use
elaboration as a memory strategy (Pressley 1982; Schneider and Pressley 1997).
Older adolescents are more likely than grade-school children to report using elab-
oration during study. Interestingly, however, the classification of participants with
respect to strategy use has been shown to be more predictive of associative learn-
ing performance than age (Pressley 1982).
Effects of imagery strategy instruction. Unlike rehearsal and organizational
strategies, elaboration strategies are not acquired during the natural course of
Encoding Strategies During the Elementary and Middle-School Years 207

development by most individuals of a given age cohort. Even many adults need to
be prompted to use elaboration on paired-associate tasks (Rohwer 1980). However,
as with the other two mnemonics, young children can be trained to use elabora-
tion strategies with corresponding increases in memory performance (Pressley and
Hilden 2006; Siaw and Kee 1987). When younger children generate elaborations,
they are typically not as effective as those produced by older children (Reese 1977).
For instance, in a series of studies, Levin and Pressley and their colleagues produced
a substantial body of evidence demonstrating that the effects of imagery instruction
were, at best, limited for preschoolers and children in the early elementary grades.
Some of this work was done with paired associates, some with vocabulary, and some
with simple prose (see Pressley and Van Meter 1993; Schneider and Pressley 1997,
for more detailed accounts). With paired associates, the children were taught to find
a linking relationship for pairs of items. The vocabulary research focused on the key-
word method (Atkinson 1975; see Pressley et al. 1982; Schneider and Pressley 1997,
for comprehensive reviews). This two-stage method for learning foreign vocabulary
words involves generating interactional images between definition referents and the
referents of words that sound like parts of the to-be-learned vocabulary item (i.e., the
keyword). For instance, cart is a good keyword for the Spanish word carta (postal
letter), which is generated in a first step. The second stage consists of generating
an interactive visual image. Thus, for carta, a reasonable image would be that of
a postal letter inside a shopping cart. Alternatively, a meaningful sentence could be
used to link the keyword to the vocabulary word’s definition, such as “the cart trans-
ported the letter” (Pressley et al. 1982).
In general, the potency of elaborative imagery as a facilitator of children’s
associative learning increases between the ages of 5 and 11 years. The results for
the paired-associate and keyword methods tend to be very similar. That is, for 3-
to 8-year-olds, the effects of imagery training were either not obtained or were
obtained only in situations that supported imagery formation. Thus, these young
children could generate images to mediate their learning of paired associates that
involved strong associations between the to-be-learned items (e.g., needle-balloon)
but could not do so on a regular basis when the noun pairs were not so obviously
related. In the latter case, imagery instructions were useful only with a fairly slow
rate of presentation (Pressley and Levin 1977). By 5–6 years of age, children ben-
efit from simple imagery-generation instructions when learning, and by 7–8 years
of age, children can apply the strategy with purely verbal materials (Begg and
Anderson 1976; Levin and Pressley 1978). By 11 years of age, children appear to
be no more constrained in their use of imagery mnemonics than adults are. Verbal
elaborations can be effectively trained in even younger children and can improve
children’s learning performance from the kindergarten level on.
Once grade-school children are taught to use elaborative strategies, they con-
tinue to use them if they are presented with the same task later (Borkowski et al.
1976; Schneider and Pressley 1997). Maintenance effects are not limited to just
grade-school children, nor are they limited to simple elaboration strategies. Even
6-year-olds can maintain rather complex elaborative strategies (Kestner and
Borkowski 1979). The available data also indicate that elaborative transfer can be
208 7  The Development of Encoding and Retrieval Strategies

obtained at least some of the time even in young grade-school children (Pressley
1982). For instance, in the study by Kestner and Borkowski (1979), there was only
a small difference between the training and generalization tasks. Children who had
been trained to use an elaboration strategy to learn paired associates were also able
to cope with the new task to learn and remember triads of pictures instead of pairs
and remembered more of these items than did control children.
Nonetheless, elementary schoolchildren do not always generalize elaboration
strategies. In a study by Pressley and Dennis-Rounds (1980), 10- to 13-year-old
children and 16- to 19-year-old adolescents performed two associative tasks: In
the first, they learned the products of cities, and in the second, they learned Latin
definitions. The study comprised three experimental conditions that are relevant
to the current discussion. Participants in two of these conditions were taught to
use the mnemonic keyword method to mediate city-product learning. Participants
in the control condition were left to their own devices to learn the cities and their
products. The critical manipulation with respect to transfer occurred before the
Latin words were presented. That is, the participants who were given control
instructions for city-product learning were also given control instructions for Latin
learning, as was one of the two groups that were given keyword training for the
city-product task (hereafter, those in the keyword no-transfer training condition).
Participants in the other keyword condition received additional keyword training
before being presented with the Latin words (keyword complete reinstruction con-
dition). Transfer would be apparent in this design if the performance of the no-
transfer training participants exceeded that of the control participants on the Latin
task. The more performance in the no-transfer training group resembled perfor-
mance in the complete reinstruction group, the greater the transfer.
The most critical data in the study are presented in Fig. 7.4. There was no evi-
dence of transfer among 10- to 13-year-old participants given that the no-transfer
training and the control group means were virtually identical and significantly
lower than those of the complete reinstruction condition. Consistent with the per-
spective that the propensity to transfer increases during the adolescent years, there
was significant transfer among the 16- to 19-year-olds. That is, complete reinstruc-
tion in the keyword method was not necessary to induce keyword-method use on
the Latin task. On the other hand, however, performance in the no-transfer train-
ing condition was far below performance in the complete reinstruction condition
at both age levels, indicating that spontaneous transfer was far from complete.
Although these data support the assumption that the transfer of elaborative strat-
egy use increases during the adolescent years, transfer is obviously far from per-
fect during middle to late adolescence.
In sum, the conclusions reached about the development of elaboration are not
very different from those reached in other areas of memory strategy development
(see Brown 1978; Flavell 1970; Schneider and Pressley 1997). That is, production
deficiencies are often overcome as children develop, and generalized strategy use
does not always occur in children. Moreover, systematic instruction in strategy use
can dramatically increase children’s learning and memory performance if the chil-
dren do not spontaneously use the techniques they have been taught to use.
Encoding Strategies During the Elementary and Middle-School Years 209

80

70

60
percent learned

50

40

30

20

10

0
10 to 13 years 16 to 19 years
AGE
control no transfer instruction complete reinstruction

Fig. 7.4  Mean percent learned on the transfer task as a function of age and transfer instruction.
(Adapted from Pressley and Dennis-Rounds 1980) (Schneider and Pressley 1989, p. 151, Fig. 6.10)

Elaboration strategies from a life span perspective. One of the few studies
that directly compared the effects of a systematic strategy training in children and
older adults was carried out by Brehmer et al. (2007). The authors used a mul-
tiphase training design consisting of a baseline assessment, strategy instruction,
and strategy practice to compare what they labeled “memory plasticity” from mid-
dle childhood to old age. Younger children aged 9–10, older children aged 11–12,
young adults aged 20–25, and older adults aged 65–78 learned and practiced the
Method of Loci, an imagery-based mnemonic technique, to encode and retrieve
words using location cues. Children and older adults performed at similar levels
at baseline and after being instructed in how to use the strategy. However, children
profited considerably more than older adults from subsequent practice in using the
strategy, yielding increasing differences between these age groups over the course
of the training study. According to the authors, these findings suggest that 9- to
12-year-old children and older adults do not differ much in their baseline perfor-
mance and their baseline plasticity. However, children possess greater develop-
mental plasticity than older adults.
These and other findings led Shing et al. (2008) to introduce a two-component
model of episodic memory development across the life span. This model posits two
evolving components of episodic memory—one strategic and the other associative—
and it portrays the ontogeny of episodic memory as the interaction between the two
(cf. Chap. 5). The model posits that the associative component is relatively mature
210 7  The Development of Encoding and Retrieval Strategies

by middle childhood and undergoes senescent decline in late adulthood and old age.
Second, it also posits that the strategic component of episodic memory develops later
than the associative component (e.g., from late childhood on) and also undergoes a
decline in late adulthood and old age (see also Shing and Lindenberger 2011).

Strategies for Remembering Complex Materials

Most studies of strategy development have focused on processes that improve rote
recall of words or pictures. As already noted by Brown et al. (1983), this is not
the only form of learning. Deliberate strategies can also be applied to understand,
store, and retrieve complex meaningful information such as texts. Learning to read
and write requires strategic activities rarely observed in elementary schoolchil-
dren. In particular, early research carried out by Ann Brown and her colleagues
has shown that processes such as identifying, underlining, and summarizing the
main ideas of text passages develop during the high school years but usually only
with explicit instructions (Brown et al. 1983; Pressley and Hilden 2006).
Although these strategies develop rather late, the developmental pattern found
between Grades 6 and 12 resembles that found for rote-recall strategies between
Grades 1 and 6. That is, these activities are only sporadically observed, followed
by more frequent robust application in a broader variety of learning situations. As
is true for simpler strategies, such as rehearsal and organization, the initial use of
strategies for understanding and remembering texts is often accompanied by produc-
tion deficiencies and poor performance. It is only after prolonged periods of prac-
tice that mature, successful strategy application can be observed. As will be shown
later in some detail, strategies related to text processing include metacognitive activi-
ties (e.g., comprehension monitoring), which require skills to integrate and evaluate
information across larger segments of texts. Even college students may have diffi-
culty with such demanding tasks (Brown et al. 1983; Pressley and Hilden 2006).
Training experiments have shown that complicated text processing strategies
can be taught to normal elementary schoolchildren (e.g., Gaultney 1995; Paris and
Oka 1986; Pressley et al. 1988). Instructional programs that include reading strat-
egy components are beneficial to poor readers as well (Palincsar and Brown 1984;
Short and Ryan 1984). In the seminal study by Palincsar and Brown (1984), a set of
four comprehension strategies was developed. Seventh graders who were adequate
decoders but poor comprehenders were taught to (a) make predictions before read-
ing, (b) question as they read, (c) seek clarification when confused, and (d) sum-
marize after reading a section of text. In the reciprocal teaching condition, the
participants were taught these four strategies, with the teacher first discussing the
topic of the day’s text, calling for predictions about the contents of the passage if the
text was completely new, or calling for a review of the main points covered thus far
for passages that had been started the previous day. Next, the teacher assigned one
or two students to be the “teacher.” The student teacher(s) imitated the adult teach-
er’s strategic behavior, and the adult teacher would make corrections if necessary.
Encoding Strategies During the Elementary and Middle-School Years 211

Students took turns as the student teacher with a session lasting about 30 min.
Throughout the approximately 20 days of the reciprocal teaching intervention, the
students were explicitly informed that questioning, summarization, prediction, and
seeking clarification were strategies that would help them understand better and
that they should always try to use these strategies. The intervention was followed by
5 days of post-testing. A long-term follow-up was provided 8 weeks later.
Relative to the control condition, reciprocal teaching positively affected all
measures of text comprehension and memory that were collected. Given the large
effect sizes, the reciprocal teaching paradigm had a wide-ranging impact on prac-
tice. For instance, the outcome of this research stimulated textbook publishers to
include strategy instruction and inspired additional study of the method. Ten years
later, Rosenshine and Meister (1994) came up with a meta-analysis of the impact
of reciprocal teaching. They concluded that reciprocal teaching was most suc-
cessful when there was more direct teaching of the four strategies. Obviously, the
approach produced consistent and sometimes striking effects on cognitive process
measures even though these effects were reduced when standardized comprehen-
sion measures were used. Positive effects were also reported by Scott Paris and
his colleagues (e.g., Paris et al. 1984; Paris and Oka 1986) who used a somewhat
different methodological approach by devising and evaluating a 30-h approach
to teaching a variety of comprehension strategies in classrooms. This approach
emphasized students’ understanding of strategic processes, facilitated by a lot of
explanations from teachers and modeling of the relevant strategic activities. Given
that the teaching took place in classroom groups and that there was only a lim-
ited application of the relevant strategies, this approach lacked the intensity of the
reciprocal teaching procedure. Although the outcome of this “informed strategy
training” approach was generally positive, the findings were less impressive than
those obtained by Palincsar and Brown (cf. Pressley and Hilden 2006).
The brief overview of this literature showed that the spontaneous production
of strategies for learning and remembering complex text material is the exception
rather than the rule, and even adults may fail to use such strategies routinely. It
appears that traditional summaries of the development of memory strategies focus-
ing on rote-recall memory tasks frequently underestimate the variation in the age-
related development of natural strategic competence. Whereas simple strategies
can already be detected in early childhood, more sophisticated and complex pro-
cedures are not commonly observed even in bright and mature participants such as
university students (Pressley and Afflerbach 1995; Pressley and Van Meter 1993).
The development of memorability-based strategies. One line of recent
research has investigated the strategies children use to remember events, and
in particular, the strategies they use to reject false memories (e.g., Ghetti and
Alexander 2004; Ghetti et al. 2006; Roebers and Schneider 2005). For instance,
Ghetti and her colleagues investigated a strategy that they named the memora-
bility-based strategy. Such a strategy refers to the expectation of the memora-
bility of a specific event. When children are asked whether they recall a specific
event, they may rely in part on their estimate of the salience or probability of
such an event. Some events, if they happened, would surely be recalled (“Do I
212 7  The Development of Encoding and Retrieval Strategies

remember being hit by a car? No, and if I had been, I am sure I would remember
it!”), whereas others may be less salient (“Did I eat sausage for dinner last week?
Maybe.”). Ghetti and Alexander (2004) informed 5-, 7-, and 9-year-old children
about possible events that may have happened to them, some with high memora-
bility (e.g., taking a trip to the Grand Canyon) and others with low memorability
(e.g., wearing a bandage after hurting an eye). Children were then told that peo-
ple typically remember high-memorability events, whereas they probably forget
low-memorability events more easily. When children were asked whether they
had experienced certain events, only the 9-year-olds rejected the high-memorabil-
ity events more than the low-memorability events. In a subsequent training study
(Ghetti et al. 2006), 6- to 7-year-old children rejected bizarre (high-memorability)
events, which indicates that even young schoolchildren can use a memorability-
based strategy under favorable conditions.
As found for rehearsal and organizational strategy development, most strategy
development related to elaboration, text processing, and estimates of memorabil-
ity has been attributed to encoding factors. However, there is evidence indicating
that some developmental differences are tied to retrieval, particularly in younger
children (Pressley and Levin 1980). In the next section, the importance of retrieval
strategies for memory performance will be explored in more detail.

Retrieval Strategies

Although the strategies described so far have been described as encoding strate-
gies, their effects do not become apparent until retrieval. Sometimes children fail
to recall all that they encode. Something has to be done to get the relevant infor-
mation out of long-term memory. It is possible to increase the probability of recall
through strategic efforts at the time of testing. Retrieval strategies refer to planful
operations that individuals undertake in order to access stored information. As we
have shown above, there is a slow developmental progression in the emergence of
retrieval strategies in young children (e.g., Hudson and Fivush 1983). Even older
children seem to have a lot to learn about the use of deliberate strategies for cueing
and searching memory. However, with increases in age, they become better able to
more effectively implement these strategies (Ackerman 1985; Kobasigawa 1977;
Schneider and Bjorklund 1998).

Developmental Trends in the Use of External and Internal

Retrieval Cues

In contrast to the voluminous literature on encoding strategies, relatively little

work has been carried out on retrieval strategy development in children and ado-
lescents (Ornstein and Light 2010). Like encoding strategies, retrieval strategies
Retrieval Strategies 213

vary considerably in complexity and sophistication. Age differences in the effec-

tive use of retrieval strategies have been frequently explored. Kobasigawa (1977)
provided a detailed analysis of potential deficiencies in children’s retrieval: (a)
Children may fail to perceive that internal or external memory cues are potential
memory aids; (b) They may lack strategies for locating target information in mem-
ory; and (c) They may not have had enough experience with the problem to evalu-
ate when the search process should be considered finished.
Regarding the first potential deficit cited by Kobasigawa, several studies have
demonstrated that young schoolchildren have trouble using external memory cues.
For instance, Kobasigawa (1974) presented lists of categorizable items to children
in Grades 1, 3, and 6 to learn for subsequent free recall. He also provided picture
cards to the children during learning and recall. The cards served to classify the
objects (e.g., pictures of animals went with a cue card of a zoo). After study, chil-
dren were assigned to one of three different assessment conditions. Children in a
free-recall condition were instructed to remember as many of the pictures as they
could, in any order. By contrast, those in an available cue condition were given
the pictorial cue cards and told that they could use these cards to facilitate remem-
bering. However, no instructions were provided concerning the order in which the
items should be recalled. Finally, children in the directive cue condition were
given the pictorial cue cards one at a time, told that three items had been paired
with each picture card, and were then instructed to recall as many items that were
related to the first picture card as they could remember, then the second picture
card, and so on.
The findings indicated that the recall performance of the children in the Free-
recall condition increased with age and that age differences were eliminated for
the children in the directive cue condition. Thus, the instructions to recall accord-
ing to the semantic categories facilitated the performance of all participants, but
the younger children benefitted more than the sixth graders. By contrast, the pres-
entation of pictorial cues in the available cue condition facilitated performance
only for the sixth graders, who used the cues rather systematically, but not for
the two younger age groups. Thus, a major result was that the younger children
in the sample did not spontaneously use retrieval cues at testing to guide their
recall, and they also displayed no benefits when told that the retrieval cues could
be used. However, their recall was much improved and comparable to that of
the older children when they were explicitly told to use the cues at recall. These
findings suggest that the younger children had stored as much information about
the target stimuli as the older children. Their major problem was getting it out
(Howe et al. 1985; see Halperin 1975, for similar findings). Thus, developmen-
tal improvements in recall were interpreted by Kobasigawa (1974, 1977) to
mean that there was a significant developmental difference in the efficient use of
retrieval strategies.
Retrieval deficits in grade-school children are also evident with regard to the
use of internal cues in the form of category names. An early example was provided
by Scribner and Cole (1972). Second, fourth, and sixth graders were presented
20 categorizable nouns in a random order. As in Kobasigawa (1974), the children
214 7  The Development of Encoding and Retrieval Strategies

were aware of the category names during learning and recall. The participants in
the control condition were simply instructed to recall the items when given the
category cues. By contrast, the participants in the experimental condition had to
recall all the items from one category before they were allowed to proceed to the
next category (constrained recall). For all ages, the children in the experimental
condition recalled more items than the children in the control condition. However,
none of the children transferred the constrained strategy when given a transfer
list consisting of to-be-learned items from new categories. Generally comparable
results were reported by Lange (1973), whose procedure was similar except that
he used pictures rather than verbal materials and used the same categories for the
transfer task.
How should these findings be interpreted? Methodological problems in the
studies by Scribner and Cole (1972) and Lange (1973) preclude a definitive con-
clusion. First, including a free-recall condition without retrieval cues would have
improved both studies. Such a condition would have made it possible to better
estimate the age-specific effect of cue presentation, particularly during the transfer
phase. In addition, nothing was known about how participants behaved during the
encoding phase. Age-correlated differences may have occurred because the older
children had already created better conditions for retrieval, that is, by organizing
the learning material into categories during the encoding phase. Finally, the to-be-
remembered lists were presented several times in succession in these studies, with
items presented in a random order on each trial. Such a design increases the prob-
ability of encoding variability (Waters and McAlaster 1983; Waters and Waters
1979) because participants encounter each item in the encoding context and then
again in a subject-determined retrieval context (i.e., the item is retrieved with other
items such that the order of the retrieved items is often quite different from the
order at encoding). A great deal of encoding variability affects younger children
more negatively than older children. Thus, these experiments may have provided
poorer conditions for effective learning by younger children than for older chil-
dren. In short, it is difficult to differentiate between the quality of encoding and
retrieval processes in the experiments conducted by Kobasigawa (1974), Lange
(1973), and Scribner and Cole (1972), even though all of these studies suggest
retrieval deficiencies in young children.

Age Differences in the Flexibility of Retrieval Strategies

One solution to this problem is to minimize age-correlated differences in encod-

ing strategies, for example, by having the items sorted exactly according to their
taxonomic groups during the learning phase. A number of studies (Ceci and Howe
1978; Ceci et al. 1980; Worden 1974) have used such a procedure. The study by
Ceci and Howe (1978) is especially relevant. These authors used to-be-learned
lists that could be organized thematically and taxonomically and determined that
Retrieval Strategies 215

25

20

15

10

0
4 7 10
AGE

Cued recall Free recall

Fig. 7.5  Mean numbers of pictures cue recalled and free recalled along with the proportion of
mode switching as a function of age. (Schneider and Pressley 1997, p. 172)

4-, 7-, and 10-year-old children could organize the learning materials perfectly in
both modalities. A cued-recall test given immediately after sorting used taxonomic
cues for half of the children at each age level and thematic cues for the remaining
children. Free recall was measured 1 day later.
The main results are summarized in Fig. 7.5. Please note that cued recall varied
little as a function of age, suggesting that about the same number of items were
available in memory for each of the age groups. The free-recall results were quite
different. Older children spontaneously generated many of the taxonomic and
thematic cues and used them to mediate free recall, but younger children did not
do so. A “modal switching index” was calculated to indicate the degree to which
children switched between taxonomic and thematic modalities during free recall.
Although the 10-year-olds made such a shift in 66 % of all possible cases, the cor-
responding percentages for the 4- and 7-year-olds were 14 and 29 %, respectively,
which shows that the two younger age groups did much less switching. Within
each age level, there was a reliable positive relation between this index and the
number of items recalled. Additional analyses led Ceci and Howe (1978) to con-
clude that the age differences in free recall observed in this study were primarily
due to the use of more flexible retrieval strategies by older children.
Age differences in the flexibility of retrieval strategies were also demonstrated
by Salatas and Flavell (1976b). They attempted to control for developmental dif-
ferences in encoding strategies by having the participants (6- and 9-year-olds
and college students) learn a set of items to a high level of performance, that is,
to the point that they could recall all categorizable list items during constrained
recall (i.e., when category cues were provided and an exhaustive recall of category
216 7  The Development of Encoding and Retrieval Strategies

members was required). At the end of this phase, participants were asked indirect
retrieval questions that required an optimal retrieval plan. For instance, to answer
the question, “which of the items on the list are small enough to be put into this
box?” the optimal retrieval plan would be to recall the category names, search in
each category exhaustively, and evaluate each item in each category to see whether
it meets the requirements. Given the complexity of this retrieval strategy, it is not
surprising that only a few schoolchildren used it. However, most of the college stu-
dents applied this optimal strategy for retrieval.
A study by Keniston and Flavell (1979) explored developmental trends in another
example of a complex retrieval heuristic. The approach taken by these authors illus-
trates another retrieval problem of younger children referred to by Kobasigawa
(1977), namely lack of adequate search strategies. First, third, and seventh graders
as well as college students were read a list of 20 different alphabet letters, which they
wrote down. Unexpectedly, participants were asked to recall these letters. An effec-
tive retrieval strategy is to go through the entire alphabet and identify the letters that
have been recently seen, thus transforming the difficult free-recall task into a much
easier recognition task. If items are recalled in alphabetical order, one can infer that
such a retrieval strategy was used. The results showed that the younger children could
use this intelligent retrieval strategy effectively and skillfully when told to do so,
although they did not use the strategy spontaneously. By contrast, no specific feed-
back was required for the older participants to use the alphabetic retrieval strategy.
These and other-related findings suggest that young children benefit more
from being instructed to apply retrieval cues than older children and adolescents,
presumably because the older participants spontaneously generate appropriate
retrieval plans. The findings summarized in this section indicate clear trends in the
development of efficient retrieval processes from 4 to 12 years of age. Although
the spontaneous use of simple retrieval strategies can be observed in the late
kindergarten and early school years (cf. Wellman 1985), complex retrieval strat-
egies such as the ones explored in the study by Salatas and Flavell (1976b) are
not observed before the late grade-school or adolescent years. As noted by Flavell
et al. (2002), much of what develops in the area of memory retrieval consists of
the ability to search the long-term memory system intelligently, that is, system-
atically, exhaustively, and selectively—whatever the retrieval problem at hand
demands. In general, this ability develops rather late in childhood, particularly
when the use of internal memory cues is required. On the other hand, we have
already shown that even preschool children demonstrate the ability to effectively
use retrieval cues when searching the external world.

Multiple and Variable Strategy Use

From the discussion of strategy development provided above, one might get the
impression that children use one strategy at a time, progressing stagelike from less
to more effective strategies. Research carried out during the past two decades has
Multiple and Variable Strategy Use 217

made it clear, however, that this is not the case. Rather, there is substantial inter-
and intrapersonal variability in strategy use, with children using different strategies
and combinations of strategies on any given memory problem. This position has
been emphasized by Robert Siegler (e.g., Siegler 1996, 2006), who views cogni-
tive development not as a series of stagelike steps, but rather as changes in the
frequency with which children use different strategies. Accordingly, Siegler views
developmental change as a series of overlapping waves: Strategies occur with dif-
ferent frequencies over time and compete with one another for use. With time, the
modal strategy children use to solve a problem will typically increase in efficiency,
replacing less effective strategies. It seems important to note, however, that simple
and more complicated strategies may coexist in a child’s cognitive repertoire and
compete for use.
Regarding the development of memory strategies, several studies have con-
firmed the view that children make use of more than one strategy at a time. For
instance, Coyle and Bjorklund (1997) reported multiple and variable strategy use
for children in Grades 2, 3, and 4 on a multitrial sort-recall task with categorized
sets of items. These authors assessed the strategies children used at each trial (i.e.,
sorting items into groups, category naming, rehearsal, and clustering). Although
the average number of strategies children used increased over the five trials and
with age, there was substantial variability in the number of strategies children used
per trial in all grades, with more than one-third of the second-grade children using
two or more strategies on a majority of the trials. Children at all ages also varied
the strategies they used. No age differences were found in whether the children
switched strategies (i.e., used one combination of strategies on one trial and a dif-
ferent combination on the next trial). Children in all grades switched strategies
on about half of the trials. Strategy effectiveness was also related to age such that
older children typically improved their memory performance when they increased
the number of strategies they used on adjacent trials, whereas younger children
actually showed a decrement in recall when they increased the number of strat-
egies they used (for similar findings, see Cox et al. 1989; DeMarie and Ferron
2003; DeMarie et al. 2004; Schneider et al. 2009; Shin et al. 2007).
Several studies also revealed that even young children (e.g., kindergarten-
ers) sometimes use more than one strategy on memory tasks (e.g., DeMarie and
Ferron 2003; Schneider et al. 2004; Schwenck et al. 2009). The study carried out
by Schwenck et al. (2009) was the first to systematically investigate multiple strat-
egy use on a memory task with children as young as 4 years of age. Even these
young children used more than one strategy on average (particularly in the prompt
condition) when performing the memory tasks, although the older children in the
study used more strategies and also used the more effective ones. Moreover, recent
research based on the Würzburg-Göttingen longitudinal study indicated that the
first graders who used two or more strategies in a sort-recall task at the beginning
of the longitudinal study outperformed those who did not use a strategy or used
only a single one (Schneider et al. 2004). Subsequent assessments of strategy use
on the sort-recall task included in this study showed that using only one strategy
(either sorting or clustering or rehearsal) led to worse recall performance than
218 7  The Development of Encoding and Retrieval Strategies

Table 7.2  Approximate ages by which most children display spontaneously various memory

strategies effectively for children in schooled culturesa. (Adapted from Schneider and Bjorklund
2003) (Bjorklund et al. 2009, Table 7.1)
6–7 years 8–10 years 11–14 years
Single-item rehearsal X
Cumulative rehearsal X
Organization with highly associated items X
Organization with less highly associated items X
Self-testing/category naming X
Effective allocation of study time X
Retrieval strategies X
Memorability-based strategy X
Elaborationb X
Strategies for remembering complex textb X
Notes
a Younger children than listed often display effective strategies when prompted
b These strategies may not be displayed until later, and many adults fail to use these strategies

effectively

using two or more strategies (see Schneider et al. 2009). Moreover, as shown by
Lehmann and Hasselhorn (2007), effective multiple strategy use was also found
for the rehearsal tasks used in this longitudinal study.
What does the literature on memory strategy development tell us about the
approximate ages by which most children display spontaneously varied memory
strategies? Table 7.2 presents estimates of the typical age of first strategy appear-
ance as a function of strategy type for children from cultures in which they are
schooled (see Bjorklund et al. 2009; Schneider and Bjorklund 2003). As can be
seen from this table, simple memory strategies are already available at the begin-
ning of elementary school, and most strategies are acquired between the ages of
8 and 10 years. Although young elementary schoolchildren can also be trained to
use more complex techniques such as elaboration and text recall strategies, sponta-
neous use of such strategies is not observed before late childhood.

Factors Affecting Children’s Strategy Use and Development

Research in the development of memory strategy presents a relatively clear pic-

ture. Although young children use simple strategies, they rarely spontaneously
use the more complicated strategies employed by older children. Children can be
trained to use such strategies, but they often fail to transfer such training (produc-
tion deficiency), and they sometimes show little gain in memory performance as a
result of using the strategies (utilization deficiency). Charting the development of
these different strategies and documenting the conditions under which strategies
Multiple and Variable Strategy Use 219

can be successfully trained is important, but the underlying factors that are respon-
sible for these age-related patterns are even more critical from a theoretical per-
spective. In the following sections, research that has sought to discover the factors
that influence age differences in children’s use of memory strategies and their
effectiveness will be reviewed, including efficiency of processing, knowledge
base, encoding, and metamemory.
The impact of memory capacity. As noted in Chap. 6, young children’s prob-
lems with the deployment of memory strategies may be caused by the fact that
memory capacity demands are too high. By definition, strategies are effort-
ful cognitive operations. The use of memory strategies results in a cost to mental
resources. Several theorists have proposed that the principal factor underlying chil-
dren’s strategy use is age differences in efficiency of processing. Younger children
require more mental effort to execute most cognitive operations relative to older
children (e.g., Case 1985, 1992), and because of this age difference, younger chil-
dren are less likely to implement a strategy and experience less of a benefit from
its execution than older more cognitively efficient children (e.g., Bjorklund 1987;
Cowan et al. 2005; Guttentag 1997; Kee 1994).
To test this hypothesis, researchers have used dual-task procedures in which
children are asked to perform two tasks, both separately and together. This
approach is based on the simple idea that it is difficult to do two things at once.
The amount of performance reduction on the secondary task (often a simple motor
task such as tapping one’s finger as fast as possible) as a result of performing the
primary task (usually a measure of strategy performance) is used as an indica-
tion of the capacity of the mental resources the primary task requires. The dual-
task methodology has been used to assess mental capacity constraints for various
encoding strategies such as rehearsal, organization, and elaboration. For instance,
Guttentag instructed children in the second, third, and sixth grades to utilize a
cumulative-rehearsal strategy. While these children rehearsed a list of unrelated
words, they were required to perform the finger-tapping task as quickly as pos-
sible. This tapping rate during rehearsal was compared to a baseline rate of tap-
ping only. Guttentag found that children in all grades were able to execute the
cumulative-rehearsal task, and he found that dual-task problems decreased with
age, a finding that he interpreted as showing that younger children allocated more
mental effort toward executing the strategy than older children. In accordance with
this assumption, Guttentag et al. (1987) found that some children who used a pas-
sive rehearsal strategy under normal presentation conditions switched to an active
cumulative-rehearsal strategy under conditions in which the resource demands
of the active strategy were reduced (e.g., when the items remained visible during
study).
Similar developmental differences in dual-task performance were obtained
for organizational strategies. For example, Bjorklund and Harnishfeger (1987,
Experiment 2) asked 9- and 13-year-old children to tap their index fingers as rap-
idly as possible on the space bar of a keyboard (secondary task). They were then
asked to perform two free-recall memory tasks while tapping their fingers. On the
first task, children were asked to recall a set of categorized items in any order they
220 7  The Development of Encoding and Retrieval Strategies

wished (free recall). On the second task, children were instructed in the use of an
organizational memory strategy. Children of both ages significantly increased their
strategy use on the trained task and displayed a significant reduction in tapping
rate on the more effortful trained task relative to the less effortful free-recall task.
However, only the older children showed a corresponding increase in recall rela-
tive to the free-recall task. This demonstrates that strategy use is effortful, espe-
cially for younger children, and, because it consumes a substantial portion of
young children’s limited mental capacity, strategy use may not always result in
enhanced performance (i.e., a utilization deficiency). Bjorklund and Harnishfeger
(1987) interpreted these findings as reflecting the fact that the younger children
expended too much of their limited mental resources executing the strategy and
thus had too few resources left to devote to retrieving individual items.
Finally, there is evidence that capacity limits also influence the acquisition
and use of elaboration strategies. For example, Kee and Davies (1988) examined
sixth graders’ and college students’ use of rehearsal and elaboration strategies on a
paired-associate learning task. As noted above, this is the age range during which
elaborative propensities have been concluded to mature (Kee 1994; Pressley 1982;
Schneider and Pressley 1997). Kee and Davies used a dual-task methodology,
with participants required to press a key as they processed the paired associates.
Instructions on memory trials requested that participants either repeat the names of
the word pairs aloud during the interval (rehearsal) or generate a sentence aloud to
describe the pair in an interaction. Recall results showed the typical improvements
associated with age and elaboration. As indicated by an age by instruction interac-
tion, the advantage of elaboration was larger in the sixth-grade group than in the
college group. Although finger-tapping interference (mental effort) results did not
predict grade and instruction effects in the first experiment, which required par-
ticipants to execute their rehearsal and elaboration of pairs aloud, finger-tapping
interference with elaboration declined with age in the second experiment, which
required participants to perform their strategy application silently. Kee and Davies
(1988) concluded from their findings that age-related recall differences that remain
after successful elaboration of pairs can be attributed to less of an availability of
the information processing resources needed to ensure sufficient encoding (see
Miller et al. 1991, for similar developmental differences in dual-task performance
related to the use of a selective attention strategy).
Further support for the assumption that short-term memory capacity differences
may account for age-related improvements in the ability to profit from elabora-
tion strategy instruction comes from Pressley et al.'s (1987) study, which was
not based on a dual-task paradigm. Children in Grades 1 through 6 heard com-
plicated sentences specifying relations between concrete elements (e.g., “The fat
bay ran with the gray balloon”). In one experimental condition, children were
instructed to try hard to remember the sentences because a memory test would be
given after the presentation of sentences. In the second condition, children were
instructed to construct images depicting the meaning of the sentences. Their short-
term memory was assessed using a battery of different measures. Pressley and col-
leagues assumed that functional short-term memory would be highly predictive of
Multiple and Variable Strategy Use 221

Fig. 7.6  Mean percent 80
learned as a function of 70
grade level and condition.
(Constructed from Pressley 60

Percent learned
et al. 1987) (Schneider and
50
Pressley 1997, p. 249, Fig. 7.3)
40

30

20

10

0
1,2,3 4,5,6
GRADE
Imagery Control

sentence learning when the children used an imagery strategy to code sentences
but would not be similarly predictive of learning when the children learned any
way they wanted.
As in previous research, the imagery versus control difference was significant
for students in Grades 4–6 but not for the younger participants. Thus, there was
a clear developmental effect in that the imagery instructions promoted sentence
learning in the older half of the sample but not in the younger half (see Fig. 7.6).
Much more interesting, however, was that children with higher functional short-
term memory were more successful in the imagery condition than children with
poorer short-term memory capacity. Short-term memory capacity differences
between students were more predictive of performance in the imagery instructions
condition than in the control condition. Causal modeling analyses made clear that
even the predictability of sentence learning from verbal competence was mediated
in part by functional short-term memory. Cariglia-Bull and Pressley (1990) car-
ried out a constructive replication of this study and obtained results consistent with
the earlier study. This increased their confidence in the earlier finding that at least
some of the developmental differences in improved performance under imagery
instructions can be explained by short-term capacity differences in children.
Effects of the knowledge base. Research conducted during the last four dec-
ades has convincingly shown that age-related effects in the frequency of use and
quality of children’s strategies play a large role in memory development between
the early school years and adolescence. However, there is now an increasing real-
ization that the use of encoding and retrieval strategies largely depends on chil-
dren’s strategic as well as nonstrategic knowledge. There is now a broad consensus
that the narrow focus on developmental changes in strategy use should be replaced
by an approach that also takes into account the effects of various forms of knowl-
edge on strategy execution.
The impact of children’s knowledge base on their strategy use has been dem-
onstrated in numerous studies (see reviews by Bjorklund 1987; Bjorklund et al.
1990; Chi and Ceci 1987; Ornstein et al. 1988; Schneider and Bjorklund 1998,
222 7  The Development of Encoding and Retrieval Strategies

2003). We usually define the knowledge base as children’s long-term represen-

tations of world knowledge, particularly as it relates to how sets of “facts” are
organized in a modified semantic memory network. Knowledge base refers to
what children know about typical (or perhaps atypical) events that occur in their
culture (e.g., going to restaurants, going to church, going camping) and about
relations among items within a domain (e.g., football, chess, dogs, Pokemon
cards). The more people know about a topic, the easier it is for them to learn and
remember new information about it. With age, the number of items in a child’s
knowledge base expands, as do the number and strength of associations and the
number and type of features that characterize items (see Bjorklund 1987). We use
the term elaborated knowledge base to refer to a domain of knowledge that is
highly organized with many strong connections among the items within the base
(i.e., so that the activation of one item in the domain will make the activation of
related items easier). The primary effect that an elaborated knowledge base has
on memory is to increase the speed of processing for domain-specific information
(Bjorklund et al. 1990, 2009). The various studies using the dual-task paradigm
described above have shown that younger children require greater mental effort to
use memory strategies than older children. Mental effort is reduced when highly
familiar stimuli are used, leaving more of one’s limited capacity to search for indi-
vidual items or to execute a complex strategy (Guttentag 1989; Kee 1994). If that
faster processing can be equated with more efficient processing, the result is a
greater availability of mental resources.
Age and individual differences in children’s knowledge base have been shown
to be important factors in their memory strategy use and effectiveness (see
Alexander and Schwanenflugel 1994; Bjorklund et al. 1990; Hasselhorn 1995; Kee
1994; Schneider 1993). What children know affects how they process information,
and older children generally know more about the things they are asked to remem-
ber than younger children.
Research using a variety of materials (e.g., categorized words, texts) has con-
sistently demonstrated that children with more detailed knowledge of the to-be-
remembered information recall that information better and typically implement
strategies more effectively than children with less detailed knowledge (e.g., Best
1993; Bjorklund and Zeman 1982; Schneider 1986; Schneider et al. 1989).
According to Bjorklund (1987), age differences in domain knowledge can
affect memory in three different ways: (a) by increasing the accessibility of spe-
cific items, (b) by the relatively effortless activation of relations among sets of
items, and (c) by facilitating the use of deliberate strategies. Regarding the latter,
domain knowledge can also generalize strategy use to related domains or even
diminish the need for strategy activation (Schneider 1993).
As already noted above, several developmental studies have shown substantial
relations between conceptual knowledge and strategy use. Experimental manipu-
lations have often examined children’s knowledge of categorical relations among
items, such as category typicality or interitem associativity (e.g., Frankel and
Rollins 1985; Schneider 1986; Hasselhorn 1992). This research has clearly shown
that differences in the meaningfulness of words have a considerable influence on
Multiple and Variable Strategy Use 223

the processing of verbal information in semantic categorization tasks. One reason

proposed for the conceptual knowledge effect is that semantic relations between
high associates are activated with relatively little mental effort, resulting in
retrieval that appears organized and strategic even in young children but does not
require the use of a deliberate strategy. Whereas young children’s strategic compe-
tencies may be underestimated in situations in which complex effort-consuming
strategies such as cumulative rehearsal are required, they may actually be over-
rated in situations in which high associativity among items automatically guides
the structure of recall (see Bjorklund 1987; Roebers 2014; Weinert et al. 1988).
There are also examples demonstrating the fact that knowledge generalizes
strategy use to related domains. For instance, Bjorklund and Jacobs (1985) pro-
posed that semantic organization initially seen in recall by young schoolchildren is
not mediated by a deliberately imposed strategy but rather by the relatively auto-
matic activation of well-established semantic memory relations. As children auto-
matically process highly related items in a categorical fashion, some of them may
notice categorical relations in their recall. They may then realize that categoriza-
tion is a useful learning technique. Hence, they may initiate categorization strate-
gies even when materials are not highly related (see Best and Ornstein 1986, for
another example of this generalization effect). These findings support the hypoth-
esis that there is a carryover from the processing of materials that “elicit” more
sophisticated processing because of semantic characteristics (e.g., relatedness) that
are readily accessible in the knowledge base to materials that are not so obviously
consistent with prior knowledge. How does one explain such carryover effects?
Bjorklund (1987) referred to the process of reflective abstraction when discuss-
ing this phenomenon. At some point in their development, children are able to
reflect upon the outcomes of their own cognitions and to examine the products of
their recall efforts. In this process, they may recognize previously unnoticed rela-
tions among the recalled items. Such carryover experiences are believed to be crit-
ical in the development of strategy use (see also Ornstein et al. 1988). Apparently,
the deliberate activation of category knowledge as a retrieval strategy does not
occur before the late elementary school years (Bjorklund et al. 1990; Hasselhorn
1990; Schneider 1986).
There is also evidence in the literature that there are nonstrategic effects of the
knowledge base. In some situations, the greater memory shown by more knowl-
edgeable children is not accompanied by increased strategy use in terms of high
levels of organization (e.g., Bjorklund and Bjorklund 1985; Bjorklund and Zeman
1982; Gaultney et al. 1992; Schneider and Bjorklund 1992; Schneider et al. 1996).
As an example of this, in research by Bjorklund and Zeman (1982), children were
asked to recall the names of their current school classmates (class-recall task).
They were also asked to memorize and recall a list of taxonomically clusterable
items. Although the usual age differences in recall and clustering were found for
the traditional list learning task, children of all ages could remember the names
of their classmates equally well. Recall was generally high, and most children
showed relatively high levels of clustering based on seating arrangements, read-
ing groups, or children’s sex. However, there was little relation between degree
224 7  The Development of Encoding and Retrieval Strategies

of strategy use and recall. And in a subsequent experiment when children were
required to recall their classmates by a specified category (seating arrangement or
sex of child), children of all the ages that were tested (6, 8, and 10 years) did so
almost perfectly but did not recall a greater number of names than the children
who organized their recall less well (Bjorklund and Bjorklund 1985). For these
children, it seems that most of their elevated levels of recall were attributed to
nonstrategic factors associated with an elaborated knowledge base. One conclu-
sion from this research is that the likelihood of nonstrategic effects of knowledge
increases with increases in the richness of the knowledge base.
The impact of metamemory on strategy use. There is also evidence that the
knowledge that individuals possess regarding the functioning and contents of their
memory (i.e., metamemory) may influence their deployment of strategies and
developmental changes in their strategy use (Bjorklund et al. 2009; Ornstein and
Light 2010; Schneider and Pressley 1997). The development of metamemory in
children and adolescents will be described in more detail in Chap. 9. This section
briefly summarizes what children and adolescents know about memory strategies
and how this knowledge affects strategy use and memory performance.
In fact, much of the research on the development of metamemory has been moti-
vated by the assumption that children’s knowledge about memory influences their
selection of specific strategies on tasks that require remembering (e.g., Brown et al.
1983). Flavell (1971), who coined the term, was interested in metamemory as a par-
tial determinant of production deficiencies in children’s strategy use. Kreutzer et al.
(1975) came up with the first comprehensive questionnaire for assessing children’s
metamemory. Kindergarteners and first, third, and fifth graders were asked a series
of questions regarding memory. Children’s knowledge about organizational strate-
gies was tapped by a single item. The participants were shown nine picture cards
(three from each of three semantic categories), and they were told to imagine that
these items had to be learned in a few minutes. They were also told that they could
do anything they wished to learn the items. There was a clear age trend. Whereas
most fifth graders in the sample referred to a categorization strategy, only one kin-
dergarten child did so. Findings from several other studies have confirmed that meta-
memory concerning strategies develops steadily over the elementary school years
but is still incomplete in early adolescence (for details, see Chap. 9).
One of the main motivations for studying metamemory has been the assump-
tion that there are important relations between knowing about memory and
memory strategies (cf. Brown 1978; Flavell and Wellman 1977; Weinert 1986).
However, early investigations did not find substantial links between the two com-
ponents (Cavanaugh and Perlmutter 1982). More recent research has shown that
the relation one finds between memory and metamemory is considerably stronger
than previously assumed. The findings of several studies are in accord with theo-
retical assumptions that emphasize a bidirectional relation between metamemory
and strategy use (see Chap. 9 for an overview). Accordingly, metamemory can
influence memory behavior, which in turn leads to enhanced metamemory.
Taken together, research on the role of metamemory in memory development
has created a large body of evidence supporting the utility of the concept. Mainly
Multiple and Variable Strategy Use 225

due to methodological improvements, more recent work on the metamemory-

memory link has provided evidence for rather strong relations between meta-
memory, memory strategies, and memory performance (see Schneider 2010, for
a review of this literature). Overall, the available evidence suggests that children
who understand how memory strategies work show significantly higher levels of
recall than children who cannot provide explanations for their strategic behav-
ior, regardless of age. Increases in capacity and domain knowledge may permit
the execution of strategic behavior, but an understanding of the causal relations
between such behaviors and desired outcomes may be necessary for their effective
use (Joyner and Kurtz-Costes 1997).
The impact of environmental factors and schooling on strategy use.
Evidence has accumulated from cross-cultural studies that strategy acquisition
does not occur naturally but is broadly situated in the cultural context and cultural
institutions in which memory skills are practiced (e.g., Kurtz 1990; Mistry 1997;
Rogoff and Mistry 1990; Wagner 1981). These studies have shown that cultures
clearly differ in the extent to which they support and encourage particular strategic
activities. More specifically, they have demonstrated a powerful effect of school-
ing on strategy use and memory performance. Given that differences between
schooled and nonschooled children do not usually show up until the schooled chil-
dren have received several years of formal instruction, it appears that experience at
school influences the acquisition of memory strategies. Mnemonic skills typically
develop in the context of school learning in which remembering is both expected
and valued (Cole 1992; Cole and Scribner 1977; Mistry 1997).
The research strategy most frequently used in cross-cultural studies of memory
consists of measuring memory performance in American and non-Western sam-
ples. Much of this work has been conducted with the Kpelle and Vai in Liberia
(Scribner and Cole 1981), in Morroco (Wagner 1974, 1978), Mexico (e.g., Rogoff
1981), and Guatemala, and with the aborigines in Australia (e.g., Kearins 1981,
1983). Several cross-cultural studies were motivated by the fact that memory strat-
egies such as rehearsal and organization are especially helpful for children from
Western-industrialized societies where school activities involve rote memorization
and list learning. It was assumed that these strategies might not be similarly useful
to nonschooled children from nonindustrialized societies where the most impor-
tant memory tasks might involve recalling the location of objects or landmarks in
a natural setting or remembering instructions passed along in a context of proverbs
and stories (Bjorklund 2012).
For instance, Cole et al. (1971) studied the development of the free recall of
lists of categorically related words, providing some of the most complete cross-
cultural information on memory development. Although the to-be-learned mate-
rials were carefully selected to ensure that the participants (unschooled children
and adults of the Kpelle tribe from Nigeria) would be quite familiar with them,
the memory differences between the African sample and a comparison group from
California were substantial. Whereas the usual adult versus child differences were
obtained in the American sample, the adults and children did not differ in the
Kpelle sample. The latter group also did not improve much over trials, and there
226 7  The Development of Encoding and Retrieval Strategies

was no indication that the African sample organized materials according to seman-
tic categories. Taken as a whole, these results indicate that semantic organization
strategies are not spontaneously employed by the Kpelle.
Similarly, Rogoff (1981) reported that nonschooled children generally do not
make use of organizational techniques for remembering lists of items and that
school appears to be necessary for the acquisition of these skills (Rogoff and
Mistry 1990). Moreover, Wagner (1974, 1978), using a serial memory task, found
that intentional rehearsal processes could not be detected in nonschooled samples
but were evident in schooled participants. Subsequent work (Wagner and Spratt
1987) supported the assumption that culturally related experiences affect memory
processing. For example, children in Morocco who are enrolled in Quranic pri-
mary school are asked to practice rote learning to a great extent, and this explains
the superiority of their serial learning (i.e., memory for information in a specific
order). The Quranic students have developed strategies for implementing the
memory task that is demanded of them the most.
Although nonschooled children and adults are unlikely to spontaneously
engage in memory strategies, they are able to make use of them when interitem
associations are made explicit (Sharp et al. 1979). Moreover, cross-cultural differ-
ences in memory performance disappear when information can be integrated in a
meaningful context and is linked to everyday experiences (and not to the amount
of schooling) as has been demonstrated for the recall of stories and visuospatial
memory (e.g., Kearins 1981; Mandler et al. 1980; Rogoff and Waddell 1982).
Nonschooled children and adults perform at the same level or even better than
Western samples on personally relevant memory tasks in which the memory feat
is accomplished in the service of a culturally important nonmnemonic goal. Thus,
meaningful purposes integrate the memory task in an appropriate cultural activity
(Mistry 1997).
Moreover, cross-cultural differences in stimulus familiarity—and thus, in
knowledge—appear to be a critical factor. For instance, Kearins (1983) reported
differences favoring aboriginal children over civilized children in the verbal recall
of lists of wild animals and assumed that aboriginal children’s interest in and
knowledge about wild animals promoted their memory in this task. The data sug-
gested that Anglo-Australian children were more likely to attempt verbal rehearsal
strategies, whereas aboriginal children seemed to rely more on spatial–imaginal
elaboration strategies. In summary, Kearins’ data clearly support the hypothesis
that the knowledge base and culturally supported visuospatial skills can more than
compensate for verbal strategies. Findings from cross-cultural studies have also
shown that individual differences in strategy use and memory performance are not
determined solely by maturational factors but depend greatly on prior familiarity
with the to-be-remembered materials. Moreover, this research consistently sug-
gests that formal schooling contributes significantly to children’s increasing skill
in the use of memory strategies.
Educational effects on strategy use. As noted above, there is evidence
that the development of memory strategies is linked to the effects of school-
ing. Interestingly, educational effects on strategy use have also been observed in
Multiple and Variable Strategy Use 227

cross-national studies in which the amount of formal schooling was comparable

across the participants in the study. Recall that I already mentioned systematic
age differences in the spontaneous use of organizational strategies above, with
German children typically acquiring the strategy earlier than American children.
The available evidence suggests that individual differences in the task demands
of different school systems play an important role. Several cross-national train-
ing studies have revealed systematic and reliable differences between American
and German schoolchildren’s use of organizational strategies (Carr et al. 1989;
Kurtz et al. 1990; Schneider et al. 1986). German second and third graders showed
higher levels of sorting and clustering than their American peers before training.
However, levels of sorting, clustering, and recall were comparable between the
two groups after training, indicating that the pretraining difference was mainly due
to a production deficiency on the part of the American children. Results from par-
ent and teacher self-report questionnaires showed that differences in the degree to
which strategies were taught by parents and teachers paralleled the strategic dif-
ferences between the American and German children. Furthermore, the degree to
which parents taught strategies was also associated with children’s metamemory.
Accordingly, differences in educational practices and instructional systems have
an impact on strategy acquisition and may account for some cross-cultural differ-
ences in memory performance.
Moreover, several studies by Moely and her colleagues (Moely et al. 1992,
1986, 1995) conducted within different schools in the USA indicate that teach-
ers within a given country vary enormously in the amount of strategy instruction
they provide. Classroom observations have revealed that generally little teaching
of strategies was found for a broad range of activities even though the frequency
of strategy teaching varied with the subject matter. That is, strategies were more
likely to be offered when math problem solving was taught than during language
lessons. Furthermore, it was shown that children from high-strategy-instruction
classrooms benefited more from subsequent strategy instruction on an experi-
mental memory task than children from low-strategy-instruction classrooms did
(Moely et al. 1995). These and other findings (e.g., Folds et al. 1990; Kurtz 1990)
illustrate that individual differences in school experiences not only influence the
acquisition of memory strategies but also affect the development of children’s
metacognitive knowledge base.
One of the most important outcomes of the studies carried out by Moely and
colleagues was that mnemonic strategies are not generally taught by teachers
throughout the elementary school grades. Nonetheless, differences in teachers’
memory-relevant behaviors may have a long-lasting impact on children’s strat-
egy development. In an ongoing longitudinal study, Ornstein, Coffman, McCall,
and colleagues evaluated first-grade teachers’ use of “mnemonic orientation” in
their classrooms and related it to children’s strategic memory behavior (Coffman
et al. 2003, 2008; Ornstein et al. 2010). However, even though these authors found
that mnemonic strategies are not generally taught in an explicit fashion, they
observed that first-grade teachers employed a considerable amount of “memory
talk” over the course of their whole-class instruction, including indirect requests
228 7  The Development of Encoding and Retrieval Strategies

for deliberate remembering, strategy suggestion, and metacognitive questioning

(Coffman et al. 2003, 2008). Moreover, children in classes taught by teachers who
used more of this memory-related language displayed higher levels of strategy use
on memory tasks than the children of teachers with a low mnemonic orientation,
and this difference was maintained at least into the second grade when children
were in new classrooms with different teachers. Ornstein et al. (2011) found that
teacher’s mnemonic style in the first grade was linked to the children’s organ-
ized sorting patterns and study skills on more difficult tasks administered in the
fourth grade. Ornstein and colleagues concluded from their findings that aspects
of the language used by teachers may be relevant for the development of children’s
memory strategies. This work thus suggests that just as “parent talk” about events
can impact preschoolers’ developing abilities to remember (e.g., Boland et al.
2003; Reese et al. 1993), “teacher talk” may also be relevant for the emergence
and refinement of mnemonic skills.
Models of Good Strategy User and Good-Information processing. On the
basis of the assumption that good memory is the product of a number of interact-
ing factors such as strategies, capacity, knowledge base, metamemory, and motiva-
tion, Pressley et al. (1987, 1989) developed two different models. Whereas their
earlier model (“The Good Strategy User”) focused primarily on the importance of
sophisticated strategy use for good and efficient thinking, their more recent model
(“good-information processing”) emphasizes the fact that strategies never operate
in isolation but must be considered in the light of other factors.
One of the merits of the good-information processing model is that it provides
a detailed account of interactional processes in the acquisition of strategies and
metacognitive knowledge. Pressley et al. (1989) conceive of metamemory as a
number of interacting mutually dependent components. They suggest that there is
a developmental orderliness to strategy development. At one extreme are young
children who possess only general action control tendencies and the beginnings of
world knowledge. By contrast, mature learners find it easy to appropriately deploy
higher order, goal-specific, and monitoring strategies. The process of knowledge
acquisition induced by strategy instruction is characterized by three steps:
1. Children are taught by parents and teachers to use a strategy, and with repeti-
tion, come to learn about the attributes and advantages of the strategy (specific
strategy knowledge). If children’s environments at home and in school are
stimulating, more strategies will be encountered, leading to an expanded knowl-
edge of specific strategies.
2. Teachers may demonstrate similarities and differences across the many strat-
egies in a domain, allowing children to structure strategies according to their
shared features (relational knowledge).
3. At this point, children recognize the general utility of being strategic, leading
to general strategy knowledge. Children then learn to attribute their success-
ful learning outcomes to the effort they expended in strategy deployment. They
also acquire higher order skills such as selecting and monitoring strategies that
are appropriate for the task (metacognitive acquisition procedures).
Multiple and Variable Strategy Use 229

The core assumption of the model is that such higher order skills and also attribu-
tional beliefs about the utility of strategies (the motivation component) are highly
important for producing generalized thinking and problem solving. The good-
information processing model highlights the impact of educational experiences
on the development of memory skills and thus emphasizes the role of parents
and teachers in the acquisition of strategies and knowledge. One of the obvious
implications of this work is to transfer strategy instruction from the laboratory
to real-world educational settings. Subsequent work by Pressley and colleagues
(e.g., Pressley 1995; Pressley and Woloshyn 1995) illustrates how this goal can be
accomplished.

Summary

About 50 years ago, the modern area of memory development began with the
study of strategies. My overview of the literature has shown that scientific inter-
est in the development of memory strategies is still alive even though our view
of the role of strategies in memory development has changed somewhat since the
early days. Much of the early research in strategy development focused on the
factors responsible for production deficiencies, the subsequent failure to transfer
an acquired strategy to a new situation, and the ways in which children’s strat-
egy effectiveness can be improved (see reviews by Pressley and Van Meter 1993;
Schneider and Pressley 1997). Several studies showed that insufficient mental
capacity and lack of domain-specific knowledge were partially responsible for
production deficiencies in young children. Research based on dual-task procedures
in which children are asked to perform two tasks both separately and together
demonstrated that young children require more mental effort than older ones for
implementing and executing memory strategies. It was also found that young chil-
dren who try to employ memory strategies for the first time are at risk of expe-
riencing a utilization deficiency such that strategy use is not accompanied by
superior memory performance. Moreover, more recent research has indicated that
individual differences in domain knowledge contribute to age differences in strat-
egy use (Bjorklund and Schneider 1996; Schneider and Bjorklund 2003). Young
children typically lack domain knowledge, which increases the speed at which
they can process domain-related information. As a consequence, more mental
energy is needed to execute a strategy based on this information, making strategy
use very effortful and less attractive. Overall, the research literature suggests that
domain knowledge, metamemory, motivation, and educational experiences can be
viewed as mediators of children’s strategy use and memory performance.
Furthermore, numerous studies have shown that memory strategies develop
most rapidly over the elementary school years. Older children are more likely to
actively rehearse items and to group items on the basis of meaning and to study
same-category items together, with higher levels of sorting and clustering yielding
higher levels of recall. However, the ages of strategy acquisition are relative and
230 7  The Development of Encoding and Retrieval Strategies

variable within and between strategies. Even preschoolers and kindergarten chil-
dren are able to use intentional strategies, both in ecologically valid settings such
as hide-and-seek tasks and in the context of a laboratory task. There is also no
doubt that memory strategies can be effectively taught to young children.
Another interesting finding derived from recent research is that memory strat-
egies do not develop in as straightforward a way as we once thought. Instead,
findings from longitudinal research indicate that strategies do not develop gradu-
ally but may actually increase abruptly and in combination with other strategies.
Longitudinal research conducted during the last two decades has convincingly
shown that individual differences in the frequency of use and quality of children’s
strategies play a large role in memory development between the early school
years and adolescence. Recent longitudinal findings concerning multiple strategy
use have provided evidence for considerable individual differences in the intra-
individual development of organizational and rehearsal strategies, indicating that
the course of children’s strategy acquisition may be more complex and variable
than previously assumed. These new findings are helpful for understanding and
explaining the discrepancy in findings reported in developmental studies on the
course of memory strategy acquisition. They contribute to our understanding of
developmental changes in strategic memory across the entire elementary school
period by showing that the process of strategy acquisition is not one of sim-
ply replacing the ineffective with the effective, even though the majority of chil-
dren are competent strategy users when they finish elementary school. Strategies
do not always help performance, at least not immediately, and simple and inef-
ficient strategies reside alongside more sophisticated and efficient ones. Finally,
differences in educational contexts have been shown to affect children’s strategy
acquisition, suggesting that a teacher’s focus on “memory talk” and teacher–child
conversations in the classroom are of great relevance for the development of the
repertoire of strategies that can be skillfully deployed in the service of goals of
remembering.
Chapter 8
Effects of the Knowledge Base on Memory
Development

It has been recognized for some time that memory performance is highly dependent
on the development of a person’s knowledge base. Flavell (1985) eloquently described
the situation:
Thus, what the head knows has an enormous effect on what the head learns and remem-
bers. But, of course, what the head knows changes enormously in the course of develop-
ment, and these changes consequently make for changes in memory behavior (p. 213).

There is reason to believe that one of the major factors that influence memory
development is children’s acquisition of knowledge. In fact, some researchers
view the role of age differences in knowledge as central to memory development.
For instance, according to Carey (1985), the acquisition and reorganization of
domain knowledge probably account for most of the cognitive differences between
young children and adults.

Conceptualizations of Knowledge

Unfortunately, numerous conceptualizations of knowledge can be found in both the

developmental and educational literatures (Dochy and Alexander 1995; Schneider and
Weinert 1990). The long list of knowledge components includes terms such as semantic
knowledge, episodic knowledge, conceptual knowledge, prior knowledge, the knowl-
edge base, content knowledge, and domain-specific knowledge. Given this impressive
inconsistency in terminology, it is understandable that even experts in the field get con-
fused about the possible denotations and connotations of the concept. The definitions
found in the literature suggest that world knowledge, semantic knowledge, and con-
ceptual knowledge are perceived as synonymous concepts and that the same is true for
terms such as prior knowledge, content knowledge, domain-specific knowledge, and
the knowledge base. But how are these two classes of terms related to each other?

© Springer International Publishing Switzerland 2015 231

W. Schneider, Memory Development from Early Childhood Through
Emerging Adulthood, DOI 10.1007/978-3-319-09611-7_8
232 8  Effects of the Knowledge Base on Memory Development

We posit that the common denominator of all these definitions is the belief that
knowledge should be conceptualized in terms of information contained in a modi-
fied network model of semantic memory (see Bjorklund 1987; Bjorklund et al.
1990). Each item in semantic memory is defined by a node that is connected to
other nodes. In addition to connections with other items (concepts), each node is
also associated with the features that characterize it. In development, the number of
nodes changes as do the number and strength of connections among items. Also, the
number of features associated with an item changes (increases) as a function of eve-
ryday experiences. Terms such as “world knowledge” or “conceptual knowledge”
concern the total amount of information stored in an individual’s semantic network.
Modified network models (e.g., Anderson 1976; Collins and Loftus 1975) have been
adopted by cognitive psychologists to describe and explain the functions of seman-
tic memory. An example of such a network is displayed in Fig. 8.1. The network
consists of concepts stored in long-term memory or nodes (the ellipses) that are
linked together. The connections between the nodes of the network are determined
by various relations, with interitem associativity identified as a fundamental one.
When we speak of a knowledge “domain,” we are referring to semantic knowl-
edge that is highly interrelated (e.g., knowledge of the game of tennis or knowl-
edge in the fields of physics or medicine). Research has shown that experts in a
given domain retrieve information from a semantic network that contains an enor-
mous number of nodes for that domain and also strong and frequent interconnec-
tions among these nodes. An important developmental hypothesis is that young
children’s representations of knowledge are different from the representations of
older children, adolescents, and adults. Consequently, younger children cannot use
their knowledge in the same way as older children (Chi and Ceci 1987).
The density of the associative network depends on children’s experiences
with the core concepts and thus on age. Let us assume that the conceptual knowl-
edge about dogs illustrated in Fig. 8.1 belongs to a 12-year-old. Corresponding
networks for a 5-year-old and a 17-year-old would include fewer nodes for the
5-year-olds but more for the 17-year-olds. Children’s knowledge base—defined as
their long-term representations of world knowledge and organized in a semantic
memory network—thus changes with age and experience. Developmental changes
in the network affect the number of available items and their accessibility. It is not
only the number of items (concepts) in the knowledge base that expands with age
and experience but also the number and strengths of associations and the number
and types of features that characterize concepts (Bjorklund 1987; Schneider and
Bjorklund 2003).
As will be shown below in more detail, the richness of the knowledge base and
thus the “density” of the semantic network significantly influence the speed and
accuracy of memory processes. Individual items related to a familiar domain can
be accessed more quickly from the long-term store as can relations among seman-
tically related items in the knowledge base. Faster processing means more efficient
processing, which results in the greater availability of mental resources. These
mental resources can then be applied to retrieve specific items, facilitate strategy
use, or induce metacognitive processes (Bjorklund et al. 1990). Thus, there is
Conceptualizations of Knowledge 233

Animal has
Skin 4,12;5,13,…86;99

can isa
looks like
isa Elephant
Move

has
Dog Trunk

has
looks like

can Tail
2,6;8,12,…,99;95

Bark can
isa

Spots Be walked
isa
has

Dalmation 1. Get leash

2. Go outside
3. Find spot
looks 4. Clean up mess
like Dachshund 5. Return home

isa
1,1;2,5,…,90;82
looks
like Otto

3,5;6,9,…,90;92A

Fig. 8.1  A portion of a person’s knowledge of animals. The ellipses correspond to nodes in the
network that are linked by qualitatively different types of associations: isa, has, can, looks like. The
node denoting a mental image contains pairs of numbers. These refer to x, y coordinates of points
that would be used to generate an image, much as particular elements on a television screen are
activated to generate a picture. (Source Kail and Pellegrino 1985). (Adapted from Kail 1990, p. 80)

reason to assume that the semantic network of experts in a given domain is par-
ticularly dense for that domain, making it very easy for experts to access domain-
relevant information. Please note, however, that the picture can be totally different
when these experts try to remember information from domains about which they
have little knowledge. In such cases, there are fewer nodes and interrelations
234 8  Effects of the Knowledge Base on Memory Development

among nodes, implying that the information processing speed and accuracy of
memory performance may be considerably lower. Accordingly, information about
the sheer amount of available “world knowledge” a person possesses may prove
to be less predictive of memory performance than information about the person’s
specific knowledge about a particular domain. For the purpose of the present chap-
ter, it seems important to keep in mind that the semantic network is comprised of a
single set of encoding and storage processes, indicating that world knowledge and
domain-specific knowledge are based on the same mechanisms.

Effects of the Knowledge Base on Memory

Age differences in the knowledge base can affect memory in three main ways
(Bjorklund 1987; Schneider 1993). First, the knowledge base can render specific
items more accessible because they are more richly represented (e.g., they have
more features in semantic memory) and are therefore more vivid. Second, the
knowledge base can influence memory performance by the relatively effortless
activation of relations among sets of items. Third, it can facilitate the use of delib-
erate memory strategies.

Item-Specific Effects

The first conclusion regarding item-specific effects was based on the assumption
that a particular item may be more apt to be activated in certain contexts than other
items because it has been activated frequently in the past in those contexts and
thus has a relatively low threshold for activation. Moreover, an item may be more
apt to be activated because it is associated with a large number of features, most of
which can be easily activated; this ensures that the item itself will be remembered
later (Bjorklund 1987). These item-specific effects are generally not influenced
by the use of deliberate strategies, even though item-specific and organizational
effects can covary.
This conclusion has been supported by experiments using lists of unrelated
items that were equally meaningful for children of different ages, eliminating age
differences for the information that was to be remembered. Such findings were
obtained when the influence of capacity, strategy use, and awareness on memory
was considerably reduced as is the case in incidental memory tasks. For example,
Ghatala (1984) compared 7- and 11-year-old children as well as adults on such an
incidental learning task. In her experiment, she presented words and asked ques-
tions that were designed to evoke the participants’ knowledge about the words. For
instance, when the word “chair” was presented, participants may have been asked
“Is this a piece of furniture?” Later, the participants were asked to recall as many
of the words as they could. The typical result was that older children and adults
Effects of the Knowledge Base on Memory 235

recalled more words than the younger children. Ghatala (1984) found that 11-year-
olds recalled about twice as many words as the 7-year-olds. However, this effect
was quite specific to words for which the participants’ knowledge was evoked.
Recall was comparable across the two age groups when participants judged words
purely in terms of their sounds. Obviously, the sound judgments did not evoke
much knowledge in the various age groups.
Item-specific effects are not limited to the retention of unrelated information but
may also be found for sets of categorically related items. For example, Ceci (1980)
showed that 4-year-olds failed to generate multiple semantic features for items,
resulting in reduced levels of recall compared with older children and to other
4-year-olds who were provided with multiple features for encoding the materials
that were to be remembered. Ceci suggested that young children’s deficient knowl-
edge base prevents them from easily generating multiple semantic features to be
used in encoding. This may explain their generally poor performance in list-learning
tasks. Similarly, Ackerman (1985, 1987), who used cued-recall tasks, was able to
show that 7-year-olds needed more of the original encoding environment to be rein-
stated for retrieval cues to be effective than older children and adults. This was par-
ticularly obvious when category labels served as retrieval cues. Ackerman suggested
that one reason for young children’s dependence on the encoding environment in
determining the effectiveness of a retrieval cue is their deficient semantic memory.
Although item-specific effects of the knowledge base typically produce age dif-
ferences in performance, the pattern of findings is completely different in situa-
tions in which the content of semantic memory is comparable across age groups
or when children’s semantic knowledge base is even richer than that of adults.
Evidence confirming this conclusion comes from a study by Lindberg (1980). In
a first experiment, kindergarten children, sixth graders, and college students were
compared on an incidental learning task for which they analyzed words accord-
ing to physical, acoustic, semantic, and visual featural systems, and large devel-
opmental differences in recall were found in the visual and semantic conditions.
However, only small differences were detected in the acoustic condition, and
no differences were found in the physical condition. These findings confirmed
Lindberg’s hypothesis that the physical and acoustic conditions would show lit-
tle or no developmental differences in recall because all participants, regardless of
age, would encode the items according to just a few features. On the other hand,
reliable developmental differences in recall were expected for the semantic and
imagery conditions because the knowledge-base system develops with experience
and encodes information according to its features such that the number of these
features is critical to retrieval on such tasks.
Results from a second experiment revealed that children’s recall exceeded
adults’ recall when the stimuli consisted of names of television characters and
teachers who were well-known to the children but not to the adults. Two word lists
using items from different semantic categories were constructed and presented to
the participants (third graders and college students) who were given free-recall
instructions. Whereas one list was generated by a separate group of third graders,
the other one was composed of items chosen from a popular category norm list.
236 8  Effects of the Knowledge Base on Memory Development

Fig. 8.2  a Amount recalled 16

as a function of age (third (a)
14
grade or college) and types of B&M
items (categories and items 12 WORDS
generated by third-grade
10

RECALL
raters, and corresponding
Battig and Montague 8
categories). b Clustering
6
measured in terms of the
repetition ratio (RR) as a 4
function of age (third grade
2
or college) and type of
items (categories and items 0
generated by third-grade 3 C
raters and corresponding GRADE
Battig and Montague
categories). (Adapted from 0.6
Lindberg 1980, Figs. 2 and 3) (b)
0.5
CLUSTERING "RR"

0.4 B&M
WORDS

0.3

0.2

0.1

0
3 C
GRADE

The categories selected from the latter corresponded roughly to those included
in the list generated by the third graders. As a main result of this experiment, an
Age  × List interaction was found, with the third graders clustering and recall-
ing more words from the third-grader-generated list and college students cluster-
ing and recalling more words from the other list (cf. Fig. 8.2). This experiment
thus demonstrated that when children were able to use a knowledge base that was
richer for them than for adults, they were superior to the adults in terms of both
recall and clustering. For similar findings, see Chechile and Richman (1982).

Effortless Activation of Semantic Memory Relations

As already illustrated in the preceding chapter, a well-developed knowledge base

facilitates recall by activating associations among sets of items in long-term memory
in a rather effortless and automatic way. As noted above, older children’s more elab-
orate semantic knowledge shows more and also stronger associations among items
Effects of the Knowledge Base on Memory 237

or concepts. Items are more easily accessed because they can be triggered by more
stimuli. As a consequence, highly organized retrieval is possible without the need for
effortful deliberate strategies. Effects of nonstrategic activation increase with age as
knowledge becomes better integrated in semantic memory. Work by Bjorklund and
Bjorklund (1985; see also Bjorklund and Zeman 1982) concerning the recall of chil-
dren’s classmates provides a nice example. Although children’s recall of classmates’
names was highly organized, most children were unaware that they had used any
specific strategy. When they were induced to use a specific retrieval strategy, they
followed this strategy almost perfectly. However, there was no increase in recall as a
function of strategy use. Consequently, the use of intentional memory strategies had
only a minimal effect on performance on this type of task.

The Knowledge Base and Strategy Use

According to the third conclusion stated above, well-developed content knowledge

can support the use of strategies and metacognitive processes, which in turn help
recall. Bjorklund and colleagues (Bjorklund 1987; Bjorklund and Harnishfeger
1990) argued that older children have a greater capacity for these activities
because familiar items with many associations are processed with little effort. As
already illustrated in the preceding chapter, content knowledge can also facilitate
strategy by providing the conceptual foundation for organizational strategies (e.g.,
Frankel and Rollins 1985; Rabinowitz 1984; Schneider 1986). That is, when the
degree of category typicality and interitem associativity was systematically var-
ied in developmental studies, younger children were found to display category
clustering only when the typicality and associativity of items were high. By con-
trast, older elementary schoolchildren showed high levels of organization regard-
less of the amount of interitem associativity and category typicality of items. Thus,
only the older children could be considered strategic. By contrast, a conceptual
knowledge effect was assumed for the younger children in that semantic rela-
tions between highly associated items were activated with relatively little mental
effort, resulting in retrieval that appeared organized and strategic even in younger
children.
As already illustrated in the preceding chapters on capacity and strategy devel-
opment, age and individual differences are important factors with regard to the
knowledge base. For instance, the effect of knowledge on memory span was dem-
onstrated earlier in studies of child chess experts for the positions of chess pieces
on a chessboard (Chi 1978; Schneider 1993; Gruber et al. 1993). Similarly, rela-
tions between knowledge, mental effort, strategy use, and memory performance
were repeatedly demonstrated in experiments that used the dual-task procedure in
paired-associate learning tasks (Kee 1994). Moreover, age and individual differ-
ences in knowledge have been shown to influence the effectiveness of children’s
strategy use (e.g., Alexander and Schwanenflugel 1994; Guttentag 1989, 1997;
Hasselhorn 1995; Ornstein and Naus 1985; Rabinowitz and Chi 1987).
238 8  Effects of the Knowledge Base on Memory Development

Evidence That World Knowledge Mediates Memory

and Learning

Effects of the knowledge base are not restricted to strategic memory and list-learn-
ing tasks but have been similarly observed for other types of memory tasks. One
interesting example concerns the impact of constructive memory, which refers to
the phenomenon that people may use their world knowledge to embellish, elabo-
rate, and otherwise go beyond the information given as they represent informa-
tion in memory. As noted by Kail (1990), one important function of knowledge
is to allow a person to “fill gaps” that may exist in the information that is to be
remembered.
The idea that prior knowledge can affect children’s memory was hypothesized
in studies of inferential memory by Paris and his colleagues (e.g., Paris 1978;
Paris and Lindauer 1976, 1982; see also Trabasso and Nicholas 1980). These stud-
ies have shown that the tendency to make inferences develops considerably dur-
ing the elementary school years. For instance, in a study reported by Paris and
Lindauer (1976), 7-, 9-, and 11-year-olds were presented isolated sentences in
which instruments were either implied or explicitly stated (e.g., “The workman
dug a hole” or “The workman dug a hole with a shovel”). Each sentence described
an action performed by an individual. Half of the sentences included the instru-
ments typically used to perform the activity; in the other half, the instruments
were deleted. At testing, the instruments were provided as cues for recall for both
instrument-explicit and instrument-implicit sentences. For example, “shovel”
was used as the test cue for the workman sentence. The most important finding
was that, for younger children, explicitly stated instruments were more effective
retrieval cues than implied instruments. For older children, however, instruments
were equally effective cues for both types of sentences. Apparently, only the older
children routinely inferred the presence of the instruments as the sentences were
presented and incorporated the instruments into their mnemonic representations of
the sentences. Paris and Lindauer concluded that the implied instruments, relative
to explicit instruments, failed to serve as retrieval cues for their younger partici-
pants because the 7-year-olds did not make instrumental inferences at encoding
(i.e., online). They also assumed that implied instruments were equally effective as
test cues when compared with explicit instruments for the older children because
the older children had inferred the instruments at encoding. That is, according to
Paris and Lindauer (1976), implied instruments were part of the mental represen-
tations of the instrument-implicit sentences for the older but not the younger par-
ticipants. Much the same pattern of developmental difference was reported in a
subsequent study by Paris et al. (1977).
As noted by Schneider and Pressley (1997), doubts about the conclusion that
older children make instrumental inferences online arose in the light of subsequent
analyses of adults’ retrieval of instrument-reference sentences when given instru-
ments as test cues as these analyses showed that possible inferences were not
always generated by adults (e.g., Corbett and Dosher 1978; McCoon and Ratcliff
Evidence That World Knowledge Mediates Memory and Learning 239

1992). Van Meter and Pressley (1994) re-examined whether the spontaneous encod-
ing of instruments would occur when 10- to 14-year-olds read instrument-implicit
sentences. On the basis of the pattern of their findings, Van Meter and Pressley con-
cluded that children’s spontaneous generation of instruments was anything but cer-
tain by the time the children entered the late elementary- and middle-school grades.
Overall, the studies on constructive memory described above indicate that meaning
making while processing words, sentences, or longer text units involves interactions
between externally presented information and prior knowledge. These studies also
demonstrate that, on the basis of prior knowledge, people often construct inferences
that go beyond the information they are presented. The point that van Meter and
Pressley (1994) wanted to make is that when people do this, it is situationally con-
strained and is not just a simple function of development. Although older children
tend to go beyond the information given in texts, they do not always use their rel-
evant prior knowledge when reading sentences or a text.
Other examples demonstrating how world knowledge can be used by children
when processing texts were provided in developmental studies by Brown et al.
(1977) and Landis (1982). In the study by Brown et al. (1977), prior knowledge
was developed in children. Some of the second-, fourth-, and sixth-grade partici-
pants received information about a topic a week before they were presented with
a text passage that could be interpreted in light of this information. A key fact dic-
tated the manner in which the text passage was remembered. The story was about a
young man of the fictitious Targa people. Some children had been told before that
the Targa were Eskimos; others had been led to believe that the Targa were desert
Indians. After listening to the story twice, the children were asked to recall it in
their own words. In order to assess the impact of the participants’ knowledge of the
Targa on their recall of the story, the frequency with which they “recalled” infor-
mation that was not presented in the text passage was recorded. For instance, the
passage included the sentences “The weather was bad.” Children who assumed that
the Targa were Eskimos were likely to recall “It was cold and icy…,” whereas par-
ticipants who understood the Targa to be desert Indians recalled instead that “It was
hot and dry.” The memory of even the youngest children in the sample was affected
by this prior knowledge-building manipulation. Intrusion of such knowledge of the
Targa during recall occurred at all ages but increased from about 50 % for second
graders to almost 80 % for seventh graders. Particularly relevant here, all children
made many constructive errors in recalling the text by making inferences that were
consistent with the prior knowledge they had acquired the week before.
The study by Landis (1982) illustrated much the same phenomenon. Text pas-
sages were read aloud to 7- and 10-year-old children. For half of the children, the
passage described a well-known historical figure (e.g., Abraham Lincoln). The
other children were read the same passage but the name of the well-known figure
was replaced with the name of a fictitious person (i.e., Robert Baker). The memory
task was performed in two sessions scheduled 1 week apart. In the first session, par-
ticipants listened to the two story passages. In the second session, participants were
given a recognition memory test. Individual sentences were presented including (a)
some sentences from the passage, (b) sentences that presented known facts about
240 8  Effects of the Knowledge Base on Memory Development

the historical figure but that had not been presented in the passage, and (c) some
sentences unrelated to the topic of the passage. The most interesting finding was
that the children who had heard the passage about the historical figure had great dif-
ficulty distinguishing between the sentences about the historical person presented in
the passage from the facts about the person that had not been presented in the pas-
sage. Landis (1982) was able to demonstrate that the confusion was not due to poor
recognition memory because the children were quite accurate at judging that sen-
tences in the final category had not been presented. By contrast, the children who
had heard the sentences as part of a paragraph about a fictitious person had almost
no difficulty identifying the sentences they had actually heard. Taken together, these
studies show that older children and adolescents have more pathways by which to
retrieve information. Although even younger children can use their knowledge to
elaborate on information that is to be remembered, older children and adolescents
are more likely to do so. This elaboration sometimes implies that the recall of the
past is somewhat distorted: Children tend to forget specific events and instead sim-
ply remember more typical acts related to these events (Kail 1990).

Evidence for the Limited Effects of Domain Knowledge

on Subsequent Learning

Although most of the studies summarized above provide clear evidence for
positive effects of the knowledge base on memory performance, there are also
exceptions to this rule. For instance, DeMarie-Dreblow (1991) found that the
development of domain knowledge did not mediate learning and performance in
a list-learning task. The criterion task in DeMarie-Dreblow’s study was to recall
of a list of birds. Before presentation of the list, children from Grades 2–5 in a
prior knowledge-building condition received a great deal of instruction intended
to increase their knowledge base concerning some of the birds that appeared in
the subsequently presented list. This was accomplished by using videotapes and
experimenter-provided explanations about attributes of birds (e.g., type of feet,
diet, nesting, migration patterns). In fact, this instruction improved children’s
knowledge about birds.
Contrary to expectations, this knowledge did not promote the subsequent learn-
ing of lists of birds. That is, recall was not increased by the knowledge manipula-
tion. Moreover, the use of strategies such as rehearsal, categorization of presentation
cards, and self-testing was not affected by the knowledge manipulation. In order
to explore whether the knowledge manipulation may have been too difficult for
young children, DeMarie-Dreblow (1991) replicated the most important features of
her study with adults and produced the same results. The knowledge manipulation
increased participants’ knowledge about birds but did not affect list learning.
Although the results of DeMarie-Dreblow’s study challenge the idea that rich
domain knowledge mediates memory either through automatic associations with
the knowledge base or by enabling the use of strategies, there are alternative
Evidence for the Limited Effects of Domain Knowledge on Subsequent Learning 241

interpretations of the data. For instance, in order for knowledge related to the
listed items to have an impact on list learning, this knowledge has to be acquired
to a high degree (e.g., by exposure to multiple learning trials) so that associations
with prior knowledge occur automatically rather than with conscious effort. This
may not have been the case in DeMarie-Dreblow’s study because the connections
between the newly established knowledge and the birds’ names were not so auto-
matic. Furthermore, it may be the case that the particular information taught to the
participants was not viewed as useful by the participants for classifying birds or
for restructuring rehearsal patterns (e.g., rehearsing birds clustered according to
characteristics such as diet or migration pattern). In short, although the problem
DeMarie-Dreblow chose to study is important, there are alternative interpretations
of her null results that mandate caution in interpreting the outcomes. Obviously,
within-person longitudinal studies examining the progression of knowledge acqui-
sition from initial teaching to the point at which this knowledge begins to medi-
ate memory performance would aid the understanding of how the knowledge base
develops to the extent that knowledge induces memory improvement.

Interrelations Between the Knowledge Base and Strategies

As already noted above, an important issue to consider here is that the knowledge
base and strategies are not independent but interact with one another. Not surpris-
ingly, retention is most accurate when participants have considerable task-relevant
knowledge and when they use task-relevant strategies in a perfect way (Weinert et al.
1988). The interactive nature of knowledge and strategies also implies that children
who use an appropriate strategy but have limited task-relevant knowledge will per-
form rather poorly. Several studies have made it clear that children’s success with
strategies depends on their task-relevant knowledge. For instance, Tarkin (cited in
Ornstein et al. 1988) found that 8-year-olds rehearsed and recalled nearly twice
as much when the words were highly meaningful than when the words were less
meaningful.
In order to further illuminate the interaction between knowledge and strategies,
it is necessary to use experimental designs that permit separation of the effects of
the knowledge base alone, the strategy in question alone, and the knowledge base
and the target strategy operating together. Pressley and Brewster (1990) gener-
ated such a study and provided evidence that teaching prior knowledge that ena-
bles the use of a strategy can certainly have a substantial effect on memory. They
asked Canadian middle-school students to learn some facts about Canadian prov-
inces (e.g., “Canada’s first museum was in Ontario”), with half of the children in
the study instructed to use a strategy involving imagery and the other half left to
their own devices to learn the facts. The imagery instructions were to imagine the
fact occurring in a setting unique to the province in question. Thus, for the previ-
ous example, imagining a museum in downtown Toronto would be appropriate.
As most of the students lacked prior knowledge of the geographical features that
242 8  Effects of the Knowledge Base on Memory Development

distinguished the Canadian provinces, it was necessary to build some appropriate

background knowledge before students were able to construct useful images. Only
half of the students were given the chance to develop this background knowledge,
resulting in a 2 (imagery instructions) by 2 (prior knowledge developed vs. not
developed) design. The students who were given prior knowledge were presented
pictures that represented some prominent setting in the various provinces. The stu-
dents studied these picture-province associations until they knew them all.
As a major finding, neither the strategy instructions nor the picture learning
manipulation alone increased learning much over what was obtained in the con-
trol condition. The best performance, which was clearly better than recall in the
control and comparison conditions, was in the condition in which participants had
been given the prior knowledge images before being asked to use the associative
imagery strategy. Possessing an image of Nova Scotia permitted the construction
of an image of some fact occurring in Nova Scotia. That is, prior knowledge ena-
bled the use of a strategy that could not be executed profitably in the absence of
prior knowledge. The results provided by Pressley and Brewster (1990) are impor-
tant because much of good thinking involves the coordinated use of strategies and
prior knowledge rather than the isolated effects produced by one or the other (e.g.,
Pressley et al. 1989; Borkowski et al. 1989). Strategies often encourage people to
use their prior knowledge more completely than they would otherwise.
There is evidence that people do not always apply their prior knowledge to new
information that they are intentionally trying to learn and remember. Thus, there
has been interest in the development of strategies that encourage the access and
use of prior knowledge during intentional learning. Elaborative interrogation
is one strategy that has received a great deal of attention in studies conducted by
Pressley (for reviews, see Pressley et al. 1992; King and Menke 1992; Schneider
and Pressley 1997). It is based on the assumption that not all people use their
knowledge base as much as they can. There are many demonstrations that even
adolescents and university students often fail to elaborate on associations that are
to be learned (e.g., Beuhring and Kee 1987; Pressley 1982), ignoring many pieces
of factual knowledge that could be elaborately embellished in ways that would
promote memory performance.
Pressley and colleagues suggested that, because students often fail to access rel-
evant knowledge when they attempt to learn facts, such embellishment often needs
to be prompted. One mechanism for prompting such elaboration was explored
in several studies and labeled elaborative interrogation because “why” questions
were posed to learners to stimulate elaborative activity. Although most studies
involved adult learners (see reviews by Pressley 1982; Pressley 1990; Wood and
Woloshyn 1990), some also focused on children of different ages. Several experi-
ments used a “man” sentence task first developed by Stein et al. (1982). Stein and
colleagues had found that children and adults experience difficulties with remem-
bering man–action associations such as “The hungry man got into the car” or “The
brave man ran into the house” but did much better when the action was specified
in detail such as “The hungry man got into the car to go to the restaurant” or
“The brave man ran into the house to save the baby from the fire.” Pressley et al.
Evidence for the Limited Effects of Domain Knowledge on Subsequent Learning 243

(1987) were able to replicate the positive effect of providing precise elaborations
during incidental learning for adults. They also identified a mechanism that facili-
tated incidental learning even more: In their elaborative interrogation condition,
participants had to answer “why” questions for each fact (“Why did that particu-
lar man do that?”). The results were impressive given that elaborative interroga-
tion produced more than twice as much incidental learning of man sentences in
university students than providing precise elaborations did. Similar findings were
obtained in intentional learning situations when participants were instructed that
learning would be followed by a memory test.
Wood et al. (1988) performed an experiment using the “man” sentences
with children in Grades 4 through 8 to test the assumption that having children
answer why-questions as they learned “man” sentences would improve their
learning relative to when children were given the base sentences without elabo-
ration instructions. Four experimental conditions were compared in Experiment
1. In the base sentence (control) condition, children were given 18 “man” sen-
tences with no elaboration instructions. Participants in the imagery and elabora-
tive interrogation conditions were also given the base sentences but were also
instructed to generate images or to answer why-questions, respectively. The
experiment also included a condition in which precise elaborations of the “man”
sentences were provided.
The results are presented in Fig. 8.3, separately for the younger (9–11.5
years) and older (11.5–14 years) age groups. They were very similar to those
obtained for adults in that the differences between the elaborative interrogation
versus the base sentence (control) conditions were large (more than a stand-
ard deviation at the younger age level; more than two standard deviations at
the older age level). The elaborative interrogation versus precise elaboration

100
90
80
Percent Remembered

70
60
50
40 Younger

30 Older

20
10
0
Base Precise Elaboration Imagery Elab Int
Provided
Axis Title

Fig. 8.3  Intentional learning of man sentences, as a function of learning procedure. (Source

Wood et al. 1988, Experiment 1). (Adapted from Pressley et al. 1990, p. 213)
244 8  Effects of the Knowledge Base on Memory Development

difference was also large for the older group. Although most information was
remembered in the elaborative interrogation condition, the difference between
the elaboration interrogation and imagery conditions was not significant. Similar
findings were obtained in Experiment 2 of the study that examined elaborative
interrogation of more valid facts about animals in Canada.
As shown above, elaborative interrogation effects are often quite large. Nevertheless,
some researchers doubted whether the strategy was effective because it oriented learn-
ers toward the use of prior knowledge. Theoretically, more general factors such as
attentional effort and time on task may also be responsible for elaborative interroga-
tion effects. Martin and Pressley (1991) produced data consistent with the interpreta-
tion that not just any type of “why” question can promote the learning of facts. In their
study, recall was increased only by “why” questions that led specifically to the recall of
long-term knowledge that was supportive of the province–fact associations under study
(e.g., “The first Canadian-based farm protest organization was formed in Manitoba”).
For instance, questions of the form “Why would this fact be true of this province?”
led to increased performance, whereas questions such as “Why would this fact be sur-
prising for this province?” did not result in learning gains. According to the authors,
this is mainly because the latter type of question is not supportive of the province–fact
association.
A subsequent study by Woloshyn et al. (1992) also provided clear evidence that
supported the conclusion that elaborative interrogation leads learners to access and
use their knowledge base to mediate learning. Their experiment included Canadian
and German university students who were asked to learn facts about Canadian
provinces and German states. They did so either using elaborative interrogation
or by carefully rereading the facts (reading-to-understand condition). After being
presented with the facts, elaborative interrogation and reading-to-understand par-
ticipants were required to match the names of German states and Canadian prov-
inces to facts associated with them. Performance in these two conditions was then
compared to that of a third condition, a no-exposure control, which reflected how
well participants could perform the task without an opportunity to learn the facts.
This last condition thus provided an appropriate baseline for evaluating whether
and to what extent learning occurred in the elaborative interrogation and reading-
to-understand conditions.
As can be seen in Table 8.1, there was a huge effect of prior knowledge such that
learning about factual associations of one’s homeland was much better than learning
about factual associations of a foreign land. There was also a clear effect of strategy.
Most critically, however, the best performance was obtained when elaborative interro-
gation was used to learn facts about one’s homeland. Results of a multiple regression
analysis revealed that prior knowledge status and instructional condition in combina-
tion accounted for 68 % of the variability in learning Canadian facts and 73 % of the
variability in learning German facts. Thus, optimum learning occurred when partici-
pants high in knowledge used the elaborative interrogation strategy.
Similar findings were also obtained for Canadian schoolchildren who were
provided with either facts that were consistent with their knowledge base or
Evidence for the Limited Effects of Domain Knowledge on Subsequent Learning 245

Table 8.1  Mean number of Canadian and West German facts matched correctly as a function of the
nationality of the participants and experimental condition. (Adapted from Wolosyn et al. 1994, Table 1)
Facts/subjects Elaborative interrogation Reading to understand No-exposure control
M SD Mean P M SD Mean P M SD Mean P
Canadian facts
Canadian 26.25 4.23 0.8 17.1 7.15 0.52 4.1 2.23 0.12
subjects (high prior
knowledge)
West German subjects 11.1 5.4 0.34 5.8 3.07 0.18 3 1.83 0.09
(low prior knowledge)
West German facts
West German subjects 23.4 4.54 0.71 17.2 6.71 0.52 8.2 5.05 0.25
(high prior knowledge)
Canadian subjects 7.05 3.79 0.21 4.6 2.19 0.14 3.5 2.07 0.11
(high prior knowledge)
Note M and SD are raw scores; P  = proportion; n  = 20 for the elaborative interrogation and
reading-to-understanding conditions at each nationality level; n = 10 for the no-exposure control
condition; maximum matching score = 33

facts that were inconsistent with their prior knowledge (Woloshyn et al. 1994).
Again, effects of the elaborative interrogation strategy were compared with
those of a reading-to-understand condition. In general, instructions to use elab-
orative interrogation produced better learning performances than did instruc-
tions to read for understanding regardless of whether the factual information
was consistent or inconsistent with participants’ prior knowledge. The elabo-
rative interrogation learning gains were durable in this study and were main-
tained for up to 6 months after the experimental session. These findings thus
show that elaborative interrogation can promote the short-term and long-term
learning of facts, even in situations in which the facts clash with prior knowl-
edge. The authors explained this finding by suggesting that students might
question the validity of their incorrect beliefs when activating prior knowledge
that is supportive of the new content, thus promoting the increased understand-
ing of new information.
Overall, there have been many demonstrations that elaborative inter-
rogation facilitates factual learning when children or adults have relevant
background knowledge, providing confirmation that (a) people do not acti-
vate prior knowledge as much as they might when processing to-be-learned
facts and (b) for “why” questions to work, relevant prior knowledge is criti-
cal in most cases (Schneider and Pressley 1997). As shown by Woloshyn et
al. (1994), the elaborative interrogation strategy also seems to facilitate the
acquisition of new content that is inconsistent with prior knowledge, probably
by increasing the awareness of discrepancies between prior knowledge and
new content.
246 8  Effects of the Knowledge Base on Memory Development

Evidence for Strong Effects of the Knowledge Base

on Subsequent Learning and Memory: The Case
of Expertise

Since the late 1970s, a large number of developmental studies have demonstrated
that the amount of knowledge in a particular domain such as chess, physics, or
sports determines how much new information from the same domain can be stored
and retrieved (see the reviews by Bjorklund and Schneider 1996; Chi and Ceci 1987;
Rabinowitz and Chi 1987; Schneider 2000b). Evidence for the powerful effects of
domain knowledge on memory performance comes from studies using the expert–
novice paradigm. These studies compared experts and novices in a given domain
(e.g., baseball, chess, or soccer) on a memory task related to that domain.
From a developmental perspective, the major advantage of the expert–novice
paradigm is that knowledge and chronological age are not necessarily confounded,
a problem inherent in most studies addressing knowledge-base effects. The develop-
mental findings for the domain of chess were already discussed in the context of mem-
ory capacity constraints (see Chap. 6). These studies demonstrated that rich domain
knowledge enabled a child expert to perform much like an adult expert and better
than an adult novice—thus showing an elimination of and sometimes a reversal of the
usual developmental trends (e.g., Chi 1978; Schneider et al. 1993; Gruber, et al. 1993).
The superior memory performance of child experts has been demonstrated not
only in the domain of chess but also with regard to dinosaurs and snakes (Chi and
Koeske, 1983; Gobbo and Chi 1986; Pearson et al. 1979), music (Oura and Hatano
1988), “Star Wars” (Means and Voss 1985), and sports such as baseball (Recht and
Leslie 1988; Walker 1987) and soccer (Schneider et al. 1989, 1990). For instance,
Schneider et al. (1989, 1990) demonstrated a reversal of the usual age effect for text
learning. In this study, about 500 third-, fifth-, and seventh-grade participants who
could be classified as soccer experts and novices were presented with a story about
a soccer game. The text was rather easy to understand for both experts and novices.
Another important characteristic of the text that the children in this study were asked
to read was that it contained some contradictions. In addition to soccer knowledge,
the children’s IQ, metamemory, and memory capacity were also assessed.
The expected differences between experts and novices were especially evident
in the recall and comprehension of the soccer-related passage. A reversal of devel-
opmental trends was demonstrated such that third-grade soccer experts recalled
significantly more text units than seventh-grade soccer novices. An analysis of
covariance (ANCOVA) using memory for text as the criterion variable and meas-
ures of metamemory, domain-specific knowledge (i.e., knowledge about soccer),
memory monitoring (importance ratings of text units), and memory capacity as
covariates provided impressive results. That is, age differences in recall were com-
pletely eliminated after these covariates had been taken into account. Apparently,
age differences in text recall could be explained by differences in information pro-
cessing speed, strategy use, domain-specific knowledge, and metamemory, with
individual differences in domain knowledge accounting for most of the variance
Evidence for Strong Effects of the Knowledge Base... 247

in the criterion variable. When the sample was retested on the same measures 1
year later, this pattern of findings was replicated, confirming the robustness of the
results (Schneider and Körkel 1989).
These findings can be generalized to other domains. For instance, Gobbo and
Chi (1986) inferred from their analysis of children’s production protocols that
dinosaur experts’ knowledge structures are more integrated, cohesive, differenti-
ated, and complete than those of dinosaur novices. Thus, their knowledge struc-
tures differed mainly in terms of the quantity of their knowledge. With increasing
expertise, information stored in the knowledge base becomes more accessible
because the number of concepts as well as the number of attributes related to each
concept increases and because there are more links between the various concepts
and attributes that children already have.
On the other hand, there is evidence that experts and novices differ not only
with regard to the quantity of their knowledge but also the quality of their knowl-
edge, that is, in the way their knowledge is represented in the mind. As already
shown above, qualitative differences between experts and novices were observed
in the chess studies. Schneider (1993) and Gruber et al. (1993) found that most
experts tended to pick up chess pieces in a similar order when reconstructing
meaningful chess positions and created similar patterns (“chunks”). There were no
similar findings for the novices who seemed to pick up the chess pieces randomly.
Moreover, Means and Voss (1985) found that their “Star War” experts differed
from novices in that the experts constructed a more complete hierarchical struc-
ture of “Star Wars” that contained high-level goals, subgoals, and basic actions.
Even more interestingly, there were age-related representational differences within
the sample of “Star War” experts. Whereas the older experts seemed to interpret
“Star Wars” in relation to an “international conflict” schema involving interrelated
political–moral–military components, the younger experts tended to interpret “Star
Wars” in reference to a military-oriented “good guy–bad guy” schema.
Based on a secondary analysis of the Schneider et al. (1989, 1990) soccer exper-
tise data, qualitative differences in the recall structures of child experts and novices
were also confirmed by Körkel and Schneider (1992). For example, experts and
novices differed considerably in the importance ratings they gave to individual parts
of the text. The youngest experts were clearly superior to the oldest novices in iden-
tifying the important aspects of the text. Comparisons within the sample of experts
revealed that the recall patterns of seventh-grade experts differed qualitatively
from those obtained for the two younger groups such that the mean proportion of
recall decreased almost linearly with decreasing importance level for the seventh-
grade experts. As can be seen in Fig. 8.4, the pattern observed for the two younger
expert groups did not show the same linear trend even though the most important
information was recalled best. Recall patterns for all expert groups clearly differed
from those of soccer novices who recalled as much important as unimportant text
information, regardless of age. Taken together, the findings from the developmen-
tal studies using the expert–novice paradigm corroborated those from studies on
expert–novice differences in adults in that performance differences could be attrib-
uted to both quantitative and qualitative differences in information processing.
248 8  Effects of the Knowledge Base on Memory Development

35 third graders
fifth graders
30 seventh graders

25
Recall

20

15

10
1 2 3 4
Levels of importance

Fig. 8.4  Mean proportion of text units recalled by soccer experts as a function of age and impor-
tance level (reconstructed from data reported by Körkel and Schneider 1992)

How is such a rich knowledge base acquired? The few available longitudinal
studies indicate that expertise is based on a long-lasting process of motivated learn-
ing. Building up a rich knowledge base requires not only cognitive abilities but
also high levels of interest and motivation. In several domains, it is the amount of
practice and not so much the level of general aptitude that determines exceptional
performance (see Ericsson 1996; Ericsson et al. 1993, for reviews). The available
evidence is in accordance with the view that expertise has to be conceived of as an
acquired competency, that is, as an outcome of extensive training and practice in
the domain of expertise. This competency was labeled skilled memory by Chase
and Ericsson (1982) who assumed that encoding and retrieval processes can be
accelerated considerably through extensive practice. Overall, recent research on
adult expertise performance has convincingly shown that exceptional performance
in many fields is primarily due to a vast body of acquired knowledge and experi-
ence as well as to acquired skills (e.g., Ericsson 1996; Ericsson et al. 1993).
This view is also supported by findings of the few available longitudinal stud-
ies on children, indicating that expertise is based on a very long-lasting process of
motivated learning. Building up a rich knowledge base requires not only cognitive
abilities but also high levels of interest and motivation. In several domains, it is the
amount of practice and not so much the level of general aptitude that determines
exceptional performance (Schneider 1997, 2000b). For instance, the importance
of practice was illustrated by Gruber et al.’s (1994) study, which was a follow-
up of Schneider (1993) and Gruber et al. (1993). In the follow-up study, children
were presented with the same chessboard reconstruction tasks they had worked
on 3 years earlier. The most interesting outcome was that those children who had
lost their expert status because of a lack of interest and motivation reconstructed
fewer chess positions from memory than those children who had shifted from nov-
ice to expert status. Whereas the latter group had improved considerably over the
Evidence for Strong Effects of the Knowledge Base... 249

years, only small gains in memory for chess positions were observed for the for-
mer group. Similarly, in a longitudinal study of young tennis talents, Schneider
et al. (1993) found that the amount and intensity of practice as well as the level
of achievement motivation significantly predicted tennis rankings obtained 5 years
later. Together, these findings indicate that cognitive ability seems less important
than noncognitive factors such as interest, endurance, dedication, and concentra-
tion when the major goal is to build a rich knowledge base (Renninger et al. 1992).

Intelligence, Knowledge, and Memory Performance

One of the most intriguing findings in the literature on expertise and its devel-
opment concerns the role of psychometrically measured intelligence. Several
researchers have documented that the knowledge base exerts a greater influence
on memory development than any other single measure. For instance, Alexander
and Schwanenflugel (1994) assessed the effects of the knowledge base, metacog-
nition, and IQ on young elementary schoolchildren’s performance on a sort-recall
task. They reported that the knowledge base was the primary driving force behind
strategic memory performance and that metacognition and IQ were of minor
importance. Similar findings were reported by researchers who employed the
expert–novice paradigm.
The idea that domain-specific expertise can compensate for low overall ability
on domain-related cognitive processing tasks was first confirmed in studies that
focused on adult populations. For example, Ceci and Liker (1986) demonstrated
that adults who appeared to have low levels of intellectual functioning (e.g., IQs in
the 1980s) were capable of complex classification and reasoning processes when the
stimuli were highly familiar. Walker (1987) compared high- and low-aptitude adults
who were either baseball experts or novices. When presented with a baseball text
passage, low-aptitude/high-knowledge participants recalled more information than
high-aptitude/low-knowledge participants. Recht and Leslie (1988) also reported
that junior-high students high in domain knowledge about baseball learned more
from reading a passage about baseball than students low in domain knowledge.
Several developmental studies confirmed Ericsson and colleagues’ proposition
that rich domain knowledge determines performance and can even compensate
for low overall aptitude on domain-related memory tasks. The aforementioned
longitudinal study by Schneider et al. (1989, 1990) provided detailed research on
the power of domain knowledge relative to general intelligence. They measured
both the soccer expertise and the general intelligence of their participants and the
relative impacts of these constructs on memory performance and text comprehen-
sion. The data were striking in that at every age level (ages 8 to 12) and on every
measure, soccer experts showed higher performance than soccer novices. Most
critically, general intelligence was not a strong determinant of performance. For
example, 8-year-olds high in soccer expertise detected about 10 times more con-
tradictions in the text than 8-year-olds low in soccer knowledge. By contrast, there
250 8  Effects of the Knowledge Base on Memory Development

was a minimal difference in the number of contradictions detected by high-ability

and low-ability domain experts. The corresponding difference for novices was also
minimal. The same patterns of results were found when the children were retested
1 year later (Schneider and Körkel 1989).
As impressive as these findings are, they are not universal. Studies reporting
strong expertise effects and null effects of IQ have typically used text processing
tasks for which few deliberate encoding strategies are used. It seems important to
note that Schneider, Bjorklund, and their colleagues never dismissed general apti-
tude as an important determinant of memory performance when exploring the gen-
eralizability of their findings on text processing across different domains. When
examining the effects of expertise on more deliberate strategic tasks, they found
that the effects of knowledge continued to be strong, but the influence of IQ was
not eliminated. This was most clearly illustrated in Schneider and Bjorklund’s
(1992) study, which used an expert–novice experimental design similar to that used
by Schneider et al. (1989, 1990) except that it employed a sort-recall rather than
a text processing task. Second- and fourth-grade German children classified as
experts or novices in soccer and as having high or low IQ were presented with soc-
cer and nonsoccer word lists, using the same scales and procedures used in the text
processing experiments by Schneider et al. (1989, 1990). Schneider and Bjorklund
found significant effects of expertise on recall of the soccer list, thus confirming the
results of the previous studies. No group differences between experts and novices
were observed for the nonsoccer lists. However, unlike the text processing experi-
ments, the effects of IQ were not eliminated. For both the soccer and nonsoccer
lists, and for both the second- and fourth-grade children, high-IQ children remem-
bered significantly more information than low-IQ children.
Schneider and Bjorklund (1992) interpreted these findings as indicating that
when a memory task involves deliberate strategies, as the sort-recall task does,
expertise reduces the influence of IQ but, unlike less obviously deliberate text
processing tasks, does not eliminate them. This interpretation received support
from a subsequent study using a within-subjects design (Schneider et al. 1996).
In two experiments, both text-recall and sort-recall measures related to the game
of soccer were presented to high- and low-aptitude fourth-grade children who
were either soccer experts or soccer novices. The most important findings are sum-
marized in Fig. 8.5. Both experiments confirmed previous research findings that
effects of aptitude on text recall and comprehension were eliminated when the
effect of expertise was considered simultaneously. Schneider et al. (1996) also rep-
licated the earlier finding that this pattern of results does not generalize to strategic
memory tasks such as semantic categorization. Here, main effects of expertise and
aptitude were found. The authors concluded that expertise reduces but does not
eliminate the relation between IQ and memory tasks that involve deliberate strat-
egies (see also Alexander and Schwanenflugel 1994). Rich knowledge seems to
fully compensate for the effect of low IQ on some tasks (e.g., text recall) but not
on others (e.g., sort recall).
One of the problems found in studies on (child) experts is that usually the studies
do not involve “true” (in the sense of exceptional) experts but rather knowledgeable
Intelligence, Knowledge, and Memory Performance 251

0.7

0.6

0.5
Proportion recalled

0.4

0.3

0.2

0.1

0
Sort recall Text recall

High-IQ experts Low-IQ experts High-IQ novices Low-IQ novices

Fig. 8.5  Mean performance on sort-recall and text-recall tasks, as a function of expertise and IQ

(slightly modified after Schneider et al. 1996)

individuals who instead may be classified as “intermediate, semi-experts” or “sub-

experts” according to various attempts at classification (e.g., Patel and Groen 1991).
This problem becomes obvious in studies that have used the median of the distri-
bution of some competence as the critical boundary for expert–novice distinctions.
Despite this validity problem, it appears that findings from developmental studies on
“semi-experts” can be generalized to “true” experts. In particular, the two longitu-
dinal studies on “true” child experts mentioned above (Horgan and Morgan 1990;
Schneider et al. 1993) provided additional evidence for the importance of practice
and motivation. Both samples were exceptional in that they included elite subgroups.
Most of the child elite chess players in Horgan and Morgan’s study had skill ratings
of 1,300 or above (the mean for all U.S. tournament players of all ages is 1,500).
Most of the 106 tennis talents in the Schneider et al. (1993) study had very success-
ful careers, with about 10 % of the sample in the top 100 in the world (a few play-
ers even belonged in the top 10 in the world). Results of both longitudinal studies
revealed that improvement in chess or tennis skill was significantly related to experi-
ence. In sum, the more improved players played more and won more. Thus, both
studies confirmed the theoretical framework developed by Ericsson and colleagues,
thus highlighting the importance of deliberate practice and motivation in developing
domain-specific expertise in children.
However, a closer analysis of knowledge versus aptitude effects reported in
these studies indicates that the position taken by Ericsson and colleagues may be
too extreme. That is, whereas Ericsson and colleagues believe that the amount of
deliberate practice is a sufficient predictor of subsequent expert performance, the
developmental findings suggest that individual differences cannot be completely
ignored when it comes to predicting the development of expertise. In the case of the
252 8  Effects of the Knowledge Base on Memory Development

longitudinal chess study, individual differences in nonverbal intelligence explained a

small but significant amount of variance in the dependent variable, that is, improve-
ment in chess skill within a year. Similarly, the tennis talent study showed that the
relative impact of basic motor skills on performance was indirect but reliable. As
illustrated by a causal modeling analysis predicting tennis ratings several years later,
the model fit the data only after individual differences in basic motor abilities had
been taken into account. Thus, the results of these studies seem to indicate that expe-
rience, although extremely important, cannot completely substitute for aptitude or
talent. Support for this position was also provided in a recent review article on delib-
erate practice. Hambrick et al. (2013) presented empirical evidence showing that
deliberate practice is not sufficient for explaining individual differences in chess and
music, the two most widely studied domains in expertise research. One important
conclusion drawn by these researchers is that the task is now to rise above discipli-
nary and ideological differences and develop and rigorously test theories that take
into account as many potentially relevant explanatory factors as possible.
One potentially relevant approach regarding the development of expertise is to
test what Schneider (1993, 1997) labeled the “threshold” or “partial compensa-
tion” models, which, from a theoretical point of view, seem more appropriate than
the deliberate practice approach for describing the interplay of expert knowledge
and basic abilities in predicting exceptional performance. The “threshold” model
can be described as follows: If the ability parameter of a person is close to or
beyond a critical or “threshold” value of ability (typically assumed to be slightly
above the population mean), individual differences in predominantly noncognitive
variables such as commitment, endurance, concentration, or motivation determine
peak performance. In comparison, the “partial compensation” model assumes
that expertise reduces the contribution of general aptitude more and more as the
amount of domain-specific knowledge increases. Although the available evidence
based on developmental studies supports such a view, more research is clearly
needed to further explore its validity.

Conclusions

Overall, research on the effects of domain knowledge conducted during the last 30
years has convincingly shown that any explanation of memory development must
reserve a large place for increasing knowledge of specific contents (Siegler 1998).
Taken as a whole, the findings presented in the chapter provide impressive evidence
that a child’s knowledge base can have a great effect on memory and performance
even though prior knowledge is not always used automatically. What is also apparent
is that declarative knowledge is very complex and can be thought of as comprised of
a number of levels ranging from neural connections to complicated schemata. These
levels are all connected to one another as evidenced by the flexibility of spreading
activation. There are still many challenges that remain in elucidating the effects of
the knowledge base. For instance, as the knowledge base develops, there are more
Conclusions 253

concepts and stronger interconnections among concepts, and thus, long-term mem-
ory should become more accessible with increasing age. With increased accessibil-
ity, both automatic and strategic processes should be facilitated. Nonetheless, at all
age levels, there are important differences between people in the degree of accessi-
bility to what is stored in long-term memory (Brown et al. 1983).
The finding that there are individual differences in accessibility is one more
indication that knowledge alone can never provide a sufficient explanation for
memory performance. The conscious and systematic search of long-term memory
for possible knowledge that could mediate memory is an important learning strat-
egy (e.g., elaborative interrogation). The finding that the ability to access knowl-
edge often depends on strategy highlights the idea that it is wrong to conceptualize
thinking as either a kind of knowledge or a kind of strategy. Thinking often
involves the articulation of knowledge and strategies, and both components are
assumed to interact in several everyday situations even though many of the effects
reviewed in this chapter were probably mediated directly by the knowledge base,
with the conscious use of memory strategies playing a minor role.
Domain knowledge is a powerful determinant of memory and learning. It
increases steadily from infancy to adulthood and contributes to the development
of other sources of memory competencies, such as basic capacities, strategies, and
metacognitive knowledge. As shown in the last section of this chapter, studies using
the expert–novice paradigm have highlighted the importance of domain knowl-
edge for exceptional performance in different fields such as memory or sports. The
developmental evidence clearly supports Ericsson and colleagues’ (e.g., Ericsson
1996; Ericsson et al. 1993) suggestion that individual differences in the amount of
deliberate practice and motivation are key variables for predicting individual differ-
ences in the level of expertise in a given domain. However, the available evidence
has failed to rule out the possibility that innate talent or aptitude is necessary for
high achievement (Schneider 1997; Sternberg 1996; Winner 1996).
Research on the development of domain knowledge has illustrated that the
sources of memory development interact in numerous ways, and this fi ­ nding
sometimes makes it very difficult to disentangle the effects of specific sources
from those of other influences. The importance of these interactions was
­highlighted in the model of good-information processing (Pressley et al. 1989;
Borkowski et al. 1989; Pressley and Hilden 2006), which emphasizes the interplay
of intact neurological (basic capacities), strategic, knowledge-base, and motiva-
tional components in determining cognitive performance.
Chapter 9
The Development of Metamemory

Metamemory is knowledge about memory. It resembles constructs advanced by

earlier theoretical traditions (Cavanaugh and Perlmutter 1982) such as systematic
introspection as defined by the “Würzburg School of Thought” (Ach 1905). Selz
(1913) made assumptions about evaluation and selection as self-regulation pro-
cesses that resemble other processes that are central to contemporary metacogni-
tive theory. Despite reservations about introspective data throughout the twentieth
century (e.g., Lyons 1986), the potential importance of modern research on meta-
cognition was enthusiastically anticipated by several experimental psychologists
including Tulving and Madigan (1970):
Why not start looking for ways of experimentally studying and incorporating into theories
and models of memory one of the true characteristics of human memory: Its knowledge
of its own knowledge… We cannot help but feel that if there is ever going to be a genuine
breakthrough in the psychological study of memory… it will, among other things, relate the
knowledge stored in the individual’s memory to his knowledge of that knowledge. (p. 477)

Although Tulving and Madigan did not use the label “metamemory,” their state-
ment stimulated research on different introspection issues that particularly affected
the study of memory in the field of developmental psychology. Shortly after the
publication of the Tulving and Madigan paper, Flavell (1971) introduced the
term metamemory to refer to knowledge about memory processes. He gave it
a special status in the taxonomy of memory phenomena that he developed with
Henry Wellman (Flavell and Wellman 1977). From a developmental perspective,
this concept seemed well suited to explain young children’s production deficien-
cies across a broad variety of tasks. Whereas young children usually do not know
much about the advantages of memory strategies, schoolchildren are regularly
confronted with various memory tasks that eventually help them to discover the
advantages of strategies by which to improve their strategy knowledge (meta-
memory), which in turn should positively affect their memory behavior (i.e., strat-
egy use) and performance on future memory tasks. Thus, the expectation was that

© Springer International Publishing Switzerland 2015 255

W. Schneider, Memory Development from Early Childhood Through
Emerging Adulthood, DOI 10.1007/978-3-319-09611-7_9
256 9  The Development of Metamemory

clear developmental trends in the relation between metamemory and memory

would be found across childhood and adolescence. As a consequence, there was a
great expansion of metamemory theory in the ensuing decade and the initiation of
a research program that has been influential for the last three decades.

Conceptualizations of Metacognitive Knowledge

Research on metacognitive development was initiated in the early 1970s by Ann

Brown, John Flavell, and their colleagues (for reviews, see Brown et al. 1983;
Flavell et al. 2002). At the very beginning, this research focused on knowledge
about memory, which was coined metamemory by Flavell (1971) as mentioned
above. Although most developmental studies classified as metacognitive have
explored children’s metamemory (i.e., their knowledge about memory), the term
has also been applied to studies investigating children’s comprehension, commu-
nication, and problem-solving skills (Flavell 2000; Schneider and Pressley 1997).
John Flavell and his colleagues were the first to carry out systematic research on
memory knowledge in children and also the first to develop a conceptualization of
metamemory. In their taxonomy of metamemory, Flavell and Wellman (1977) dis-
tinguished between two main categories: sensitivity and variables. The sensitivity
category, also called procedural metamemory, refers primarily to knowledge of
when implicit unconscious behavioral memory activity is necessary. For instance, it
refers to the awareness that a particular task in a particular setting requires the use
of memory strategies. By contrast, the variables category corresponds to declara-
tive metamemory and refers to explicit, conscious, factual knowledge that perfor-
mance in a memory or problem-solving task is influenced by a number of different
factors or variables. It was divided into three subcategories: (a) person variables, (b)
task variables, and (c) strategy variables, that is, knowledge about potentially appli-
cable memory strategies. In brief, the person subcategory concerns knowledge about
the limitations and capacity of one’s own memory. Examples of the person varia-
ble include the child’s mnemonic self-concept and knowledge about intraindividual
differences in memory performance (e.g., “I am better at recalling numbers than at
recalling poems”). Task variables refer to the child’s awareness that task demands or
properties of input organization can influence memory performance, thus making a
memory task easier (e.g., by using highly familiar items) or harder (e.g., by present-
ing long item lists and/or by limiting study time). The strategy variables include all
changes in knowledge about the advantages and problems of certain memory strate-
gies. Although Flavell and Wellman (1977) distinguished between these three cat-
egories, they argued that most metacognitive knowledge is actually a combination
of two or three categories. They also argued that metamemory might be inaccurate,
might not be activated when needed, and might not be beneficial when activated.
The scope of the concept was later broadened. In a seminal paper, Flavell (1979)
emphasized the fact that metamemory was not isolated from other aspects of the
mind, and he generalized the metamemory taxonomy to metacognition in general.
Conceptualizations of Metacognitive Knowledge 257

Although various definitions of this term have been used in the literature on cognitive
development, the concept has usually been broadly and rather loosely defined as any
knowledge and/or cognitive activity that takes as its object, or regulates, any aspect
of any cognitive enterprise (cf. Flavell et al. 2002). This conceptualization refers to
people’s knowledge of their own information processing skills as well as to their
knowledge of the nature of cognitive tasks and the strategies they use to cope with
such tasks. Moreover, it also includes executive skills related to the monitoring and
self-regulation of one’s own cognitive activities. Flavell (1979) described three major
facets of metacognition, namely, metacognitive knowledge, metacognitive experi-
ences, and metacognitive skills, that is, strategies that control cognition. According
to Flavell et al. (2002), declarative metacognitive knowledge refers to the segment of
world knowledge that has to do with the human mind and its doings. Metacognitive
experiences refer to a person’s awareness and the feelings elicited in a problem-
solving situation (e.g., feelings of knowing). Metacognitive skills are also believed
to play a role in many types of cognitive activity such as the oral communication of
information, reading comprehension, attention, and memory.
The taxonomy of metamemory was not intended to be exhaustive. Indeed,
Flavell and Wellman (1977) did not attempt to define the concept precisely. A
number of other theorists have since contributed to the development of metacogni-
tive theory (for useful reviews and critiques, see Borkowski et al. 1988; Joyner and
Kurtz-Costes 1997; Schneider 2010; Schneider and Pressley 1997). For instance,
Paris and colleagues (e.g., Paris and Lindauer 1982; Paris and Oka 1986) intro-
duced a component called “conditional metacognitive knowledge,” which referred
to children’s ability to justify or explain their decisions concerning memory
actions (see also Brown 1987; Schraw and Moshman 1995). Whereas declara-
tive knowledge as defined by Flavell and Wellman focuses on “knowing that,” the
component added by Paris and colleagues addresses “knowing why.”
An impression that was gleaned from some of Flavell’s early research was that
a lot of metacognitive development was complete by age 8 or 9 (e.g., Kreutzer
et al. 1975) and that this was true for both declarative and procedural metacogni-
tive knowledge. One reason Ann Brown was motivated to reconceptualize meta-
memory was to counteract this impression. Ann Brown and her colleagues (Brown
1978, 1987; Brown et al. 1983) elaborated on Flavell and Wellman’s work. The
frame of reference used by Brown and colleagues was the competent informa-
tion processor, one possessing an efficient “executive” that regulated cognitive
behaviors. In their view, this regulatory component is responsible for selecting and
implementing strategies, monitoring their usefulness, and modifying them when
necessary. It was assumed that children do not monitor and regulate their perfor-
mance as well as metacognitively mature adults do. Overall, Brown et al. (1983)
adopted the perspective that memory monitoring and regulation processes play
a large role in complex cognitive tasks such as comprehending and memorizing
text materials. They also argued that the two aspects of metamemory (i.e., the
declarative and procedural components) complicate its definition (see also Joyner
and Kurtz-Costes 1997). That is, they are not only closely related but also fun-
damentally different in nature. Whereas the declarative knowledge component is
258 9  The Development of Metamemory

MONITORING

JUDGMENTS OF FEELING-OF-
LEARNING KNOWING
JUDGMENTS

EASE-OF- CONFIDENCE IN
LEARNING RETRIEVED
JUDGMENTS ANSWERS

ACQUISITION RETENTION RETRIEVAL

IN ADVANCE ON-GOING MAINTENANCE SELF-DIRECTED OUTPUT OF
OF LEARNING LEARNING OF LEARNING SEARCH RESPONSE

ALLOCATION OF
STUDY TIME SELECTION OF
SEARCH
STRATEGY
TERMINATION
SELECTION OF OF STUDY
TERMINATION
KIND OF
CONTROL OF SEARCH
PROCESSING

Fig. 9.1  Main components of metacognitive monitoring and metacognitive control. (Source

Nelson 1996). (Adapted from Schneider and Pressley 1997, p. 197)

primarily verbalizable, stable, and late-developing, the procedural knowledge

component is not necessarily verbalizable, rather unstable, relatively age-inde-
pendent, and depends on the specific task or situation.
Although the metamemory concept has been criticized for being fuzzy and
vague (e.g., Cavanaugh and Perlmutter 1982), the basic distinction between
declarative and procedural metacognitive knowledge already apparent in Flavell
and Wellman’s (1977) taxonomy appears to be widely accepted in the develop-
mental literature (cf. Alexander et al. 1995; Kuhn 2000a, b; Schneider 2008;
Schneider and Bjorklund 1998). Subsequent conceptualizations of metacognition
expanded the scope of this theoretical construct. For example, Nelson and Narens
(1990, 1994) developed a general theoretical procedural metamemory model (see
Fig. 9.1) that was supposed to illustrate the interplay between monitoring and con-
trol processes. Various monitoring and control activities are depicted as a function
of the stages of information acquisition, retention, and retrieval. This model has
been used frequently by both cognitive and developmental researchers.
Another important extension of the conceptual framework was provided by
Wellman (1990) who linked the declarative metacognitive component to the
broader concept of children’s theory of mind. This framework, which will be
described in more detail below, focuses on classes of knowledge about the inner
Conceptualizations of Metacognitive Knowledge 259

mental world and cognitive processes that develop during the preschool years.
Finally, as already described above, Pressley, Borkowski, Schneider, and their
colleagues (e.g., Pressley 1985; Pressley et al. 1987, 1989) proposed an elabo-
rate model of metacognition, the good-information processing model, which
not only includes aspects of procedural and declarative metacognitive knowl-
edge, but also links these concepts to other features of successful information
processing. According to this model, sophisticated metamemory is closely
related to the learner’s strategy use, motivational orientation, general knowl-
edge about the world, and automated use of efficient learning procedures. All
of these components are assumed to interact. For instance, specific strategy
knowledge influences the adequate application of memory strategies, which
in turn affect knowledge. As the strategies are carried out, they are monitored
and evaluated, which leads to the expansion and refinement of specific strategy
knowledge.
More recent models of metacognition and self-regulated learning have
emphasized multiple and bidirectional relations between cognitive and affec-
tive-motivational factors located on different levels of learning behavior. For
instance, Borkowski et al. (2000) extended the good-information processing
model by including learning experiences and related feedback loops through
which people gradually build their metacognitive functions. It was assumed
that such higher order learning-related skills, motivation, and knowledge can
be used to further improve the quality of learning and memorization processes
in subsequent learning situations. More recently, Efklides (2011) emphasized
the importance of specific learning-related “metacognitive experiences,” which
should play a crucial role in the development of metacognitive skills and were
also considered to influence the efficiency of future learning processes. Efklides
thus assumed that learning outcomes are influenced not only by conscious and
explicit information processing skills but also by the implicit metacognitive
experiences that arise from a person’s affective and motivational reactions to a
given task.
A similar view of the development of metacognition was provided by Kuhn
(1999, 2000a) who assumed that the acquisition of both declarative and proce-
dural metacognitive skills is accompanied by predominantly unconscious asso-
ciative processes during learning. Whereas the emergence of metacognitive skills
is implicit in the beginning, it gradually becomes explicit and can eventually be
accessed consciously. One interesting aspect of the developmental model proposed
by Kuhn is that it identified the meta-level as having the privilege of being the
locus of developmental change. According to this model, the construct of meta-
cognition can be conceptualized in a developmental framework, with young chil-
dren’s dawning awareness representing one end of a developmental progression
that ends in complex metacognitive capabilities that many adults do not master.
Although all theoretical approaches agree that a variety of metacognitive skills
should be available in older children, adolescents, and adults, this does not ensure
that experienced learners always apply their metacognitive knowledge in the right
way at the right time (Veenman 2011).
260 9  The Development of Metamemory

Children’s “Theory of Mind” and Metacognitive Knowledge

From the early 1980s on, a series of studies have focused on young children’s
knowledge about the mental world better known as “theory-of-mind” research. This
wave is still very much in motion and may have produced more than 1,000 publica-
tions within the last 30 years or so (Sodian 2005; Wellman et al. 2001). It deals with
very young children’s understanding of mental life and age-related changes in this
understanding, for instance, their knowledge that mental representations of events
need not correspond to reality. In a now classic study, Wimmer and Perner (1983)
tested young children’s understanding of false belief and confirmed that children
younger than the age of about 4 years, find it impossible to believe that another per-
son could hold an assertion that the child knows to be false. A little later, beginning
at about 4 years of age, children come to recognize assertions as the expression of
someone’s belief, which is not necessarily true. Subsequent theory-of-mind research
has addressed young children’s understanding of mental states such as desires,
intentions, emotions, attention, consciousness, and so forth.
How does the “theory-of-mind” approach differ from research on metamemory
and metacognition? Although researchers in both traditions share the same gen-
eral objective, that is, to explore children’s knowledge about and understanding
of mental phenomena, the research literatures have been distinct and unconnected
because they have focused on different developments. For instance, whereas the-
ory-of-mind researchers have investigated children’s initial knowledge about the
existence of various mental states such as desires and intentions, metacognitive
researchers have focused more on task-related mental processes such as strategies
for improving performance on various tasks or attempts to monitor improvements.
Flavell (2000) conceived of this approach as “problem-centered” and suggested
that it be labeled “applied theory of mind.” A second distinction between the two
research paradigms concerns the age groups under study. Because theory-of-mind
researchers are mainly interested in the origins of knowledge about mental states,
they predominantly study infants and young children. On the other hand, meta-
cognitive researchers investigate knowledge components and skills that already
require some understanding of mental states; thus, these researchers focus primar-
ily on older children and adolescents. A further distinction concerns the fact that
developmental research on metacognition deals with what a child knows about his
or her own mind rather than somebody else’s. As noted by Flavell (2000), how and
how often other people use their minds in similar situations is not of primary inter-
est. By contrast, it is the participant’s understanding of some other person’s mind
that is usually of central concern in theory-of-mind studies.
Clarification of the terminology issue seems important. Figure 9.2 contains an
overview of the various theoretical perspectives on metacognitive knowledge popu-
lar in the field of developmental psychology and links the various taxonomies and
terminologies that have been used in different lines of research. It should be noted
that the conceptualizations of metacognition outlined above and originally elabo-
rated on by developmental psychologists actually differ in several respects from the
Conceptualizations of Metacognitive Knowledge 261

Metacognition (Knowledge about cognition)

‘‘ Meta-Knowing’’ (Kuhn, 1999, 2000)

Knowledge about the mental world Knowledge about memory (Metamemory, Flavell, 1971)

“ Metacognitive Knowing” (Kuhn, 1999, 2000) “ Metastrategic Knowing” (Kuhn, 1999, 2000)

“ Theory-of-mind’’-research Declarative metamemory Procedural metamemory

“ variables’’ category “ sensitivity” category

Understanding of false (Flavell & Wellman, 1977) (Flavell & Wellman, 1977)
belief

Knowledge about
Understanding of mental
person, task, and Monitoring component Control and self
-
states such as desires,
strategy variables regulation component
emotions, attention,
(Nelson & Narens,
consciousness etc.
Understanding of 1990, 1994) (Nelson & Narens, 1990,
mental verbs 1994)
Understanding of mental ease-of-learning (EOL)
verbs judgments Knowledge of
recall readiness
judgments of learning
(JOL) Allocation of study
time
feeling-of-knowing
(FOK)

Fig. 9.2  Taxonomy of metacognition components (Schneider 2008, p. 116)

models developed in other fields of psychology such as social psychology, general

cognitive psychology, and the psychology of aging. In general, the models from
the latter group are narrower in scope. For instance, popular conceptualizations
of metacognition in the field of cognitive psychology exclusively elaborate on the
procedural component, focusing on the interplay between monitoring and self-con-
trol (see Koriat 2007; Nelson 1996). On the other hand, when issues of declarative
metacognitive knowledge are analyzed in the fields of social psychology and ger-
ontology, the focus is on a person’s beliefs about cognitive phenomena rather than
on veridical knowledge (e.g., Herrmann 1982; Dixon and Hertzog 1988). More
recent conceptualizations of metacognition that have been developed in the field of
educational psychology have subsumed the procedural and declarative components
of metacognition under the super-construct of self-regulation (e.g., Efklides 2001,
2011; Schunk and Zimmerman 1998).
Although more recent developments also include cognitive neuroscience
models of metacognition and emphasize the impact of brain maturation (cf.
Shimamura 2000), a controversial point of discussion has been the view that as
children develop, neural maturation permits more efficient monitoring, which then
262 9  The Development of Metamemory

leads to improved memory. For instance, Ceci et al. (2010) recently presented a
comprehensive developmental-representational theory that takes issue with the
claim that neurodevelopmental constraints play the primary role in accounting for
age differences in metacognitive monitoring. This theory posits that although the
neural systems involved in memory undergo substantial development throughout
childhood and adolescence, the levels needed for basic monitoring, binding, com-
paring, and organizing representations already exist by early childhood.
Taken together, the available literature suggests that the issue of metacognition
has attracted a lot of attention in various social science fields and continues to do
so. Obviously, the popularity of the metacognition construct is mainly due to the
fact that it seems crucial for concepts of everyday reasoning and those assessing
scientific thinking as well as social interactions.

Assessment of Metamemory

Measures of Declarative Metamemory

There are a variety of measures that have been used to capture what children know
about memory. As noted by Cavanaugh and Perlmutter (1982), measures assess-
ing declarative metamemory are taken without a concurrent assessment of mem-
ory (independent measures), whereas measures of procedural metamemory are
collected simultaneously with the measurement of memory activity (concurrent
measures). Most measurements of declarative metamemory in children have used
interviews or questionnaires that feature questions about memory processes. One
of the earliest and best-known interview studies on declarative metamemory was
carried out by Kreutzer et al. (1975) who assessed children’s knowledge about
person, task, and strategy variables relevant to memory performance in different
settings. The interview technique was aimed at measuring children’s knowledge
about everyday memory phenomena relevant to their own lives, including planning
for future events (e.g., remembering to bring ice skates to school the next day) and
recalling past events (e.g., remembering how long one has owned a dog). Overall,
the findings indicated that even kindergarteners and first graders possessed a fairly
sophisticated understanding of memory concepts, even though third and fifth grad-
ers had better metamemory than younger children.
Although no information on reliability was provided for this metamemory bat-
tery of 14 items, subsequent replications and extensions carried out in the 1980s
showed that reliability was not a major problem (see Belmont and Borkowski
1988; Cavanaugh and Borkowski 1980; Kurtz et al. 1982; Schneider et al. 1986).
Subsequent interview and questionnaire construction procedures seemed even
better from a psychometric perspective, however. For instance, Schlagmüller
et al. (2001) began with an extensive item pool and then conducted pilot test-
ing, dropping items that were not sufficiently reliable or did not seem to be oth-
erwise valid. The resulting questionnaire tapped metamemory related to everyday
memory situations, semantic categorization tasks, and memory for text. Overall,
Assessment of Metamemory 263

the internal consistency was sufficient (alpha = 0.77 for the total scale), and the
test–retest correlation after 4 months was 0.71. For similar construction principles,
see Hasselhorn (1994). One of the advantages of the questionnaires developed
by Belmont and Borkowski (1988) and Schlagmüller et al. (2001) was that they
could be used in group settings and also administered to relatively young children
between 6 and 12 years of age.

Measurement Problems

It has been pointed out in previous reviews of the metamemory literature that assess-
ing declarative metacognitive knowledge in children is not always easy and that there
are measurement problems (Joyner and Kurtz-Costes 1997; Schneider and Pressley
1997). Problems with the assessment of declarative metamemory via interviews
and questionnaires could be explained by the reliance on verbal self-report, which
may be particularly difficult for young children. To avoid such problems, alterna-
tive nonverbal assessment procedures such as videotaped illustrations of memory
strategies have been used (for details, see the overviews in Joyner and Kurtz-Costes
1997; Schneider and Pressley 1997). For instance, Wellman (1977, 1978) and
Yussen and Bird (1979) developed a procedure in which children were presented
pictures of actors engaged in various memory tasks. The children were asked about
the effects of age, time, hair color, and clothing on memory. Importantly, the chil-
dren were asked to select pictures to indicate their responses rather than to give a
verbal response. In the procedure developed by Justice (1985, 1986), children were
presented with various memory strategies (e.g., looking, naming, rehearsing, and
grouping) shown by another child on a videotape. After watching the videotapes and
naming the strategies, children made pairwise comparisons of the strategies, rank-
ing the strategies from best to worst. Justice and other investigators who have used
this approach (Schneider 1986; Schneider et al. 1987) have produced consistent data.
Kindergarten and young elementary schoolchildren recognize differential effec-
tiveness only when strategies produce dramatic differences in performance. Older
elementary schoolchildren, however, can detect more subtle differences in strategy
effectiveness. The conclusion that older children have better knowledge of the advan-
tages or disadvantages of a particular strategy was validated in these studies.
To ensure that children would provide all of their available metacogni-
tive knowledge in a test situation, Best and Ornstein (1986) used a peer-tutoring
assessment procedure in which older children (e.g., third or sixth graders) were
asked to teach a memory strategy such as sorting items into semantic categories
to younger children (e.g., first graders). The tutors were videotaped while giving
their instructions, and the tapes were subjected to content analyses. The measure of
metamemory was the extent to which the instructions included appropriate strategy
instructions (e.g., appropriate use of organizational strategies for a sort-recall task).
Overall, these alternative methods alleviated some of the problems usually
related to the use of questionnaire measures. However, other problems tend to
264 9  The Development of Metamemory

arise when these kinds of assessment procedures are used. For instance, the task
to rank order several strategies may be easier for older than for younger children
because older children are more familiar with classification tasks. Also, although
the peer-tutoring approach seems attractive because children are more likely to use
their metamemory when communicating with another child than when communi-
cating with an adult, this method is not infallible as a measurement tool given that
child tutors do not necessarily use all their metacognitive knowledge in a single
tutoring setting. One likely hypothesis is that they might bias their explanations of
strategies downward to match their perceptions of the cognitive levels of the tutees
(Joyner and Kurtz-Costes 1997; Schneider and Pressley 1997).
Moreover, questionnaire measures that cannot be used with younger children
create difficulties when applied to older children and adolescents, particularly
when knowledge about text processing is assessed. For instance, when I was asked
to construct such a metacognition measure for the first international OECD PISA
study in 2000, the pilot data were disappointing. Although the measure was per-
fectly reliable, it lacked validity. One of the reasons for this was that not only did
the children who were good text information processors identify suitable strategies
and indicate that they used such strategies most of the time, even the poor text
processors intuitively selected the better strategies and pretended that they used
these strategies all of the time. The surprising outcome was that the actual text
processing data yielded nonsignificant correlations between metamemory, strategy
use, and memory performance. In fact, one paradoxical outcome of this research
was that low achievers produced higher metamemory scores than high achievers.
To avoid such problems, more sophisticated measures of metacognition must
be used with older children and adolescents. Schlagmüller and Schneider (2007)
came up with a standardized measure of metacognition that was based on a revised
test instrument developed for PISA 2000 (see Artelt et al. 2001) and was later
included in the PISA 2009 study. This instrument taps adolescents’ knowledge of
strategies that are relevant during reading and tests their comprehension as well
as their recall of text information. For each of six scenarios, students have to eval-
uate the quality and usefulness of five different strategies available for reaching
the intended learning or memory goal. The rank order of strategies obtained for
each scenario is then compared with an optimal rank order provided by experts
in the field of text processing. The correspondence between the two rankings is
expressed in a metacognition score that indicates the degree to which students are
aware of the best ways to store and remember text information.

Measures of Procedural Metamemory

Concurrent measures of metamemory are characterized by the presence of

simultaneous memory activity (Cavanaugh and Perlmutter 1982; Meichenbaum
et al. 1985). Here, children and adolescents are asked to judge their memory
performance either shortly before, while, or after they work on a memory task.
Assessment of Metamemory 265

The most-studied type of procedural metamemory is self-monitoring, that is,

evaluating how well one is progressing (see Brown et al. 1983; Schneider and
Lockl 2002, 2008). The developmental literature has focused on performance
prediction or ease-of-learning (EOL) judgments, judgments of learning (JOLs),
and feeling-of-knowing (FOK) judgments and also explored some aspects of
postdiction, control, and self-regulation such as confidence judgments, alloca-
tion of study time, and termination of study (recall readiness).
Ease-of-learning (EOL) judgments occur in advance of the learning process,
are largely inferential, and refer to items that have not yet been learned (Nelson and
Narens 1994). The corresponding memory paradigm is performance prediction. A clas-
sic EOL task refers to the prediction of one’s own memory span. Individuals are pre-
sented incrementally longer lists of materials to be learned, such as pictures, words, or
figures and are asked to indicate whether they could recall a list that long. Children’s
memory is then tapped using the same lists. Comparisons of the predictor value with
actual memory span yields the metamemory indicator. Performance-prediction accu-
racy can be measured for a variety of memory tasks, including list-learning paradigms
and text-learning tasks (cf. Schneider et al. 1990; Schneider and Uhl 1990).
Judgments of learning (JOLs) occur during or soon after the acquisition of mem-
ory materials and are predictions about future test performance on recently studied (and
probably still recallable) items. Typically, paired-associate learning tasks are used in this
context. After completing a learning trial, participants are shown a given pair as a stimu-
lus and have to indicate how confident they are whether they will remember the correct
item response either immediately or 10 min later. Some studies have also assessed chil-
dren’s postdictions (e.g., Pressley et al. 1987). After memorizing a list of items, chil-
dren were asked how many items they had correctly recalled or how sure they were that
they correctly answered specific items (confidence judgments, see below). Overall, the
database concerning children’s performance in JOL tasks is still rather small compared
with the large body of literature addressing JOLs in adults.
A number of developmental studies have explored children’s feeling-of-knowing
(FOK) judgments (e.g., Butterfield et al. 1988; Lockl and Schneider 2002a). These
judgments occur either during or after a learning procedure and are judgments about
whether a currently unrecallable item will be remembered on a subsequent retention
test. Typically, children are shown a series of items and asked to name them. When
children cannot recall the name of an object when shown its picture, they are asked
to indicate whether they would recognize the name if the experimenter provided it.
These FOK ratings are then related to subsequent performance on the recognition
test that included nonrecalled items.
Several studies exploring procedural metamemory in naturalistic contexts (e.g.,
in eyewitness memory situations) have also assessed participants’ confidence
about given answers. These confidence judgments (CJs) are frequently based on
“confidence scales,” particularly when the participants are young children. Here,
the experimenter usually provides a few examples, showing children how to make
confidence judgments about their previous answers using a Likert scale on which
higher values indicate higher confidence (e.g., Roderer and Roebers 2010; Roebers
2002; Roebers and Schneider 2005).
266 9  The Development of Metamemory

A concurrent measure tapping memory behaviors as they occur is the verbal

protocol analysis developed by Ericsson and Simon (1980, 1993). This technique
requires participants either to verbalize all thoughts that come to mind while per-
forming a task (i.e., think aloud) or to verbalize between experimental trials. Such
“online” methods have been used in studies with older children, adolescents, and
adults and seem to be preferable to “offline methods” such as questionnaires and
interviews in these age groups given that such online methods are more predictive
of learning performance relative to the offline methods (Bannert and Mengelkamp
2008; Veenman 2006).
Whereas self-monitoring involves knowing where you are with regard to your
goal of understanding and memorizing task materials, self-regulation includes
planning, directing, and evaluating your own mnemonic activities (cf. Flavell et al.
2002). Some developmental studies have addressed aspects of children’s control
and self-regulation processes such as termination of study (recall readiness) and
allocation of study time (see the reviews by Roebers 2014; Schneider and Lockl
2008). Recall readiness assessments are made after learning materials have been
studied at least once. Typically, participants are asked to continue studying until
their memory of the materials to be learned is perfect. After children indicate that
they reached this goal, their memory performance is tested.
Another example of self-regulation skills concerns the allocation of study
time. This research observes how learners deploy their attention and effort when
studying lists of items. As already noted by Brown et al. (1983), the ability to
attend selectively to relevant aspects of a memory task is a traditional index of
a learner’s understanding of the task. Developmental studies on the allocation of
study time have examined whether schoolchildren and adults were more likely to
spend more time on less well-learned material. For instance, after a first free-recall
trial, participants had to distinguish between recalled and nonrecalled items (moni-
toring component) and were then asked to select half of the items for additional
study (self-regulation component). Other developmental studies using paired-asso-
ciate learning tasks have compared study times for objectively easy and difficult
pairs and have related this information to learning outcomes.

Measurement Problems

A number of concerns have been raised about some measures of procedural

metamemory, and measurement problems have been discussed in the relevant
literature. For instance, problems in assessments of EOL judgments seem
related to the fact that the competence of young children can be overesti-
mated if the researcher stops taking estimates once the child indicates that a
list is too long. Young children often respond inconsistently, and even when
they indicate that they would not be able to recall a list composed of five or
six items, they might claim that seven- or eight-item lists would be no problem
(Brown et al. 1978). The accuracy of span estimates seems to vary with material
Assessment of Metamemory 267

and task familiarity (Markman 1973; Schneider et al. 1987) or whether the chil-
dren received training on the task (Cunningham and Weaver 1989; Markman
1973). Also, mode of assessment seems to affect span estimations. For instance,
Cunningham and Weaver (1989) found that preschoolers’ span predictions were
more realistic when they listened to items on a tape and used a stop key to indi-
cate whether an item was too long rather than responding to verbal probes.
Finally, research focusing on the impact of motivation has shown that young
children’s overestimations of their own performance on memory-span-pre-
diction tasks may not be due to metacognitive deficiencies but may instead be
primarily affected by wishful thinking (Schneider 1998; Stipek 1984). That is,
young children may be estimating the performance they would like to achieve
rather than how they think they will perform when predicting task performance.
Similarly, one problem with the paradigm of the allocation of study time is
that it might not only tap metacognitive processes but may also be influenced by
motivational variables (see Schneider and Lockl 2008).
Recall readiness measures can also be criticized. Is it possible that young chil-
dren indicate that they are ready for a test because they understand the task to be
one of learning quickly? It is known that younger and older children use differ-
ent response criteria in evaluating their readiness such that younger children apply
more liberal criteria than older children (Worden and Sladewski-Awig 1982). A
similar problem plagues feeling-of-knowing data. Nelson and Narens (1980) have
shown that the feeling-of-knowing paradigm confounds participants’ threshold for
“know” versus “do not know” responses with their knowledge about unrecalled
items. This is a problem in developmental data because knowledge about unre-
called material varies with development.
Methodological concerns have also been discussed in relation to thinking-
aloud techniques. The main issue is whether verbalizing one’s own thoughts when
working on a memory task may influence memory behavior and performance.
Also, individual differences in the ability to verbalize ongoing thoughts and thus
the level of language development may influence results. However, most studies
investigating these issues have found that the think-aloud method does not inter-
fere with metacognitive processes, although it may slow down the speed of these
processes to some extent (Bannert and Mengelkamp 2008; Ericsson and Simon
1993; Veenman 2006).
In summary, there is no perfect index to measure any aspect of metamemory.
The candor of metamemory researchers about the limitations of their measures
is commendable. However, many of the measurement problems that metamem-
ory researchers confront are similar in other areas of psychology. The forthright
approach to these methodological problems has led to considerable methodo-
logical progress (see examples below). It is impossible to read the metamemory
research produced in the last three decades without being impressed by the meth-
odological savvy of investigators in overcoming what could have been viewed as
insurmountable problems. Even more striking, diverse methodologies often pro-
duce converging data. See the reviews by Wagoner (1983) or Garner (1987) on
monitoring during text learning for especially impressive evidence on this point.
268 9  The Development of Metamemory

The Development of Metamemory in Children

and Adolescents

Precursors of Metamemory: Knowledge of “Mental Verbs”

and the Acquisition of a “Theory of Mind”
A basic requirement for the acquisition of (declarative) metamemory is an appro-
priate understanding of mental verbs such as thinking, forgetting, or knowing.
Although Kreutzer et al. (1975) provided evidence that the youngest participants
in their study (kindergarten children) could properly apply mental verbs, it has
proven more difficult to determine preschoolers’ knowledge of mental verbs. Early
studies with preschoolers on the issue (e.g., Miscione et al. 1978; Johnson and
Wellman 1980) as well as subsequent research (e.g., Astington and Olson 1990;
Schwanenflugel et al. 1994) have all demonstrated that young children’s compe-
tent use of mental verbs was highly constrained. Obviously, acquiring this kind
of knowledge is a long-term development, and children’s understanding is limited
compared with adults’ understanding.
Since the early 1980s, studies of preschoolers’ metacognition have been moti-
vated by Perner’s (1991) and Wellman’s (1990) conceptualizations of children’s
theory of mind (see also Sodian 2005), which emphasizes important classes of
knowledge about the inner mental world that children acquire by the ages of 3–4.
From this age on, children develop a rudimentary understanding of mental verbs
such as “thinking” or “remembering” and can separate mental processes from the
external behaviors associated with them. For instance, Moore et al. (1989) demon-
strated that 4-year-olds can reliably distinguish between “knowing” and “thinking”
or between “knowing” and “guessing.” However, children of that age are still una-
ble to distinguish between “thinking” and “guessing” on a word comprehension
task. Preschool children are relatively able to act in accordance with mental verbs
indicating certainty and access to knowledge such as “know” or “remember” but
find it difficult to act in accordance with mental verbs indicating uncertainty such
as “think” or “guess” (Cherney 2003). They then gradually learn to recognize that
the mental world can be differentiated into processes such as remembering, know-
ing, and guessing (i.e., they acquire knowledge about distinct mental processes).
Although 3- to 4-year-olds are not generally capable of differentiating between
these processes, older preschoolers already make distinctions that are very similar
to those of adults.
The relation between the development of language, theory-of-mind devel-
opment, and metamemory development has only recently been systematically
explored in longitudinal studies. For instance, Astington and Jenkins (1999) as
well as de Villiers and Pyers (2002) found that language competence predicted
theory-of-mind development but not the reverse. Whereas these studies did not
focus on longitudinal relations among knowledge of mental verbs, theory of mind,
and subsequent metamemory, this issue was carefully analyzed in a subsequent
longitudinal study carried out in our lab (see Lockl and Schneider 2006, 2007;
Schneider et al. 2005). Here, the prediction was that both early theory-of-mind
The Development of Metamemory in Children and Adolescents 269

competence and metacognitive vocabulary (i.e., knowledge of mental verbs)

would affect subsequent knowledge of memory.
Our findings showed that metacognitive vocabulary, theory of mind, and general
metamemory improved considerably over the preschool years. In accordance with
previous results, mean performance on the mental verb and metamemory assess-
ments was far from the ceiling, indicating that knowledge of mental verbs and
knowledge about memory strategies is not particularly rich before children enter
school. Another interesting finding was that the test–retest stabilities were moder-
ate for the three constructs under study (about 0.50). This suggests that individual
differences in the development of metacognitive vocabulary, theory of mind, and
metamemory already exist at an early age. Last but not least, results of hierarchi-
cal regression analyses indicated that both early theory of mind and metacognitive
vocabulary made reliable and substantial contributions to the prediction of meta-
memory even when individual differences in nonverbal intelligence and general
vocabulary were taken into account. Given that prior metamemory also predicted
subsequent metacognitive vocabulary, the findings seemed to indicate a reciprocal
association and did not support the assumption of a clear-cut cause-effect relation.

Development of Declarative Metamemory

from Early Childhood to Adolescence

In the classic interview study by Kreutzer et al. (1975), children in kindergarten

and Grades 1, 3, and 5 were asked about person, task, and strategy variables. For
example, children were asked whether they ever forgot things, if it was easier to
recall the gist of a story than to recall it verbatim, and what they could do to find
a jacket they had lost while at school. Overall, the results of this study and related
assessments (e.g., Myers and Paris 1978; Schneider et al. 1986, 2004) indicated
substantial improvements on most of the variables as a function of age. Regarding
person variables, Kreutzer et al. (1975) included only one item on the individual
stability of memorization skills. The 9- and 11-year-olds in the study were better
able to conceptualize memory abilities than the younger children in that the older
children realized that memorization skills vary from person to person and from
situation to situation. The 9- and 11-year-olds also knew that they did not have
an equally good memory in all situations and that it was quite possible that their
friends had better memories than they did. By contrast, most of the kindergarten
and first-grade children were convinced that they always remembered well and
that they were better at remembering than their friends. This finding is consistent
with other demonstrations that the memory-related self-concept of young children
is unrealistic (Flavell et al. 1970; Schneider 1998).
However, preschool children can be quite accurate in other aspects of person-
related metamemory. For instance, Wellman (1977) demonstrated that more than
75 % of the 3- to 5-year-olds in his sample knew that person characteristics such
as hair color, weight, or clothing would not influence memory performance, a
270 9  The Development of Metamemory

finding basically replicated by Weinert and Schneider (1986) using data from the
Munich Longitudinal Study (LOGIC). Even more impressive, most of the young
children were also able to provide appropriate reasoning for their decisions. By
contrast, however, only about 50 % of the children in Wellman’s sample realized
that chronological age is an important variable and that memory improves with
increasing age. The modest knowledge about the effects of age on memory pos-
sessed by preschool children is only evident when age is a salient variable in the
metamemory assessment items. Much less awareness of the age-memory associa-
tion is obtained by using problems in which age is a less obvious factor (Yussen
and Bird 1979). Preschool children clearly have a great deal of difficulty in deter-
mining the importance of relatively stable personal characteristics that determine
memory performance.
However, this does not mean that young children do not possess any adequate
knowledge about memory. There is some knowledge about the importance of
task variables for memory performance. For instance, the kindergarteners in the
Kreutzer et al. (1975) study knew that remembering many items is more dif-
ficult than remembering just a few, and the majority of these children also knew
that using external devices (e.g., writing telephone numbers down) is helpful for
remembering information. According to Wellman (1977), even younger children
(3- to 5-year-olds) know about the importance of this task factor: About 82 %
of the children in his sample were convinced that 18 items are more difficult to
remember than three items. However, subsequent studies (Weinert and Schneider
1986; Yussen and Bird 1979) produced data suggesting that Wellman (1977) may
have overestimated preschoolers’ knowledge of task variables. In both the Munich
longitudinal study and Yussen and Bird’s (1979) research, only about 40 % of the
4-year-old children were able to correctly judge the effect of the number of items
on the difficulty of memory problems (see Beal 1985; O’Sullivan 1993; Schneider
and Sodian 1988, for confirmatory findings). Yussen and Bird (1979) reported that
6-year-olds’ understanding of the significance of task characteristics is much more
complete than the understanding of preschool children. Almost 80 % of the 6-year-
old first graders in their study knew that the number of items affected memory per-
formance. The first graders also knew about the importance of other task-relevant
aspects such as the influence of noise on memory performance. Even the most dif-
ficult item, that is, the relevance of learning time for performance, was answered
correctly by about half of the 6-year-olds. Apparently, knowledge about the impact
of task variables on memory develops quite rapidly between the ages of 4 and 6.
Although these results seem to indicate that by the beginning of school, chil-
dren know a lot about the importance of some task characteristics on memory
performance, it must be emphasized that other variables are poorly understood by
young elementary schoolchildren. For instance, 7-, 9-, and 10-year-old students in
Moynahan’s (1978) study were asked to judge which of two lists would be easier
to learn: one composed of taxonomically organized items or one composed of con-
ceptually unrelated words. Although both older groups recognized the advantages
conferred by the taxonomic structure, this did not hold true for the 7-year-olds.
Older children are more likely than younger children to notice relations among
The Development of Metamemory in Children and Adolescents 271

items and are also more likely to understand that if items “go together” in some
meaningful way, such items will be easier to learn (e.g., Hasselhorn 1990). This
result has been confirmed in several follow-up studies (Schneider 1986; Schneider
et al. 1987). Grade-1 children also lack knowledge of retrieval factors and their
effects on memory: The majority of first graders in the Speer and Flavell (1979)
study did not know that recognition memory tasks are easier than recall tasks.
Moreover, even though many kindergarten children and first graders know that
memory tasks are easier when retrieval cues are available (see below), a sophis-
ticated understanding of retrieval cues and how they work develops during the
grade-school years (Beal 1985; Schneider and Sodian 1988).
One area in which preschoolers show elementary metacognitive knowledge is
with regard to the usefulness of a strategic approach to remembering. In particular,
adequate metacognitive knowledge can be found in memory situations that young
children are familiar with, for instance, in the context of memory-for-location
tasks. Justice (1989) found that 5-year-olds were able to identify an effective strat-
egy for finding a hidden object. Four- and 5-year-old children were first familiar-
ized with a task in which a colored chip was hidden in a movable display. Next,
the children watched a videotape that showed actors who either looked away when
the hiding place was rotated or marked the place with a colored chip. Although
all children identified the least effective strategy, only the 5-year-olds in this study
were able to identify the marking strategy as the more effective memory technique.
Similarly, Beal (1985) demonstrated that preschoolers have a basic but incomplete
understanding of strategy use as a memory aid. All children in her study (pre-
schoolers, kindergarteners, and third graders) understood that a retrieval cue (using
a paperclip to mark the cup under which a penny was hidden) should be helpful.
However, the preschoolers did not understand the problems of multiple retrieval
cues and how they may present problems in the future. For example, about half of
the preschoolers believed that the paper-clip marker would help them even when
three other markers were present. Accordingly, these young children did not recog-
nize that ambiguous cues would interfere with accurate retrieval even though they
correctly realized that the strategy of marking the cup should be effective.
Several studies have reported that a basic knowledge of retrieval strategies is
already available in young children. Kreutzer et al. (1975) included one item in
their battery to assess knowledge of strategies that could be used to prepare for
future retrieval. Children were asked to explain everything they could do to ensure
that they would not forget to take their ice skates to school the next day. There
was a significant increase in the number of strategies reported, and even kindergar-
ten children were able to come up with at least one strategy each. All age groups
reported more external (e.g., putting the skates next to the door) than internal
strategies (e.g., rehearsal of the fact that the skates need to be taken to school).
In their replication work, Borkowski and his colleagues (Borkowski et al. 1983;
Cavanaugh and Borkowski 1980; Kurtz and Borkowski 1984) obtained generally
comparable data in response to the “skate” question.
Another item in the Kreutzer et al. (1975) metamemory battery assessed knowl-
edge of retrieval strategies by asking “What would one do to find a jacket that
272 9  The Development of Metamemory

was lost at school?” Children of all age levels suggested looking in places where
the jacket would most likely be found (e.g., the cloak room) and asking people
who would likely be helpful (e.g., the teacher). Every kindergarten child offered
at least one meaningful solution, and first-grade children averaged two solutions.
Suggestions to search systematically and elaborately were offered more often by
older (Grade 5) than younger participants. In a follow-up study, Yussen and Levy
(1977) found that the number of reported strategies for retrieving a lost object
increased up until about 10 years of age.
To find out what young children know about memory strategies, Justice (1985)
showed children videotaped examples of other children using different strategies
(e.g., looking, naming, rehearsing, and grouping). Children were then presented with
each strategy paired with each other one, one pair at a time, and were asked which
of the two strategies was better for remembering. This research and similar stud-
ies (Justice 1986; O’Sullivan 1993; Schneider 1986) produced a consistent picture
of children’s developing strategy knowledge. In general, preschoolers considered
“looking at” to be the best strategy, with taxonomic sorting, rehearsal, and naming
trailing in popularity. For instance, O’Sullivan (1993) confirmed that preschoolers
have great faith in the “looking at” strategy even though it is less effective than taxo-
nomic grouping. Kindergarten children were more likely to view all four strategies
as equally effective. Second and fourth graders preferred grouping and rehearsal but
did not differentiate between the two. By Grade 6, there was a clear understanding
that semantic grouping strategies are superior to rehearsal, naming, and looking.
An analysis of strategy-related metamemory using data from the Munich
Longitudinal Study (LOGIC) by Schneider and Sodian (Schneider and Sodian
1991; Sodian and Schneider 1999) indicated that the developmental pattern seems
different for German children. At 4 years of age, participants in this study were
given a short metamemory interview addressing the utility of various memory
strategies. This interview was repeated when the children were 6, 8, 10, and 12.
As can be seen in Fig. 9.3, only a few preschoolers judged conceptual (seman-
tic organizational) strategies to be “best” for memorization. Most preschool-
ers judged organization by color to be the best strategy. From the age of 8 years
onward, the large majority of the children chose conceptual organization as the
best strategy, and most could provide memory-related reasons for their judgments.
There was also some evidence that individual differences in metamemory were
related to further stability in strategy use in the sense that early metamemory
was associated with high stability in strategy use. Please note that metamemory
regarding strategies was observed at an earlier stage in German children (see
also Schneider 1986) in comparison with the American children in the studies
by Justice (1986) and Corsale and Ornstein (1980). In the Corsale and Ornstein
study, seventh but not fourth graders seemed to know what to do in the context of
a sort-recall task that would help them remember. Apparently, the fourth graders
studied by Corsale and Ornstein did not understand that their semantic knowledge
could be used effectively when confronted with a memory goal, whereas most
of the fourth graders in Schneider’s (1986) study knew about the advantages of
semantic-organizational strategies.
The Development of Metamemory in Children and Adolescents 273

120

100

80

60

40

20

0
4 years 6 years 8 years 10 years 12 years
pref for conceptual pref for color other

Fig. 9.3  The development of task-related metamemory (Sodian and Schneider 1999, p. 69,

Fig. 4)

Data on knowledge of categorization strategies has also been generated by

making pairwise comparisons of videotaped presentations. For instance, Sodian
et al. (1986) presented lists containing items that could be categorized taxonomi-
cally or by color. The 4- and 6-year-olds made pairwise comparisons between
taxonomic-sorting, color-sorting, random-sorting, and looking strategies. The
judgments of the two age groups differed only a little with the exception that the
taxonomic-sorting strategy was judged more positively by the older children.
However, they did not see a difference between the taxonomic-sorting and the
color-sorting strategies.
Although young children do have a basic understanding of memory, factual
knowledge about the importance of task characteristics and memory strategies
develops more rapidly after children enter school. Knowledge about the useful-
ness of memory strategies was tapped in several studies that focused on organiza-
tional strategies (see the reviews by Schneider 1999; Schneider and Lockl 2002).
Preferences for the most appropriate (sorting and clustering) strategies were not
found before the ages of 8 or 10 years, and reasonable justifications for such pref-
erences were not always provided.
The studies summarized above have focused primarily on children’s understand-
ing of the variables that influence how likely a person is to remember or to forget
something. In a more recent study, Jaswal and Dodson (2009) explored young chil-
dren’s understanding of why people occasionally misremember things. The authors
included 5- and 6-year-olds in their study to ensure that the children already had a
good grasp of the distinction between remembering, forgetting, and guessing and also
understood the difference between actually seeing something and just thinking about
it. Participants heard a story about a fictitious boy who had earlier played a mem-
ory game and who was given a recognition test consisting of pictures he had seen
274 9  The Development of Metamemory

before and a few distractors. The distractors differed in their degree of similarity to
the target items. Thus, sometimes the boy claimed to have seen an item that was simi-
lar to the ones he had seen before, and sometimes he claimed to have seen an item
that was unrelated. The critical question was whether the participants would show an
understanding of the link between the similarity of the distractor and the boy’s false
memories. As a main result, the 6-year-olds consistently made different attributions
for errors involving similar and dissimilar lures, attributing more similar than dissimi-
lar lures to false memories and not to guesses. This was not true for the 5-year-olds,
whose understanding of the link between similarity and false memories was very lim-
ited. Thus, the authors concluded that understanding this link improves significantly
between 5 and 6 years of age.
Substantial age trends were also observed when the interaction of memory
variables was considered. In a classic study, Wellman (1978) presented memory
problems to 5- and 10-year-olds. Each problem consisted of ranking three picture
cards, each of which contained a memorizing scenario. Whereas all of the chil-
dren solved the simple problems that involved a single-task variable such as the
impact of the number of items on memory performance, substantial developmental
differences were found for the complex memory problems in which two aspects
were varied (e.g., number of items and type of strategy). Only a very small propor-
tion of children in the younger age group were able to judge the complex memory
problems appropriately. In comparison, almost all 10-year-olds performed per-
fectly on this task. Thus, these data indicate that interactive memory knowledge
develops considerably during the elementary school years.
However, it may be premature to assume that preschool children are not capa-
ble of considering and comparing two different features of memory problems
at the same time. Wellman et al. (1981) focused on knowledge of the combined
effects of a task variable (i.e., the number of items to be remembered) and a per-
son variable (i.e., the amount of effort expended). Three values of each of these
variables were combined factorially to create nine different picture cards. The par-
ticipants in this study (5-, 8-, 10-, and 19-year-olds) were asked to predict how
many items the boy on the card would remember. The prediction of memory per-
formance served as the dependent variable. The most important result was that
all of the children were aware of the combined influence of the number of items
and effort on the probability of success although there were differences between
the age groups. Compared with older children and young adults, younger children
did not weigh the amount to be learned as heavily as effort. Thus, the differences
between age groups were due to the differential weighting of the two factors. Only
the 19-year-olds made performance predictions that appropriately reflected a bal-
anced consideration of the two variables in combination.
How can the apparent inconsistencies between Wellman (1978) and Wellman
et al. (1981) be explained? Wellman (1978) required participants to rank three
scenes that varied with respect to two variables. Wellman et al. (1981) predicted
that preschoolers in that situation would concentrate on one variable and ignore
the other one entirely. As a consequence, Wellman et al. (1981) presented only one
value of each of the two variables at a time, and this should have been an easier
The Development of Metamemory in Children and Adolescents 275

task, especially for younger children. Although Wellman et al. (1981) succeeded
in showing that 5-year-olds possess at least elementary knowledge of the com-
bined influence of two variables, the data also made clear that interactive memory
knowledge develops very slowly. This development continues well into adoles-
cence (see Schneider and Pressley 1997).
Taken together, the empirical evidence illustrates important changes in declara-
tive metamemory over time. Using sensitive methods that minimize the demands
on the child, it has been possible to demonstrate some rudimentary knowledge
about memory functioning in preschoolers. Knowledge of facts about memory
develops quickly during the course of elementary school and is already impres-
sive by 11 or 12 years of age (cf. Pressley and McCormick 1995; Schneider and
Lockl 2008). Nonetheless, declarative metamemory is not complete by the end
of childhood. It seems important to note that even though metacognitive knowl-
edge increases substantially between young childhood and late adolescence, there
is also evidence that many adolescents and adults (including college students)
frequently fail to be strategic, demonstrating little knowledge of powerful and
important memory strategies when the task is to read, comprehend, and memo-
rize complex text materials (cf. Brown et al. 1983; Garner 1987; Pressley and
Afflerbach 1995). Moreover, adolescents and adults also occasionally fail to moni-
tor their performance and change their strategy use when it would be appropriate
to do so (e.g., Shaughnessy 1981). Developmental trends in skills involving moni-
toring and self-regulation will be described in more detail in the next section.

The Development of Monitoring and Self-Control

Early research focusing on monitoring showed that even young children seem
to possess the relevant skills, particularly when memory tasks are not very diffi-
cult (see reviews by Joyner and Kurtz-Costes 1997; Roebers 2014; Schneider and
Pressley 1997). However, the evidence regarding developmental trends has not
been consistent, with some studies showing better performance in younger than in
older children and others illustrating age-related improvement. More recent stud-
ies exploring developmental trends in monitoring and self-regulation (as well as
the interaction between these two components) have been helpful for clarifying the
situation and will be summarized next.
According to Nelson and Narens (1990, 1994), self-monitoring and self-regu-
lation correspond to two different levels of metacognitive processing that interact
very closely. Self-monitoring refers to keeping track of where you are with your
goal of understanding and remembering (a bottom-up process). In comparison,
self-regulation or -control refers to central executive activities and includes plan-
ning, directing, and evaluating your behavior (a top-down process).
What are the determinants of metacognitive judgments and their accuracy?
Most researchers have adopted a “cue-utilization” view, according to which
metacognitive judgments are inferential in nature and are based on a variety of
276 9  The Development of Metamemory

heuristics and cues that have some degree of validity in predicting memory perfor-
mance (e.g., Benjamin and Bjork 1996; Dunlosky and Nelson 1992; Koriat 2007).
An important distinction has been made between “theory-based” and “experience-
based” metacognitive judgments (Koriat 1997). Whereas theory-based judgments
rely on the deliberate application of metacognitive beliefs or theories about one’s
competencies and skills, experience-based judgments are assumed to rely on mne-
monic cues that derive from online information processing. So far, developmen-
tal research on procedural metacognition has hardly examined the contributions
of mnemonic cues and heuristics to children’s judgments. Even among adults, the
contribution of people’s theories and knowledge to their monitoring and control
seems to be quite limited (see Koriat et al. 2004). Thus, there is reason to assume
that children’s judgments are predominantly guided by the online implicit utiliza-
tion of subtle experiential cues.
The most-studied type of procedural metamemory is that of self-monitoring,
which is the evaluation of how well one is progressing (cf. Borkowski et al. 1988;
Brown et al. 1983; Schneider 1998). The developmental literature has focused on
monitoring components such as ease-of-learning (EOL) judgments, judgments of
learning (JOLs), feeling-of-knowing (FOK) judgments as well as confidence judg-
ments (CLs) and has also explored some aspects of control and self-regulation
such as the allocation of study time and termination of study.

Ease-of-Learning (EOL) Judgments

Ease-of-learning judgments occur in advance of the learning process, are largely

inferential, and refer to items that have not yet been learned (Nelson and Narens
1994). The corresponding memory paradigm is performance prediction. A form of
performance prediction first introduced by Flavell et al. (1970) refers to the predic-
tion of one’s own memory span. Individuals are presented incrementally longer
lists of materials to be learned, such as pictures, words, or figures and are asked
to indicate whether they could recall a list that long. Children’s memory is then
tapped using the same lists. A comparison of the predicted value with the person’s
actual memory span yields the metamemory indicator. Performance-prediction
accuracy can be measured for a variety of memory tasks, including list-learning
paradigms and text-learning tasks (cf. Schneider et al. 1990).
Almost all of the studies on EOL judgments that have used list-learning para-
digms have found that preschool and kindergarten children overestimate their
memory performance, whereas grade-school children are much more accurate
(e.g., Schneider 1998; Worden and Sladewski-Awig 1982; Yussen and Levy 1975).
Although this phenomenon has been observed repeatedly, the underlying mech-
anisms are not yet clear. Several studies have tried to identify young children’s
difficulties in making accurate performance predictions. It was found that their
predictions tended to be more accurate in familiar than in unfamiliar laboratory-
type situations (Justice and Bray 1979). Moreover, young children’s predictions
The Development of Metamemory in Children and Adolescents 277

were more accurate when they were tested using nonverbal as opposed to more
traditional verbal measures (e.g., Cunningham and Weaver 1989). Also, preschool-
ers and kindergarteners were found to be more accurate in predicting other chil-
dren’s performance than their own (Schneider 1998; Stipek 1984).
Overall, the evidence does not support the original assumption that young
children’s overestimations of future performance are due to metacognitive defi-
ciencies as indicated by more recent work on the issue (Schneider 1998; Visé and
Schneider 2000). For instance, a study by Visé and Schneider (2000) explored
possible reasons for young children’s unrealistic predictions. In particular, the
study examined whether overestimations in performance predictions are due
to deficits in metacognitive monitoring or to motivational factors, for instance,
wishful thinking. Four-, 6- , and 9-year-old children were asked to predict their
own performance on motor tasks (ball throwing and jumping) and memory tasks
(memory-span and hide-and-seek tasks). Children in the “wish condition” were
asked to declare what they would like their performance to be on the next trial,
whereas children in the “expectation condition” were asked to indicate what they
expected their scores to be on the next trial. A comparison of children’s perfor-
mance and their “postdictions” (i.e., their estimates of performance assessed after
they completed the task) indicated that all children were able to monitor their
performance regardless of the task even though they did not use this knowledge
to make further predictions. Accordingly, the memory monitoring deficiency
hypothesis could not account for the overestimation phenomenon. Furthermore,
4- and 6-year-old children did not differentiate between their wishes and their
expectations, thus replicating and extending the findings of motivational
researchers (e.g., Stipek 1984).
Taken together, such findings provided at least partial support for the wishful
thinking hypothesis and also clear evidence that overestimations made by pre-
schoolers and kindergarten children were linked to their belief (causal attribution)
that effort has a powerful effect on performance. However, because such motiva-
tional processes are not similarly influential in schoolchildren, performance on
EOL tasks indeed reflects memory monitoring in this population. Although EOL
judgments have already been found to be accurate in young elementary school-
children, there are subtle improvements across the elementary school years (see
Pressley and Ghatala 1990; Schneider et al. 1990).
A few other studies have also evaluated children’s postdictions (Bisanz et al.
1978; Pressley et al. 1987; Levin et al. 1987). For instance, Pressley et al. (1987),
Levin et al. (1987) compared 7- and 10-year-olds’ postdictions for entire word lists
and individual items. There were two major findings: (a) although rather accu-
rate postdictions were found even for the younger age group, the older children
were significantly more accurate; and (b) those children who were most accurate
with regard to estimating their performance on individual items were not simi-
larly accurate when asked to postdiction their performance on the entire list and
vice versa. Overall, the findings of these studies are in accordance with Visé and
Schneider’s (2000) results, which indicated that even young children are able to
monitor their performance.
278 9  The Development of Metamemory

Learning (JOLs)

Whereas numerous developmental studies have addressed differences in the pre-

diction of memory performance, only a few studies have examined judgments of
learning (JOLs) that occur during or soon after the acquisition of memory materi-
als and are predictions of future test performance for recently studied (and proba-
bly still recallable) items. The database concerning children’s performance on JOL
tasks is particularly small compared with the large body of literature addressing
JOLs in adults. Clearly, the tools developed by cognitive psychologists that have
been incorporated into developmental research have helped in the investigation of
important questions concerning memory monitoring in a developmental context.
In the first study that explicitly addressed developmental trends in children’s
JOLs, Schneider, Visé, Lockl, and Nelson (2000) used a paired-associate learning
task to assess JOL judgments in 6-, 8-, and 10-year-old children. A major goal of
the study was to explore whether the delayed-JOL effect that has been repeatedly
confirmed in the adult literature could be observed in children of different ages.
The children were asked to study 24 pairs (presented in two trials of 12 pairs) of
unrelated objects depicted on picture cards. In the immediate condition, children
were shown the stimulus picture immediately after studying the item pair and were
asked to predict whether approximately 10 min from now they would still be able
to recall the second picture of the pair when prompted with the stimulus picture. In
the delayed condition, children first studied all of the item pairs. Next, they were
presented with the stimulus picture of each item pair and made their JOLs. The
average delay between studying and making JOLs was about 2 min. As a main
result, the findings indicated that the delayed-JOL effect was also present in chil-
dren regardless of age. That is, JOL accuracy was significantly higher when JOLs
were delayed than when they were assessed immediately after studying the items.
This finding indicates that even 6-year-old children show accurate metacognitive
monitoring under delayed conditions.
A second goal of the study was to compare individual-item JOLs with aggre-
gate JOLs based on all of the items from a given list. After the children studied
the entire list, their aggregate JOLs were assessed by asking them how many of
the 12 items they would be able to remember correctly. It was found that overcon-
fidence was typically larger for item-by-item JOLs than for aggregate-item JOLs
for all age groups, thus replicating the aggregation effect obtained with adults. As
a matter of fact, the pattern of findings for the older schoolchildren was very simi-
lar to that found for adults. In accordance with the findings reported by Pressley
et al. (1987), Levin et al. (1987), however, only low to moderate correlations were
found between the two estimation procedures, thus suggesting that they tap differ-
ent aspects of the estimation process (see also Mazzoni and Nelson 1995; Nelson
and Narens 1994).
Taken together, the findings of this study suggest that even young children can
effectively monitor their learning progress under certain circumstances. On the
one hand, the results indicate that immediate JOLs are typically inaccurate and
The Development of Metamemory in Children and Adolescents 279

also represent overestimations of one’s actual performance. Remarkably, this is

true not only for children of different ages but also for adults. Immediately after
studying new information, judgments about its future recall seem severely biased
by the false belief that information currently in STM can be easily recalled some
minutes later. Obviously, this bias operates similarly in participants of different
ages. On the other hand, however, even young children can make rather accurate
assessments of the subsequent recallability of items when this judgment is some-
what delayed; that is, when it takes place about 2 min after studying the item. In
other words, even young children seem to have a good feeling for which items will
be recallable and which will not when LTM information has to be accessed for the
JOL. Our findings show not only that developmental trends are negligible but also
that the accuracy of young children’s delayed JOLs is close to that of adults.
In another study concerning JOLs in children, Koriat and Shitzer-Reichert
(2002) addressed the question of the basis of JOLs in children. On the basis of
the “cue-utilization model” of JOLs proposed by Koriat (1997), the study aimed
to examine whether different classes of cues that affect JOLs in adults also influ-
ence children’s metacognitive judgments. In particular, Experiment 1 focused on
two different factors: item difficulty, which is regarded as an intrinsic factor and
practice, which is viewed as an extrinsic factor according to the cue-utilization
model (Koriat 1997). Second and fourth graders were instructed to study easy
and difficult pairs of words so that they would be able to recall the response word
when cued with the stimulus word. After studying each word pair for 5 s, children
rated their JOLs on a 5-point scale. To investigate the effect of practice, the study-
test cycle was administered four times. Concerning the effect of item difficulty, it
was found that children gave higher JOLs for easy than for difficult word pairs.
However, there was an Age x Difficulty interaction, which indicated that the fourth
graders more strongly differentiated between easy and difficult pairs and showed
less overconfidence than the second graders. Moreover, the results revealed that
children’s JOLs increased with practice for both age groups. Thus, two factors that
were found to influence adults’ JOLs—namely, item difficulty and practice—simi-
larly affected children’s metacognitive judgments. With regard to JOL accuracy,
the results were not entirely consistent with those reported by Schneider et al.
(2000). Significant age effects emerged with a reliably higher mean Gamma corre-
lation between JOLs and recall for the fourth graders than for the second graders.
In a second experiment, Koriat and Shitzer-Reichert (2002) aimed to replicate
the delayed-JOL effect in second- and fourth-grade children by comparing the
delayed-JOL effect to the effect of practice on JOLs. It was predicted that either
delaying JOLs or allowing sufficient practice with the learning material would
result in an improved JOL accuracy but that combining the two would have lit-
tle effect beyond the effect of delaying JOLs. This is basically what they found.
Koriat and Shitzer-Reichert concluded from these results that delaying JOLs
and practice are two different ways to reach the same result of enhancing JOL
accuracy.
Whereas most developmental research on JOLs has focused on laboratory-
type paired-associate learning tasks, several other studies have assessed children’s
280 9  The Development of Metamemory

Immediate JOLs Delayed JOLs

Correct
answers, answerable
7 7
questions
6 6
Commission
5 errors, answerable 5
questions
4 4

3 invented answers 3
unanswerable
2 questions 2

1 1
"Don't know" answers
0 unanswerable 0
questions
8-year-olds 10-year-olds Adults 8-year-olds 10-year-olds Adults

Fig. 9.4  Mean JOLs as a function of delay, age, and question type (Roebers et al. 2007, p. 122)

monitoring skills in naturalistic contexts by investigating children’s eyewitness

memory (cf. Chap. 6). Most of these studies began by showing children a video
of some event, having them observe some activity, or involving them personally in
some activity. Usually, children were not told that they would be asked to remem-
ber what they viewed. Several hours, days, or weeks later, the children were then
asked what they remembered (e.g., “Tell me what happened in the video you saw”).
There were, of course, many variations depending on the purpose of the study.
For instance, a recent study by Roebers et al. (2007), von der Linden et al.
(2007) aimed to investigate children’s monitoring abilities in the context of such a
complex everyday memory task. The important goals of this study were to explore
developmental differences in children’s JOLs and to examine the effects of delay
and the role of retrievability with respect to JOLs. The sample consisted of 8- and
10-year-old children as well as adults who all watched a short event on a video.
Afterwards, participants rated on a 7-point scale how certain they were that they
would later be able to correctly recall specific details about the event. The mem-
ory test took place 2 weeks after the video presentation and the JOL interview.
To investigate the role of retrievability, the JOL interview as well as the memory
test included not only answerable questions about details that really occurred in
the video but also questions that could not be answered on the basis of the video.
For example, participants were asked “What did the boys have in their backpacks
when they arrived at the farmhouse?” but in the video, no information was given
about the contents of the backpacks. Roebers et al. (2007), von der Linden et al.
(2007) assumed that if JOLs are partly based on the retrievability of information,
JOLs about details that were not encoded should be lower than JOLs for encoded
information. In other words, individuals should be able to make a metacognitive
distinction between answerable and unanswerable questions.
As can be seen in Fig. 9.4, JOLs clearly differed as a function of the appro-
priateness of the answer, and JOLs were also influenced by the question type.
JOLs were highest for answerable questions that were answered correctly on the
memory test. In comparison, JOLs were significantly lower for incorrect answers
to answerable questions, and these in turn were significantly higher than JOLs for
The Development of Metamemory in Children and Adolescents 281

incorrect answers to unanswerable questions. The lowest level of JOLs was found
for unanswerable questions that were subsequently appropriately answered with
“I don’t know.” Remarkably, the three age groups did not differ in their mean lev-
els of JOLs, and no interactions between age and appropriateness of answer or
question type emerged. Thus, even children in the youngest age group were able
to appropriately differentiate between correct and incorrect answers. The same
applied for the ability to differentiate between answerable and unanswerable ques-
tions. JOL accuracy was also comparable across age groups with mean Gamma
correlations between JOLs and recall performance ranging between 0.53 and 0.70,
thus indicating moderate to high interrelations.
Furthermore, the comparison of JOLs for answerable and unanswerable ques-
tions showed that JOLs were higher for potentially answerable than for unan-
swerable questions. This suggests that participants based their JOLs on their
evaluations of information retrievability regardless of age. Thus, the amount
of information that comes to mind during the retrieval process seems to have an
impact on children’s and adults’ JOLs. Overall, the results are in accordance with
the view that JOLs are based on, among others, memory characteristics or mne-
monic cues, such as the ease with which information is retrieved or the accessibil-
ity of pertinent partial information about the memory target (Koriat 1993).

Knowing (FOK) Judgments

A number of developmental studies have explored children’s feeling-of-knowing

accuracy (e.g., Cultice et al. 1983; DeLoache and Brown 1984; Wellman 1977).
FOK judgments occur either during or after a learning procedure and are judg-
ments about whether a currently unrecallable item will be remembered on a subse-
quent retention test. Typically, children are shown a series of items and are asked
to name them. When children cannot recall the name of an object when shown its
picture, they are asked to predict whether they would recognize the name if the
experimenter provided it. These FOK ratings are then related to subsequent perfor-
mance on the recognition test.
Overall, most of the available evidence on FOK judgments suggests that FOK
accuracy improves continuously across childhood and adolescence (e.g., Wellman
1977; Zabrucky and Ratner 1986). However, the pattern of developmental trends
is not entirely clear. In a study that avoided a methodological problem apparent in
previous research on FOK judgments, Butterfield et al. (1988) showed that 6-year-
olds’ FOK judgments were actually more accurate than those of 10-year-olds and
18-year-olds. Obviously, this finding did not square well with the results of previ-
ous research.
A subsequent study by Lockl and Schneider (2002a) was based on a meth-
odologically improved design similar to that used by Butterfield et al. (1988)
but included different age groups (i.e., 7-, 8-, 9-, and 10-year-olds). Although
the major goal of this study was to replicate the findings by Butterfield and
282 9  The Development of Metamemory

colleagues, another aim was to explore the basis of FOK judgments by comparing
the traditional “trace-based” view with the “trace accessibility” model developed
by Koriat (1993). Whereas the former assumes a two-stage process of monitor-
ing and retrieval, the latter proposes that FOK judgments are based on retrieval
attempts and are determined by the amount of information that can be spontane-
ously generated regardless of its correctness. A prediction derived from the trace
accessibility view is that FOK judgments for correctly recalled items and incor-
rect answers (commission errors) should be comparably high and also consider-
ably higher than FOK judgments for omission errors.
As a main result regarding the first goal, no developmental trends in the accu-
racy of FOK judgments were found. Overall, FOK accuracy was low but signifi-
cantly greater than chance for all age groups. The main difference between these
findings and Butterfield and colleagues’ (1988) findings concerned the perfor-
mance of the youngest age group (i.e., first graders). Whereas FOK accuracy was
rather high for the American first graders, it was lower for the German first grad-
ers. Although there is no truly convincing reason for the differences between the
two studies regarding the youngest age groups, the findings suggest that there are
no significant developmental trends in FOK accuracy over the course of the ele-
mentary school years. However, given inconsistencies in the findings for the young
elementary schoolchildren, more research is needed here.
Furthermore, Lockl and Schneider’s (2002a) findings provided support for the
“trace accessibility” view and the assumption that a feeling of knowing can be dissoci-
ated from knowing. That is, the magnitude of FOK judgments given after commission
errors did not differ much from that of FOK judgments provided after correct recall.
In comparison, FOK judgments were considerably higher after commission than after
omission errors. This is in sharp contrast with the finding that recognition performance
for commission and omission errors was comparable (about 50 % correct), whereas it
was nearly perfect when items had already been recalled correctly before.
Contrary to earlier assumptions, recent research assessing monitoring abili-
ties in JOL or FOK tasks shows that even young children are able to monitor their
progress (Butterfield et al. 1988; Roebers et al. 2007; von der Linden et al. 2007;
Schneider et al. 2000). Although monitoring skills seem to improve continuously
across childhood and adolescence, it is important to note that developmental trends
in self-monitoring are less pronounced than those observed for declarative meta-
memory. Thus, these studies have contributed to a modification of the view that
children are generally overconfident and possess deficient monitoring abilities. By
contrast, previous work on procedural metamemory that was largely based on the
performance-prediction paradigm indicated that young children tend to be overly
optimistic and overestimate their memory performance (e.g., Flavell et al. 1970;
Schneider et al. 1986; Worden and Sladewski-Awig 1982; Yussen and Levy 1975).
The discrepancy between these two lines of research may for the most part be due
to the fact that different indicators of metacognitive abilities were used (e.g., Koriat
and Goldsmith 1996): On the one hand, studies including performance-prediction
tasks address absolute metacognitive accuracy or calibration, that is, the match
between a person’s predicted and actual overall memory performance. On the other
The Development of Metamemory in Children and Adolescents 283

hand, studies based on the JOL or FOK paradigms focus on relative accuracy or
resolution, that is, the accuracy with which a person can monitor the relative recalla-
bility of different items. This aspect of monitoring accuracy is especially important
in self-paced learning situations in which students have to differentially allocate their
study time to the materials that are to be learned. JOL or FOK studies have typically
used Gamma correlations as indicators of relative accuracy and found no (e.g., Lockl
and Schneider 2002a; Schneider et al. 2000) or only small age differences (e.g.,
Koriat and Shitzer-Reichert 2002, Experiment 1).
A closer look at the data obtained in these studies reveals that younger children
tend to be more optimistic about their future recall. That is, they seem to have a
more liberal response criterion and produce more false alarms than older children
(and presumably also adults). Because both types of errors (i.e., overestimation
and underestimation) are similarly taken into account when Gamma correlations
are computed, the resulting magnitude of the Gamma correlations may be com-
parable even though the types of underlying errors made by younger children are
rather different from the ones made by older children. In sum, there is converg-
ing evidence that younger children are more optimistic than older children and
adults and often overestimate their future memory performance. This tendency
to be overconfident, however, does not seem to affect children’s relative monitor-
ing accuracy, which seems comparable across age groups. Besides, other studies
concerning children’s overconfidence (Schneider 1998; Visé and Schneider 2000)
have not supported the original assumption that young children’s overestimations
of future performance are due to metacognitive deficiencies. Rather, these stud-
ies suggest that motivational factors such as wishful thinking and effort attribution
bias the recall estimates of young children. Bjorklund and Bering (2002) inter-
preted the general overestimation of one’s competencies as a protective factor for
cognitive development in general, helping individuals to maintain motivation and
task persistence and thereby fueling developmental progression in a broader sense.
Overall, taking experimental designs that were already frequently used in general
cognitive psychology and applying them to developmental studies has added to our
knowledge about age differences in children’s monitoring proficiency. It has also
provided important insights into the origins of metacognitive judgments in children.

Retrieval Monitoring: The Case of Confidence

Judgments (CJs)

Several developmental studies have been carried out in naturalistic contexts, par-
ticularly those exploring children’s eyewitness memory using the paradigms
outlined above, to assess children’s confidence judgments (i.e., their memory
monitoring after retrieval). Confidence judgments are thought to reflect a substan-
tive sense of certainty that arises from the strength of the memory that is being
retrieved, and this sense of certainty has been interpreted as an indicator of mem-
ory accuracy (Ghetti et al. 2008; Roebers 2002).
284 9  The Development of Metamemory

Several studies focusing on eyewitness memory have used the paradigm out-
lined above to assess children’s and adults’ confidence in their ability to recall
information accurately. In this procedure, participants watch a brief video and
are questioned about the contents of the video several weeks later. Participants
are encouraged to try as hard as possible to give only correct answers. They are
also encouraged to use the “I don’t know” option whenever they are uncertain.
After answering each question, participants are asked to apply a Likert-type scale
to rate how confident they are that their answer is correct. Using this experimental
paradigm, Roebers (2002) found that the participants in her study (8- and 10-year-
old children and adults) tended to be overly optimistic regarding the accuracy of
their answers regardless of whether the questions were unbiased or misleading,
even though the degree of overestimation was smaller for adults than for children.
Overall, the children showed appropriate retrieval monitoring when questions
were asked in a neutral interview context (unbiased questions) but did not fare
well in a context involving misleading questions.
Subsequent studies have basically confirmed these findings. Roebers and
Howie (2003), using the same paradigm and the same age groups as Roebers
(2002), found that children and adults gave higher confidence judgments (CJs)
after correct than after incorrect answers. However, children’s retrieval monitoring
ability was limited to an unbiased question format. When asked misleading ques-
tions, children’s ability to differentiate was undermined, as reflected in equally
high-confidence judgments after correct and incorrect answers. Roebers et al.
(2007) and von der Linden et al. (2007; Study 2) replicated this pattern of results
using the same age groups and a similar paradigm. They also compared JOLs and
CJs for the three age groups and found that there was moderate consistency across
all groups. From the age of 8 years onwards, relatively appropriate monitoring
abilities could be found with both indicators, particularly for correct answers and
contexts involving unbiased questions (for confirming evidence, see also Ghetti
et al. 2002; von der Linden and Roebers 2006). Using a vocabulary learning task
as the experimental paradigm, Roderer and Roebers (2010) also found evidence
for improvements in uncertainty monitoring and CJs during the elementary school
years, with 9-year-olds clearly outperforming 7-year-olds.
Overall, research focusing primarily on the extent to which children’s CJs
discriminate between accurate and inaccurate responses has provided important
insight into children’s developing ability to monitor the quality of their memory
representations. However, accurate and inaccurate memories may differ on a num-
ber of dimensions, for instance, memory strength and ease of retrieval. Thus, we
do not yet know the degree to which children are capable of introspecting on spe-
cific memory features that may indicate that a potential response is likely to be
accurate. In a series of studies, Ghetti and colleagues (e.g., Ghetti and Alexander
2004; Ghetti et al. 2006, 2008, 2010) investigated children’s ability to monitor
their memories by examining confidence ratings associated with accurate mem-
ories whose strength was experimentally manipulated. For instance, Ghetti et al.
(2008) explored the ability to monitor memory strength and memory absence at
retrieval in 7- and 10-year-old children and adults. Memory strength was defined
The Development of Metamemory in Children and Adolescents 285

in terms of memory vividness. It was hypothesized that memories for enacted

actions would produce stronger traces and be experienced as more vivid than
memories for imagined actions. Moreover, memories for bizarre actions were
expected to be more vivid than memories for commonly enacted actions. As a
main result, Ghetti et al. (2008) found that by 7 years of age, children appear to
be relatively competent at monitoring differences in memory strength between
classes of memories in which the difference in strength is relatively great. That is,
the 7-year-olds successfully monitored differences in strength between memories
for enacted versus imagined actions. However, compared with 10-year-olds and
adults, 7-year-olds exhibited deficits when they attempted to monitor differences
in the strength of memories for different imagined actions as well as when they
attempted to monitor memory absence. The results of this study also indicate that
critical changes in the ability to rely on metacognitive cues in monitoring memory
absence occur between the 7th and 8th years of life. Taken together, these find-
ings as well as those of Roebers and colleagues all indicate that relatively accu-
rate retrieval monitoring is possible for 7- and 8-year-old children under favorable
conditions and that this ability improves considerably until the end of elementary
school. They also suggest that memory monitoring is not a unitary process and
that future studies should investigate developmental trends in its critical compo-
nents and consider developments in related domains.

The Relation Between Monitoring and Control Processes

in Children and Adolescents

It appears that the considerable developmental improvements in procedural met-

amemory observable in elementary schoolchildren are mainly due to an increas-
ingly better interplay between monitoring and self-regulatory activities. That is,
even though young children may be similarly capable of identifying memory prob-
lems than older ones, in most cases, only the older children will effectively regu-
late their behavior in order to overcome these problems.
An important reason to study metacognitive monitoring processes is that moni-
toring is supposed to play a central role in directing how people study. Numerous
studies of adult participants have shown that individuals use memory monitoring,
especially judgments of learning (JOLs), to decide which items to study and how
long to spend on them (e.g., Metcalfe 2002; Nelson et al. 1994; Nelson and Narens
1990; Son and Metcalfe 2000). However, little is know about how children use
monitoring to regulate their study time.
A classic paradigm appropriate for exploring this issue further refers to the
­allocation of study time. Research on study-time allocation observes how learners
deploy their attention and effort. As already noted by Brown et al. (1983), the ability
to selectively attend to relevant aspects of a memory task is a traditional index of a
learner’s understanding of the task. Developmental studies on the allocation of study
time have examined whether schoolchildren and adults were more likely to spend
286 9  The Development of Metamemory

more time on less well-learned material. A few studies examined whether schoolchil-
dren are more likely to spend more time on less well-learned material. For example,
Masur et al. (1973) asked 7-year-olds, 9-year-olds, and college students to learn a
list of pictures for free recall. After the first study trial, participants were instructed to
select half of the pictures for additional study. Whereas 9-year-olds and college stu-
dents tended to select items not recalled correctly on the first trial, 7-year-olds did not
seem to consider their first-trial performance in selecting items for additional process-
ing. Similar findings were reported by Bisanz et al. (1978) for a paired-associate task.
Bisanz and colleagues found that Grade-5 and college students were more likely than
Grade-1 or Grade-3 students to select items that they failed to learn on the first trial.
However, children in the Masur et al. (1973) study were forced to be selective. Thus,
we do not know how young children might behave in a spontaneous learning situation.
A study by Dufresne and Kobasigawa (1989) investigated how children of
different ages spontaneously allocated their study time. In this study, 6-, 8-, 10-,
and 12-year-old children were asked to study booklets containing either “easy”
(highly related) or “hard” (unrelated) paired-associate items until they were
sure they could remember all of the pairs perfectly. As a main result, Dufresne
and Kobasigawa (1989) reported an age-related improvement in the efficiency of
study-time allocation. That is, 10- and 12-year-olds spent more time studying the
hard items than they spent studying the easy items. However, 6- and 8-year-olds
spent about the same amount of time on hard pairs as they spent on easy pairs
(see Fig. 9.5, upper part). At the same time, young schoolchildren were more
optimistic about being ready for the test, although only a small number of par-
ticipants achieved perfect recall. Children’s subsequent answers to metacognitive
knowledge questions showed that even many of the 6-year-old children were able
to distinguish between hard and easy pairs, so they were aware of which materi-
als were easy or hard to learn. Thus, developmental differences were not so much
observed in the children’s metacognitive knowledge itself but in the efficient appli-
cation of such knowledge to self-regulation strategies. However, being able to gen-
erally distinguish between easy and difficult word pairs does not necessarily imply
that judgments about past performance on easy and difficult word pairs are cor-
rect. As shown by Bisanz et al. (1978), first graders showed less accurate postdic-
tions than older children and adults when asked to indicate which item pairs they
answered correctly and which they did not master. The young children obviously
experienced considerable difficulty in selecting the items that were incorrect (but
see Pressley and Ghatala 1990, for discrepant findings).
A subsequent study by Kobasigawa and Metcalf-Haggert (1993) indicated that
young children’s use of regulatory skills depends on the difficulty and complexity
of the memory task. First- and third-grade children were asked to learn the names
of familiar and unfamiliar objects until they were sure they could name all of the
items correctly. Both first- and third-grade children spontaneously spent more
time studying the unfamiliar items than they spent studying the familiar items.
According to Kobasigawa and Metcalf-Haggert (1993), differences in item diffi-
culty were particularly salient, which was probably necessary for young children
to adjust their use of study time. The assumption that performance on study-time
The Development of Metamemory in Children and Adolescents 287

(a) 14

12
Study time (in seconds)

10

difficult
6
easy
4

0
6 8 10 12
Age

(b) 5

4.5
Study time (in seconds)

3.5
difficult
easy
3

2.5

2
7 9
Age

Fig.  9.5  a Developmental trends in the allocation of study time. (Adapted from Dufresne and
Kobasigawa 1989). b Monitoring and self-control processes for children’s allocation of study
time. (Lockl and Schneider 2003)

apportionment tasks strongly depends on the difficulty of the task has also been
confirmed by studies using text materials. For example, Brown et al. (1978) dem-
onstrated that the sophisticated selection of text material for further study develops
somewhat later than the grade-school years.
One problem with the studies by Dufresne and Kobasigawa (1989) and
Kobasigawa and Metcalf-Haggert (1993) is that study times were only roughly esti-
mated. To overcome this problem, Lockl and Schneider (2002b) presented a series
of easy and hard paired-associate pictures via computer, which allowed for a more
288 9  The Development of Metamemory

precise measurement of study times. The main findings of the previous studies were
confirmed. Again, young schoolchildren spent about the same amount of time on easy
pairs as they spent on hard pairs, whereas older schoolchildren devoted more time to
studying the hard items than the easy ones (see Fig. 9.5, lower part). In a follow-up
study, Lockl and Schneider (2004) explored whether 7- and 9-year-olds would benefit
from incentive conditions (5 cents for every correct answer) or accuracy instructions
that emphasized the importance of mastering each pair before going on to the next.
Contrary to expectations, the incentives did not affect study times in either age group.
Only the 9-year-olds studied for significantly longer time periods when accuracy
was emphasized. Thus, the findings of this study underline the developmental trends
obtained in previous studies, with young children demonstrating considerable difficul-
ties in regulating their study time (Bisanz et al. 1978; Masur et al. 1973).
Although there is evidence of clear increases in self-regulation skills from mid-
dle childhood to adolescence, the existing database does not provide us with any
detailed information about the relation between monitoring and control processes.
Accordingly, we do not know exactly whether children use the output of monitor-
ing processes to regulate their study time. For instance, Dufresne and Kobasigawa
(1989) demonstrated that many of the younger children were able to discriminate
between hard and easy pairs (see also Lockl and Schneider 2002b). However, in
this study, different materials were used for the metacognitive knowledge ques-
tions as well as for measuring study times. Thus, it remains unclear whether a par-
ticular item that is judged as difficult to recall will be studied for a longer time
than an item that is judged as rather easy to recall.
To investigate the relation between monitoring processes and self-regulation
processes more analytically, Lockl and Schneider (2003) asked 7- and 9-year-
old children to study easy (highly related) and difficult (unrelated) paired-asso-
ciate pictures. After a first learning phase with a fixed presentation time (3 s),
judgments-of-learning (JOL) was assessed on a 5-point scale. Subsequently, the
same pairs were presented again for self-paced study. As a first result, both 7- and
9-year-old children were able to differentiate between easy and difficult pairs.
Thus, the age groups did not differ in their mean level of JOLs, and there was no
Age x Difficulty interaction. When JOLs were made, children of both age groups
seemed to consider the degree of associative relatedness between the members
of the pairs. Put differently, item difficulty as an intrinsic cue affected children’s
JOLs regardless of age (see also Koriat and Shitzer-Reichert 2002).
To examine whether children used monitoring output to guide their learning,
Gamma correlations between JOLs and subsequent study time were computed.
The resulting mean Gamma correlations were G  =  −0.22 for 7-year-olds and
G = −0.40 for 9-year-olds, and both values as well as the difference between the
two values were significantly different from zero. Thus, even though children of
both age groups studied item pairs with lower JOLs for a longer period of time
than item pairs with higher JOLs, the relation between monitoring and control was
significantly stronger for 9-year-olds than for 7-year-olds. Accordingly, 9-year-
olds’ regulation of their study time was more in accordance with their preceding
JOLs than 7-year-olds’ regulation was.
The Development of Metamemory in Children and Adolescents 289

To examine the impact of self-regulation on recall performance, subgroups

were formed within each age group including children with either high or low self-
regulation. It was found that recall performance was significantly better for the
children with high self-regulation than for those with low self-regulation regard-
less of age. Overall, the data indicated that the ability to translate monitoring
into adequate self-regulation strategies reliably improved with age. Furthermore,
the quality of self-regulation significantly contributed to the learning outcome.
However, because JOLs in this study were generally high and the task was rela-
tively easy, further research including different and perhaps more complex learn-
ing materials is needed to analyze the relation between monitoring and control in
children and adolescents in more detail.
Interestingly, similar patterns of results were found in studies including older
adults (e.g., Dunlosky and Connor 1997; Murphy et al. 1987). For instance,
Dunlosky and Connor (1997) found that although older adults (mean age = 67 years)
and younger adults (mean age = 22 years) showed equivalent monitoring accuracy,
the correlations between JOLs or recall on one trial and study times on the next trial
were lower for older than for younger adults. The results of this study indicate that
older adults might not optimize their use of their monitoring output to control how
much they study. Furthermore, the study revealed that age-related differences in
people’s allocation of study time reliably mediated age-related differences in recall
performance (see also Hertzog 2002). Thus, the findings of this study are similar to
those obtained with younger children. However, the observation that both younger
children and older adults seem to experience difficulties translating monitoring
into adequate self-regulation does not imply that the causes are similar in the two
age groups. In fact, the causes could be quite different. For example, Dunlosky and
Connor (1997) discussed the possibility that the older participants in their study may
have tried to compensate for possible declines in their memory. That is, older adults
may have spent too much time studying already well-known items, thereby neglect-
ing less well-known items.
The relation between monitoring accuracy and strategic self-regulation as well
as its development has also been investigated in studies on children’s eyewit-
ness memory. As already mentioned above, research on eyewitness memory has
focused on memory accuracy (i.e., quality of recall) and not so much on recall
quantity (Koriat and Goldsmith 1996). Whereas pronounced age differences are
generally found for recall quantity, the developmental literature on age differences
in quality of recall indicates that these differences are comparably small (Cassel
et al. 1996; Peterson and Whalen 2001). Nonetheless, age-related differences in
recall accuracy have been found repeatedly, particularly when participants were
forced to give an answer or when they were asked misleading questions.
In the context of children’s eyewitness memory, individuals have to decide
which pieces of information to provide (and to whom) and which pieces to with-
hold. In contexts in which memory accuracy is emphasized, efficient control pro-
cesses will consist of skipping uncertain answers by responding with “I don’t
know” so as to avoid making incorrect responses and to maintain a high level of
memory accuracy. In other words, an “I don’t know” response in such situations
290 9  The Development of Metamemory

mirrors efficient control processes. When children of different ages are questioned
about an experienced or observed event, age differences in the ability to provide
“I don’t know” answers are to be expected. For instance, Roebers and Schneider
(2000) found that 6-year-olds’ memory accuracy in such a situation was about
55 %, whereas the memory accuracy of 10-year-olds and adults was about 80 %.
The main reason for such differences is that younger children see an obligation in
the social situation (elicited by the influence exerted by the adult interviewer) and
try to answer every question, even when they are very uncertain about the cor-
rect response (Schwarz and Roebers 2006). This is typically indicated by very low
percentages of “I don’t know” answers. Moreover, suggestive questions typically
result in lower numbers of “I don’t know” answers compared with unbiased ques-
tions (Cassel et al. 1996; Roebers and Schneider 2000; Shrimpton et al. 1998).
Several developmental studies have investigated whether it is possible to
increase memory accuracy in children by using incentives for correct recall. These
studies were originally stimulated by Koriat and Goldsmith’s (1996) finding that
adults who are rewarded for each correct answer on a list-learning task end up
with high accuracy scores. At the same time, recall quantity drops significantly.
The authors manipulated accuracy motivation by varying the bonus-to-penalty
ratio for correct and incorrect answers. The size of the quantity–quality trade-off
was found to depend on the response criteria or participants’ accuracy motivation.
When the criterion was raised to a maximum by taking all incentives away from
participants when only one incorrect answer was given, the amount of information
provided dropped to the floor level.
The few empirical developmental studies using incentives for correct recall and
looking at children’s ability to strategically regulate their memory performance
indicated that when the threshold to provide or withhold answers was manipulated
in extreme ways, children seemed to be able to increase their memory accuracy,
giving significantly more “I don’t know” answers. For instance, Roebers et al.
(2001) found that children as young as 6 years of age were able to withhold uncer-
tain answers and thus increased their recall accuracy when their motivation to be
accurate was high. This finding was also confirmed in a similar experimental study
by Koriat et al. (2001) with children who were 8 years old and older. In both stud-
ies, children won one token for every correct response and lost one token for every
incorrect response. The older children in the Koriat et al. (2001) study seemed to
be able to also adjust their strategic memory reporting to higher threshold lev-
els. That is, when the penalty for incorrect responding was increased, children’s
memory accuracy also increased significantly because they gave very few answers
altogether.
However, this pattern of findings was not replicated in Roebers and Fernandez’s
(2002) subsequent study, which differed from the previous studies in that it
not only used unbiased questions but also included misleading questions.
Furthermore, Roebers and Fernandez omitted the penalty for incorrect answers
and rewarded only correct answers. In this situation, children achieved signifi-
cantly higher memory accuracy scores by giving fewer incorrect and more correct
answers for both the unbiased and misleading questions than children in a typical
The Development of Metamemory in Children and Adolescents 291

80

70

60

50

40

30

20

10

0
6-year-olds 7-year-olds 8-year-olds Adults
Feedback + incentives Feedback only Incentives only Free report

Fig. 9.6  Mean percentages of correct answers to the misleading questions and recall accuracy as
a function of age and experimental condition in Study 2 (Roebers and Schneider 2005)

free-report condition. Contrary to the previous studies, the percentage of “I don’t

know” answers was not affected by the experimental manipulation. Roebers and
Fernandez concluded from their findings that not only can situational demands
affect control processes that build on monitoring but also they can influence the
effort that children put into their long-term memory searches.
In a later study, Roebers and Schneider (2005) used a similar design to test
the prediction that the combination of immediate feedback and incentives would
account for increases in children’s accuracy motivation. They compared the accuracy
of children’s memory reports for an observed event when answering unbiased and
misleading questions across three interview conditions: (a) the “usual” free-report
condition using accuracy instructions (including the emphasis on the “I don’t know”
option), (b) a “reminder” condition in which the suggestion to focus on accuracy
by considering the “don’t know” options was repeated several times throughout the
interview, and (c) an incentive interview condition that was based on Roebers and
Fernandez’s (2002) reward structure, which provided immediate feedback about the
correctness of answers. The findings for unbiased questions are given in Fig. 9.6. As
can be seen in this figure, there were significantly more correct answers in the incen-
tives plus feedback condition than in the other two conditions regardless of age.
Overall, the youngest children in the sample gave more incorrect answers
than the other three age groups. Interestingly, the percentages of “I don’t know”
answers ranged between 25 and 35 %, independent of age and condition. Repeated
reminders to focus on accuracy did not seem to have an effect given that there
were no significant differences between the free-report and reminder conditions.
A second experiment also revealed that feedback alone or incentives alone did not
yield positive effects as found in the incentives plus feedback condition, regardless
of age and question format (unbiased versus misleading). Thus, both immediate
292 9  The Development of Metamemory

feedback and incentives were necessary to improve memory accuracy. For chil-
dren to exhibit sophisticated accuracy-oriented memory behavior, they seem to
need a game context that motivates them to carefully monitor and control their
retrieval processes but simultaneously encourages them to provide useful answers
rather than to suppress them. We can conclude that it is not sufficient simply to
remind children, give feedback alone, or provide incentives alone.
Subsequent studies have explored whether these findings can be generalized to
educational contexts. Using a short educational film as the observed event, Krebs and
Roebers (2010) found that 8- to 12-year-olds showed relatively adequate memory
monitoring when distinguishing between easy, difficult, and unanswerable questions
even though there was a developmental increase in monitoring uncertainty. When
provided with the option of subsequent self-control (i.e., withdrawing answers), all
children were able to adjust their test-taking behavior to benefit their memory accu-
racy by withdrawing primarily incorrect answers. Taken together, the results suggest
that metacognitive skills seem to become increasingly accurate between the ages of
9 and 12 and that control strategies can be used in a flexible way by late childhood.
As demonstrated by Krebs and Roebers (2010), high achievers were more accurate in
their memory monitoring and self-regulation regardless of age. These findings thus
validate those obtained in laboratory-type conditions (see above) and indicate that the
latter can be generalized to more naturalistic school-related contexts.
The studies on the monitoring-control relation described so far all suggest that
monitoring affects subsequent self-regulation and thus supports a monitoring-
control (MC) model. One basic conclusion drawn from these developmental stud-
ies is that self-regulated learning behavior is based on monitoring processes. In
this sense, item difficulty is monitored before actually investing study time. The
outcome of monitoring processes then serves as a basis for subsequent study-time
allocation. However, recent work on adult metacognition indicates that sometimes
monitoring follows control operations and is based on feedback from them (see
Koriat 2008, for a review). For instance, Koriat et al. (2006) found evidence for the
control-monitoring (CM) model, indicating that in the case of self-paced learn-
ing, the allocation of study time (ST) to different items is generally data-driven;
that is, it is determined by the cues experienced during the learning process con-
cerning learning fluency or retrieval difficulty. Contrary to the developmental find-
ings regarding the allocation of study time described above, Koriat and colleagues
found that adults’ JOLs and recall decreased with increases in study time, thus
running counter to the common belief that study effort should enhance memory.
In fact, more recent developmental studies have demonstrated that the sequence
of monitoring and control can also be reversed. Koriat et al. (2009a, b) investi-
gated what they labeled the “data-driven function” of study time: According to this
approach, learners use self-paced study time as a basis for subsequent judgments of
learning (JOLs). Consistent with this view, in self-paced learning, older elementary
schoolchildren’s JOLs (but not younger one’s) made at the end of each study trial
decreased with the amount of time spent studying the item, suggesting that older
children’s JOLs are based on the memorizing-effort heuristic that easily learned
items are more likely to be remembered. Remarkably, recall decreased with ST in
The Development of Metamemory in Children and Adolescents 293

(a) 100
Below Above

80
Percentage Recall

60

40

20

0
1 2 3 4
Age group

(b) 100
Below Above

80
Percentage Recall

60

40

20

0
1 2 3 4
Age group

Fig. 9.7  Mean JOL (a) and mean percentage recall (b) for below-median and above-median ST
for each of the four age groups (Hoffmann-Biencourt et al. 2010, p. 9)

all age groups under study. This finding strengthens the validity of the memorizing-
effort heuristic. Furthermore, children from all grades showed evidence for accu-
rate monitoring accuracy, which is reflected by the positive relation between JOLs
and recall performance. In a follow-up study using a similar self-paced study-time-
allocation paradigm but different learning materials (picture pairs instead of word
pairs) and also a more comprehensive sample of children ranging from Grades 1
to 8, Hoffmann-Biencourt et al. (2010) were able to confirm the previous findings.
Overall, children’s JOLs decreased with increasing ST, suggesting that JOLs were
based on the memorizing-effort heuristic that easily learned items are more likely
to be remembered. As can be seen in Fig. 9.7, JOLs were higher for items with
294 9  The Development of Metamemory

below-median STs than for items with above-median STs. There was an ST × Age
interaction, which was due to the fact that the difference in JOLs between below-
median and above-median STs was more pronounced for older than for younger
students. These findings thus underline the importance of mnemonic cues in shaping
metacognitive feelings, not only in adults but also in (older) children.
Taken together, the research addressing the MC and CM models suggests
bidirectional links between monitoring and control during learning. As noted by
Koriat (2008), monitoring affects control when self-regulation is goal-driven;
that is, when the intention to reach a particular goal is motivated by incentives or
rewards. In such a situation, increased ST enhances JOLs. However, when self-
regulation is data-driven, JOLs are based on the feedback from control, and
therefore, JOLs decrease with ST under the heuristic that ease of encoding is diag-
nostic of successful recall. A recent study by Koriat et al. (2014) explored whether
children and adults are able to respond differentially to data-driven and goal-
driven variation in ST within the same paired-associate learning task. Samples of
fifth and sixth graders as well as ninth graders and college students were included
in this study. The youngest subsamples (i.e., the fifth- and sixth-grade children)
were chosen as the target population for the various experiments because children
in these grades not only exhibit evidence for data-driven regulation in self-paced
learning but also rely on the feedback they receive from this regulation as they
monitor their own knowledge while they study (Hoffmann-Biencourt et al. 2010;
Koriat et al. 2009a). At the same time, they also show evidence for goal-driven
regulation. The results of five experiments in this study indicate that fifth and
sixth graders can exhibit sensitivity to both goal-driven and data-driven variation
in ST but cannot respond simultaneously to both types of variation on the same
task. Ninth graders and college students, by contrast, are able to react differentially
to the two types of variation, demonstrating a positive ST–JOL relation for goal-
driven variation and a negative relation for data-driven variation. Whereas primary
schoolchildren behave as if the MC and CM models are mutually exclusive, ado-
lescents and adults are able to tune their JOLs differentially to differences in ST
according to the source (i.e., data-driven vs. goal-driven self-regulation). These
findings on the whole point to developmental changes in monitoring that occur
well beyond the primary school years, as indicated by the impressive nontrivial
achievement demonstrated by the ninth graders.

Metamemory–Memory Relations

From a developmental and educational perspective, the metamemory concept

appears to be appropriate for explaining children’s “production deficiencies” on
a broad variety of memory tasks. Early empirical research on metamemory was
stimulated by the belief that young children do not spontaneously use memory
strategies because they are not familiar with memory tasks and are unable to
judge the advantages of memory strategies such as rehearsal or categorization.
Metamemory–Memory Relations 295

Metamemory researchers assumed that this situation changes after children

enter school and are confronted with numerous memory tasks. Experience with
such tasks should improve strategy knowledge, which in turn should exert a
positive influence on memory behavior (e.g., strategy use). Thus, a major moti-
vation behind studying metamemory and its development was the assumption
that although links between metamemory and memory may be weak in early
childhood, they should become much stronger with increasing age (Flavell and
Wellman 1977; Weinert 1986).

Evidence from Correlational Studies

Given the theoretical assumptions about possible linkages between children’s met-
amemory and their strategy use, it is somewhat surprising that the correlational
evidence has been quite mixed (Schneider 1985, 1999; Schneider and Pressley
1997). Whereas a first series of investigations of the metamemory-memory link
yielded only weak support (see reviews by Brown et al. 1983; Cavanaugh and
Perlmutter 1982), subsequent analyses have resulted in more positive outcomes
(Schneider 1985; Schneider and Pressley 1997; Wellman 1983).
Overall, it appears that metamemory-strategy associations are generally weak
in preschoolers and increase with age (Joyner and Kurtz-Costes 1997; Schneider
2011). For instance, Lange, Guttentag, and Nida (1990) tested 4- to 7-year-old
children and did not find reliable relations between declarative metamemory, use
of organizational strategies, and recall in children of this age (for similar findings,
see Wimmer and Tornquist 1980). Although the findings reported by Sodian et al.
(1986) were somewhat more positive, there was generally little evidence of rela-
tions between metamemory, strategy use, and memory performance in sort-recall
tasks in children younger than 7. However, this does not mean that such relations
never exist for young children. For example, Schneider and Sodian (1988) used a
retrieval-cue (hide-and-seek) task and found rather strong relations between 4- and
6-year-olds’ knowledge of the utility of retrieval cues, their actual use of retrieval
cues, and their memory performance. That is, those young children who decided to
hide a policeman in a house that was cued by a police car or to hide a soccer player
in a house that had a soccer ball on it not only remembered more items but were
also able to answer metamemory questions correctly and to provide convincing rea-
sons for their strategic approach (see Justice 1989, for confirming evidence).
However, although it is possible to demonstrate links between young children’s
metamemory and performance on certain simple memory tasks that pertain to eve-
ryday life, it is very unlikely that such relations would occur on traditional memory
tasks such as the recall of categorizable lists (Cornoldi et al. 1991). As demonstrated
by Justice et al. (1997), the causal relation between strategy use and recall is not
yet understood by most preschool and first-grade children. In a first session, the
authors tested the prediction that preschoolers’ as well as first- and third-grade chil-
dren’s understanding of their own strategic behaviors would be related to strategy
296 9  The Development of Metamemory

use and recall. In a second session, children’s metamemory concerning the causal
relation between strategy use and recall level was investigated. As a first main result,
children who gave mentalistic explanations for their study behaviors in the first ses-
sion recalled more than those who gave nonmentalistic explanations. In other words,
children who reported that their recall was affected by a memory strategy performed
better than those who did not identify a strategy when they referred to their recall
performance. Moreover, analyses concerning children’s metamemory assessed in the
second session showed a developmental trend in that only third graders who were
aware of the strategy’s usefulness also employed it more effectively, resulting in
increased recall. It was only in this age group that children’s understanding of the
causal relation between strategy use and recall performance was significantly related
to an understanding of their own strategic behavior, even though a trend for such
a relation was also observed for the younger children. Overall, those children who
understood how the strategy worked had higher levels of recall than the children
who could not provide explanations for their strategic behavior regardless of age.
Justice et al. (1997) concluded from their findings that increases in capacity
and domain knowledge may permit the execution of strategic behavior but that an
understanding of the causal relation of such behaviors to desired outcomes may
be necessary for their effective use. Although these results are basically compa-
rable to Fabricius and Cavalier’s (1989) results, Fabricius and Cavalier found that
children with mentalistic explanations increased their strategy use on a subsequent
trial but did not show higher recall levels than nonmentalistic children. This dis-
crepancy may be accounted for by procedural differences.
More positive and reliable relations between metamemory about semantic-
organizational strategies, use of these strategies, and memory performances can
be observed in older children (Cavanaugh and Borkowski 1980; Hasselhorn 1992;
Justice 1985; Schneider 1986). However, it cannot be taken for granted that these
children clearly prefer semantic organization to alternative procedures such as
rehearsal. For instance, Cox and Paris (1979) found preferences for rehearsal in
fourth-grade children. Although Justice (1985) and Schneider (1986) noted that chil-
dren at this age level tended to favor semantic organization over rehearsal, nontrivial
proportions of the children actually used rehearsal while claiming that they preferred
semantic organization on such memory tasks. More positively, by Grade 6, students
are more likely to generally prefer semantic organization over rote rehearsal and also
to be able to justify their choice appropriately (e.g., Justice 1985). Similarly, robust
relations between strategy use and metamemory have been found for younger chil-
dren (fourth graders) in more recent longitudinal studies that provided the opportu-
nity to practice memory strategies several times (e.g., Kron-Sperl et al. 2008).
Comprehensive statistical meta-analyses examining the relation between strat-
egy knowledge and memory performance and involving more than 60 studies and
7,000 children and adolescents (Schneider 1985; Schneider and Pressley 1997) con-
firmed that correlations between these variables tended to be low for kindergarten
children and first and second graders but were quite robust and of moderate size from
Grades 3 to 4 and beyond. Findings indicated a robust relation between metamemory
and memory even though the associations were not extremely strong. The statistical
Metamemory–Memory Relations 297

Table 9.1  Overall correlations between metamemory and memory. Classified by kind of study

and school grade of subjects (Schneider and Pressley 1997, p. 220)
School grade
k 1/2 3/4 5/6 7+ Average
Memory monitoring 0.39 0.45 0.35 0.42 0.59 0.39
(Lab tasks) (5) (7) (10) (8) (2) (16)
Memory monitoring 0.24 – 0.28 0.49 0.41 0.44
(Text processing) (2) (3) (10) (4) (15)
Memory monitoring – 0.52 0.37 0.37 0.28 0.40
(Training studies) (4) (10) (10) (1) (13)
Organizational strategies 0.12 0.15 0.41 0.47 – 0.33
(Clustering) (1) (6) (14) (5) (43)
Organizational strategies – 0.39 0.32 – – 0.37
(Training studies) (10) (19) (36)
Note Number of studies providing data in parentheses

meta-analysis produced an average correlation of 0.41 (Schneider and Pressley 1997).

The size of the correlation seemed to depend on factors such as type of task, age of
children, task difficulty, and timing of metamemory assessment (before or after the
memory task). The details of this meta-analysis are given in Table 9.1.
As can be seen in Table 9.1, children’s metamemory makes a reliable and mod-
erately strong contribution to their strategic behavior and performance on a variety
of memory tasks. Obviously, the correlations for monitoring observed in labora-
tory tasks are greater than those for organizational strategies at the younger age
levels but not for older schoolchildren. Interestingly, whereas clear developmental
trends were found for the correlation between metamemory and strategy use, simi-
lar age trends for monitoring on laboratory tasks were not observed.
Such a conclusion was confirmed in numerous other studies exploring the
metamemory-memory association in elementary schoolchildren. In most cases,
such studies focused on rehearsal and organizational strategies (see reviews by
Borkowski et al. 1988; Joyner and Kurtz-Costes 1997; Schneider 2010; Waters
and Kunnmann 2010). A similar pattern was also reported in the few develop-
mental studies that explored the relation between metamemory and the use of
elaboration strategies. For instance, Beuhring and Kee (1987) examined the
relations between metamemory and the use of elaboration strategies in sam-
ples of fifth and 12th graders. Participants verbalized their strategies while
studying noun pairs; their cued-recall memory of those pairs was then tested.
Metamemory was assessed before or after this task. As a main result, Beuhring
and Kee (1987) found that the developmental increase in the use of elaboration
strategies could be attributed to an increase in metamemory, especially task-spe-
cific knowledge about the relative effectiveness of different associative memory
strategies. Taken together then, numerous correlational studies on the linkage
between metamemory and the use of various memory strategies clearly show that
the relation is significant and practically important.
298 9  The Development of Metamemory

Evidence from Training Studies

Further evidence supporting the metamemory-memory strategy link comes from

several short-term training studies (e.g., Cox et al. 1991; Ghatala et al. 1986;
Ringel and Springer 1980). These studies demonstrated that memory monitor-
ing is an important aspect of metamemory that could be related to both children’s
strategy use and strategy effectiveness. Memory monitoring can be described as a
bottom-up process that involves keeping track of one’s progress toward goals con-
cerning understanding and remembering (Schneider and Bjorklund 1998, 2003).
For instance, Ringel and Springer (1980) assessed first, third, and fifth graders’
spontaneous use of an organizational strategy in a first phase of their study. Next,
most participants were instructed in how to use and were allowed to practice an
appropriate semantically based sorting strategy, whereas the control group was not
instructed. The participants demonstrated increases in memory performance as a
result of this training. Immediately following the training period, some participants
were given explicit feedback regarding their improved performance on the task,
whereas others were not. Interestingly, fifth graders demonstrated successful trans-
fer of the organizational strategy in both conditions, whereas first graders did not
show any sign of transfer. However, third graders who received positive feedback
transferred the strategy to a new task, whereas those who did not receive feedback
failed to demonstrate any significant transfer. The authors’ explanation for this find-
ing was that the older children did not need the feedback because they were capa-
ble of assessing their own progress, but this was not true for the younger children.
Results of a metamemory interview confirmed the authors’ basic assumption that
only the older children were aware of the effectiveness of organizational strategies.
The main findings of this study are consistent with those of other research-
ers who demonstrated that transfer of strategy training is most effective when the
training includes a significant metamemory (self-testing) component (e.g., Ghatala
et al. 1986; Lodico et al. 1983; Melot 1998; Paris et al. 1982). For instance, in a
training program carried out by Ghatala et al. (1986), elementary schoolchildren
were presented with paired-associate learning tasks. Before studying these lists,
some children were trained to apply three components. They were taught (a) to
assess their performance with different types of strategies, (b) to attribute differ-
ences in performance to the use of different strategies, and (c) to use the informa-
tion from their assessments and attributions to help them select the best strategy
for a task. As a major result, it was shown that even children as young as 7–8 years
of age can be taught to monitor the relative efficacy of the strategies that they use
and to use the utility information gained from monitoring their strategies to make
future strategy selections.
All of these training studies showed that metacognitive information about the
value of being strategic increased the probability that children would learn the strat-
egy and use it later, demonstrating that strategy instruction alters metamemory.
However, these studies also illustrated that increasing metacognitive knowledge may
facilitate strategy use. For instance, Schlagmüller and Schneider (2002) reported that
Metamemory–Memory Relations 299

children who acquired an organizational strategy over the course of a microgenetic

longitudinal study showed increases in declarative metamemory well ahead of actu-
ally showing the strategy (see Grammer et al. 2011, for confirming evidence).
In general, the research evidence points to a bidirectional relation between met-
amemory and strategy use (Brown et al. 1983; Hasselhorn 1995; Schneider 1985).
Metamemory can influence memory behavior, which in turn can lead to enhanced
metamemory. It should be noted, however, that there are other factors in addi-
tion to metamemory, domain knowledge, and memory capacity that influence the
development of children’s memory strategies. These include intelligence and moti-
vation (see reviews by Bjorklund and Schneider 1996; Guttentag 1995; Pressley
and Hilden 2006; Roebers 2014).

Evidence from Multivariate Analyses

From the early 1980s on, multivariate experimental designs and also compre-
hensive field studies have been used to examine the complex relations among
metamemory, domain knowledge, memory behavior (strategy use), and mem-
ory performance as well as their relations with other important variables such as
intelligence, memory capacity, and motivation (e.g., see Borkowski et al. 1983;
DeMarie et al. 2004; Körkel and Schneider 1992; Kurtz et al. 1982; Schneider
et al. 1987; Schneider et al. 1998). These studies used multivariate statistical anal-
yses to explore the interactions among various variables in predicting memory per-
formance in sort-recall tasks. Thus, they were not limited to analyses of simple
intercorrelations but typically used causal modeling procedures (e.g., LISREL).
In one of the first of these investigations carried out as part of the Heidelberg
Study on Memory and Metacognition, Schneider et al. (1987) specified a causal
model that included IQ, motivation, metamemory, memory monitoring, and strat-
egy use as predictors of memory performance in a sort-recall task (see Schneider
2014, for more details on this study). Identical structural models were specified for
large samples of third and fifth graders. According to these causal models, IQ had
both direct and indirect effects on recall performance, influencing metamemory
and strategy use. This finding supports the view outlined by Cornoldi (2010) that
intelligence and metacognition are related and that metacognitive knowledge is
an important facet of human intelligence. In both samples, metamemory affected
strategy use, which in turn influenced memory performance. Memory monitoring
directly influenced recall for all children but had an additional impact on strat-
egy use only in the older sample. Findings also differed with regard to the impact
of motivation on recall performance such that this impact was negligible in the
younger sample but rather substantial for the fifth graders (the findings for the lat-
ter subgroup are provided in Fig. 9.8, upper part).
Overall, although significant mean differences between age groups were
observed for most of the variables included in the causal models, the structural
patterns were rather similar with the exception of the motivation construct. The
300 9  The Development of Metamemory

.40

Memory
monitoring

.18

.25 Memory
.75 performance
IQ Metamemory

.43
.30
.18
.22 Strategy
use

Success motivation
.43

Verbal
IQ

.29

.19
Metamemory Recall

.74

.27 .81
Memory
capacity Strategic
behavior

Fig.  9.8  Examples of causal modeling procedures illustrating metamemory–memory relation-

ships. a Causal model illustrating impacts of IQ, metamemory, and motivation on memory per-
formance in a sort-recall task. (Adapted from Schneider et al. 1987). b Causal model depicting
metamemory—memory relationships in a sort-recall task. (Adapted from Schneider et al. 1998)
Metamemory–Memory Relations 301

developmental trend found for the motivation variable indicates that individual
differences in intrinsic motivation become increasingly important with increas-
ing age, exerting a rather strong influence on memory performance from advanced
elementary school age on. When Flavell and Wellman (1977) introduced their tax-
onomy of metamemory, they already pointed out that one cannot always expect
to find a strong connection between memory knowledge and memory behavior
because individual differences in the learners’ motivation to carry out the task may
be a critical factor. Obviously, the findings outlined above support this assumption
(for similar findings in the area of text processing and recall, see Artelt et al. 2001;
van Kraayenoord and Schneider 1999).
In a subsequent study, Schneider et al. (1998) assessed relations between ver-
bal IQ, memory capacity, domain knowledge, declarative metamemory, use of a
semantic-organizational strategy, and recall in a sort-recall task in a sample of 155
third and fourth graders. As can be seen in Fig. 9.8 (lower part), metamemory was
affected by both verbal IQ and memory capacity. Although there was only a mod-
est direct contribution of metamemory to the prediction of recall, the indirect link
via strategic behavior was much stronger (about 0.60). As a consequence, indi-
vidual differences in declarative metamemory explained a large proportion of the
variance in the recall data. A somewhat different pattern of findings was reported
by Schneider et al. (1998) when the sort-recall task was based on soccer items and
when children’s knowledge of soccer was used as an additional predictor varia-
ble. In this case, soccer knowledge turned out to be the most powerful predictor,
explaining the lion’s share of the variance in children’s recall. However, meta-
memory still kept its indirect influence via strategy use (sorting) even though the
respective path coefficients were considerably lower than those obtained with the
first model. See Körkel and Schneider (1992) for similar findings using a memory
for text paradigm.
The importance of declarative metamemory for explaining individual differ-
ences in strategy use and memory performance was also illustrated by DeMarie
et al. (2004) who recruited different age groups ranging from kindergarteners to
fifth graders for their study. DeMarie and colleagues investigated how mental capac-
ity, knowledge base, and metamemory transact to produce changes in children’s
memory performance in 5- to 11-year-old children. They compared several causal
models. The model that provided the best fit to the developmental data represented
capacity and metamemory as interacting across age to produce patterns of strategic
memory performance. Also in this model, both strategy production and strategy effi-
ciency were represented as important factors for memory development. DeMarie
et al. (2004) also concluded from their findings that multiple strategy use must be
considered (see Chap. 7) and that metamemory plays a significant role in predict-
ing the use of multiple strategies and strategy effectiveness. Individual differences
in mental capacity seemed to be more important in influencing memory strategies in
young children (see also Woody-Dorning and Miller 2001), whereas individual dif-
ferences in metamemory seemed more important for older children.
Grammer et al. (2011) carried out a longitudinal study with 107 first-grade
children and followed them through the beginning of second grade. Children’s
302 9  The Development of Metamemory

organizational strategy use and metamemory were repeatedly assessed to examine

the development of these skills and their interrelations across time. Results from
latent growth curve models revealed that children’s metamemory at the beginning
of the study was linked to child- and home-level factors, whereas the develop-
ment of both strategies and metamemory was linked to maternal educational level.
Further modeling of the longitudinal relations between organizational strategies
and metacognitive knowledge showed that these links were best described by time-
lagged associations rather than concurrent ones. That is, the findings suggested
that the acquisition of metamemory precedes strategic sorting and influences
the amount of strategy use that children exhibit at subsequent time points. These
results not only confirm the reports of Schlagmüller and Schneider (2002) but also
allow for a more specific description of the process underlying the development of
memory strategies and metamemory and their interrelations across time.
Overall, the findings of various multivariate analyses described above con-
firm that the causal relation between metamemory and memory is complex.
Metamemory sometimes has an indirect effect on recall such as when knowl-
edge about categorization strategies leads to semantic grouping during the study
period, which in turn produces better recall (see Hasselhorn 1995; Weinert et al.
1988). Moreover, the influence seems to be bidirectional (see Flavell et al. 2002;
Schneider and Pressley 1997). That is, metamemory can influence memory
behavior, which in turn leads to enhanced metamemory. Finally, many other rel-
evant variables such as IQ, domain knowledge, memory capacity, and motivation
significantly contribute to the explanation of memory differences. Obviously,
these empirical findings are in accordance with the core assumptions of the good-
information processing model (Pressley et al. 1989; Borkowski et al. 1989),
which provides a detailed theoretical account of the complexity of these interac-
tional processes.

The Importance of Metacognition for Education

Several studies on the development of memory and metamemory have empha-

sized the importance of educational contexts for developmental changes, particu-
larly when the acquisition of memory strategies is in question. Most of memory
development is not so much a product of age but of education and practice. For
instance, in a recent longitudinal study on memory development from kindergarten
to the end of elementary school, Kron-Sperl et al. (2008) repeatedly presented the
children in their sample with a sort-recall task without giving any specific strategy
cues. When the performances of these children were compared with those of ran-
dom samples of schoolchildren of the same age who received this task for the first
time, substantial practice effects were found. Children in the longitudinal sample
not only outperformed the control children regarding strategy use and memory
performance but also showed better task-specific memory. Obviously, it does not
require much effort to improve children’s strategy knowledge.
The Importance of Metacognition for Education 303

There is no doubt that memory strategies can be effectively taught to young

children. For example, young children trained to use an active (cumulative)
rehearsal strategy substantially increased their levels of recall (Cox et al. 1989),
and even preschool children used organizational strategies and demonstrated
enhanced levels of recall when instructions emphasized the importance of group-
ing (Carr and Schneider 1991; Lange and Pierce 1992). The available intervention
studies clearly show that memory strategies can be easily acquired and transferred
to different learning contexts by elementary schoolchildren over the course of a
few training sessions (Carr et al. 1989; Schneider et al. 1986). Training effects
are particularly stable and impressive when children are allowed to compare their
performance when using the strategy to their previous performance without it and
when theoretical (metacognitive) explanations for the effects are provided.
During the 1980s and 1990s, numerous studies explored the efficiency of strat-
egy training approaches in school (for reviews, see Pressley and Hilden 2006;
Schneider and Pressley 1997). The basic assumption was that although children in
most cases do not efficiently monitor the effectiveness of the strategies they use,
they can be trained to do so (e.g., Ghatala et al. 1986; Paris and Oka 1986). As
described earlier in this chapter, all of these studies have shown that knowledge
about the usefulness of strategies can be easily and quickly acquired by elemen-
tary schoolchildren, in particular when information about why these strategies
work is provided (see also Waters and Kunnmann 2010).
There is now broad agreement that one way in which parents and teachers can
facilitate cognitive development is by developing children’s metacognition (see
Carr et al. 1989; Ornstein et al. 2008). Both classic and recent studies on this issue
have shown that there is still room for improvement in this regard but that con-
siderable progress has been made. During the past three decades or so, several
attempts have been made to apply metacognitive theory to educational settings
(cf. Moely et al. 1986, 1995; Palincsar 1986; Pressley 1995; Pressley et al. 1989).
Observational studies carried out by Moely and colleagues as well as by Ornstein
and colleagues reported that teachers vary widely in the extent to which they fos-
ter children’s metacognitive development. Individual differences in teachers’ mne-
monic style were found to have a long-lasting impact on children’s acquisition of
memory strategies. Moreover, individual differences in educational settings within
the same cultural/national context seem to influence strategy development.
As already mentioned in Chap. 7, recent research by Ornstein, Coffman,
and their colleagues (e.g., Coffman et al. 2008; Grammer et al. 2013; Ornstein
et al. 2010) focused on aspects of the classroom setting, investigating the rela-
tion between teachers’ memory-related instructions and children’s acquisition
of memory strategies. Ornstein and colleagues observed that first-grade teachers
tend to engage in “memory talk,” including strategy suggestions and metacogni-
tive questioning (see also Moely et al. 1992). First-grade children taught by “high-
mnemonic” teachers who used more of this type of memory talk benefitted more
from a memory-strategy training than those children with “low-mnemonic” teach-
ers (Coffman et al. 2008). Interestingly, the long-term impact of the first-grade
teachers was still observed in second grade when children were taught by other
304 9  The Development of Metamemory

teachers (Ornstein et al. 2010). Also, as noted by Grammer et al. (2013), first and
second graders exposed to memory-rich teaching exhibited greater levels of stra-
tegic knowledge and used strategies more efficiently on a memory task involv-
ing instructional content than did students exposed to low memory instructions.
Thus, this research suggests that “teacher talk” may be relevant for the emergence
of mnemonic skills and that there is a causal link between teachers’ language and
children’s strategy use.
One interesting and effective approach to teaching knowledge about strategies
was developed by Palincsar and Brown (1984). Here, teachers and students take
turns executing the reading strategies that are being taught, and instruction occurs
in true dialogue. Strategic processes are made very overt, with plenty of exposure
to the modeling of strategies and opportunities to practice these techniques over
the course of a number of lessons. The goal is for children to discover the util-
ity of reading strategies and for teachers to convey strategy-utility information as
well as information about when and where to use particular strategies. Teachers
using reciprocal instruction assume more responsibility for strategy implementa-
tion early in their instruction and gradually transfer that control over to the student
(see Palincsar 1986, for an extensive description of the implementation of recipro-
cal instruction; see Rosenshine and Meister 1994, for a realistic appraisal of its
benefits).
Another more large-scale approach concerns the implementation of compre-
hensive evaluation programs that are aimed at assessing the systematic instruc-
tion of metacognitive knowledge in schools. As emphasized by Holland Joyner
and Kurtz-Costes (1997), both Moely and Pressley and their colleagues have
conducted very ambitious programs of evaluating effective instruction in public
school systems. For instance, Pressley and colleagues found that effective teachers
regularly incorporated strategy instruction and metacognitive information about
effective strategy selection and modification as a part of their daily teaching. It
seems important to note that strategy instruction was not carried out in isolation
but was rather integrated into the curriculum and taught as part of language arts,
mathematics, science, and social studies. In accordance with the assumptions of
the good-information processing model outlined above (cf. Pressley et al. 1989;
Borkowski et al. 1989), effective teachers did not emphasize the use of single
strategies but taught the flexible use of a range of procedures that corresponded to
the subject matter, time constraints, and other task demands. On most occasions,
strategy instruction occurred in groups, with the teachers modeling appropri-
ate strategy use. By comparison, the work by Moely and colleagues (e.g., Moely
et al. 1995) illustrated that the effective teaching process described by Pressley
and coworkers does not necessarily constitute the rule and that effective teachers
may be the minority in elementary school classrooms. Taken together, the care-
ful documentation of instructional procedures carried out by Pressley, Moely, and
their research groups has shown that there is a lot of potential for metacognitively
guided instructional processes in children’s everyday learning.
Other researchers have focused on the relations between measures of metacogni-
tive knowledge and children’s school performance. For instance, Geary et al. (1989)
The Importance of Metacognition for Education 305

explored the relation between declarative metamemory and academic performance

in second and fourth graders. The sample included not only “normal” elementary
schoolchildren but also learning-disabled children. Not surprisingly, Geary et al.
(1989) found that fourth graders performed better than second graders on the meta-
memory battery and that metamemory–memory relations increased with age, even
though the link was moderate at best. Contrary to expectations, however, the perfor-
mance of children with learning disabilities did not differ from that of academically
normal children.
Although most early research on metacognition and performance in educational
contexts focused on memory and problem solving, numerous recent studies have
addressed the relations between metacognitive knowledge and school subjects
such as reading and mathematics. For instance, Roebers and colleagues (Roebers
et al. 2012, 2014) explored the impact of motivation, executive functioning, and
procedural metamemory (i.e., monitoring and self-control) on students’ school
achievement. For instance, Roebers et al. (2014) used structural equation mode-
ling procedures to test the prediction that elementary schoolchildren’s motivation
would predict their use of monitoring and control, which in turn would positively
affect test performance. Overall, individual differences in motivation did not con-
tribute much toward explaining the results. However, the models, which were
specified separately for 9-year-olds and 11-year-olds, clearly showed that mem-
ory monitoring was a substantial predictor of test performance in both samples,
whereas the impact of metacognitive control on test performance was generally
less impressive. Roebers et al. (2014) found a significant effect of this predictor
variable on test performance for the 11-year-olds but not for the younger children.
This finding indicates a developmental trend, suggesting that the structural rela-
tions between metacognitive processes and test performance were stronger for
11-year-olds than for 9-year-olds.
Using even younger children (first graders) as participants, Roebers et al.
(2012) explored relations among executive functioning, academic self-concept,
metacognitive monitoring, and control in the context of literacy and mathemat-
ics tasks. Children were about 7 years old at the beginning of the study and
were tested twice within a period of 1 year. Again, structural equation mode-
ling procedures were used to assess the impact of the various predictor variables
on achievement in mathematics and literacy. The findings of the causal model
explaining school achievement at the end of Grade 2 are given in Fig. 9.9. As
can be seen in this figure, individual differences in executive functioning pre-
dicted achievement differences in both mathematics and literacy. Whereas
metacognitive control had a substantial direct impact on literacy achievement,
the influence of monitoring was less impressive and only indirect (via control).
Individual differences in children’s self-concept did not affect performance
outcomes. Further longitudinal analyses indicated that executive functioning
assessed at the end of Grade 1 had a significant impact on metacognitive control
at the end of Grade 2, whereas early self-concept significantly affected monitor-
ing at the end of Grade 2. In sum, this study suggests significant links between
early executive functioning and subsequent metacognitive control in young
306 9  The Development of Metamemory

End of 2nd Grade

(8 years; 6 months)
Self Concept -.02

.32** .02
.23
.63

Executive .66**
Mathematics
Functioning

.51** .12
.49** .79
.24 .48**

Metacognitive Language Arts

.56**
Control
-.08
.46**
.09

Metacognitive -.15
Monitoring

Fig. 9.9  Structural equation model explaining mathematics and literacy achievement with exec-
utive functioning, self-concept, metacognitive monitoring, and control skills at the end of 2nd
grade (cross-sectional analyses), solid lines represent significant paths, dashed lines represent
nonsignificant paths (Roebers et al. 2012, p. 164)

schoolchildren in predicting school-related performance. Early metacognitive

skills appear to be more important in the context of literacy than in the context
of mathematics.
Evidence for the central role of metacognition during reading also stems from
numerous intervention studies. One core finding of the report from the National
Institute of Child Health and Human Development (NICHD; 2000) was that when
readers are given cognitive and metacognitive strategy training, they make sig-
nificant gains on measures of reading comprehension over students trained with
conventional instructional procedures (Pressley et al. 1989; Borkowski et al.
1989; Pressley et al. 2006; Rosenshine and Meister 1994; Rosenshine et al. 1996).
Thus, policy makers should be aware that, according to evidence from interven-
tion studies (National Institute of Child Health and Human Development, NICHD
2000; see also Pressley 2002), the assessment of metacognition can be applied to
improve reading literacy.
Other more recent research has explored the utility of the metacognition concept
in research with older children and adolescents by assessing the predictive potential
of metacognitive knowledge and skills in reading and math (e.g., Artelt et al. 2001;
Carr 2010; Mevarech and Kramarski 1997; Mevarech and Fridkin 2006; Schneider
The Importance of Metacognition for Education 307

and Artelt 2010; van der Stel and Veenman 2008, 2010; Veenman et al. 2005; see
also the contributions in Desoete and Veenman 2006). Overall, these studies con-
firm the view that metacognitive knowledge and the self-regulated insightful use
of learning strategies are not only influential in elementary schoolchildren but also
predict math performance and reading comprehension in secondary school settings,
even after differences in intellectual abilities have been taken into account. They
also provide evidence that metacognitive knowledge relevant for school-related
domains can still be effectively trained in late childhood and early adolescence.
Given the impressive evidence for the strong impact of metacognitive knowl-
edge on academic performance, it seems highly regrettable that current classroom
curricula still pay little attention to strategy development and the acquisition of
metacognition. The research outlined above clearly shows that elaborate strategy
instruction is not necessary for improving students’ metamemory. There is reason
to assume, however, that the situation will improve in the future because metacog-
nitive training programs that provide long-term strategy instruction and promise
long-lasting success are now available (cf. Pressley and Hilden 2006; Schneider
and Pressley 1997). One precondition for increasing the use of such metacogni-
tion-related teaching is to increase teachers’ understanding of the conceptual
foundations of effective learning. As long as teachers do not think in information
processing terms, it will be difficult to establish progress in this field. However,
recent changes in teacher education indicate that teachers will soon have a much
better understanding of information processing, enabling them to implement in
their classrooms strategy training programs that will pay off for most students.

Summary

Research on the development of metacognitive knowledge began more than four

decades ago, struggling with several fuzzy definitions of the concept, poor concep-
tualizations, and suboptimal measurement procedures. There was broad agreement
in the beginning that metamemory represents knowledge about and the regula-
tion of memory processes but not much more (Joyner and Kurtz-Costes 1997).
As explained in this chapter, remarkable progress has been made since the early
period. There is evidence that metamemory can be observed in young children and
that theory-of-mind development is an important prerequisite for the emergence of
metacognitive knowledge. Although research on metacognitive development was
quickly divided into two basic categories—declarative metacognitive knowledge
and procedural metacognitive knowledge—the field seems ready for an “integra-
tion” phase and appears concerned about the ramifications of how and why declar-
ative and procedural knowledge interact (Waters and Schneider 2010).
Recent reviews have shown that declarative and procedural metacognitive
knowledge seems to follow different developmental trajectories. On the one hand,
there is clear evidence that declarative metacognitive knowledge increases sub-
stantially from the early elementary school period on, with gains still observed
308 9  The Development of Metamemory

in adolescence and young adulthood. On the other hand, one of the major out-
comes of developmental studies on procedural metacognition concerns the lack
of clear-cut developmental trends in children’s and adolescents’ monitoring skills.
Although monitoring accuracy tends to improve over the school years, even pre-
schoolers show remarkable monitoring in learning situations they are familiar
with. By contrast, the available evidence on the development of self-regulation
skills demonstrates that there are clear increases from middle childhood to adoles-
cence. Effective self-regulation occurs only in highly constrained situations dur-
ing the elementary school years and continues well into adolescence. One of the
major developmental trends concerns the integration of monitoring outcomes and
self-regulation skills. Whereas young children may be able to accurately monitor
comprehension problems, they usually do not know how to solve them. In com-
parison, older schoolchildren and adolescents know how to benefit from monitor-
ing by implementing effective control strategies. One interesting new aspect of
research on the monitoring–self-regulation relation is that it cannot be taken for
granted that the monitoring of development always precedes self-regulation devel-
opment. Recent findings presented by Koriat et al. (2009a, b, 2013) clearly dem-
onstrate that self-regulation processes can affect monitoring in children and adults.
The links between metamemory, memory behavior (strategies), and memory
performance have also been frequently explored during the past three decades.
Overall, the findings point to nontrivial quantitative associations between meta-
memory and memory. Several multivariate studies have shown that metamemory
about strategies is an extremely important type of metacognition. For instance, it
has been repeatedly found that there are often stronger relations between meta-
memory about a strategy and strategy use than between metamemory and memory
because strategy use directed by metamemory is only one of several determinants
of performance (e.g., memory capacity and the nonstrategic knowledge base).
Nonetheless, metamemory has been a particularly useful construct when study-
ing children’s strategy use, and many investigators have incorporated metamemory
information into training programs designed to enhance children’s strategy use and
memory performance.
One of the most striking features of recent research is the increasing promi-
nence of metacognition in other areas, for instance, research on reading, math, and
science. In addition, the move into the classroom has forced the issue of transfer to
the forefront, and with that, more sophisticated discussions of the role of metacog-
nitive knowledge in promoting transfer in academic settings have ensued (Waters
and Schneider 2010).
Chapter 10
Conclusions and Future Directions

Most reviews of children’s memory development emphasize that memory devel-

opment has been one of the most-studied topics in all of cognitive development
(e.g., Bauer 2006; Schneider and Bjorklund 1998). As was documented in Chap. 2,
experimental research on memory development is as old as scientific psychology
as both began before the end of the nineteenth century. The early studies carried
out between 1885 and 1930 focused on individual differences in memory develop-
ment, long-term memory, and forgetting and were mostly applied. For instance,
there were studies that explored memory skills in school and focused on children’s
testimonial competence. With the exception of the comprehensive Austrian study
on the development of “general memory” (Brunswik et al. 1932), most of the
research on memory development conducted between 1930 and 1965 took place
in Russia. In particular, the outcome of this research improved our knowledge
about preschoolers’ memory activities, foreshadowing later investigations of chil-
dren’s implicit and explicit memory. Other major accomplishments of this period
included the analysis of developmental differences in children’s text processing
and recall.
Since the early 1960s, a truly impressive number of scientific studies have
been published on the topic of memory development. How does the modern era
of research on memory development differ from the historical approaches? One
of the crucial differences concerns a shift from describing development to iden-
tifying the underlying mechanisms of change. Another main difference concerns
the theoretical framework used: Since the mid-1960s, research on memory devel-
opment has been strongly influenced by theoretical models derived from infor-
mation processing and neuroscience approaches (Bauer 2007; Schneider 2011).
From this perspective, memory development can be seen as reflecting either
hardware (the capacity of memory systems and the speed with which informa-
tion can be processed) or software (e.g., the use of strategies). As noted by Miller
(2014), hundreds of studies carried out during the 1970s and 1980s investigated

© Springer International Publishing Switzerland 2015 309

W. Schneider, Memory Development from Early Childhood Through
Emerging Adulthood, DOI 10.1007/978-3-319-09611-7_10
310 10  Conclusions and Future Directions

age differences in the spontaneous production of memory strategies such as

rehearsal and clustering and the impact of task and instructional differences on
the outcomes. The development of strategic competence in children was probably
attractive to memory researchers because it seemed at that time that there were
qualitative differences in strategy use as a function of age. There is no doubt that
the “cognitive revolution” of the 1960s as well as Flavell’s pioneering work on
memory strategy development led to a dramatic increase in research on children’s
memory. As noted in Chap. 2, there are several ways to prove that interest in this
research paradigm is still growing. For instance, the recently published two-vol-
ume handbook on memory development edited by Bauer and Fivush (2014) con-
sists of 44 chapters reviewing different aspects of memory development, thus
indicating the need for and a perspective on future research in the area. Another
demonstration comes from an analysis of the PsycINFO abstracting and indexing
database in which Baker-Ward and Ornstein (2014) identified all peer-reviewed
English-language publications from between 1950 and 2010 that included the key-
words “memory” and “children” (0–12). Baker-Ward and Ornstein were able to
show that the number of publications assessed in 5-year intervals increased con-
siderably from the mid-1960s on, with another rapid increase in production after
1990. The number of topics addressed in these publications has also increased
tremendously. As already mentioned in the introduction to this book, it is impos-
sible to provide a representative summary of recent developments in all fields of
memory development. The review presented in this book was not intended to be
exhaustive. I decided to focus on verbal memory development, giving only limited
space to developmental trends in other information processing channels such as
visuospatial memory.
How can we summarize the progress that has been made in research on mem-
ory development during the past three or four decades? A short review of the
major findings described in the various chapters of this book will serve the pur-
pose of providing this kind of information in a condensed format.

Development of Implicit Memory

Developmental memory research carried out in the 1970s and 1980s was strongly
influenced by multistore memory models, which proposed that information can
be stored in two ways in the long-term memory store (LTS). Whereas explicit
memory refers to the capacity for conscious recollection, implicit memory repre-
sents a variety of nonconscious abilities. Modern memory researchers believe that
implicit or procedural (nondeclarative) memory refers to an evolutionarily older
memory system, which can be contrasted with the explicit declarative memory
system (Bjorklund and Sellers 2014). Implicit memory is “memory without aware-
ness,” demonstrating memory for some information without being consciously
aware that one is remembering (Jacoby 1991; Schacter 1992). The findings on
the development of implicit memory presented in Chap. 3 strongly suggest that
Development of Implicit Memory 311

implicit memory is largely distinct from explicit memory, operating through dif-
ferent processes in the brain. All studies comparing the development of implicit
and explicit memory have shown clear-cut differences, with considerable age
trends found on explicit but not on implicit memory tests. It seems that the brain
systems mediating perceptual and conceptual priming are fully developed early
in life, which is in clear contrast with the continuous development of the explicit
memory system (cf. Squire et al. 1993). There is substantial evidence in cogni-
tive neuroscience that perceptual and conceptual priming does not depend on the
medial temporal and diencephalic brain structures that mediate intentional declar-
ative memory. Although there are considerable differences in the developmental
patterns of implicit and explicit memory, indicating that the two memory systems
are relatively independent and distinct, they have also been shown to interact, with
priming experiences influencing explicit memory decisions, particularly in older
children and adults.
Whereas most studies on age differences in perceptual priming have by and
large provided evidence that this type of implicit memory is developmentally
invariant, the evidence regarding conceptual priming has been rather mixed (Lloyd
and Miller 2014). Some work on conceptual priming suggests that it is age invari-
ant, whereas other work suggests developmental changes. The reason for this
inconsistency in findings is not entirely clear. Task difficulty seems to play a role.
That is, age effects on conceptual priming tasks have been obtained only when the
tasks were particularly difficult. Obviously, more research is needed to explore this
issue further.
One important conclusion from more recent research on implicit memory is
that it constitutes a rather broad theoretical construct. Many different tasks (e.g.,
priming, conditioning, sequence learning) can be said to be implicit in that they
demonstrate influences of past experience on memory without a role for conscious
awareness. Implicit memory is thus an umbrella term, and the behavioral as well
as physiological profiles of different implicit memory measures are quite diverse.
One of the promises of recent neuroscience approaches is that their measures have
the potential to overcome the problems of many behavioral tools, allowing for
more rigorous tests of the developmental invariance hypothesis across a wide age
range and also offering the option to explore developmental patterns in subtypes of
implicit memory. There is no doubt that future research on implicit memory will
benefit from cognitive neuroscientific studies, which should be able to identify dif-
ferent developmental trajectories for different kinds of implicit memory.

Infancy and Early Childhood

The distinction between implicit and explicit memory outlined above helps to
explain why the assumption that very young infants do not remember anything has
been held so tenaciously and has survived for almost 100 years. Methodologically,
in the beginning, it was very difficult to elicit the behaviors that illustrate the
312 10  Conclusions and Future Directions

function of declarative as opposed to implicit memory from infants. As was shown

in Chap. 4, technologies such as the habituation–dishabituation or the preference-
for-novelty procedures as well as techniques adapted from operant-conditioning
paradigms certainly changed the situation. There is no doubt that the study of
memory development in infants and very young children has been very active and
productive since the 1970s. Research on infant memory conducted by pioneers
of the new era such as Bauer, Diamond, Fagan, Fantz, Mandler, Meltzoff, Rose,
Rovee-Collier, and their research groups produced fascinating results that illus-
trated amazing competencies in very young children. In fact, numerous methodo-
logical advancements have stimulated so many interesting research questions that
it is fair to state that infant memory has been one of the most active areas in the
field of developmental memory research in the last three decades.
Several studies focusing on infants’ short-term memory and using innova-
tive and creative experimental procedures have yielded important new findings.
This research suggests that infants have a limited-capacity system that they use
to temporarily maintain information, with capacity increasing quickly during the
first year of life. Whereas young infants (e.g., 6 months) can retain only one item
in short-term memory, older infants (e.g., 12 months) can already retain three or
four objects (Oakes and Luck 2014). Experimental findings also suggest that these
dramatic developmental changes in short-term memory are accompanied by the
fast development of neural systems. In particular, the development and maturation
of anterior (prefrontal) and posterior (temporal and parietal) brain systems is very
important for driving memory development in infancy. For instance, several EEG
studies have demonstrated that frontal lobe maturation contributes significantly
to individual differences in the development of short-term and working memory
around the end of the first year of life (see Bell and Morasch 2007). More research
is needed to further clarify the relations between brain and memory development.
In particular, there is a need for research to validate these new methodological
accomplishments and to provide better controls for possible influences on per-
formance including attention, interest, vulnerability to distraction, and aspects of
temperament (Reznick 2014).
Findings concerning infants’ long-term memory are similarly exciting, provid-
ing plenty of evidence of remarkable progress in the field. Although infants and
toddlers cannot tell us what they remember, the use of innovative techniques has
shown that young infants can readily form new and relatively permanent associa-
tions between objects and events that are physically present, activate memory rep-
resentations of objects and events and a physically present object or event, and
also activate memory representations of absent objects and events. Research based
on operant-conditioning techniques such as the mobile conjugate reinforcement
task as well as research based on deferred and elicited imitation techniques has
shown remarkable strides in recall across the first year of infancy and beyond.
Whereas the temporal extent of memory seems limited in the first 6 months of
life, 6-month-olds already exhibit evidence of recall for 24 h, and 9-month-olds
recall multistep sequences after a delay of 1 month. This research has also dem-
onstrated the important roles of encoding and memory consolidation processes for
Infancy and Early Childhood 313

the long-term retention of objects and events, related to developmental trends in

neural “hardware.” There is ample evidence that re-exposure to an event can facili-
tate long-term recall in infants and young children and that multiple reinstatements
can extend memories for very long time intervals, with the timing of a reminder
exposure also affecting recall. Reminders that are introduced later when a memory
is no longer active are more effective at enhancing the recall of older children (see
Hudson and Grysman 2014).
Recent research in cognitive neuroscience indicates that there is correspond-
ence between the time course of changes in memory behavior and performance
during the first year of life and what is known about the development of aspects
of the temporal cortical network that supports declarative memory (e.g., Bauer
2009; Richmond and Nelson 2008). Overall, older infants retain information about
objects, people, and events across longer intervals than younger infants do, and
older toddlers integrate more information across longer time periods than younger
toddlers, allowing them to use what they have learned after increasingly longer
delays (Howe et al. 2009; Hayne and Simcock 2009). As the neural structures
subserve memory mature, memories become more durable and less vulnerable
to decay. Interestingly, infant memory experts disagree about whether the clas-
sic experimental tasks used to assess long-term retention actually tap declarative
memory. Overall, the experimental findings related to this issue seem to indicate
that although the brain structures subserving implicit procedural memory may be
developed very early in life, explicit memory systems may be available much ear-
lier than generally assumed. The prolonged development of encoding, consolida-
tion of memory traces, retention, and retrieval processes significantly contributes
to age-related changes in declarative memory performance throughout infancy and
early childhood.

Event Memory and Autobiographical Memory

Research on memory in infants and toddlers has shown that these very young chil-
dren can remember information for several weeks and months. This implies that
long-term memory for episodes and events develops early in life. Although event
memory is explicit in that a person is aware that he or she is remembering, the
encoding of most events is unintentional, meaning that a person does not specifi-
cally try to remember the event when it is experienced. The term “event” is rather
broadly defined, and event memory can be conceived of as an amalgam of inciden-
tal and deliberate memory, with information usually encoded without the intent
to remember but with deliberate forces operating at the level of reconstructing
the experience (Ornstein et al. 2006). When such event memory is about personal
experiences, it has been described as autobiographical memory. As noted in
Chap. 5, although the mechanisms underlying event memory and autobiographical
memory appear to be rather similar, autobiographical memories differ from most
other event memories in that they are infused with a sense of personal involvement
314 10  Conclusions and Future Directions

in the event, implying that one has emotions, special thoughts, and reactions with
regard to these events (Bauer 2007).
Research on event and autobiographical memory has been very active during
the past three decades, and findings have improved our understanding of its devel-
opment in childhood and adolescence. Numerous studies have not only illustrated
that even young children can memorize salient events for a long time but have
also identified important sources of development and individual differences. For
instance, language development and maternal reminiscing style as well as indica-
tors of socioemotional development have been found to be highly relevant in this
respect. The hypothesis that children learn to narrate through dialogue experi-
ences has been confirmed in numerous studies (see reviews by Fivush 2014; Reese
2014). This impressive research program has also helped to evaluate traditional
perspectives and new hypotheses concerning the infantile amnesia phenomenon,
which has been discussed in the literature for more than 100 years; that is, the
fact that the events that one experiences during the first 2 or 3 years of life are no
longer accessible later on. Although we are still far from completely understanding
this phenomenon, there is reason to assume that several factors significantly con-
tribute to its end, such as the development of neural structures related to declara-
tive memory occurring during this period of time, the emergence of the cognitive
self, and the beginning of verbal mother–child dialogues about children’s experi-
ences (Bauer 2006; Howe et al. 2009).
Recent research on the development of autobiographical memory has also
enriched our understanding of the problems that young children have when they
are asked to testify in legal contexts. On the one hand, the findings of early and
recent eyewitness memory research confirm the view that age is positively cor-
related with the recall of events and that young children are more suggestible and
thus more vulnerable to the effects of misleading questions and repeated question-
ing. On the other hand, however, recent studies do not support the assumption held
by the early eyewitness memory researchers that young children are generally
unreliable witnesses. Although young children’s eyewitness memory is typically
less accurate than that of older children and adults, young children are capable
of providing accurate eyewitness memories about personally significant events in
their lives. Children’s memory for emotional stimuli seems to be influenced by a
variety of cognitive and individual factors and appears to be quite accurate even in
young children (Paz-Alonso et al. 2009). One merit of contemporary research on
this complicated issue is that it has identified conditions that increase the probabil-
ity of obtaining accurate memory reports from young children. One example is the
development of new and promising interview techniques that avoid the problems
of previous interrogations and provide conditions that support optimal recall, par-
ticularly in young children (Pipe and Salmon 2009; Goodman et al. 2014).
An interesting finding of recent research on eyewitness memory is that false
memories do not necessarily decrease with age but can actually increase as a
function of age under specific circumstances (i.e., the use of semantically related
materials). Dual-process theories such as fuzzy-trace theory provide a convincing
explanation for this developmental reversal, pointing to the relevance of processes
Event Memory and Autobiographical Memory 315

of recollection and familiarity and their developmental changes. As noted by

Ghetti and Lee (2014), the developmental cognitive neuroscience of recollection
and familiarity is just emerging, indicating that developmental changes in these
processes might occur across multiple cortical and subcortical regions. There is
reason to assume that future behavioral and neuropsychological research will fur-
ther improve our knowledge about the emotional, social, and cognitive factors that
influence children’s reports and event memories and our understanding of which
factors affect memory accuracy in childhood and adolescence.

Development of Memory Capacity and Working Memory

When Schneider and Pressley (1997) summarized the state of the art of research
on memory development at the end of the last century, they emphasized the fact
that the focus has shifted over the years from laboratory research on the impact
of memory capacity, strategies, and knowledge to more applied field research on
autobiographical and eyewitness memory. Regarding the development of basic
memory capacities and working memory, they concluded that there is a lack of
consensus in the domain concerning the developmental functions of these con-
structs. One conclusion from the findings described in Chap. 6 is that the situa-
tion has improved during the past two decades. On one hand, this is due to the
fact that theoretical models such as those developed by Baddeley, Case, Cowan,
and Pascual-Leone attracted a lot of attention and stimulated empirical research
to test basic assumptions. On the other hand, this research also demonstrated that
the concept of working memory is important for other domains. Meanwhile, most
researchers agree that the “total processing space” assumed to be invariant across
developmental stages by Case (1985) is in fact not invariant but increases stead-
ily with age (e.g., Cowan 2014; Halford and Andrews 2011; Pascual-Leone 2000).
Several factors such as knowledge and strategy use have been found to influence
performance on memory span tasks. If precautions are taken to prevent the use
of strategies such as rehearsal and chunking, the resulting memory spans vary
between two and four items and are thus much lower than the magic number 7
suggested by Miller (1956).
Numerous studies have demonstrated that developmental changes in informa-
tion processing speed affect developmental changes in memory capacity, suggest-
ing a rather global mechanism or a “cognitive primitive,” which has been assumed
to be domain-general by several researchers. Overall, however, the various evalu-
ations of working memory models do not confirm the view that a single set of
resources influences all aspects of working memory performance. Different devel-
opmental patterns for simple and complex span tasks were found, and there is also
plenty of evidence from behavioral and neuropsychological studies that verbal
working memory and visuospatial working memory show different developmen-
tal trends. Regarding the latter component, it appears that visual working mem-
ory develops faster than spatial working memory, indicating that the processing
316 10  Conclusions and Future Directions

of visual and spatial information may be maintained by separable subsystems of

the visuospatial sketchpad in Baddeley’s model and suggesting a fractionation of
these two subcomponents. Recent advances in working memory modeling show
that individual differences in both processing speed and memory trace decay
account for developmental differences on complex span tasks (Barrouillet et al.
2009). There is also evidence for the assumption that developmental improve-
ments in executive control and feature-binding abilities contribute to increases in
both verbal and nonverbal working memory. Apparently, different factors such
as brain growth, knowledge increases and strategy acquisition, developmental
changes in information processing speed, and rates of memory trace decay all con-
tribute to this developmental process. Overall, then, age differences in basic mem-
ory capacity and working memory are influenced by maturational and experiential
factors. Maturational factors place biological limits on how quickly children can
process information and retain items in their STS. However, speed of processing
is also influenced by experiential factors (e.g., knowledge base), making it clear
that the development of basic memory abilities is a result of the dynamic inter-
action between biological and experimental factors that vary over time. The fact
that working memory is central to all of cognition and important for the practical
understanding of developmental and educational disorders has resulted in numer-
ous studies that have investigated the specific role of working memory in different
educational settings with the goal of better understanding how to treat develop-
mental and educational disorders (Cowan and Alloway 2009).

Development of Memory Strategies

The modern era of memory development began with the study of strategies and
dominated the field well into the 1980s. The overview of the literature in Chap. 7
demonstrates that scientific interest in the development of memory strategies is
still alive, even though our view of the role of strategies in memory development
has changed somewhat since the early days. Strategies are still important, particu-
larly for people growing up in information-age societies (Bjorklund et al. 2009;
Ornstein and Light 2010; Roebers 2014). Much of the early research in strategy
development focused on factors responsible for production deficiencies, the sub-
sequent failure to transfer an acquired strategy to a new situation, and how to
improve children’s strategy effectiveness (see reviews by Pressley and Van Meter
1993; Schneider and Pressley 1997). Several studies showed that insufficient men-
tal capacity and a lack of domain-specific knowledge were partially responsible
for production deficiencies in young children and that these deficiencies are over-
come by most children after they gain experience with memory tasks in typical
school settings.
Research based on dual-task procedures in which children are asked to per-
form two tasks both separately and together has demonstrated that young children
Development of Memory Strategies 317

require more mental effort than older ones to implement and execute memory
strategies. It was also found that young children who try to employ memory strate-
gies for the first time are at risk of experiencing a utilization deficiency such that
their strategy use is not accompanied by superior memory performance. Moreover,
more recent research has indicated that individual differences in domain knowl-
edge contribute to age differences in strategy use (Bjorklund and Schneider 1996,
2003). Young children typically lack domain knowledge, which increases speed
when processing domain-related information. As a consequence, young children
require more mental energy to execute a strategy based on this information, mak-
ing strategy use very effortful and less attractive. Overall, the literature suggests
that domain knowledge, metamemory, motivation, and educational experiences
can be viewed as mediators of children’s strategy use and memory performance.
Recent research indicates that memory strategies do not develop in as straight-
forward a way as we once thought. Numerous studies have shown that memory
strategies develop most rapidly over the elementary school years and that older
children are more likely to use them than younger ones. However, findings from
longitudinal research indicate that strategies do not develop gradually but may
actually increase abruptly and in combination with other strategies. Thus, the ages
of strategy acquisition are relative and vary within and between strategies. Even
preschoolers and kindergarten children are able to use intentional strategies, both
in ecologically valid settings such as hide-and-seek tasks and in the context of a
laboratory task.
Longitudinal research conducted during the last two decades has convinc-
ingly shown that individual differences in the frequency of use and the quality of
children’s strategies play a large role in memory development between the early
school years and adolescence. Recent longitudinal findings concerning multiple
strategy use have provided evidence of considerable individual differences in the
intra-individual development of organizational and rehearsal strategies, indicat-
ing that the course of children’s strategy acquisition may be more complex and
variable than previously assumed. These new findings are helpful for understand-
ing and explaining the discrepancies in findings reported in developmental stud-
ies on the course of memory strategy acquisition. These new findings contribute
to our understanding of developmental changes in strategic memory across the
entire elementary school period by showing that the process of strategy acquisi-
tion is not one of simply replacing the ineffective with the effective, even though
the majority of children end up as competent strategy users when they complete
elementary school (Siegler 2006). Strategies do not always help performance, at
least not immediately, and simple and inefficient strategies reside alongside more
sophisticated and efficient ones. Finally, there is no doubt that memory strategies
can be effectively taught to young children. Differences in the educational context
have been shown to affect children’s strategy acquisition, suggesting that teachers’
focus on “memory talk” and teacher–child conversations in the classroom have
great relevance for the development of the repertoire of strategies that can be skill-
fully deployed in the service of goals of remembering (Coffman et al. 2008).
318 10  Conclusions and Future Directions

The Impact of the Knowledge Base

on Memory Development

Overall, research on the effects of domain knowledge conducted during the last
30 years has convincingly shown that any explanation of memory development
must reserve a large place for increases in children’s knowledge of specific content
(Schneider and Bjorklund 2003; Siegler 1998). Taken as a whole, the findings pre-
sented in Chap. 8 provide impressive evidence that a child’s knowledge base can
have a great effect on memory and performance, even though prior knowledge is
not always used automatically. What is also apparent is that declarative knowledge
is very complex and can be thought of as reflecting a number of levels ranging
from neural connections to complicated schemata—all of which are connected to
one another as evidenced by the flexibility of spreading activation. There are still
many challenges that remain in elucidating knowledge-base effects. For instance,
as the knowledge base develops, there are more concepts and stronger intercon-
nections among concepts, and thus, long-term memory should be more accessi-
ble with increasing age. With increased accessibility, both automatic and strategic
processes should be facilitated. Nonetheless, at all age levels, there are important
differences between people in the degree of accessibility to what is stored in long-
term memory (Brown et al. 1983).
The fact that there are individual differences in the accessibility of informa-
tion in the long-term store is one more indication that knowledge alone can never
provide a sufficient explanation of memory performance. The conscious and sys-
tematic search of long-term memory for possible knowledge that could mediate
memory is an important learning strategy (e.g., elaborative interrogation). The fact
that knowledge access often depends on strategy highlights the idea that it is wrong
to conceive of thinking as either knowledge or strategies. Thinking often involves
the articulation of knowledge and strategies, and both components are assumed to
interact in several everyday situations even though many of the effects of strategies
on memory performance were probably mediated directly by the knowledge base
with the conscious use of memory strategies playing only a minor role.
Domain knowledge is a powerful determinant of memory and learning. It
increases steadily from infancy to adulthood and contributes to the development
of other sources of memory competencies, such as basic capacities, strategies, and
metacognitive knowledge. For a long time, researchers believed that most develop-
mental improvements in memory reflect changes in the extent and accessibility of
the knowledge base. Studies using the expert–novice paradigm have highlighted
the importance of domain knowledge for exceptional performance in differ-
ent fields such as memory, music, or sport. Although the memory advantage of
experts was also demonstrated for strategic memory tasks, this advantage was not
mediated by more elaborated strategy use. Several studies have demonstrated that
expert children’s enhanced performance on strategic tasks was primarily medi-
ated by item-specific effects associated with their more elaborated knowledge base
rather than through their more effective use of strategies (Bjorklund et al. 2009).
The Impact of the Knowledge Base on Memory Development 319

The developmental evidence clearly supports the assumption that individual dif-
ferences in the amount of deliberate practice and motivation are key variables for
predicting individual differences in the level of expertise in a given domain (e.g.,
Ericsson 1996). However, the available evidence fails to rule out the possibility
that innate talent or aptitude plays a necessary role in high achievement (Hambrick
et al. 2013; Schneider 1997; Sternberg 1996; Winner 1996).
In recent years, the strong knowledge-base hypothesis has yielded to the idea
that cognitive development reflects multiple developments in basic capacities,
strategies, domain knowledge, and metamemory and the increasing articulation of
these components. Thus, research on the development of domain knowledge has
illustrated the fact that the sources of memory development interact in numerous
ways, and this sometimes makes it very difficult to disentangle the effects of spe-
cific sources from that of other influences. The importance of these interactions
was highlighted by the model of good-information processing (Pressley and
Borkowski et al. 1989; Pressley and Hilden 2006), which emphasizes the interplay
of intact neurology (basic capacities), strategic, knowledge base, and motivational
components in determining cognitive performance.

The Development of Metamemory

Research on the development of metacognitive knowledge began more than four

decades ago, struggling with several fuzzy definitions of the concept, poor concep-
tualizations, and suboptimal measurement procedures. In the beginning, there was
broad agreement that metamemory is knowledge about and the regulation of mem-
ory processes but not much more (Joyner and Kurtz-Costes 1997). As was shown
in Chap. 9, remarkable progress has been made since the early period. There is
evidence that both declarative and procedural metamemory can be observed in
young children and that theory-of-mind development is an important prerequisite
for the emergence of metacognitive knowledge. Although research on metacogni-
tive development was quickly divided into two basic groupings (i.e., declarative
and procedural metacognitive knowledge), the field seems ready for an “integra-
tion” phase and concerned about the ramifications of how and why declarative and
procedural knowledge interact.
Recent reviews have shown that declarative and procedural metacognitive
knowledge components follow different developmental trajectories (Schneider
2010). On the one hand, there is clear evidence that declarative metacognitive
knowledge increases substantially from the early elementary school period on, with
gains still observed in adolescence and young adulthood. On the other hand, one of
the major outcomes of developmental studies on procedural metacognition concerns
the lack of clear-cut developmental trends in children’s and adolescents’ monitor-
ing skills. Although monitoring accuracy tends to improve over the school years,
even preschoolers show remarkable monitoring in learning situations they are famil-
iar with. By contrast, the available evidence on the development of self-regulation
320 10  Conclusions and Future Directions

skills shows that there are clear increases from middle childhood to adolescence.
Effective self-regulation occurs only in highly constrained situations during the
elementary school years and continues well into adolescence. One of the major
developmental trends concerns the integration of monitoring outcomes and self-
regulation skills. Whereas young children may be able to monitor their compre­
hension problems accurately, they usually do not know how to address them. In
comparison, older schoolchildren and adolescents know how to benefit from
monitoring by implementing effective control strategies. One interesting new
aspect of research on the monitoring–self-regulation relation is that it cannot be
taken for granted that the development of monitoring always precedes the devel­
opment of self-regulation. Recent findings presented by Koriat et al. (2009a, b)
clearly demonstrate that self-regulation processes can affect monitoring in children
and adults.
The link between metamemory, memory behavior (strategies), and memory
performance has also been frequently explored during the past three decades.
Overall, the findings point to nontrivial quantitative associations between meta-
memory and memory. Several multivariate studies have shown that metamemory
about strategies is an extremely important type of metacognition. For instance,
studies have repeatedly found that there are often stronger relations between meta-
memory about a strategy and strategy use than between metamemory and memory
performance (e.g., recall) because strategy use directed by metamemory is only
one of several determinants of performance (e.g., memory capacity, the nonstra-
tegic knowledge base). Nonetheless, metamemory has been a particularly use-
ful construct when studying children’s strategy use, and many investigators have
incorporated metamemory information into training programs designed to enhance
children’s strategy use and memory performance. One of the most striking features
of recent research is the increasing prominence of metacognition in other areas,
for instance, research on reading, math, and science. In addition, the move into the
classroom has forced the issue of transfer to the fore, and, with that, more sophisti-
cated discussions of the role of metacognitive knowledge in promoting transfer in
academic settings have ensued (Waters and Schneider 2010).

Future Directions

Although memory development is a rather mature field, there is still much to learn,
and the centrality of memory in relation to all other aspects of cognition makes it
likely that the ontogeny of memory, in one form or another, will continue to be a
primary focus of research on cognitive development.
One area that has already attracted much interest during the past two decades
and certainly needs more exploration concerns the brain–memory relation. As
shown in the various chapters of this book, many cognitive developmentalists have
meanwhile recognized that the development of information processing cannot be
isolated from the development of the brain (Bauer 2009; Cycowicz 2000; Johnson
Future Directions 321

1998). In the field of memory development, there is an increasing emphasis on

the need for a greater awareness of and interaction with neuroscience perspectives.
However, we still know very little about the neuropsychology of memory develop-
ment and have to learn more at the levels of both description and explanation.
Neuropsychological research can influence memory development research in
a number of ways. One example concerns the developmental course of implicit
and explicit memory. For instance, De Haan et al. (2006) investigated memory
development following bilateral hippocampal damage early in life. Individuals
with such damage did not show the severe deficits in semantic memory that might
be expected if the hippocampus was crucial for both episodic and semantic mem-
ory. However, their performance on episodic memory tasks such as the delayed
recall of information was markedly impaired. One explanation for the relative
preservation of semantic memory following early hippocampal injury is that tran-
sient projections exist during normal development between the rhinal cortex and
hippocampus, and these projections are retained if the hippocampal lesions occur
before the time when these projections are usually retracted. This explanation thus
reflects the plasticity of developing memory systems. De Haan and colleagues
concluded from their research that the hippocampus and related medial temporal
lobe (MTL) structures contribute significantly to memory development in infancy
and childhood. They proposed that normal memory development involves a
sequence in which semantic-like memory emerges first, whereas episodic memory
develops later with progressive development of the hippocampus. Accordingly,
the development of memory abilities appears to unfold in a sequence, beginning
with novelty preferences and/or familiarity based recognition, followed by recall,
then by flexible memory, and ultimately by source memory. More research of this
kind is needed to enhance our knowledge about developmental changes in brain–
behavior relations (see de Chastelaine et al. 2007, for an interesting ERP study on
the development of source memory from childhood to adulthood; and Ofen et al.
2007, for the involvement of the prefrontal cortex and MTL in the development of
declarative memory).
Another line of research that already shows great promise (at least in the area
of implicit memory) concerns parallel memory studies with human infants and
age-matched nonhuman primates or the development of infant tasks that are
based on tasks that are already being used with human adults (cf. Nelson 1997).
Although it seems important for memory development researchers to become
familiar with neuropsychological techniques and research findings, it is also impor-
tant not to lose sight of the corresponding cognitive and behavioral levels while
examining brain functions. Accordingly, biological levels of explication should not
replace psychology-level theories but should rather be integrated with them to yield
a proper perspective on memory development (Bjorklund 1997). As noted by Bauer
(2009), the young field of the cognitive neuroscience of memory development has
already demonstrated its potential to move beyond description to explanation.
As can be seen from the various contributions to the recent handbook edited by
Bauer and Fivush (2014), the impact of emotion and stress on memory and its
development has attracted considerable attention lately. Although research with
322 10  Conclusions and Future Directions

adult samples still dominates the field, several studies with children carried out in
the laboratory or under naturalistic conditions have demonstrated the influence of
emotion and stress (both at encoding and at retrieval) on memory performance.
Overall, the findings seem to indicate that negative emotions and stress have a
negative impact on children’s retrieval of explicit memories. Thus, the effects of
emotions and stress on performance frequently found for adults are also evident
in children and can directly affect the neural structures implicated in memory. The
available evidence from the few child studies that have used neuroimaging and
neuropsychological approaches suggests that similar key brain regions and neural
mechanisms are recruited in children and adults during the successful encoding of
memory for emotional stimuli (Carver 2014; Hamann and Stevens 2014; Quas and
Klemfuss 2014). Overall, however, relatively little is understood about the impact
of emotion systems in the brain on children’s memory development. The available
evidence points to a relatively early-developing system, comprised of the amyg-
dala and other subcortical systems, which allows for fast responses to emotions
and acts by increasing arousal. Findings also seem to suggest that young children’s
memory is more affected by negative than neutral or positive emotion (Carver
2014). In particular, research on the effects of traumatic experiences in childhood
such as abuse, violent assaults, or accidents has shown that these negative events
tend to be remembered after extensive delays unless the events were experienced
during infancy or toddlerhood (Greenhoot and Sun 2014).
There is broad agreement that more research is needed to investigate how
emotion, stress, and trauma influence memory function and how we remember
emotional events. It is important to note that such developmental studies of mem-
ory–emotion interaction should measure the correlates of both behavior and brain.
Undoubtedly, a complete view of emotion–memory interactions in development
requires both perspectives. Practically, such research is relevant to eyewitness con-
texts, which often require children to recount highly emotional experiences, as
well as to clinical settings in which children are asked to discuss negative emo-
tions or are learning to cope with potentially arousing and emotionally challenging
situations. This research may also be relevant to educational settings, which often
require children to store and retrieve information in sometimes arousing contexts
(Quas and Klemfuss 2014).
Another promising line of research on memory development that focuses on
social factors rather than on biological/neuropsychological factors has already
been described above in detail. Research by Fivush, Haden, Nelson, Reese, and
their colleagues has convincingly illustrated that early memory skills develop in
the context of social interaction (Fivush 2014; Nelson 2014). This research has
illustrated the importance of sociocultural factors for early memory development,
indicating that differences in parents’ reminiscing styles (low elaborative versus
high elaborative) during children’s early years are associated with later differ-
ences in children’s ability to recall personally experienced events. Moreover,
recent research by Ornstein, Coffman, and their colleagues (Coffman et al. 2008;
Grammer et al. 2011; Ornstein et al. 2010) supplemented this work on the social
origins of memory skills by focusing on the classroom setting. They observed that
Future Directions 323

first grade teachers’ engagement in “memory talk,” including strategy sugges-

tions, and metacognitive questioning predicted children's strategy use. Thus, this
work suggests that “teacher talk” may be relevant for the emergence of mnemonic
skills. Finally, work by Güler et al. (2010) and Haden et al. (2001) indicates that
mother–child interactions during the solution of a deliberate memory task have
positive influences on children’s memory development. Thus, one interesting ave-
nue for future research would be to further specify the causal connection between
social and cognitive factors in the memory domain, for instance, by linking delib-
erate and event memory. Given the demonstrations of parent–child interactions
for enhancing children’s understanding and remembering of events, it seems
important to ask about the ways in which these social factors are related to the
emergence of memory skills in general (Haden et al. 2014). The importance of
this link could be explored via correlational longitudinal studies that also include
the impact of the preschool context and preschool teachers’ behaviors but could
also be done via intervention studies in which teachers or parents are trained in
the use of verbal and nonverbal behaviors that emphasize mnemonic skills. This
approach could be easily extended by considering cross-cultural differences in the
mnemonic orientation and mnemonic language of teachers (Larkina and Güler
2014). We know from previous cross-cultural research on memory development
that differences in socialization practices and parental behavior influence chil-
dren’s strategic behavior and acquisition of metacognitive knowledge (Schneider
and Pressley 1997). Ideally, correlational as well as intervention-based longitu-
dinal studies exploring links between parent–child interactions and preschoolers’
development of event memory and deliberate memory skills should also include a
cross-cultural component to explore the generalizability of findings. The same is
true for studies investigating the impact of teacher behaviors on schoolchildren’s
memory development.
Related to this, it would be interesting to explore the impact of changes in
the sociocultural environment such as computerization and other media changes
on the contents and organization of memory (Nelson 2014). Memory researchers
have the unique opportunity to explore the implications of media changes such
as the loss of the dominance of print media and the increasing popularity of the
internet on episodic and semantic memory. From a sociocultural developmental
perspective, it is important to know how episodic memory for everyday situations
is affected by individual differences in online socializing, gossiping, and internet
self-representation, all of which have replaced face-to-face meetings to a greater or
lesser extent across many ages and socioeconomic groups.
Moreover, research on memory–knowledge interactions could benefit from
an approach that explores the role of conversations and social interactions for
knowledge acquisition and transfer. As noted by Haden (2014), despite the sub-
stantial gains in our understanding of the links between knowledge and remem-
bering, we still do not exactly understand what it means to say that knowledge
and memory interact. We still know very little about how knowledge changes with
age and experience and about the mechanisms by which these changes operate to
affect memory. As Haden (2014) suggested, more within-person, longitudinal, and
324 10  Conclusions and Future Directions

microanalytic examinations of the process of knowledge acquisition and transfer

should be included in future research because they possess the potential to gener-
ate valuable information about how knowledge and memory interact. For instance,
such studies should carefully examine the role of parent–child conversation during
and after novel events and analyze how such conversations contribute to children’s
understanding and subsequent remembering. The first illustrations of the utility of
such an approach are already available (e.g., Bauer and San Souci 2010; Benjamin
et al. 2010).
Many suggestions outlined above emphasize the need for longitudinal stud-
ies. In their evaluation of the state of the art, Schneider and Pressley (1997)
concluded that perhaps the most frustrating aspect of the literature on memory
development available at that time was that very little was known about the mech-
anisms that account for the development of memory strategies and which memory
activities can be viewed as precursors of the later use of strategies. Schneider and
Pressley complained about the fact that no clear data were available on how early
nonstrategic knowledge and metamemory contribute to later development. They
emphasized the fact that these gaps in the literature cannot be filled by additional
cross-sectional research but can be tackled only by new longitudinal work. Their
point was that whether early person and environmental factors account for later
memory characteristics can be evaluated only when it is possible to describe how
particular individuals change and differ from one point in development to another.
So they asked for more longitudinal studies in the field, particularly for long-term
multivariate longitudinal designs that would allow for sophisticated statistical
analysis.
Almost 20 years later, a closer look at this issue shows that the situation has
changed considerably. The reviews and progress reports on research carried out in
different fields such as long-term memory in infants, event memory, and eyewit-
ness memory, but also in more traditional fields such as working memory, strategy
acquisition, as well as metamemory clearly indicate that the number of longi-
tudinal studies has increased substantially (see reviews by Bauer 2009; DeMarie
and Lopez 2014; Ornstein and Light 2010; Reese 2014; Roebers 2014; Schneider
2014). Altogether, this longitudinal research has provided valuable insights into
interindividual differences in intra-individual changes regarding numerous memory
domains, thus considerably broadening the perspective that was held around the
end of the last century. However, according to Ornstein and Haden (2001, 2009),
this does not necessarily mean that the focus of longitudinal studies on issues of
memory development has also yielded relevant information about the develop-
ment of memory. To ensure the latter, it is required that we (a) characterize the
memory skills of different age groups, (b) map age-related changes by tracing
the development of individual children longitudinally, and (c) search for media-
tors of developmental change, that is, for biological, cognitive, or social factors
that influence typical developmental trajectories. Although Ornstein and Haden
acknowledge that the past two decades have witnessed a substantial increase in our
knowledge of both the early mnemonic competencies of young children and age-
related differences in memory performance, they still see room for improvement
Future Directions 325

with regard to the mediator issue. They recommend the implementation of

longitudinal research that examines the development of multiple tasks that vary
in their information processing requirements. In my view, a promising new lon-
gitudinal research program should include identical sets of memory tasks for sev-
eral age cohorts in an effort to assess developmental changes for large portions
of the life span and to provide a standard metric for evaluating the performance
of children and adults on the same scale. Such comprehensive longitudinal studies
should be combined with microgenetic longitudinal studies and parallel experimen-
tal investigations in which hypothesized mediators of developmental change are
brought under experimental control (e.g., by using intervention designs). Although
a few recent longitudinal studies have included several of these components,
more research is needed to improve our knowledge regarding the development of
memory.
On a more general level, an interesting question is whether the study of mem-
ory needs to be situated in a number of broader conceptual and research contexts
or whether future research activities should focus on more specific memory func-
tions. The position that future research on memory development should be embed-
ded within the context of broader topics such as comprehension, knowledge,
context, and strategy was proposed by Kuhn (2000b) some time ago. The basic
claim is that although memory research has contributed significantly to our under-
standing in each of these four areas, the study of memory development in turn
needs to profit from what has been accomplished in each of those areas and to be
enriched by integration with each of them. Kuhn advocated for an approach that
could connect the studies on memories and memorizing to the four broader topics
described above and that would interpret the results within the frameworks pro-
vided by those topics.
However, there is also research indicating that the assumption of general
memory concepts may oversimplify the case. For instance, several studies have
concerned the issue of whether the concept of verbal memory represents a
domain-general skill or consists of domain-specific verbal abilities (for a review,
see Schneider and Bjorklund 1998). The way to assess this question empirically
is to administer different verbal memory tasks and assess the inter-task correla-
tions. Overall, both cross-sectional and longitudinal studies have shown that inter-
task correlations tend to be rather low regardless of the children’s ages and that the
low correlations are not due to reliability problems (Schneider and Weinert 1995).
Findings from the Munich Longitudinal Study indicate that relations among vari-
ous facets of verbal memory such as strategic memory, memory capacity, and text
recall are only moderate in early childhood and continue to be so until early adult-
hood (Schneider and Knopf et al. 2009). These findings are consistent with the
position that there is no “unitary” verbal memory construct and that deliberate
verbal memory may be better thought of as a set of specific abilities (“modules”)
rather than a domain-general concept.
Overall, although the field of memory development constitutes one of the old-
est and most active research areas in the field of cognitive development, there are
still many open issues that need to be clarified in future research. We still need
326 10  Conclusions and Future Directions

more precise cross-sectional experimental studies to help us better understand the

role of mediators of memory development. We also need more longitudinal studies
that should be combined, when possible, with parallel experimental studies that
bring hypothesized mediators of developmental change under experimental con-
trol. As emphasized by Ornstein and Haden (2009), it should be possible to study
both memory development and the development of memory by integrating these
methodological approaches.
References

Abravanel, E., & Gingold, H. (1985). Learning via observation during the second year of life.
Developmental Psychology, 21, 614–623.
Ach, N. (1905). Über die Willenstätigkeit und das Denken. Göttingen: Vandenhoeck & Ruprecht.
Ackerman, B. P. (1985). Children’s retrieval deficits. In C. J. Brainerd & M. Pressley (Eds.),
Basic processes in memory development (pp. 1–46). New York: Springer.
Ackerman, B. P. (1987). Descriptions: A model of nonstrategic memory development. In H. W.
Reese (Ed.), Advances in child development and behavior (Vol. 20, pp. 143–183). Orlando:
Academic Press.
Ackil, J. K., van Abbema, D. L., & Bauer, P. J. (2003). After the storm: Enduring differences
in mother-child recollections of traumatic and nontraumatic events. Journal of Experimental
Child Psychology, 84, 286–309.
Ackil, J. K., & Zaragoza, M. S. (1995). Developmental differences in eyewitness suggestibility
and memory for source. Journal of Experimental Child Psychology, 60, 57–83.
Adams, L. T., & Worden, P. E. (1986). Script development and memory organization in preschool
and elementary school children. Discourse Processes, 9, 149–166.
Akers, K. G., Martinez-Canabal, A., Restivo, L., Yiu, A., De Cristofaro, A., Hsiang, H.-L.,
et al. (2014). Hippocampal neurogenesis regulated forgetting during adulthood and infancy.
Science, 344, 598–602.
Alexander, J. M., Carr, M., & Schwanenflugel, P. J. (1995). Development of metacognition in
gifted children: Directions for future research. Developmental Review, 15, 1–37.
Alexander, J. M., & Schwanenflugel, P. J. (1994). Strategy regulation: The role of intelligence,
metacognitive attribution, and knowledge base. Developmental Psychology, 30, 709–723.
Alloway, T. P., & Alloway, R. G. (2013). Working memory in the lifespan: A cross-sectional
approach. Journal of Cognitive Psychology, 25, 84–93.
Alloway, T. P., Gathercole, S. E., & Pickering, S. J. (2006). Verbal and visuospatial short-term
and working memory in children: Are they separable? Child Development, 77, 1698–1716.
Alloway, T. P., Gathercole, S. E., Willis, C. D., & Adams, A. M. (2004). A structural analysis
of working memory and related cognitive skills in early childhood. Journal of Experimental
Child Psychology, 87, 85–106.
Anderson, J. R. (1976). Language, memory, and thought. Hillsdale: Erlbaum.
Anooshian, L. J. (1997). Distinctions between implicit and explicit memory: Significance for
understanding cognitive development. International Journal of Behavioral Development, 21,
453–478.
Artelt, C., Schiefele, U., & Schneider, W. (2001). Predictors of reading literacy. European
Journal of Psychology of Education, 16, 363–383.

© Springer International Publishing Switzerland 2015 327

W. Schneider, Memory Development from Early Childhood Through
Emerging Adulthood, DOI 10.1007/978-3-319-09611-7
328 References

Astington, J. W., & Jenkins, J. M. (1999). A longitudinal study of the relation between language
and theory-of-mind development. Developmental Psychology, 35, 1311–1320.
Astington, J. W., & Olson, D. R. (1990). Metacognitive and metalinguistic language: Learning to
talk about thought. Applied Psychology: An International Review, 39, 77–87.
Atkinson, R. C. (1975). Mnemotechnics in second-language learning. American Psychologist,
30(8), 821–828.
Atkinson, R. C., & Shiffrin, R. M. (1968). Human memory. A proposed system and its control
processes. In K. W. Spence & J. T. Spence (Eds.), The psychology of learning and motivation
(Vol. 2, pp. 90–197). New York: Academic Press.
Ausley, J. A., & Guttentag, R. E. (1993). Direct and indirect assessments of memory:
Implications for the study of memory development during childhood. In M. L. Howe & R.
Pasnak (Eds.), Emerging themes in cognitive development (Vol. 1, pp. 234–264)., Foundations
New York: Springer.
Bachevalier, J. (2014). The development of memory from a neurocognitive and comparative per-
spective. In P. J. Bauer & R. Fivush (Eds.), The Wiley handbook on the development of chil-
dren’s memory (Vol. I, pp. 109–125). Chichester: Wiley.
Bachevalier, J., & Mishkin, M. M. (1994). Effects of selective neonatal temporal lobe lesions on
visual recognition memory in rhesus monkeys. Journal of Neuroscience, 14, 2128–2139.
Baddeley, A. (1986). Working memory. Oxford: Clarendon Press.
Baddeley, A. (2000). The episodic buffer: A new component of working memory? Trends in
Cognitive Sciences, 4, 417–423.
Baddeley, A. (2003). Working memory and language: An overview. Journal of Communication
Disorders, 36, 189–208.
Baddeley, A. D. (2007). Working memory, thought, and action. Oxford: OUP.
Baddeley, A. D., & Hitch, G. J. (1974). Working memory. In G. Bower (Ed.), The psychology
of learning and motivation: Advances in research and theory (Vol. 8, pp. 47–89). New York:
Academic Press.
Baddeley, A. D., & Logie, R. H. (1999). Working memory: The multiple component model. In
A. Miyake & P. Shah (Eds.), Models of working memory: Mechanisms of active maintenance
and executive control (pp. 28–61). Cambridge: Cambridge University Press.
Baddeley, A., Rawlings, B., & Hayes, A. (2013). Constrained prose recall and the assessment of
long-term forgetting: The case of ageing and the Crimes Test. Memory. doi:10.1080/0965821
1.2013.865753.
Bahrick, H. P., Bahrick, P. O., & Wittlinger, R. P. (1975). Fifty years of memory for names and
faces: A cross-sectional approach. Journal of Experimental Psychology: General, 104, 54–75.
Bahrick, L. E., & Pickens, J. (1995). Infant memory for object motion across a period of three
months: Implications for a four-phase attention function. Journal of Experimental Child
Psychology, 59, 343–371.
Bailey, H., Dunlosky, J., & Hertzog, C. (2009). Does differential strategy use account for age-
related deficits in working memory performance? Psychology and Aging, 24, 82–92.
Baillargeon, R., deVos, J., & Graber, M. (1989). Location memory in 8-month-old infants in a
non-search AB task: Further evidence. Cognitive Development, 4, 345–367.
Baillargeon, R., & Graber, M. (1988). Evidence of location memory in 8-month-old infants in a
non-search AB task. Developmental Psychology, 24, 502–511.
Baker-Ward, L., Gordon, B. N., Ornstein, P. A., Larus, D. M., & Clubb, P. A. (1993). Young chil-
dren’s long-term retention of a pediatric examination. Child Development, 64, 1519–1533.
Baker-Ward, L., & Ornstein, P. A. (1988). Age differences in visual-spatial memory perfor-
mance: Do children really out-perform adults when playing Concentration? Bulletin of the
Psychonomic Society, 26, 331–332.
Baker-Ward, L., & Ornstein, P. A. (2014). The coaction of theory and methods in the study of the
development of memory. In P. J. Bauer & R. Fivush (Eds.), The Wiley handbook on the devel-
opment of children’s memory (Vol. I, pp. 41–64). Chichester: Wiley.
Baker-Ward, L., Ornstein, P. A., & Holden, D. J. (1984). The expression of memorization in early
childhood. Journal of Experimental Child Psychology, 37, 555–575.
References 329

Bannert, M., & Mengelkamp, C. (2008). Assessment of metacognitive skills by means of instruc-
tion to think aloud and reflect when prompted. Does the verbalization method affect learning?
Metacognition and Learning, 3(1), 17–37.
Barnes, K. A., Howard, J. H, Jr, Howard, D. V., Gilotty, L., Kenworthy, L., Gaillard, W. D.,
et al. (2008). Intact implicit learning of spatial context and temporal sequences in childhood
Autism Spectrum Disorders. Neuropsychology, 22, 563–570.
Barr, R., Dowden, A., & Hayne, H. (1996). Developmental changes in deferred imitation by 6- to
24-month-old infants. Infant Behavior and Development, 19, 159–170.
Barr, R., & Hayne, H. (2000). Age-related changes in imitation: Implications for memory devel-
opment. In C. Rovee-Collier, L. P. Lipsitt, & H. Hayne (Eds.), Progress in infancy research
(Vol. 1, pp. 21–67). Hillsdale: Erlbaum.
Barrouillet, P., Bernardin, S., & Camos, V. (2004). Time constraints and resource-sharing in
adults’ working memory spans. Journal of Experimental Psychology: General, 133, 83–100.
Barrouillet, P., & Camos, V. (2001). Developmental increase in working memory span: Resource
sharing or temporal decay? Journal of Memory and Language, 45, 1–20.
Barrouillet, P., & Camos, V. (2007). The time-based resource-sharing model of working memory.
In N. Osaka, R. H. Logie, & M. D’Esposito (Eds.), The cognitive neuroscience of working
memory (pp. 59–80). Oxford: Oxford University Press.
Barrouillet, P., Gavens, N., Vergauwe, E., Gaillard, V., & Camos, V. (2009). Working mem-
ory span development: A time-based resource-sharing model account. Developmental
Psychology, 45, 477–490.
Barry, E. S. (2007). Does conceptual implicit memory develop? The role of processing demands.
The Journal of Genetic Psychology, 169(1), 19–36. doi:10.3200/GNTP.168.1.19-36.
Bartlett, F. C. (1932). Remembering: A study in experimental and social psychology. Cambridge:
Cambridge University Press.
Bauer, P. J. (1995). Recalling the past: From infancy to early childhood. Annals of Child
Development, 11, 25–71.
Bauer, P. J. (2002). Early memory development. In U. Goswami (Ed.), Blackwell handbook of
childhood cognitive development (pp. 127–146). Oxford: Blackwell.
Bauer, P. J. (2005). Developments in declarative memory: Decreasing susceptibility to storage
failure over the second year of life. Psychological Science, 16, 41–47.
Bauer, P. J. (2006). Event memory. In W. Damon & R. M. Lerner (Eds.-in-chief), D. Kuhn & R.
Siegler (Vol. Eds.), Handbook of child psychology: Vol. 2. Cognition, perception, and lan-
guage (6th edn., pp. 373−425). Hoboken: Wiley.
Bauer, P. J. (2007). Remembering the times of our lives: Memory in infancy and beyond.
Mahwah: Erlbaum.
Bauer, P. J. (2009). The cognitive neuroscience of the development of memory. In M. L. Courage
& N. Cowan (Eds.), The development of memory in infancy and childhood (pp. 115–144).
Hove: Psychology Press.
Bauer, P. J. (2014). The development of forgetting: Childhood amnesia. In P. J. Bauer & R.
Fivush (Eds.), The Wiley handbook on the development of children’s memory (Vol. II, pp.
519–544). Chichester: Wiley.
Bauer, P. J., Burch, M. M., & Kleinknecht, E. E. (2002). Developments in early recall memory:
Normative trends and individual differences. Advances in Child Development and Behavior,
30, 103–152.
Bauer, P. J., Burch, M. M., van Abbema, D. L., & Ackil, J. K. (2007). Talking about twisters:
Relations between mothers’ and children’s contributions to conversation about a devastating
tornado. Journal of Cognition and Development, 8, 371–399.
Bauer, P. J., DeBoer, T., & Lukowski, A. F. (2007). In the language of multiple memory systems,
defining and describing developments in long-term explicit memory. In L. M. Oakes & P. J.
Bauer (Eds.), Short- and long-term memory in infancy and early childhood: Taking the first
steps towards remembering (pp. 240–270). New York: Oxford University Press.
Bauer, P. J., & Fivush, R. (1992). Constructing event representations: Building on a foundation of
variation and enabling relations. Cognitive Development, 7, 381–401.
330 References

Bauer, P. J., & Fivush, R. (Eds.). (2014). The Wiley handbook on the development of children’s
memory. Chichester: Wiley.
Bauer, P. J., Hertsgaard, L. A., & Dow, G. A. (1994). After 8 months have passed: Long-term
recall of events by 1- to 2-year-old children. Memory, 2, 353–382.
Bauer, P. J., Hertsgaard, L. A., Dropik, P., & Daly, B. P. (1998). When even arbitrary order
becomes important: Developments in reliable temporal sequencing of arbitrarily ordered
events. Memory, 6, 165–198.
Bauer, P. J., Kroupina, M. G., Schwade, J. A., Dropik, P. L., & Wewerka, S. S. (1998). If mem-
ory serves, will language? Later verbal accessibility of early memories. Development and
Psychopathology, 10, 655–679.
Bauer, P. J., Larkina, M., & Deocampo, J. (2011). Early memory development. In U. Goswami
(Ed.), The Blackwell-Wiley handbook of childhood cognitive development (2nd ed., pp. 153–
179). London: Wiley.
Bauer, P. J., & San Souci, P. (2010). Going beyond the facts: Young children extend knowledge
by integrating episodes. Journal of Experimental Child Psychology, 107, 452–465.
Bauer, P. J., & Shore, C. M. (1987). Making a memorable event: Effects of familiarity and orga-
nization on young children’s recall of action sequences. Cognitive Development, 2, 327–338.
Bauer, P. J., van Abbema, D. L., Wiebe, S. A., Cary, M. S., Phill, C., & Burch, M. M. (2004).
Props, not pictures, are worth a thousand words: Verbal accessibility of early memories under
different conditions of contextual support. Applied Cognitive Psychology, 18, 373–392.
Bauer, P. J., Wenner, J. A., Dropik, P. L., & Wewerka, S. S. (2000). Parameters of remembering
and forgetting in the transition from infancy to early childhood. Monographs of the Society
for Research in Child Development, 65(263), 4.
Bauer, P. J., Wenner, J. A., & Kroupina, M. G. (2002). Making the past present: Later verbal
expressed accessibility of early memories. Journal of Cognition and Development, 3, 21–47.
Bauer, P. J., & Wewerka, S. S. (1995). One- to two-year-olds’ recall of events: The more
expressed, the more impressed. Journal of Experimental Child Psychology, 59, 475–496.
Bauer, P. J., & Wewerka, S. S. (1997). Saying is revealing: Verbal expression of event memory in
the transition from infancy to early childhood. In P. van den Broek, P. J. Bauer, & T. Bourg
(Eds.), Developmental spans in event comprehension and representation: Bridging fictional
and actual events (pp. 139–168). Hillsdale: Erlbaum.
Bauer, P. J., Wiebe, S. A., Carver, L. J., Lukowski, A. F., Haight, J. C., Waters, J. M., et al.
(2006). Electrophysiological indices of encoding and behavioral indices of recall:
Examining relations and developmental change late in the first year of life. Developmental
Neuropsychology, 29, 293–320.
Bauer, P. J., Wiebe, S. A., Waters, J. M., & Bangston, S. K. (2001). Reexposure breeds recall:
Effects of experience on 9-month-olds’ ordered recall. Journal of Experimental Child
Psychology, 80, 174–200.
Bayley, N. (1949). Consistency and variability in the growth of intelligence from birth to eigh-
teen years. Journal of Genetic Psychology, 75, 165–196.
Bayliss, D. M., Jarrold, C., Baddeley, A. D., Gunn, D. M., & Leigh, E. (2005). Mapping the
development constraints on working memory span performance. Developmental Psychology,
41, 579–597.
Beal, C. R. (1985). Development of knowledge about the use of cues to aid prospective retrieval.
Child Development, 56, 631–642.
Beal, C. R., & Fleisig, W. E. (1987, March). Preschooler’s preparation for retrieval in object
relocation tasks. Paper presented at the biennal meeting of the Society for Research in Child
Development, Baltimore, MD.
Begg, I., & Anderson, M. C. (1976). Imagery and associative memory in children. Journal of
Experimental Child Psychology, 21, 480–489.
Bell, M. A. (2001). Brain electrical activity associated with cognitive processing during a looking
version of the A-not-B object permanence task. Infancy, 2, 311–330.
Bell, M. A., & Adams, S. E. (1999). Equivalent performance on looking and reaching versions of
the A-not-B task at 8 months of age. Infant Behavior & Development, 22, 221–235.
References 331

Bell, M. A., & Fox, N. A. (1992). The relations between frontal brain electrical activity and cog-
nitive development during infancy. Child Development, 63(5), 1142–1163.
Bell, M. A., & Morasch, K. C. (2007). Individual differences in the development of working
memory during infancy. In L. M. Oakes & P. J. Bauer (Eds.), Short- and long-term memory
in infancy and early childhood: Taking the first steps towards remembering (pp. 27–50). New
York: Oxford University Press.
Belli, R. F. (1998). The structure of autobiographical memory and the event history calendar:
Potential improvements in the quality of retrospective reports in surveys. Memory, 6, 383–406.
Belmont, J. M., & Borkowski, J. G. (1988). A group-administered test of children‘s metamem-
ory. Bulletin of the Psychonomic Society, 26, 206–208.
Belmont, J. M., & Butterfield, E. C. (1969). The relations of short-term memory to development
and intelligence. In L. P. Lipsitt & H. W. Reese (Eds.), Advances in child development and
behavior (Vol. 4, pp. 30–83). New York: Academic Press.
Belmont, J. M., & Butterfield, E. C. (1977). The instructional approach to developmental cogni-
tive research. In R. V. Kail & J. W. Hagen (Eds.), Perspectives on the development of memory
and cognition (pp. 437–481). Hillsdale: Erlbaum.
Benjamin, A. S., & Bjork, R. A. (1996). Retrieval fluency as a metacognitive index. In L. M.
Reder (Ed.), Implicit memory and metacognition (pp. 309–338). Hillsdale: Erlbaum.
Benjamin, N., Haden, C. A., & Wilkerson, E. (2010). Enhancing building, conversation, and
learning through caregiver-child interactions in a children’s museum. Developmental
Psychology, 46, 502–515.
Berntsen, D., & Rubin, D. C. (2006). Flashbulb memories and posttraumatic stress reactions
across the life span: Age-related effects of the German occupation of Denmark during World
War II. Psychology and Aging, 21, 127–139.
Best, D. L. (1993). Inducing children to generate mnemonic organizational strategies: An exami-
nation of long-term retention and materials. Developmental Psychology, 29, 324–336.
Best, D. L., & Ornstein, P. A. (1986). Children‘s generation and communication of mnemonic
organizational strategies. Developmental Psychology, 22, 845–853.
Best, J. R., & Miller, P. H. (2010). A developmental perspective on executive function. Child
Development, 81, 1641–1660.
Beuhring, T., & Kee, D. W. (1987). Developmental relationships among metamemory, elabora-
tive strategy use, and associative memory. Journal of Experimental Child Psychology, 44,
377–400.
Billingsley, R. L., Smith, M. L., & McAndrews, M. P. (2002). Developmental patterns in priming
and familiarity in explicit recollection. Journal of Experimental Child Psychology, 81, 251–277.
Binet, A. (1900). La Suggestibilité. Paris: Schleicher.
Binet, H., & Henri, V. (1894a). La memoire des mots. L’Annee Psychologique, 1, 1–23.
Binet, H., & Henri, V. (1894b). La memoire des phrases. L’Annee Psychologique, 1, 24–59.
Bird, A., & Reese, E. (2006). Emotional reminiscing and the development of an autobiographical
self. Developmental Psychology, 42, 613–626.
Bisanz, G. L., Vesonder, G. T., & Voss, J. F. (1978). Knowledge of one's own responding and the
relation of such knowledge to learning. Journal of Experimental Child Psychology, 25, 116–128.
Bjorklund, D. F. (1985). The role of conceptual knowledge in the development of organization in
children’s memory. In C. J. Brainerd & M. Pressley (Eds.), Basic processes in memory devel-
opment (pp. 103–142). New York: Springer.
Bjorklund, D. F. (1987). How age changes in knowledge base contribute to the development of
children’s memory: An interpretive review. Developmental Review, 7, 93–130.
Bjorklund, D. F. (1997). In search of a metatheory for cognitive development (or, Piaget is dead
and I don't feel so good myself). Child Development, 68, 144–148.
Bjorklund, D. F. (2005). Children’s thinking. Cognitive development and individual differences
(4th ed.). Belmont: Wadsworth.
Bjorklund, D. F. (2012). Children’s thinking. Cognitive development and individual differences
(5th ed.). Belmont: Wadsworth.
332 References

Bjorklund, D. F., & Bering, J. M. (2002). The evolved child applying evolutionary developmental
psychology to modern schooling. Learning and Individual Differences, 12, 1–27.
Bjorklund, D. F., & Bjorklund, B. R. (1985). Organization versus item effects of an elaborated
knowledge base on children‘s memory. Developmental Psychology, 21, 1120–1131.
Bjorklund, D. F., Bjorklund, B. R., Douglas Brown, R. D., & Cassel, W. S. (1998). Children’s
susceptibility to repeated questions: How misinformation changes children’s answers and
their minds. Applied Developmental Science, 2, 99–111.
Bjorklund, D. F., & Coyle, T. R. (1995). Utilization deficiencies in the development of memory
strategies. In F. E. Weinert & W. Schneider (Eds.), Memory performance and competencies:
Issues in growth and development (pp. 161–180). Hillsdale: Erlbaum.
Bjorklund, D. F., & DeMarchena, M. R. (1984). Developmental shifts in the basis of organization
in memory: The role of associative versus categorical relatedness in children’s free-recall.
Child Development, 55, 952–962.
Bjorklund, D. F., Dukes, Ch., & Brown, R. D. (2009). The development of memory strategies.
In M. L. Courage & N. Cowan (Eds.), The development of memory in infancy and childhood
(pp. 145–175). Hove: Psychology Press.
Bjorklund, D. F., & Harnishfeger, K. K. (1987). Developmental differences in the mental effort
requirements for the use of an organizational strategy in free recall. Journal of Experimental
Child Psychology, 44, 109–125.
Bjorklund, D. F., & Harnishfeger, K. K. (1990). Children’s strategies: Their definition and ori-
gins. In D. F. Bjorklund (Ed.), Children’s strategies: Contemporary views of cognitive devel-
opment (pp. 309–323). Hillsdale: Erlbaum.
Bjorklund, D. F., & Jacobs, J. W. (1985). Associative and categorical processes in children’s
memory: The role of automaticity in the development of organization in free recall. Journal
of Experimental Child Psychology, 39, 599–617.
Bjorklund, D. F., Miller, P. H., Coyle, T. R., & Slawinski, J. L. (1997). Instructing children to
use memory strategies: Evidence of utilization deficiencies in memory training studies.
Developmental Review, 17, 411–442.
Bjorklund, D. F., Muir-Broaddus, J. E., & Schneider, W. (1990). The role of knowledge in the
development of strategies. In D. F. Bjorklund (Ed.), Children’s strategies: Contemporary
views of cognitive development (pp. 93–128). Hillsdale: Erlbaum.
Bjorklund, D. F., & Schneider, W. (1996). The interaction of knowledge, aptitude, and strategies
in children's memory performance. In H. W. Reese (Ed.), Advances in child development and
behavior (Vol. 26, pp. 59–89). San Diego: Academic Press.
Bjorklund, D. F., & Sellers, P. D, I. I. (2014). Memory development in evolutionary perspec-
tive. In P. J. Bauer & R. Fivush (Eds.), The Wiley handbook on the development of children’s
memory (Vol. I, pp. 126–156). Chichester: Wiley.
Bjorklund, D. F., Thompson, B. E., & Ornstein, P. A. (1983). Developmental trends in children’s
typicality judgements. Behaviour Research Methods & Instruments, 15, 350–356.
Bjorklund, D. F., & Zeman, B. R. (1982). Children’s organization and metamemory awareness in
their recall of familiar information. Child Development, 53, 799–810.
Black, M. M., & Rollins, H. A. (1982). The effects of instructional variables on young children’s
organization and free recall. Journal of Experimental Child Psychology, 33, 1–19.
Blakemore, S.-J., & Choudhury, S. (2006). Development of the adolescent brain: Implications for
executive function and social cognition. Journal of Child Psychology and Psychiatry, 47, 296–312.
Blaxton, T. (1989). Investigating dissociations among memory measures: Support for a transfer-
appropriate processing framework. Journal of Experimental Psychology: Learning, Memory,
and Cognition, 15, 657–668. doi:10.1037/0278-7393.15.4.657.
Blumberg, F. C., & Torrenberg, M. (2005). The effects of spatial configuration on preschool-
ers’ attention strategies, selective attention, and incidental learning. Infant and Child
Development, 14, 243–258.
Boland, A. M., Haden, C. A., & Ornstein, P. A. (2003). Boosting children‘s memory by train-
ing mothers in the use of an elaborative conversational style as an event unfolds. Journal of
Cognition and Development, 4, 39–65.
References 333

Borkowski, J. G., Chan, L. K. S., & Muthukrishna, N. (2000). A process oriented model of meta-
cognition: Links between motivation and executive functioning. In G. Schraw & J. Impara
(Eds.), Buros Nebraska series on measurement and testing: Issues in the measurement of
metacognition (pp. 1–41). Lincoln: Buros Institute of Mental Measurements.
Borkowski, J. G., Levers, S., & Gruenenfelder, T. M. (1976). Transfer of mediational strategies in
children: The role of activity and awareness during strategy acquisition. Child Development,
47, 779–786.
Borkowski, J. G., Milstead, M., & Hale, C. (1988). Components of children’s metamemory:
Implications for strategy generalization. In F. E. Weinert & M. Perlmutter (Eds.), Memory
development: Universal changes and individual differences (pp. 73–100). Hillsdale: Erlbaum.
Borkowski, J. G., Peck, V. A., Reid, M. K., & Kurtz, B. E. (1983). Impulsivity and strategy trans-
fer: Metamemory as mediator. Child Development, 54, 459–473.
Bornstein, M. H. (1989). Stability in early mental development: From attention and information
processing in infancy to language and cognition in childhood. In M. H. Bornstein & N. A.
Krasnegor (Eds.), Stability and continuity in mental development: Behavioral and biological
perspectives (pp. 147–170). Hillsdale: Erlbaum.
Bourdon, B. (1894). Influence de l’age sur la memoire immediate. Revue Philosophique, 38,
25–39.
Bousfield, A. K., & Bousfield, W. A. (1966). Measurement of clustering and of sequential con-
stancies in repeated free recall. Psychological Reports, 19, 935–942.
Bousfield, W. A. (1953). The occurrence of clustering in the recall of randomly arranged associ-
ates. Journal of Genetic Psychology,49, 229–240.
Bousfield, W. A., Esterson, J., & Whitmarsh, G. A. (1958). A study of the developmental changes
in conceptual and perceptual associative clustering. The Journal of Genetic Psychology, 98,
95–102.
Brainerd, C. J., Holliday, R. E., & Reyna, V. (2004). Behavioral measurement of remembering
phenomenologies: So simple a child can do it. Child Development, 75, 505–522.
Brainerd, C. J., & Mojardin, A. H. (1999). Children’s and adults’ spontaneous false memo-
ries for sentences: Long-term persistence and mere-testing effects. Child Development, 69,
1361–1377.
Brainerd, C. J., & Pressley, M. (1985). Preface. In C. J. Brainerd & M. Pressley (Eds.), Basic
processes in memory development. Progress in cognitive development research (pp. vii–x).
New York: Springer.
Brainerd, C. J., & Reyna, V. F. (1998). Fuzzy-trace theory and children’s false memories. Journal
of Experimental Child Psychology, 71, 81–129.
Brainerd, C. J., & Reyna, V. F. (2002). Fuzzy-trace theory and false memory. Current Directions
of Psychological Science, 11, 164–169.
Brainerd, C. J., & Reyna, V. F. (2005). The science of false memory. Oxford: Oxford University
Press.
Brainerd, C. J., & Reyna, V. F. (2007). Explaining developmental reversals in false memory.
Psychological Science, 18, 442–448.
Brainerd, C. J., & Reyna, V. F. (2014). Dual processes in memory development: Fuzzy-trace the-
ory. In P. J. Bauer & R. Fivush (Eds.), The Wiley handbook on the development of children’s
memory (Vol. I, pp. 480–512). Chichester: Wiley.
Brainerd, C. J., Reyna, V. F., & Ceci, S. J. (2008). Developmental reversals in false memory: A
review of data and theory. Psychological Bulletin, 134, 343–382.
Brainerd, C. J., Reyna, V. F., Harnishfeger, K. K., & Howe, M. L. (1993). Is retrievability group-
ing good for recall? Journal of Experimental Psychology: General, 122, 249–268.
Brehmer, Y., Li, S.-C., Müller, V., von Oertzen, T., & Lindenberger, U. (2007). Memory plasticity
across the life-span: Uncovering children’s latent potential. Developmental Psychology, 43,
465–478.
Brown, A. L. (1978). Knowing when, where, and how to remember: A problem of metacogni-
tion. In R. Glaser (Ed.), Advances in instructional psychology (Vol. 1, pp. 77–165). Hillsdale:
Erlbaum.
334 References

Brown, A. (1987). Metacognition, executive control, self-control, and other mysterious mecha-
nisms. In F. Weinert & R. Kluwe (Eds.), Metacognition, motivation, and understanding (pp.
65–116). Hillsdale: Erlbaum.
Brown, A. L., Bransford, J. D., Ferrara, R. A., & Campione, J. C. (1983). Learning, remember-
ing, and understanding. In J. H. Flavell & E. M. Markman (Eds.), Handbook of child psychol-
ogy (Vol. III, pp. 77–166), Cognitive development. New York: Wiley.
Brown, A. L., & DeLoache, J. S. (1978). Skills, plans, and self-regulation. In R. S. Siegler (Ed.),
Children’s thinking: What develops? (pp. 3–36). Hillsdale: Erlbaum.
Brown, A. L., Smiley, S. S., Day, J. D., Townsend, M. A. R., & Lawton, S. C. (1977). Intrusion of
a thematic idea in children’s comprehension and retention of stories. Child Development, 48,
1454–1466.
Brown, A. L., Smiley, S. S., & Lawton, S. Q. C. (1978). The effects of experience on the selec-
tion of suitable retrieval cues for studying texts. Child Development, 49, 829–835.
Brown, J., Aczel, B., Jiménez, L., Kaufman, S. B., & Grant, K. P. (2010). Intact implicit learn-
ing in autism spectrum conditions. The Quarterly Journal of Experimental Psychology, 63(9),
1789–1812. doi:10.1080/17470210903536910.
Bruce, D., Dolan, A., & Phillips-Grant, K. (2000). On the transition from childhood amnesia to
the recall of personal memories. Psychological Science, 11, 360–364.
Bruck, M., & Ceci, S. J. (1999). The suggestibility of children`s memory. Annual Review of
Psychology, 50, 419–439.
Bruck, M., Ceci, S. J., & Principe, G. F. (2006). The child and the law. In K. A. Renninger & I.
E. Sigel (Vol. Eds.) & E. Damon & R. M. Lerner (Gen. Eds.), Handbook of child psychology:
Vol. 4. Child psychology in practice (6th ed., pp. 776–816). New York: Wiley.
Brunswik, E., Goldscheider, L., & Pilek, E. (1932). Untersuchungen zur Entwicklung des
Gedächtnisses [Studies on the development of memory]. In W. Stern & O. Lippmann (Eds.),
Beihefte zur Zeitschrift für angewandte Psychologie (Beiheft 64). Leipzig: Ambrosius Barth.
Bühler, K. (1918). Die geistige Entwicklung des Kindes. Jena: Gustav Fischer.
Bühler, K. (1930). Die Erinnerungen des Kindes [Memories of children]. In K. Bühler (Ed.), Die
geistige Entwicklung des Kindes (6th ed., pp. 318–329). Jena: Gustav Fischer.
Bull, R., & Espy, K. A. (2006). Working memory, executive functioning, and children's math-
ematics. In S. J. Pickering (Ed.), Working memory and education (pp. 93–123). Oxford:
Elsevier.
Burden, M. J., & Mitchell, D. B. (2005). Implicit memory development in school-aged chil-
dren with Attention Deficit Hyperactivity Disorder (ADHD): Conceptual priming deficit?
Developmental Neuropsychology, 28, 779–807. doi:10.1207/s15326942dn2803_3.
Burgwyn-Bailes, E., Baker-Ward, L., Gordon, B. N., & Ornstein, P. A. (2001). Children’s mem-
ory for emergency medical treatment after one year: The impact of individual difference vari-
ables on recall and suggestibility. Applied Cognitive Psychology, 15, S25–S48.
Burtis, P. J. (1982). Capacity increase and chunking in the development of short-term memory.
Journal of Experimental Child Psychology, 34, 387–413.
Butterfield, E., Nelson, T., & Peck, J. (1988). Developmental aspects of the feeling of knowing.
Developmental Psychology, 24, 654–663.
Camos, V., & Barrouillet, P. (2011a). Factors of working memory development: The time-based
resource-sharing approach. In P. Barrouillet & V. Gaillard (Eds.), Cognitive development and
working memory. A dialogue between Neo-Piagetian theories and cognitive approaches (pp.
151–176). Hove, UK: Psychology Press.
Camos, V., & Barrouillet, P. (2011b). Developmental change in working memory strategies:
From passive maintenance to active refreshing. Developmental Psychology, 47, 898–904.
Campbell, B. A., & Jaynes, J. (1966). Reinstatement. Psychological Review, 73, 478–480.
Campos, J. J., Anderson, D. I., Barbu-Roth, M. A., Hubbard, W. M., Hertenstein, M. J., &
Witherington, D. (2000). Travel broadens the mind. Infancy, 1, 149–219.
Campoy, G. (2008). The effect of word length in short-term memory: Is rehearsal necessary?
Quarterly Journal of Experimental Psychology, 61(5), 724–734.
References 335

Carey, S. (1985). Conceptual change in childhood. Cambridge: MIT Press.

Cariglia-Bull, T., & Pressley, M. (1990). Short-term memory differences between children pre-
dict imagery effects when sentences are read. Journal of Experimental Child Psychology, 49,
384–398.
Carr, M. (2010). The importance of metacognition for conceptual change and strategy use
in mathematics. In H. S. Waters & W. Schneider (Eds.), Metacognition, strategy use, and
instruction (pp. 176–197). New York: Guilford Press.
Carr, M., Kurtz, B. E., Schneider, W., Turner, L. A., & Borkowski, J. G. (1989). Strategy acqui-
sition and transfer: Environmental influences on metacognitive development. Developmental
Psychology, 25, 765–771.
Carr, M., & Schneider, W. (1991). Long-termi maintenance of organizational strategies in kinder-
garten children. Contemporary Educational Psychology, 16, 61–72.
Carver, L. J. (2014). Cognitive neuroscience of emotion and memory development. In P. J. Bauer
& R. Fivush (Eds.), The Wiley handbook on the development of children’s memory (Vol. II,
pp. 709–723). Chichester: Wiley.
Carver, L. J., & Bauer, P. J. (1999). When the event is more than the sum of its parts: Nine-
month-olds’ long-term ordered recall. Memory, 7, 147–174.
Carver, L. J., Bauer, P. J., & Nelson, C. A. (2000). Associations between infant brain activity and
recall memory. Developmental Science, 3, 234–246.
Case, R. (1972). Validation of a neo-Piagetian mental capacity construct. Journal of
Experimental Child Psychology, 14, 287–302.
Case, R. (1978). Intellectual development from birth to adulthood: A neo-Piagetian interpreta-
tion. In R. S. Siegler (Ed.), Children’s thinking: What develops? (pp. 37–72). Hillsdale:
Erlbaum.
Case, R. (1985). Intellectual development: A systematic reinterpretation. New York: Academic Press.
Case, R. (1992). The mind’s staircase: Exploring the conceptual underpinnings of children’s
thought and knowledge. Hillsdale: Erlbaum.
Case, R. (1995). Capacity-based explanations of working memory growth: A brief history and
reevaluation. In F. E. Weinert & W. Schneider (Eds.), Memory performance and competen-
cies: Issues of growth and development (pp. 23–44). Mahwah: Erlbaum.
Case, R., & Serlin, R. (1979). A new processing model for predicting performance on Pascual-
Leone’s test of m-space. Cognitive Psychology, 11, 308–326.
Case, R., Kurland, D. M., & Goldberg, J. (1982). Operational efficiency and the growth of short-
term memory span. Journal of Experimental Child Psychology, 33, 386–404.
Casey, B. J., Cohen, J. D., Jezzard, P., Turner, R., Noll, D. C., Trainor, R. J., et al. (1995).
Activation of prefrontal cortex in children during a nonspatial working memory task with
functional MRI. Neuroimage, 2, 221–229.
Cassel, W. S., & Bjorklund, D. F. (1995). Developmental patterns of eyewitness memory and
suggestibility: An ecologically based short-term longitudinal study. Law and Human
Behavior, 19, 507–532.
Cassel, W. S., Roebers, C. E. M., & Bjorklund, D. F. (1996). Developmental patterns of eye-
witness responses to increasingly suggestive questions. Journal of Experimental Child
Psychology, 61, 116–133.
Cavanaugh, J. C., & Borkowski, J. G. (1980). Searching for metamemory-memory connections:
A developmental study. Developmental Psychology, 16, 441–453.
Cavanaugh, J. C., & Perlmutter, M. (1982). Metamemory: A critical examination. Child
Development, 53, 11–28.
Ceci, S. J. (1980). A developmental study of multiple encoding and its relationship to age-related
changes in free recall. Child Development, 51, 892–895.
Ceci, S. J., & Bruck, M. (1993). Suggestibility of the child witness: A historical review and syn-
thesis. Psychological Bulletin, 113(3), 403–439.
Ceci, S. J., & Bruck, M. (1995). Jeopardy in the courtroom: A scientific analysis of children‘s
testimony. Washington: American Psychological Association.
336 References

Ceci, S. J., & Bruck, M. (1998). Children's testimony: Applied and basic issues. In W. Damon
(Ed.-in-Chief) & I. Sigel & K. A. Renninger (Vol. Eds.), Handbook of child psychology: Vol.
4. Child psychology in practice (5th edn., pp. 713–774). New York: Wiley.
Ceci, S. J., Bruck, M., & Battin, D. (2000). The suggestibility of children’s testimony. In D. F.
Bjorklund (Ed.), False-memory creation in children and adults: Theory, research, and impli-
cations (pp. 169–201). Mahwah: Erlbaum.
Ceci, S. J., Fitneva, S. A., & Williams, W. M. (2010). Representational constraints on the
development of memory and metamemory: A developmental-representational theory.
Psychological Review, 117, 464–495.
Ceci, S. J., & Howe, M. J. A. (1978). Age-related differences in free recall as a function of
retrieval flexibility. Journal of Experimental Child Psychology, 26, 432–442.
Ceci, S. J., Lea, S. E. G., & Howe, M. J. A. (1980). Structural analysis of memory traces in chil-
dren from 4 to 10 years of age. Developmental Psychology, 16, 203–212.
Ceci, S. J., & Liker, J. (1986). A day at the races: The study of IQ, expertise, and cognitive com-
plexity. Journal of Experimental Psychology: General, 115, 225–266.
Ceci, S. J., Loftus, E. F., Leichtman, M., & Bruck, M. (1994). The role of source misattribu-
tions in the creation of false beliefs among preschoolers. International Journal of Clinical
and Experimental Hypnosis, 62, 304–320.
Ceci, S. J., Papierno, P. D., & Kulkofsky, S. (2007). Representational constraints on children’s
suggestibility. Psychological Science, 18, 503–509.
Ceci, S. J., Ross, D. F., & Toglia, M. P. (1987). Suggestibility of children's memory: Psycholegal
implications. Journal of Experimental Psychology: General, 116, 38–49.
Chamberlain, A. H. (1915). Memory test with school children. Psychological Review, 22, 71–76.
Chase, W. G., & Ericsson, K. A. (1982). Skill and working memory. In G. H. Bower (Ed.), The
psychology of learning and motivation (Vol. 16, pp. 1–58). New York: Academic Press.
Chase, W. G., & Simon, H. A. (1973). Perception in chess. Cognitive Psychology, 4, 55–81.
Chechile, R., & Richman, C. L. (1982). The interaction of semantic memory with storage and
retrieval processes. Developmental Review, 2, 237–250.
Chen, C., & Stevenson, H. W. (1988). Cross-linguistic differences in digit span of preschool chil-
dren. Journal of Experimental Child Psychology, 46, 150–158.
Cherney, I. D. (2003). Young children’s spontaneous utterances of mental terms and the accuracy of
their memory behaviors: A different methodological approach. Child Development, 12, 89–105.
Chi, M. T. H. (1976). Short-term memory limitations in children: Capacity or processing deficits?
Memory & Cognition, 4, 559–572.
Chi, M. T. H. (1977). Age differences in memory span. Journal of Experimental Child
Psychology, 23, 266–281.
Chi, M. T. H. (1978). Knowledge structures and memory development. In R. Siegler (Ed.),
Children’s thinking: What develops? (pp. 73–96). Hillsdale: Erlbaum.
Chi, M. T. H., & Ceci, S. J. (1987). Content knowledge: Its role, representation, and restructuring
in memory development. In H. W. Reese (Ed.), Advances in child development and behavior
(Vol. 20, pp. 91–142). Orlando: Academic Press.
Chi, M. T. H., & Koeske, R. D. (1983). Network representation of a child’s dinosaur knowledge.
Developmental Psychology, 19, 29–39.
Clubb, P. A., Nida, R. E., Merritt, K., & Ornstein, P. A. (1993). Visiting the doctor: Children’s
knowledge and memory. Cognitive Development, 8, 361–372.
Coffman, J. L., Ornstein, P. A., & McCall, L. E. (2003, July). Linking teachers’ memory-rele-
vant talk to children’s memory performance. Poster presented at the biennial meeting of the
Society for Applied Research in Memory and Cognition, Aberdeen, Scotland.
Coffman, J. L., Ornstein, P., McCall, L., & Curran, P. (2008). Linking teachers’ memory-relevant
language and the development of children’s memory skills. Developmental Psychology, 44,
1640–1654.
Cole, M. (1992). Cognitive development and formal schooling: The evidence from cross-cultural
research. In L. C. Moll (Ed.), Vygotsky and education: Instructional implications and appli-
cations of sociohistorical psychology (pp. 89–110). New York: Cambridge University Press.
References 337

Cole, M., Gay, J., Glick, J., & Sharp, D. (1971). The cultural context of learning and thinking.
New York: Basic Books.
Cole, M., & Scribner, S. (1977). Cross-cultural studies of memory and cognition. In R. V. Kail Jr
& J. W. Hagen (Eds.), Perspectives on the development of memory and cognition (pp. 239–
271). Hillsdale: Erlbaum.
Collie, R., & Hayne, H. (1999). Deferred imitation by 6- and 9-month-old infants: More evidence
for declarative memory. Developmental Psychobiology, 35, 83–90.
Collins, A. M., & Loftus, E. F. (1975). A spreading-activation theory of semantic processing.
Psychological Bulletin, 82, 407–428.
Conklin, H. M., Luciana, M., Hooper, C. J., & Yarger, R. (2007). Working memory performance
in typically developing children and adolescents: Behavioral evidence of protracted frontal
lobe development. Developmental Neuropsychology, 31, 103–128.
Conway, M. A. (2005). Memory and the self. Journal of Memory and Language, 53, 594–628.
Corbett, A. T., & Dosher, B. A. (1978). Instrument inferences in sentence encoding. Journal of
Verbal Learning and Verbal Behavior, 17, 479–491.
Cornell, E. H. (1979). Infants’ recognition memory, forgetting, and savings. Journal of
Experimental Child Psychology, 28, 359–374.
Cornoldi, C. (2010). Metacognition, intelligence, and academic performance. In H. S. Waters &
W. Schneider (Eds.), Metacognition, strategy use, and instruction (pp. 257–277). New York:
Guilford Press.
Cornoldi, C., Gobbo, C., & Mazzoni, G. (1991). On metamemory-memory relationship: Strategy
availability and training. International Journal of Behavioral Development, 14, 101–121.
Corsale, K., & Ornstein, P. A. (1980). Developmental changes in children’s use of semantic
information in recall. Journal of Experimental Child Psychology, 30, 231–245.
Courage, M. L., & Cowan, N. (2009a). Introduction: What’s new in research on the development
of memory in infants and children? In M. L. Courage & N. Cowan (Eds.), The development of
memory in infancy and childhood (pp. 1–10). Hove: Psychology Press.
Courage, M. L., & Cowan, N. (Eds.). (2009b). The development of memory in infancy and child-
hood. Hove: Psychology Press.
Courage, M. L., & Howe, M. L. (2004). Advances in early memory development research:
Insights about the dark side of the moon. Developmental Review, 24, 6–32.
Cowan, N. (1988). Evolving conceptions of memory storage, selective attention, and their mutual con-
straints within the human information processing system. Psychological Bulletin, 104, 163–191.
Cowan, N. (1994). Mechanisms in verbal short-term memory. Current Directions in
Psychological Science, 3, 185–189.
Cowan, N. (Ed.). (1997). The development of memory in childhood. Hove: Psychology Press.
Cowan, N. (1999). An embedded-processes model of working memory. In A. Miyake & P. Shah
(Eds.), Models of working memory: Mechanisms of active maintenance and executive control
(pp. 62–101). Cambridge: Cambridge University Press.
Cowan, N. (2001). The magical number 4 in short-term memory: A reconsideration of mental
storage capacity. Behavioral and Brain Sciences, 24, 87–114.
Cowan, N. (2005). Working memory capacity. New York: Psychology Press.
Cowan, N. (2007). What infants can tell us about working memory development. In L. M. Oakes
& P. J. Bauer (Eds.), Short- and long-term memory in infancy and early childhood: Taking the
first steps towards remembering (pp. 126–150). New York: Oxford University Press.
Cowan, N. (2014). Short-term and working memory in childhood. In P. J. Bauer & R. Fivush
(Eds.), The Wiley handbook on the development of children’s memory (Vol. II, pp. 202–229).
Chichester: Wiley.
Cowan, N., & Alloway, T. (2009). Development of working memory in childhood. In M. L.
Courage & N. Cowan (Eds.), The development of memory in infancy and childhood (pp. 303–
342). Hove: Psychology Press.
Cowan, N., Elliott, E. M., Saults, J. S., Morey, C. C., Mattox, S., Hismjatulli, A., et al. (2005).
On the capacity of attention: Its estimation and its role in working memory and cognitive
aptitudes. Cognitive Psychology, 51, 42–100.
338 References

Cowan, N., Elliott, E. M., Saults, J. S., Nugent, L. D., Bomb, P., & Hismjatullina, A. (2006).
Rethinking speed theories of cognitive development: Increasing the rate of recall without
affecting accuracy. Psychological Science, 17(1), 67–73.
Cowan, N., Keller, T., Hulme, C., Roodenrys, S., McDougall, S., & Rack, J. (1994). Verbal mem-
ory span in children: Speech timing clues to the mechanisms underlying age and word length
effects. Journal of Memory & Language, 33, 234–250.
Cowan, N., Morey, C. C., AuBuchon, A. M., Zwilling, C. E., & Gilchrist, A. L. (2010). Seven-
year-olds allocate attention like adults unless working memory is overloaded. Developmental
Science, 13, 120–133.
Cowan, N., Naveh-Benjamin, M., Kilb, A., & Saults, J. S. (2006). Life-span development of
visual working memory: When is feature binding difficult? Developmental Psychology, 42(6),
1089–1102.
Cowan, N., Nugent, L., Elliott, E., Ponomarev, I., & Saults, J. (1999). The role of attention in the
development of short-term memory: Age differerences in the verbal span of apprehension.
Child Development, 70, 1082–1097.
Cowan, N., Nugent, L. D., Elliott, E. M., & Saults, J. S. (2000). Persistence of memory
for ignored lists of digits: Areas of developmental consistency and change. Journal of
Experimental Child Psychology, 76, 151–172.
Cowan, N., Saults, J. S., & Elliott, E. M. (2002). The search for what is fundamental in the devel-
opment of working memory. In R. Kail & H. Reese (Eds.), Advances in Child Development
and Behavior, 29, 1-49.
Cowan, N., Wood, N. L., Wood, P. K., Keller, T. A., Nugent, L. D., & Keller, C. V. (1998).
Two separate verbal processing rates contributing to short-term memory span. Journal of
Experimental Psychology: General, 127, 141–160.
Cox, B. D., Ornstein, P. A., Naus, M. J., Maxfield, D., & Zimler, J. (1989). Children’s concurrent
use of rehearsal and organizational strategies. Developmental Psychology, 25, 619–627.
Cox, B. D., Ornstein, P. A., & Valsiner, J. (1991). The role of internalization in the trans-
fer of mnemonic strategies. In L. Oppenheimer & J. Valsiner (Eds.), The origins of action:
Interdisciplinary and international perspectives (pp. 101–131). New York: Springer.
Cox, D., & Waters, H. S. (1986). Sex differences in the use of organization strategies: A develop-
mental analysis. Journal of Experimental Child Psychology, 41, 18–37.
Cox, G. L., & Paris, S. G. (1979). Evaluation, selection, and use of mnemonic skills by chil-
dren, young adults, and the elderly. Paper presented at the annual meetings of the American
Educational Research Association, San Francisco, CA.
Coyle, T. R., & Bjorklund, D. F. (1997). Age differences in, and consequences of, multiple- and
variable-strategy use on a multitrial sort-recall task. Developmental Psychology, 33(2), 372–380.
Cultice, J. C., Somerville, S. C., & Wellman, H. M. (1983). Preschoolers' memory monitoring:
Feeling-of-knowing judgments. Child Development, 54, 1480–1486.
Cunningham, J. G., & Weaver, S. L. (1989). Young children's knowledge of their memory span:
Effects of task and experience. Journal of Experimental Child Psychology, 48, 32–44.
Cycowicz, Y. M. (2000). Memory development and event-related brain potentials in children.
Biological Psychology, 54, 145–174.
Cycowicz, Y. M., Friedman, D., Snodgrass, J. G., & Rothstein, M. (2000). A developmental tra-
jectory in implicit memory is revealed by picture fragment completion. Memory, 8, 19–35.
Daehler, M. W., & Greco, C. (1985). Memory in very young children. In M. Pressley & C. J.
Brainerd (Eds.), Cognitive learning and memory in children (pp. 49–79). New York: Springer.
Darling, S., Della Sala, S., & Logie, R. H. (2007). Behavioural evidence for separating compo-
nents within visuo-spatial working memory. Cognitive Processes, 8, 175–181.
Darling, S., Della Sala, S., & Logie, R. H. (2009). Dissociation between appearance and location
within visuo-spatial working memory. Quarterly Journal of Experimental Psychology, 62(3),
417–425.
Darwin, C. R. (1877). Biographical sketch of an infant. Mind, 2, 285–294.
Davidson, M. C., Amso, D., Anderson, L. C., & Diamond, A. (2006). Development of cognitive
control and executive functions from 4 to 13 years: Evidence from manipulations of memory
inhibition, and task switching. Neuropsychologia, 44, 2037–2078.
References 339

Davies, G. (1993). Witnessing events. In G. M. Davies & R. H. Logie (Eds.), Memory in every-
day life (pp. 367–401). Amsterdam: Elsevier.
De Chastelaine, M., Friedman, D., & Cycowicz, Y. M. (2007). The development of control pro-
cesses supporting source memory discrimination as revealed by event-related potentials.
Journal of Cognitive Neuroscience, 19, 1286–1301.
De Haan, M., Mishkin, M., Baldeweg, T., & Vargha-Khadem, F. (2006). Human memory devel-
opment and its dysfunction after early hippocampal injury. Trends in Neuroscience, 29,
374–381.
DeLoache, J. S. (1980). Naturalistic studies of memory for object location in very young chil-
dren. In M. Perlmutter (Ed.), New directions for child development: Children’s memory (pp.
17–32). San Francisco: Jossey-Bass.
DeLoache, J. S. (1984). Oh where, oh where: Memory-based searching by very young children.
Developmental Psychology, 20, 37–44.
DeLoache, J. S., & Brown, A. L. (1984). Where do I go next? Intelligent searching by very
young children. Developmental Psychology, 20, 37–44.
DeLoache, J. S., Cassidy, D. J., & Brown, A. L. (1985). Precursors of mnemonic strategies in
very young children’s memory. Child Development, 56, 125–137.
DeMarie, D., & Ferron, J. (2003). Capacity, strategies, and metamemory: Tests of a three-factor
model of memory development. Journal of Experimental Child Psychology, 84, 167–193.
DeMarie, D., & Lopez, L. M. (2014). Memory in schools. In P. J. Bauer & R. Fivush (Eds.), The
Wiley handbook on the development of children’s memory (Vol. II, pp. 836–864). Chichester:
Wiley.
DeMarie, D., Miller, P. H., Ferron, J., & Cunningham, W. R. (2004). Path analysis tests for theo-
retical models of children’s memory performance. Journal of Cognition and Development, 5,
461–492. (9).
DeMarie-Dreblow, D. (1991). Relation between knowledge and memory: A reminder that corre-
lation does not imply causation. Child Development, 62, 484–498.
Demetriou, A., Christou, C., Spanoudis, G., & Platsidou, M. (2002). The development of men-
tal processing: Efficiency, working memory, and thinking. Monographs of the Society of
Research in Child Development, 67 (Serial No. 268).
Demetriou, A., Efklides, A., & Platsidou, M. (1993). The architecture and dynamics of develop-
ing mind: Experiential structuralism as a frame for unifying cognitive developmental theories.
Monographs of the Society for Research in Child Development, 58 (5, Serial No. 234).
Demetriou, A., Kui, Z. X., Spanoudis, G., Christou, C., Kyriakides, L., & Platsidou, M. (2005).
The architecture, dynamics, and development of mental processing: Greek, Chinese, or
Universal? Intelligence, 33, 109–141.
Dempster, F. N. (1978). Memory span and short-term memory capacity: A developmental study.
Journal of Experimental Child Psychology, 26, 419–431.
Dempster, F. N. (1981). Memory span: Sources of individual and developmental differences.
Psychological Bulletin, 89, 63–100.
Dempster, F. N. (1985). Short-term memory development in childhood and adolescence. In C. J.
Brainerd & M. Pressley (Eds.), Basic processes in memory development: Progress in cogni-
tive development research (pp. 209–248). New York: Springer.
De Ribaupierre, A. (1999). Working memory and attentional control. In W. Perrig & A. Grob
(Eds.), Control of human behaviour, mental processes, and consciousness (pp. 147–164).
Mahwah: Erlbaum.
De Ribaupierre, A. (2002). Working memory and attentional processes across the lifespan. In
P. Graf & N. Ohta (Eds.), Lifespan development of human memory (pp. 59–80). Cambridge:
MIT Press.
De Ribaupierre, A., & Bailleux, C. (1994). Developmental change in a spatial task of attentional
capacity: An essay toward an integration of two models of working memory. International
Journal of Behavioral Development, 17, 5–35.
De Ribaupierre, A., & Bailleux, C. (1995). Development of attentional capacity in childhood: A
longitudinal study. In F. E. Weinert & W. Schneider (Eds.), Memory performance and compe-
tencies: Issues of growth and development (pp. 45–70). Mahwah: Erlbaum.
340 References

De Ribaupierre, A., Fagot, D., & Lecerf, T. (2011). Working memory capacity and its role in cog-
nitive development: Are age differences driven by the same processes across the lifespan. In P.
Barrouillet & V. Gaillard (Eds.), Cognitive development and working memory. A dialogue between
Neo-Piagetian theories and cognitive approaches (pp. 105–133). Hove: Psychology Press.
Desoete, A., & Veenman, M. (Eds.). (2006). Metacognition in mathematics education. Haupauge:
Nova Science.
De Villiers, J., & Pyers, J. (2002). Complements to cognition: A longitudinal study of the rela-
tionship between complex syntax and false-belief understanding. Cognitive Development, 17,
1037–1060.
Dew, I. T. Z., Bayen, U. J., & Giovanello, K. S. (2007). Implicit relational memory in young and
older adults. Journal of Psychology, 215, 25–34.
Diamond, A. (1985). Development of the ability to use recall to guide action, as indicated by
infants’ performance on AB. Child Development, 56, 868–883.
Diamond, A. (1990a). The development and neural bases of memory functions as indexed by the
AB and delayed response tasks in human infants and infant monkeys. In A. Diamond (Ed.),
The development and neural bases of higher cognitive functions (Vol. 608, pp. 267–309),
Annals of the New York Academy of Sciences. New York: New York Academy of Sciences.
Diamond, A. (1990b). Rate of maturation of the hippocampus and the developmental progression
of children’s performance on the delayed non-matching-to-sample and visual-paired compari-
son tasks. In A. Diamond (Ed.), The development and neural bases of higher cognitive func-
tions (Vol. 608, pp. 394–426), Annals of the New York Academy of Sciences. New York: New
York Academy of Sciences.
Diamond, A. (1995). Evidence of robust recognition memory early in life even when assessed by
reaching behavior. Journal of Experimental Child Psychology, 59, 419–456.
Diamond, A. (2006). The early development of executive functions. In E. Bialystok & F. I.
M. Craik (Eds.), Lifespan cognition: Mechanisms of change (pp. 70–95). Oxford: Oxford
University Press.
Diamond, A. (2013). Executive functions. Annual Review of Psychology, 64, 135–168.
Diamond, A., Cruttenden, L., & Neiderman, D. (1994). AB with multiple wells: I. Why are
multiple wells sometimes easier than two wells? II. Memory or memory + inhibition?
Developmental Psychology, 30, 192–205.
Diamond, A., & Doar, B. (1989). The performance of human infants on a measure of frontal cor-
tex function, the delayed-response task. Developmental Psychobiology, 22, 271–294.
Diamond, A., & Goldman-Rakic, P. S. (1989). Comparison of human infants and rhesus mon-
keys on Piaget's A-not-B task: Evidence for dependence on dorsolateral prefrontal cortex.
Experimental Brain Research, 74, 24–40.
Diamond, A., Towle, C., & Boyer, K. (1994). Young children’s performance on a task sensitive
to the memory functions of the medial temporal lobe in adults: the delayed nonmatching-to-
sample task reveals problems that are due to non-memory-related task demands. Behavioral
Neuroscience, 108, 659–680.
Dixon, R. A., & Hertzog, C. (1988). A functional approach to memory and metamemory
development in adulthood. In F. E. Weinert & M. Perlmutter (Eds.), Memory development:
Universal changes and individual differences (pp. 293–330). Hillsdale: Erlbaum.
Dochy, F. J. R. C., & Alexander, P. A. (1995). Mapping prior knowledge: a framework for discus-
sion among researchers. European Journal of Psychology of Education, 10, 225–242.
Drummey, A. B., & Newcombe, N. (1995). Remembering versus knowing the past: Children’s explicit
and implicit memories for pictures. Journal of Experimental Child Psychology, 59, 549–565.
Drummey, A. B., & Newcombe, N. S. (2002). Developmental changes in source memory.
Developmental Science, 5, 502–513.
Dudycha, G. J., & Dudycha, M. M. (1941). Childhood memories: A review of the literature.
Psychological Bulletin, 36, 668–682.
Dufresne, A., & Kobasigawa, A. (1989). Children's spontaneous allocation of study time:
Differential and sufficient aspects. Journal of Experimental Child Psychology, 47, 274–296.
References 341

Dunlosky, J., & Connor, L. (1997). Age-related differences in the allocation of study time
account for age-related differences in memory performance. Memory and Cognition, 25,
691–700.
Dunlosky, J., & Kane, M. J. (2007). The contributions of strategy use to working memory
span: A comparison of strategy-assessment methods. Quarterly Journal of Experimental
Psychology, 60, 1227–1245.
Dunlosky, J., & Nelson, T. O. (1992). Importance of the kind of cue for judgments of learning
(JOL) and the delayed-JOL effect. Memory & Cognition, 20, 374–380.
Ebbinghaus, H. (1885). Über das Gedächtnis. [On memory]. Darmstadt: Wissenschaftliche
Buchgesellschaft.
Ebbinghaus, H. (1887). Über eine neue Methode zur Prüfung geistiger Fähigkeiten und ihre
Anwendung bei Schulkindern. Zeitschrift für Psychologie und Physiologie der Sinnesorgane,
13, 401–457.
Efklides, A. (2001). Metacognitive experiences in problem solving: Metacognition, motivation,
and self-regulation. In A. Efklides, J. Kuhl, & R. M. Sorrentino (Eds.), Trends and prospects
in motivation research (pp. 297–323). Dordrecht: Kluwer.
Efklides, A. (2011). Interactions of metacognition with motivation and affect in self-regulated
learning: The MASRL model. Educational Psychologist, 46, 6–25.
Eichenbaum, H. (1999). The hippocampus and mechanisms of declarative memory. Behavioral
Brain research, 103, 123–133.
Elischberger, H. B. (2005). The effects of prior knowledge on children‘s memory and suggest-
ibility. Journal of Experimental Child Psychology, 92, 247–275.
Elischberger, H. B., & Roebers, C. M. (2001). Improving young children’s free narratives about
an observed event: The effects of nonspecific verbal prompts. International Journal of
Behavioural Development, 25, 160–166.
Ellis, H. D., Ellis, D. M., & Hosie, J. A. (1993). Priming effects in children’s face recognition.
British Journal of Psychology, 84, 101–110.
Ellis, N. C., & Hennelly, R. A. (1980). A bilingual word-length effect: Implications for intelli-
gence testing and the relative ease of mental calculation in Welsh and English. British Journal
of Psychology, 71, 43–52.
Ellis, N. R., Katz, E., & Williams, J. E. (1987). Developmental aspects of memory for spatial
location. Journal of Experimental Child Psychology, 44(3), 401–412.
Emmerich, H. J., & Ackerman, B. P. (1978). Developmental differences in recall: Encoding or
retrieval? Journal of Experimental Child Psychology, 25, 514–525.
Engle, R. W., & Kane, M. J. (2004). Executive attention, working memory capacity, and a two-
factor theory of cognitive control. In B. Ross (Ed.), The psychology of learning and motiva-
tion (pp. 145–199). New York: Elsevier.
Engle, R. W., & Marshall, K. (1983). Do developmental changes in digit span result from acqui-
sition strategies? Journal of Experimental Child Psychology, 36, 429–436.
Engle, R. W., Tuholski, S. W., Laughlin, J. E., & Conway, A. R. A. (1999). Working memory,
short-term memory, and general fluid intelligence: A latent-variable approach. Journal of
Experimental Psychology: General, 128(3), 309–331.
Ericsson, K. A. (Ed.). (1996). The road to excellence: The acquisition of expert performance in
the arts and sciences, sports and games. Mahwah: Erlbaum.
Ericsson, K. A., Krampe, R., & Tesch-Roemer, C. (1993). The role of deliberate practice in the
acquisition of expert performance. Psychological Review, 100, 363–406.
Ericsson, K. A., & Simon, H. A. (1980). Verbal reports as data. Psychological Review, 87,
215–251.
Ericsson, K. A., & Simon, H. A. (1993). Protocol analysis: Verbal reports as data (rev ed.).
Cambridge: MIT Press.
Espy, K. (1997). The shape school: Assessing executive function in preschool children.
Developmental Neuropsychology, 13, 495–499.
Fabricius, W. V., & Cavalier, L. (1989). The role of causal theories about memory in young chil-
dren’s memory strategy choice. Child Development, 60, 298–308.
342 References

Fabricius, W. V., & Wellman, H. M. (1983). Children’s understanding of retrieval cue utilization.
Developmental Psychology, 19, 15–21.
Fagan, J. F. (1970). Memory in the infant. Journal of Experimental Child Psychology, 9, 217–226.
Fagan, J. F. (1974). Infant recognition memory: The effects of length of familiarization and type
of discrimination task. Child Development, 45, 351–356.
Fagan, J. F. (1984). Infant memory: History, current trends, relations to cognitive psychology. In
M. Moscovitch (Ed.), Advances in the study of communication and affect (Vol. IX, pp. 1–27).
New York: Plenum.
Fantz, R. L. (1956). A method for studying early visual development. Perceptual and Motor
Skills, 6, 13–15.
Farrant, K., & Reese, E. (2000). Maternal style and children‘s participation in reminiscing:
Stepping stones in children’s autobiographical memory development. Journal of Cognition
and Development, 1, 193–225.
Farrar, M. J., & Goodman, G. S. (1990). Developmental differences in the relation between
scripts and episodic memory: Do they exist? In R. Fivush & J. A. Hudson (Eds.), Knowing
and remembering in young children (pp. 30–64). New York: Cambridge University Press.
Farrar, M. J., & Goodman, G. S. (1992). Developmental changes in event memory. Child
Development, 63, 1697–1709.
Fechner, B. (1965). Zur Psychologie des Gedächtnisses III: Zur Gruppenbildung sinnvoller
sprachlicher Texte in Abhängigkeit vom Alter. Zeitschrift für Psychologie, 80, 2–21.
Feigenson, L. (2007). Continuity of format and computation in short-term memory development. In
L. M. Oakes & P. J. Bauer (Eds.), Short- and long-term memory in infancy and early childhood:
Taking the first steps towards remembering (pp. 51–74). New York: Oxford University Press.
Feigenson, L., & Carey, S. (2005). On the limits of infants’ quantification of small object areas.
Cognition, 97, 295–313.
Feigenson, L., Carey, S., & Hauser, M. (2002). The representations underlying infants‘ choice of
more: Object files versus analog magnitudes. Psychological Science, 13, 150–156.
Ferguson, A. N., & Bowey, J. A. (2005). Global processing speed as a mediator of developmental
changes in children’s auditory memory span. Journal of Experimental Child Psychology, 91,
89–112.
Ferguson, A. N., Bowey, J. A., & Tilley, A. (2002). The association between auditory memory
span and speech rate in children from kindergarten to sixth grade. Journal of Experimental
Child Psychology, 81, 141–156.
Fischer, K. W. (1980). A theory of cognitive development: The control and construction of hierar-
chies of skills. Psychological Review, 87, 477–531.
Fisk, J. E., & Sharp, C. A. (2004). Age-related impairment in executive functioning: Updating,
inhibition, shifting, and access. Journal of Clinical and Experimental Neuropsychology, 26,
874–890.
Fivush, R. (1997). Event memory in early childhood. In N. Cowan (Ed.), The development of
memory in childhood (pp. 139–161). Hove: Psychology Press.
Fivush, R. (2001). Owning experience: Developing subjective perspective in autobiographical
narratives. In C. Moore & K. Lemmon (Eds.), The self in time: Developmental perspectives
(pp. 35–52). Mahwah: Erlbaum.
Fivush, R. (2009). Sociocultural perspectives on autobiographical memory. In M. L. Courage &
N. Cowan (Eds.), The development of memory in infancy and childhood (pp. 283–301). Hove:
Psychology Press.
Fivush, R. (2014). Maternal reminiscing style: The sociocultural construction of autobiographi-
cal memory across childhood and adolescence. In P. J. Bauer & R. Fivush (Eds.), The Wiley
handbook on the development of children’s memory (Vol. II, pp. 568–585). Chichester: Wiley.
Fivush, R., Gray, J. T., & Fromhoff, F. A. (1987). Two-year-olds talk about the past. Cognitive
Development, 2, 393–409.
Fivush, R., Habermas, T., Waters, T. E. A., & Zaman, W. (2010). The making of autobiographical
memory: Intersections of culture, narratives and history. International Journal of Psychology,
46, 321–345.
References 343

Fivush, R., Haden, C., & Adam, S. (1995). Structure and coherence of preschoolers’ personal
narratives over time: Implications for childhood amnesia. Journal of Experimental Child
Psychology, 60, 32–50.
Fivush, R., Haden, C. A., & Reese, E. (2006). Elaborating on elaborations: Role of maternal remi-
niscing style in cognitive and socioemotional development. Child Development, 77, 1568–1588.
Fivush, R., & Hamond, N. R. (1990). Autobiographical memory across the preschool years:
Toward reconceptualising childhood amnesia. In R. Fivush & J. A. Hudson (Eds.), Knowing
and remembering in young children (pp. 223–248). Cambridge: Cambridge University Press.
Fivush, R., Hazzard, A., Sales, J. M., Sarfati, D., & Brown, T. (2003). Creating coherence out of
chaos? Children’s narratives of emotionally positive and negative events. Applied Cognitive
Psychology, 17, 1–19.
Fivush, R., Hudson, J. A., & Nelson, K. (1984). Children’s long-term memory for a novel event:
An exploratory study. Merrill-Palmer Quarterly, 30, 303–316.
Fivush, R., Kuebli, J., & Clubb, P. A. (1992). The structure of events and event representations: A
developmental analysis. Child Development, 63, 188–201.
Fivush, R., & Mandler, J. (1985). Developmental changes in the understanding of temporal
sequence. Child Development, 56, 1437–1446.
Fivush, R., Sales, J. M., Goldberg, A., Bahrick, L., & Parker, J. F. (2004). Weathering the storm:
Children’s long-term recall of Hurricane Andrew. Memory, 12, 104–118.
Fivush, R., & Slackman, E. (1986). The acquisition and development of scripts. In K. Nelson
(Ed.), Event knowledge: Structure and function in development (pp. 71–96). Hillsdale:
Erlbaum.
Fivush, R., & Zaman, W. (2014). Gender, subjective perspective, and autobiographical con-
sciousness. In P. J. Bauer & R. Fivush (Eds.), The Wiley handbook on the development of
children’s memory (Vol. II, pp. 586–604). Chichester: Wiley.
Flavell, J. H. (1970). Developmental studies of mediated memory. In H. W. Reese & L. P. Lipsitt
(Eds.), Advances in child development and behavior (Vol. 5, pp. 181–211). New York:
Academic Press.
Flavell, J. H. (1971). First discussant’s comments: What is memory development the develop-
ment of? Human Development, 14, 225–286.
Flavell, J. H. (1979). Metacognition and cognitive monitoring. A new area of cognitive-develop-
mental inquiry. American Psychologist, 34, 906–911.
Flavell, J. H. (1985). Cognitive development (2nd ed.). Englewood Cliffs: Prentice-Hall.
Flavell, J. H. (2000). Development of children's knowledge about the mental world. International
Journal of Behavioral Development, 24, 15–23.
Flavell, J. H., Beach, D. H., & Chinsky, J. M. (1966). Spontaneous verbal rehearsal in a memory
task as a function of age. Child Development, 37, 283–299.
Flavell, J. H., Friedrichs, A. G., & Hoyt, J. D. (1970). Developmental changes in memorization
processes. Cognitive Psychology, 1, 324–340.
Flavell, J. H., Miller, P. H., & Miller, S. A. (2002). Cognitive development (4th ed.). Englewood
Cliffs: Prentice-Hall.
Flavell, J. H., & Wellman, H. M. (1977). Metamemory. In R. V. Kail & J. W. Hagen (Eds.),
Perspectives on the development of memory and cognition (pp. 3–33). Hillsdale: Erlbaum.
Flin, R., Boon, J., Knox, A., & Bull, R. (1992). The effect of a five-month delay on children’s
and adults’ eyewitness memory. British Journal of Psychology, 83, 323–336.
Folds, T. H., Footo, M., Guttentag, R. E., & Ornstein, P. A. (1990). When children mean to
remember: Issues of context specificity, strategy effectiveness, and intentionality in the devel-
opment of memory. In D. F. Bjorklund (Ed.), Children’s strategies: Contemporary views of
cognitive development (pp. 67–91). Hillsdale: Erlbaum.
Foley, M. A., Aman, C., & Gutch, D. (1987). Discriminating between action memories: chil-
dren’s use of kinestetic cues and visible consequences. Journal of Experimental Child
Psychology, 44(3), 335–347.
Foley, M. A., Ratner, H. H., & Gentes, E. (2010). Helping children enter into another’s experi-
ences: The look and feel of it. Journal of Cognition and Development, 11, 217–239.
344 References

Fox, N. J., Kagan, J., & Weiskopf, S. (1979). The growth of memory during infancy. Genetic
Psychology Monographs, 99, 91–130.
Frank, H. S., & Rabinovitch, M. S. (1974). Auditory short-term memory: Developmental changes
in rehearsal. Child Development, 45, 397–407.
Frankel, M. T., & Rollins, H. A. (1985). Associative and categorical hypotheses of organization in
the free recall of adults and children. Journal of Experimental Child Psychology, 40, 304–318.
Freud, S. (1905/1953). Childhood and concealing memories. In A. A. Brill (Trans. & Ed.), The
basic writings of Sigmund Freud. New York: The Modern Library.
Friedman, S. (1972). Habituation and recovery of visual response in the alert human newborn.
Journal of Experimental Child Psychology, 13, 339–349.
Gaillard, V., Barrouillet, P., Jarrold, C., & Camos, V. (2011). Developmental differences in working
memory: Where do they come from?. Journal of Experimental Child Psychology, 110, 469–479.
Garner, R. (1987). Metacognition and reading comprehension. Norwood: Ablex.
Garon, N., Bryson, S. E., & Smith, I. M. (2008). Executive function in preschoolers: A review
using an integrative framework. Psychological Bulletin, 134, 31–60.
Garrison, A. (1980). Categorical and spatial modes of representation in young children’s recall.
Journal of Experimental Child Psychology, 30, 383–388.
Gathercole, S. E., & Hitch, G. J. (1993). Developmental changes in short-term memory: A
revised working memory perspective. In A. F. Collins, S. E. Gathercole, M. A. Conway, & P.
E. Morris (Eds.), Theories of memory (pp. 189–209). Hove, UK: Erlbaum.
Gathercole, S. E., Lamont, E., & Alloway, T. P. (2006). Working memory in the classroom. In S.
J. Pickering (Ed.), Working memory and education (pp. 219–240). London: Elsevier.
Gathercole, S. E., Pickering, S. J., Ambridge, B., & Wearing, H. (2004). The structure of working
memory from 4 to 15 years of age. Developmental Psychology, 40, 177–190.
Gaultney, J. F. (1995). The effect of prior knowledge and metacognition on the acquisition of a
reading comprehension strategy. Journal of Experimental Child Psychology, 59, 142–163.
Gaultney, J. F., Bjorklund, D. F., & Schneider, W. (1992). The role of children’s expertise in a
strategic memory task. Contemporary Educational Psychology, 17, 244–257.
Gavens, N., & Barrouillet, P. (2004). Delays of retention, processing efficiency, and attentional
resources in working memory span development. Journal of Memory and Language, 51,
644–657.
Geary, D. C., Bow-Thomas, C. C., Fan, L., & Siegler, R. S. (1993). Even before formal instruc-
tions, Chinese children outperform American children in mental arithmetic. Cognitive
Development, 8, 517–529.
Geary, D. C., Klosterman, I. H., & Adrales, K. (1989). Metamemory and academic achieve-
ment: Testing and validity of a group-administered metamemory battery. Journal of Genetic
Psychology, 151, 439–450.
Geis, M. F., & Lange, G. (1976). Children’s cue utilization in a memory-for-location task. Child
Development, 47, 759–766.
Ghatala, E. S. (1984). Developmental changes in incidental memory as a function of meaningful-
ness and encoding condition. Developmental Psychology, 20, 208–211.
Ghatala, E. S., Levin, J. R., Pressley, M., & Goodwin, D. (1986). A componential analysis of the
effects of derived and supplied strategy-utility information on children’s strategy selection.
Journal of Experimental Child Psychology, 41, 76–92.
Ghetti, S. (2008). Processes underlying developmental reversals in false-memory formation:
Comment on Brainerd, Reyna, and Ceci (2008). Psychological Bulletin, 134, 764–767.
Ghetti, S., & Angelini, L. (2008). The development of recollection and familiarity in childhood
and adolescence: Evidence from the dual-process signal detection model. Child Development,
79, 339–358.
Ghetti, S., & Alexander, K. W. (2004). “If it happened, I would remember it”: Strategic use of
event memorability in the rejection of false autobiographical events. Child Development, 75,
542–561.
Ghetti, S., Castelli, P., & Lyons, K. (2010). Knowing about not remembering: Developmental
dissociations in lack-of-memory monitoring. Developmental Science, 13, 611–621.
References 345

Ghetti, S., & Lee, J. K. (2014). The development of recollection and familiarity during child-
hood: Insight from studies of behaviour and brain. In P. J. Bauer & R. Fivush (Eds.), The
Wiley handbook on the development of children’s memory (pp. 309–335). Chichester: Wiley.
Ghetti, S., Lyons, K. E., Lazzarin, F., & Cornoldi, C. (2008). The development of metamemory
monitoring during retrieval: The case of memory strength and memory absence. Journal of
Experimental Child Psychology, 99, 157–181.
Ghetti, S., Papini, S., & Angelini, L. (2006). The development of the memorability-based strat-
egy: Insight from a training study. Journal of Experimental Child Psychology, 94, 206–228.
Ghetti, S., Qin, J. J., & Goodman, G. S. (2002). False memories in children and adults: Age, dis-
tinctiveness, and subjective experience. Developmental Psychology, 38, 705–718.
Gilchrist, A. L., Cowan, N., & Naveh-Benjamin, M. (2009). Investigating the childhood develop-
ment of working memory using sentences: New evidence for the growth of chunk capacity.
Journal of Experimental Child Psychology, 104, 252–265.
Globerson, T. (1983). Mental capacity and cognitive functioning: Developmental and social class
differences. Developmental Psychology, 2, 225–230.
Gobbo, C., & Chi, M. (1986). How knowledge is structured and used by expert and novice chil-
dren. Cognitive Development, 1, 221–237.
Goldman-Rakic, P. C. (1987). Development of cortical circuitry and cognitive function. Child
Development, 58, 601–622.
Goodman, G. S., Aman, C. J., & Hirschman, J. (1987). Child sexual and physical abuse:
Children’s testimony. In C. J. Ceci, M. P. Toglia, & D. F. Ross (Eds.), Children’s eyewitness
memory (pp. 1–23). New York: Springer.
Goodman, G. S., & Clarke-Stewart, A. (1991). Suggestibility in children’s testimony:
Implications for sexual abuse investigations. In J. D. Doris (Ed.), The suggestibility of chil-
dren’s recollections: Implications for eyewitness testimony (pp. 92–105). Washington, DC:
American Psychological Association.
Goodman, G. S., Hirschman, J. E., Hepps, D., & Rudy, L. (1991). Children’s memory for stress-
ful events. Merrill-Palmer Quarterly, 37, 109–158.
Goodman, G. S., Ogle, C. M., McWilliams, K., Narr, R. K., & Paz-Alonso, P. M. (2014).
Memory development in the forensic context. In P. J. Bauer & R. Fivush (Eds.), The Wiley
handbook on the development of children’s memory (Vol. II, pp. 920–942). Chichester: Wiley.
Goodman, G. S., & Quas, J. A. (1997). Trauma and memory: Individual differences in children’s
recounting of a stressful experience. In N. L. Stein, P. A. Ornstein, B. Tversky, & C. Brainerd
(Eds.), Memory for everyday and emotional events (pp. 267–294). Mahwah: Erlbaum.
Goodman, G. S., & Quas, J. A. (2008). Repeated interviews and children’s memory. Current
Directions in Psychological Science, 17, 386–390.
Gopnik, A., & Graf, P. (1988). Knowing how to know: young children’s ability to identify and
remember the source of their beliefs. Child Development, 59, 1366–1371.
Gordon, F. R., & Flavell, J. H. (1977). The development of intuitions about cognitive cueing.
Child Development, 48, 1027–1033.
Goulet, L. R. (1968). Verbal learning in children: Implications for developmental research.
Psychological Bulletin, 5, 359–376.
Graf, P., & Schacter, D. L. (1985). Implicit and explicit memory for new associations in nor-
mal and amnesic subjects. Journal of Experimental Psychology: Learning, Memory, and
Cognition, 11, 501–518.
Graf, P., Shimamura, A. P., & Squire, L. R. (1985). Priming across modalities and priming
across category levels: Extending the domain of preserved function in amnesia. Journal of
Experimental Psychology: Learning, Memory, and Cognition, 11, 386–396.
Grammer, J., Coffman, J. L., & Ornstein, P. (2013). The effect of teachers‘ memory-relevant lan-
guage on children’s strategy use and knowledge. Child Development, 84, 1989–2002.
Grammer, J. K., Purtell, K. M., Coffman, J. L., & Ornstein, P. A. (2011). Relations between chil-
dren’s metamemory and strategic performance: Time-varying covariates in early elementary
school. Journal of Experimental Child Psychology, 108, 139–155.
Greenbaum, J. L., & Graf, P. (1989). Preschool period development of implicit and explicit
remembering. Bulletin of Psychonomic Society, 27, 417–420.
346 References

Greenhoot, A. F., Ornstein, P. A., Gordon, B. N., & Baker-Ward, L. (1999). Acting out details of
a pediatric check-up: The impact of interview condition and behavioral style on children’s
memory reports. Child Development, 70, 363–380.
Greenhoot, A. F., & Sun, S. (2014). Trauma and memory. In P. J. Bauer & R. Fivush (Eds.), The
Wiley handbook on the development of children’s memory (Vol. II, pp. 774–803). Chichester:
Wiley.
Gruber, H., Renkl, A., & Schneider, W. (1994). Expertise und Gedächtnisentwicklung.
Längsschnittliche Befunde aus der Domäne Schach [Expertise and memory development:
Longitudinal findings from the chess domain]. Zeitschrift für Entwicklungspsychologie und
Pädagogische Psychologie, 26, 53–70.
Güler, O. E., Larkina, M., Kleinknecht, E., & Bauer, P. J. (2010). Memory strategies and retrieval
success in preschool children: Relations to maternal behavior over time. Journal of Cognition
and Development, 11, 159–184.
Guttentag, R. E. (1984). The mental effort requirement of cumulative rehearsal: A developmental
study. Journal of Experimental Child Psychology, 37, 92–106.
Guttentag, R. E. (1989). Age differences in dual-task performance: Procedures, assumptions, and
results. Developmental Review, 9, 146–170.
Guttentag, R. E. (1995). Mental effort and motivation: Influences on children’s memory strategy
use. In F. E. Weinert & W. Schneider (Eds.), Memory performance and competencies issues
in growth and development (pp. 207–224). Mahwah: Erlbaum.
Guttentag, R. (1997). Memory development and processing resources. In N. Cowan (Ed.), The
development of memory in childhood (pp. 247–274). London: London University College
Press.
Guttentag, R., & Dunn, J. (2003). Judgments of remembering: The revelation effect in children
and adults. Journal of Experimental Child Psychology, 86, 153–167.
Guttentag, R. E., & Lange, G. (1994). Motivational influences on children’s strategic remember-
ing. Learning and Individual Differences, 6, 309–330.
Guttentag, R. E., Ornstein, P. A., & Siemens, L. (1987). Children’s spontaneous rehearsal:
Transitions in strategy acquisition. Cognitive Development, 2, 307–326.
Habermas, T. (2007). How to tell a life: The development of the cultural concept of biography
across the lifespan. Journal of Cognition and Development, 8, 1–31.
Haden, C. A. (2014). Interactions of knowledge and memory in the development of skilled
remembering. In P. J. Bauer & R. Fivush (Eds.), The Wiley handbook on the development of
children’s memory (Vol. II, pp. 807–835). Chichester: Wiley.
Haden, C. A., Haine, R. A., & Fivush, R. (1997). Developing narrative structure in parent-child
reminiscing across the preschool years. Developmental Psychology, 33, 295–307.
Haden, C. A., Ornstein, P. A., Eckerman, C. O., & Didow, S. M. (2001). Mother-child conversa-
tional interactions as event unfold: Linkages to subsequent remembering. Child Development,
72, 1016–1031.
Haden, C. A., Ornstein, P. A., O’Brien, B. S., Elischberger, H. B., Tyler, C. S., & Burchinal,
M. J. (2011). The development of children’s early skills. Journal of Experimental Child
Psychology, 108, 44–60.
Haden, C. A., Ornstein, P. A., Rudek, D. J., & Cameron, D. (2009). Reminiscing in the early
years: Patterns of maternal elaborativeness and children’s remembering. International
Journal of Behavioral Development, 33, 118–130.
Hagen, J. W., & Stanovich, K. G. (1977). Memory: Strategiews of acquisition. In R. V. Kail &
J. W. Hagen (Eds.), Perspectives on the development of memory and cognition (pp. 89–111).
Hillsdale: Erlbaum.
Hale, S., Fry, A. F., & Jessie, K. A. (1993). Effects of practice on speed of information process-
ing in children and adults: Age-sensitivity and age-invariance. Developmental Psychology, 29,
880–892.
Halford, G. S. (1982). The development of thought. Hillsdale: Erlbaum.
Halford, G. S. (1993). Children’s understanding: The development of mental models. Hillsdale:
Erlbaum.
References 347

Halford, G. S., & Andrews, G. (2011). Information-processing models of cognitive development.

In U. Goswami (Ed.), The Wiley-Blackwell handbook of childhood cognitive development
(2nd ed., pp. 697–722). Oxford: Wiley.
Halford, G. S., Maybery, M. T., & Bennett, D. (1985). The role of strategies in the develop-
ment of memory span assessed by running probes. International Journal of Behavioral
Development, 8, 301–312.
Halliday, M. S., Hitch, G. J., Lennon, B., & Pettipher, C. (1990). Verbal short-term memory in
children: The role of the articulatory loop. European Journal of Cognitive Psychology, 2,
23–38.
Halperin, M. S. (1974). Developmental changes in the recall and recognition of categorized word
lists. Child Development, 45, 144–151.
Hamann, S., & Stevens, J. S. (2014). Memory for emotional stimuli in development. In P. J.
Bauer & R. Fivush (Eds.), The Wiley handbook on the development of children’s memory
(Vol. II, pp. 724–742). Chichester: Wiley.
Hambrick, D. Z., Oswald, F. L., Altmann, E. M., Meinz, E.J., Gobet, F., & Campitelli, G.
(2013). Deliberate practice: Is that all it takes to become an expert? Intelligence,http://dx.
doi.org/10.1016/j.intell.2013.04.001
Hamilton, C., Coates, R., & Heffernan, T. (2003). What develops in visuo-spatial working mem-
ory development? European Journal of Cognitive Psychology, 15(1), 43–69.
Hamond, N. R., & Fivush, R. (1991). Memories of Mickey Mouse: Young children recount their
trip to Disneyworld. Cognitive Development, 6, 433–448.
Han, J. J., Leichtman, M. D., & Wang, Q. (1998). Autobiographical memory in Korean, Chinese,
and American children. Developmental Psychology, 34, 701–713.
Harley, K., & Reese, E. (1999). Origins of autobiographical memory. Developmental Psychology,
35, 1338–1348.
Harnishfeger, K. K. (1995). Development of cognitive inhibition: Theories, definitions, and
research evidence. In F. N. Dempster & C. J. Brainerd (Eds.), Interference and inhibition in
cognition (pp. 175–204). San Diego: Academic Press.
Harnishfeger, K. K., & Bjorklund, D. F. (1990). Children’s strategies: A brief history. In D. F.
Bjorklund (Ed.), Children’s strategies: Contemporary views of cognitive development (pp.
1–22). Hillsdale: Erlbaum.
Harnishfeger, K. K., & Bjorklund, D. F. (1994). Individual differences in inhibition: Implications
for children’s cognitive development. Learning and Individual Differences, 6, 331–335.
Hartshorn, K., & Rovee-Collier, C. (1997). Infant learning and long-term memory at 6 months: A
confirming analysis. Developmental Psychobiology, 30, 71–85.
Hartshorn, K., Rovee-Collier, C., Gerhardstein, P. C., Bhatt, R. S., Wondoloski, T. L., Klein,
P., et al. (1998). Ontogeny of long-term memory over the first year-and-a-half of life.
Developmental Psychobiology, 32, 69–89.
Hartshorn, K., Rovee-Collier, C., Gerhardstein, P. C., Bhatt, R. S., Klein, P. J., Aaron, F., et al.
(1998). Developmental changes in the specificity of memory over the first year of life.
Developmental Psychobiology, 33, 61–78.
Hasselhorn, M. (1990). The emergence of strategic knowledge activation in categorical clustering
during retrieval. Journal of Experimental Child Psychology, 50, 59–80.
Hasselhorn, M. (1992). Task dependency and the role of category typicality and metamemory in
the development of an organizational strategy. Child Development, 63, 202–214.
Hasselhorn, M. (1994). Zur Erfassung von Metagedächtnisaspekten bei Grundschulkindern
[Assessment of metamemory in elementary school children]. Zeitschrift für
Entwicklungspsychologie und Pädagogische Psychologie, 26, 71–78.
Hasselhorn, M. (1995). Beyond production deficiency and utilization inefficiency: Mechanisms
of the emergence of strategic categorization in episodic memory tasks. In F. E. Weinert & W.
Schneider (Eds.), Memory performance and competencies: Issues in growth and development
(pp. 141–159). Hillsdale: Erlbaum.
Hayes, B. K., & Hennessy, R. (1996). The nature and development of nonverbal implicit mem-
ory. Journal of Experimental Child Psychology, 63, 22–43.
348 References

Hayne, H. (2007). Infant memory development: New Questions, new answers. In L. M. Oakes &
P. J. Bauer (Eds.), Short- and long-term memory in infancy and early childhood: Taking the
first steps towards remembering (pp. 209–239). New York: Oxford University Press.
Hayne, H., Barr, R., & Herbert, J. (2003). The effect of prior practice on memory reactivation
and generalization. Child Development, 74, 1615–1627.
Hayne, H., & Herbert, J. (2004). Verbal cues facilitate memory retrieval during infancy. Journal
of Experimental Child Psychology, 89(2), 127–139.
Hayne, H., & Simcock, G. (2009). Memory development in toddlers. In M. L. Courage & N.
Cowan (Eds.), The development of memory in infancy and childhood (pp. 43–68). Hove, UK:
Psychology Press.
Hedrick, A. M., Haden, C. A., & Ornstein, P. A. (2009). Elaborative talk during and after an
event: Conversational style influences children‘s memory reports. Journal of Cognition and
Development, 10, 188–209.
Henry, L. A. (1994). The relationship between speech rate and memory span in children.
International Journal of Behavioral Development, 17, 37–56.
Henry, L. A. (2012). The development of working memory in children. London: Sage.
Henry, L. A., & Millar, S. (1991). Memory span increases with age: A test of two hypotheses.
Journal of Experimental Child Psychology, 51, 459–484.
Henri, V., & Henri, C. (1897). Enquete sur les premiers souvenirs de l’enfance. L’Année
Psychologique, 3, 184–198.
Henri, V., & Henri, C. (1898). Earliest recollections. Popular Science Monthly,53, 108–115.
Herbert, J., Gross, J., & Hayne, H. (2007). Crawling is associated with more flexible memory
retrieval by 9-month-old-infants. Developmental Science,10, 183–189.
Herrmann, D. J. (1982). Know thy memory: the use of questionnaires to assess and study mem-
ory. Psychological Bulletin,92, 434–452.
Hertzog, C. (2002). Metacognition in older adults: implications for application. In T. J. Perfect
& B. L. Schwartz (Eds.), Applied Metacognition (pp. 169–196). Cambridge, UK: Cambridge
University Press.
Hildreth, K., & Hill, D. (2003). Retrieval difficulty and retention of reactivated memories over
the first year of life. Developmental Psychobiology, 43, 216–229.
Hill, W. H., Borovsky, D., & Rovee-Collier, C. (1988). Continuities in infant memory develop-
ment over the first half-year. Developmental Psychobiology, 21, 43–62.
Hitch, G. J. (2006). Working memory in children: A cognitive approach. In E. Bialystok & F.
I. M. Craik (Eds.), Lifespan cognition: Mechanisms of change (pp. 112–127). New York:
Oxford University Press.
Hitch, G. J., Halliday, M. S., Dodd, A., & Littler, J. E. (1989). Development of rehearsal in
short-term memory: Differences between pictorial and spoken stimuli. British Journal of
Developmental Psychology, 7, 347–362.
Hitch, G. J., Halliday, M. S., & Littler, J. E. (1989). Item identification time and rehearsal rate as
predictors of memory span in children. Quarterly Journal of Experimental Psychology, 41,
321–337.
Hitch, G. J., Halliday, M. S., & Littler, J. E. (1993). Development of memory span for spo-
ken words: The role of rehearsal and item identification processes. British Journal of
Development Psychology, 11, 159–169.
Hitch, G. J., Halliday, M. S., Schaafstal, A. M., & Heffernan, R. M. (1991). Speech, “inner
speech”, and the development of short-term memory: Effects of picture-labeling on recall.
Journal of Experimental Child Psychology, 51, 220–234.
Hitch, G. J., Towse, J. N., & Hutton, U. (2001). What limits children’s working memory span?
Theoretical accounts and applications for scholastic development. Journal of Experimental
Psychology: General, 130, 184–198.
Hoffmann-Biencourt, A., Lockl, K., Schneider, W., Ackerman, R., & Koriat, A. (2010).
Self-paced study time as a cue for recall predictions across school age. British Journal of
Developmental Psychology, 28, 767–784.
References 349

Hofstadter, M., & Reznick, J. S. (1996). Response modality affects human infant delayed-
response performance. Child Development, 67, 646–658.
Holland Joyner, M., & Kurtz-Costes, B. (1997). Metamemory development. In N. Cowan & C.
Hulme (Eds.), The development of memory in childhood (pp. 275–300). Hove: Psychology Press.
Holliday, R. E. (2003). Reducing misinformation effects in children with cognitive interviews:
Dissociating recollection and familiarity. Child Development, 74, 728–751.
Holliday, R. E., & Hayes, B. K. (2001). Automatic and intentional processes in children’s eyewit-
ness suggestibility. Cognitive Development, 16, 617–636.
Holliday, R. E., & Hayes, B. K. (2002). Automatic and intentional processes in children’s recog-
nition memory: The reversed misinformation effect. Applied Cognitive Psychology, 16, 1–16.
Horgan, D. D., & Morgan, D. (1990). Chess expertise in children. Applied Cognitive Psychology,
4, 109–128.
Howe, M. L. (2000). The fate of early memories: Developmental science and the retention of
childhood experiences. Washington, DC: American Psychological Association.
Howe, M. L. (2006). Developmentally invariant dissociations in children’s true and false memo-
ries: Not all relatedness is created equal. Child Development, 77, 1112–1123.
Howe, M. L. (2008a). Visual distinctiveness and the development of children’s false memories.
Child Development, 79, 65–79.
Howe, M. L. (2008b). What is false memory development the development of? Comment on
Brainerd, Reyna, and Ceci (2008). Psychological Bulletin, 134, 768–772.
Howe, M. L., & Brainerd, C. J. (1989). Development of children’s long-term retention.
Developmental Review, 9, 301–340.
Howe, M. L., Brainerd, C. J., & Kingma, J. (1985). Development of organization in recall: A
stages-of-learning analysis. Journal of Experimental Child Psychology, 39, 230–251.
Howe, M. L., & Courage, M. L. (1993). On resolving the enigma of infantile amnesia.
Psychological Bulletin, 113, 305–326.
Howe, M. L., & Courage, M. L. (1997). The emergence and early development of autobiographi-
cal memory. Psychological Review, 104, 499–523.
Howe, M. L., Courage, M. L., & Bryant-Brown, L. (1993). Reinstating preschoolers’ memories.
Developmental Psychology, 29, 854–869.
Howe, M. L., Courage, M. L., & Peterson, C. (1994). How can I remember when I wasn’t
there: Long-term retention of traumatic experiences and emergence of the cognitive self.
Consciousness and Cognition, 3, 327–355.
Howe, M. L., Courage, M. L., & Rooksby, M. (2009). The genesis and development of auto-
biographical memory. In M. L. Courage & N. Cowan (Eds.), The development of memory in
infancy and childhood (pp. 177–196). Hove: Psychology Press.
Howe, M. L., & O’Sullivan, J. T. (1990). The development of strategic memory: Coordinating
knowledge, metamemory, and resources. In D. F. Bjorklund (Ed.), Children’s strategies:
Contemporary views of cognitive development (pp. 129–155). Hillsdale: Erlbaum.
Howe, M. L., & O’Sullivan, J. T. (1997). What children’s memories tell us about recalling our
childhoods: A review of storage and retrieval processes in the development of long-term
retention. Developmental Review, 17, 148–204.
Hsu, V. C. (2007). Time window effects on retention over the first year-and-a-half of live.
Unpublished doctoral dissertation, Rutgers University, New Brunswick, NJ.
Hudson, J. A. (1988). Children’s memory for atypical actions in script-based stories: Evidence
for a disruption effect. Journal of Experimental Child Psychology, 46, 159–173.
Hudson, J. A. (1990). The emergence of autobiographical memory in mother-child conversation.
In R. Fivush & J. A. Hudson (Eds.), Knowing and remembering in young children (pp. 166–
196). New York: Cambridge University Press.
Hudson, J. A., & Fivush, R. (1983). Categorical and schematic organization and the development
of retrieval strategies. Journal of Experimental Child Psychology, 36, 32–42.
Hudson, J. A., & Fivush, R. (1991). As time goes by: Sixth graders remember a kindergarten
experience. Applied Cognitive Psychology, 5, 347–360.
350 References

Hudson, J. A., Fivush, R., & Kuebli, J. (1992). Scripts and episodes: The development of event
memory. Applied Cognitive Psychology, 6, 483–505.
Hudson, J. A., & Grysman, A. (2014). Extending the life of a memory: Effects of reminders on
children’s long-term event memory. In P. J. Bauer & R. Fivush (Eds.), The Wiley handbook on
the development of children’s memory (Vol. I, pp. 255–284). Chichester: Wiley.
Hudson, J. A., & Mayhew, E. M. (2009). The development of memory for recurring events. In M.
L. Courage & N. Cowan (Eds.), The development of memory in infancy and childhood (pp.
69–91). Hove: Psychology Press.
Hudson, J. A., & Nelson, K. (1986). Repeated encounters of a similar kind: Effects of familiarity
on children’s autobiographic memory. Cognitive Development, 1, 253–271.
Hughes, C. (1998). Executive function in preschoolers: Links with theory of mind and verbal
ability. British Journal of Developmental Psychology, 16, 233–253.
Hughes, C., Ensor, R., Wilson, A., & Graham, A. (2010). Tracking executive functions across the
transition to school: A latent variable approach. Developmental Neuropsychology, 35, 20–36.
Huizinga, M., Dolan, C., & van der Molen, M. W. (2006). Age-related change in executive func-
tion: Developmental trends and a latent variable analysis. Neuropsychologia, 44, 2017–2036.
Hulme, C., Thompson, N., Muir, C., & Lawrence, A. (1984). Speech rate and the develop-
ment of spoken words: The role of rehearsal and item identification processes. Journal of
Experimental Child Psychology, 38, 241–253.
Hulme, C., & Tordoff, V. (1989). Working memory development: The effects of speech
rate, word length, and acoustic similarity on serial recall. Journal of Experimental Child
Psychology, 47, 72–87.
Hunter, M. A., Ross, H. S., & Ames, E. W. (1982). Preferences for familiar or novel toys: Effects
of familiarization time in 1-year-olds. Developmental Psychology, 18, 519–529.
Hunter, W. S. (1917). The delayed reaction in a child. Psychological Review, 24, 74–87.
Huttenlocher, J., & Burke, D. (1976). Why does memory span increase with age? Cognitive
Psychology, 8, 1–31.
Issacs, E. B., & Vargha-Khadem, F. (1989). Differential course of development of spatial and
verbal memory span: A normative study. British Journal of Developmental Psychology, 7,
377–380.
Istomina, Z. M. (1977). The development of voluntary memory in preschool-age children. Soviet
Psychology, 13, 5–64.
Jack, F., Simcock, G., & Hayne, H. (2012). Magic memories: Young children’s verbal recall after
a 6-year delay. Child Development, 83, 159–172.
Jacobs, J. (1887). Experiments on “Prehension”. Mind, 12, 75–79.
Jacoby, L. L. (1991). A process dissociation framework: Separating automatic from intentional
uses of memory. Journal of Memory and Language, 30, 513–541.
Jacoby, L. L., & Dallas, M. (1981). On the relationship between autobiographical memory and
perceptual learning. Journal of Experimental Psychology: General, 110, 306–340.
Jacques, S., & Zelazo, P. D. (2001). The Flexible Item Selection Task (FIST): A measure of exec-
utive function in preschoolers. Developmental Neuropsychology, 20, 573–591.
Jarrold, C., & Hall, D. (2013). The development of rehearsal in verbal short-term memory. Child
Development Perspectives, 7(3), 182–186.
Jaswal, V. K., & Dodson, C. S. (2009). Metamemory development: Understanding the role of
similarity in false memories. Cognitive Development, 80, 629–635.
Johnson, C. N., & Wellman, H. M. (1980). Children’s developing understanding of mental verbs:
“Remember”, “know”, and “guess”. Child Development, 51, 1095–1102.
Johnson, M. H. (1998). The neural basis of cognitive development. In W. Damon (Ed.-in-Chief)
& D. Kuhn & R. S. Siegler (Vol. Eds.), Handbook of child psychology: Vol. 2. Cognition, per-
ception, and language (5th ed., pp. 1-49). New York: Wiley.
Johnson, M. K., Hastroudi, S., & Lindsay, D. S. (1993). Source monitoring. Psychological
Bulletin, 114, 3–28.
Jones, C. H., & Pipe, M.-E. (2002). How quickly do children forget events? A systematic study
of children’s event reports as a function of delay. Applied Cognitive Psychology, 16, 755–768.
References 351

Jones, H. E. (1930). The retention of conditioned emotional reactions in infancy. Journal of

Genetic Psychology, 37, 485–498.
Joyner, M. H., & Kurtz-Costes, B. (1997). Metamemory development. In N. Cowan (Ed.), The
development of memory in childhood (pp. 275–300). London: London University College
Press.
Justice, E. M. (1985). Categorization as a preferred memory strategy: Developmental changes
during elementary school. Developmental Psychology, 21, 1105–1110.
Justice, E. M. (1986). Developmental changes in judgements of relative strategy effectiveness.
British Journal of Developmental Psychology, 4, 75–81.
Justice, E. M. (1989). Preschoolers’ knowledge and use of behaviors varying in strategic effec-
tiveness. Merrill-Palmer Quarterly, 35, 363–377.
Justice, E. M., Baker-Ward, L., Gupta, S., & Jannings, L. R. (1997). Means to the goal of remem-
bering: Developmental changes in awareness of strategy use-performance relations. Journal
of Experimental Child Psychology, 65, 293–314.
Justice, E. M., & Bray, N. W. (1979, March). The effects of context and feedback on metamemory
in young children. Paper presented at the Biennial Meeting of the Society for Research in
Child Development, San Francisco, CA.
Kagan, J. (1981). The second year: The emergence of self-awareness. Cambridge: Harvard
University Press.
Kail, R. V. (1979). Use of strategies and individual differences in children’s memory.
Developmental Psychology, 15, 251–255.
Kail, R. (1986). Sources of age differences in speed of processing. Child Development, 57,
969–987.
Kail, R. (1988). Developmental functions for speeds of cognitive processes. Journal of
Experimental Child Psychology, 45, 339–364.
Kail, R. V. (1990). The development of memory in children (3rd ed.). New York: W.H. Freeman.
Kail, R. (1991). Developmental change in speed of processing during childhood and adolescence.
Psychological Bulletin, 109, 490–501.
Kail, R. V. (1992). Processing speed, speech rate, and memory. Developmental Psychology,
28(5), 899–904.
Kail, R. V. (1997). Phonological skill and articulation time independently contribute to the devel-
opment of memory span. Journal of Experimental Child Psychology, 67, 57–68.
Kail, R. V. (2004). Cognitive development includes global and domain-specific processes.
Merrill-Palmer Quarterly, 50, 445–455.
Kail, R. V., & Ferrer, E. (2007). Processing speed in childhood and adolescence: Longitudinal
models for examining developmental change. Child Development, 78(6), 1760–1770.
Kail, R. V., & Hagen, J. W. (1977). Perspectives on the development of memory and cognition.
Hillsdale: Erlbaum.
Kail, R. V., McBride-Chang, C., Ferrer, E., Cho, J.-R., & Shu, H. (2013). Cultural differences in
the development of processing speed. Developmental Science, 16(3), 476–483.
Kail, R. V., & Miller, C. A. (2006). Developmental change in processing speed: Domain specific-
ity and stability during childhood and adolescence. Journal of Cognition and Development, 7,
119–137.
Kail, R. V., & Park, Y. (1994). Processing time, articulation time, and memory span. Journal of
Experimental Child Psychology, 57, 281–291.
Kail, R. V., & Salthouse, T. A. (1994). Processing speed as a mental capacity. Acta Psychologica,
86, 199–225.
Kail, R. V., & Strauss, M. S. (1984). The development of human memory: An historical over-
view. In R. Kail & N. E. Spear (Eds.), Comparative perspectives on the development of mem-
ory (pp. 3–22). Hillsdale: Erlbaum.
Káldy, Z., & Leslie, A. M. (2005). A memory span of one? Object identification in 6.5-month-old
infants. Cognition, 97(2), 153–177.
Kearins, J. M. (1981). Visual spatial memory in Australian Aboriginal children of desert regions.
Cognitive Psychology, 13, 434–460.
352 References

Kearins, J. M. (1983). A quotient of awareness. Education News, 18, 18–22.

Keast, A., Brewer, N., & Wells, G. L. (2007). Children’s metacognitive judgments in an eyewit-
ness identification task. Journal of Experimental Child Psychology, 97, 286–314.
Kee, D. W. (1994). Developmental differences in associative memory: Strategy use, mental
effort, and knowledge-access interactions. In H. W. Reese (Ed.), Advances in child develop-
ment and behaviour (Vol. 25, pp. 7–32). San Diego: Academic Press.
Kee, D. W., & Bell, T. S. (1981). The development of organizational strategies in the storage and
retrieval of categorical items in free-recall learning. Child Development, 52, 1163–1171.
Kee, D. W., & Davies, L. (1988). Mental effort and elaboration. A developmental analysis.
Contemporary Educational Psychology, 13, 221–228.
Keeney, F. J., Cannizzo, S. R., & Flavell, J. H. (1967). Spontaneous and induced verbal rehearsal
in a recall task. Child Development, 38, 953–966.
Kemps, E., de Rammelaere, S., & Desmet, T. (2000). The development of working memory:
Exploring the complementarity of two models. Journal of Experimental Child Psychology,
77, 89–109.
Keniston, A. H., & Flavell, J. H. (1979). A developmental study of intelligent retrieval. Child
Development, 50, 1144–1152.
Keppel, G. (1964). Verbal learning in children. Psychological Bulletin, 61, 63–80.
Kestner, J., & Borkowski, J. G. (1979). Children’s maintenance and generalization of an inter-
rogative learning strategy. Child Development, 50, 485–494.
Kirkpatrick, E. A. (1894). An experimental study of memory. Psychological Review, 1, 602–609.
Klenberg, L., Korkman, M., & Lahti-Nuuttila, P. (2001). Differential development of atten-
tion and executive functions in 3- to 12-year-old Finnish children. Developmental
Neuropsychology, 20, 407–428.
Klingberg, T. (2013). The learning brain: Memory and brain development in children. Oxford:
Oxford University Press.
Klingberg, T., Forssberg, H., & Westerberg, H. (2002). Increased brain activity in frontal and
parietal cortex underlies the development of visuospatial working memory capacity during
childhood. Journal of Cognitive Neuroscience, 14, 1–10.
Knopf, M. (1999). Development of memory for texts. In F. E. Weinert & W. Schneider (Eds.),
Individual development from 3 to 12: Findings from the Munich Longitudinal Study (pp. 106–
122). Cambridge: Cambridge University Press.
Kobasigawa, A. (1974). Utilization of retrieval cues by children in recall. Child Development, 45,
127–134.
Kobasigawa, A. (1977). Retrieval strategies in the development of memory. In R. V. Kail & J.
W. Hagen (Eds.), Perspectives on the development of memory and cognition (pp. 177–201).
Hillsdale: Erlbaum.
Kobasigawa, A., & Metcalf-Haggert, A. (1993). Spontaneous allocation of study time by first-
and third-grade children in a simple memory task. Journal of Genetic Psychology, 154,
223–235.
Kobasigawa, A., & Orr, R. R. (1973). Free recall and retrieval speed of categorized items by kin-
dergarten children. Journal of Experimental Child Psychology, 15(2), 187–192.
Koffka, K. (1921). Die Grundlagen der psychischen Entwicklung. Osterwieck: Zickfeld.
Koppenaal, R. J., Krull, A., & Katz, H. (1964). Age, interference and forgetting. Journal of
Experimental Child Psychology, 1, 360–375.
Koriat, A. (1993). How do we know what we know? The accessibility model of the feeling of
knowing. Psychological Review, 100, 609–639.
Koriat, A. (1997). Monitoring one's own knowledge during study: A cue-utilization approach to
judgments of learning. Journal of Experimental Psychology: General, 126, 349–370.
Koriat, A. (2007). Metacognition and consciousness. In P. D. Zelazo, M. Moscovitch, & E.
Thompson (Eds.), Cambridge handbook of consciousness (pp. 289–325). New York:
Cambridge University Press.
Koriat, A. (2008). Easy comes, easy goes? The link between learning and remembering and its
exploitation in metacognition. Memory & Cognition, 36, 416–428.
References 353

Koriat, A., Ackerman, R., Adiv, S., Lockl, K., & Schneider, W. (2014). The effects of goal-driven
and data-driven regulation on metacognitive monitoring during learning: A developmental
perspective. Journal of Experimental Psychology: General, 143(1), 386–403.
Koriat, A., Ackerman, R., Lockl, K., & Schneider, W. (2009a). The easily learned, easily-remem-
bered heuristic in children. Cognitive Development, 24, 169–182.
Koriat, A., Ackerman, R., Lockl, K., & Schneider, W. (2009b). The memorizing-effort heuris-
tic in judgments of learning: A developmental perspective. Journal of Experimental Child
Psychology, 102, 265–279.
Koriat, A., Bjork, R. A., Sheffer, L., & Bar, S. K. (2004). Predicting one’s own forgetting: The
role of experience-based and theory-based processes. Journal of Experimental Psychology:
General, 133, 643–656.
Koriat, A., & Goldsmith, M. (1996). Monitoring and control processes in the strategic regulation
of memory accuracy. Psychological Review, 103, 409–517.
Koriat, A., Goldsmith, M., Schneider, W., & Nakash-Dura, M. (2001). The credibility of chil-
dren’s testimony: Can children control the accuracy of their memory reports? Journal of
Experimental Child Psychology, 79, 405–437.
Koriat, A., Ma’ayan, H., & Nussinson, R. (2006). The intricate relationships between monitoring
and control in metacognition: Lessons for the cause-and-effect relation between subjective
experience and behavior. Journal of Experimental Psychology: General, 135, 36–69.
Koriat, A., & Shitzer-Reichert, R. (2002). Metacognitive judgments and their accuracy. Insights
from the processes underlying judgments of learning in children. In P. Chambres, M. Izaute, &
P.-J. Marescaux (Eds.), Metacognition: Process, function, and use (pp. 1–18). Boston: Kluwer.
Körkel, J., & Schneider, W. (1992). Domain-specific versus metacognitive knowledge effects on
text recall and comprehension. In M. Carretero, M. Pope, R.-J. Simons, & J. Pozo (Eds.),
Learning and instruction: European research in an international context (Vol. 3, pp. 311–
324). Oxford: Pergamon Press.
Koski, J., Olson, I. R., & Newcombe, N. (2013). Tracking the eyes to see what children remem-
ber. Memory, 21, 396–407.
Krebs, S. S., & Roebers, C. M. (2010). Children’s strategic regulation, metacognition, metacog-
nitive monitoring, and control processes during test taking. British Journal of Educational
Psychology, 80, 325–340.
Kreutzer, M. A., Leonard, C., & Flavell, J. H. (1975). An interview study of children‘s knowl-
edge about memory. Monographs of the Society for Research in Child Development, 40
(Serial No. 159).
Kron-Sperl, V., Schneider, W., & Hasselhorn, M. (2008). The development and effectiveness of
memory strategies in kindergarten and elementary school: Findings from the Würzburg and
Göttingen longitudinal studies. Cognitive Development, 23, 79–104.
Kuebli, J., & Fivush, R. (1994). Children’s representation and recall of event alternatives. Journal
of Experimental Child Psychology, 58, 25–45.
Kuhn, D. (1999). Metacognitive development. In L. Balter & C. S. Tamis-LeMonda (Eds.), Child
psychology: A handbook of contemporary issues (pp. 259–286). Philadelphia: Psychology Press.
Kuhn, D. (2000a). Theory of mind, metacognition, and reasoning: A life-span perspective. In
P. Mitchell & K. J. Riggs (Eds.), Children's reasoning and the mind (pp. 301–326). Hove:
Psychology Press.
Kuhn, D. (2000b). Does memory development belong on an endangered topic list? Child
Development, 71, 21–25.
Kunzinger, E. L. (1985). A short-term longitudinal study of memorial development during early
grade school. Developmental Psychology, 21, 642–646.
Kurtz, B. E. (1990). Cultural influences on children’s cognitive and metacognitive development.
In W. Schneider & F. E. Weinert (Eds.), Interactions among aptitudes, strategies, and knowl-
edge in cognitive performance (pp. 177–199). New York: Springer.
Kurtz, B. E., & Borkowski, J. G. (1984). Children’s metacognition: Exploring relations among
knowledge, process, and motivational variables. Journal of Experimental Child Psychology,
37, 335–354.
354 References

Kurtz, B. E., Reid, M. K., Borkowski, J. G., & Cavanaugh, J. C. (1982). On the reliability and
validity of children’s metamemory. Bulletin of the Psychonomic Society, 19, 137–140.
Kurtz, B. E., Schneider, W., Carr, M., Borkowski, J. G., & Rellinger, E. (1990). Strategy instruc-
tion and attributional beliefs in West Germany and the United States: Do teachers foster meta-
cognitive development? Contemporary Educational Psychology, 15, 268–283.
Kwon, H., Reiss, A. L., & Menon, V. (2002). Neural basis of protracted developmental changes
in visuo-spatial working memory. Proceedings of the National Academy of Sciences, 99(20),
13336–13341.
Lah, S., Epps, A., Levick, W., & Parry, L. (2011). Implicit and explicit memory outcome in chil-
dren who have sustained severe traumatic brain injury: Impact of age at injury (preliminary
findings). Brain Injury, 25, 44–52. doi:10.3109/02699052.2010.531693.
Laible, D., & Thompson, R. (2000). Mother-child discourse, attachment security, shared positive
affect, and early conscience development. Child Development, 71, 1424–1440.
Landis, T. Y. (1982). Interactions between text and prior knowledge in children’s memory for
prose. Child Development, 53, 811–814.
Lange, G. (1973). The development of conceptual and rote recall skills among school age chil-
dren. Journal of Experimental Child Psychology, 15, 394–406.
Lange, G. (1978). Organization-related processes in children’s recall. In P. A. Ornstein (Ed.),
Memory development in children (pp. 101–128). Hillsdale: Erlbaum.
Lange, G., & Griffith, S. B. (1977). The locus of organization failures in children‘s recall. Child
Development, 48, 1498–1502.
Lange, G., Guttentag, R. E., & Nida, R. E. (1990). Relationships between study organization,
retrieval organization, and general and strategy-specific memory knowledge in young chil-
dren. Journal of Experimental Child Psychology, 49, 126–146.
Lange, G., & Jackson, P. (1974). Personal organization in children’s free recall. Child
Development, 45, 1060–1067.
Lange, G., McKinnon, C. E., & Nida, R. E. (1989). Knowledge, strategy, and motivational con-
tributions to preschool children’s object recall. Developmental Psychology, 25, 772–779.
Lange, G., & Pierce, S. H. (1992). Memory-strategy learning and maintenance in preschool chil-
dren. Developmental Psychology, 28, 453–462.
Larkina, M., & Güler, O. E. (2014). Socialization of deliberate and strategic remembering. In P.
J. Bauer & R. Fivush (Eds.), The Wiley handbook on the development of children’s memory
(Vol. II, pp. 895–919). Chichester: Wiley.
Lee, K., Bull, R., & Ho, R. M. H. (2013). Developmental change in executive functioning. Child
Development, 84, 1933–1953.
Lehmann, M., & Hasselhorn, M. (2007). Variable memory strategy use in children’s adaptive
intratask learning behavior: Developmental changes and working memory influences in free
recall. Child Development, 78, 1068–1082.
Lehmann, M., & Hasselhorn, M. (2010). The dynamics of free recall and their relation to
rehearsal between 8 and 10 years of age. Child Development, 81, 1006–1020.
Lehmann, M., & Hasselhorn, M. (2012). Rehearsal dynamics in elementary school children.
Journal of Experimental Child Psychology, 111, 552–560.
Lehto, J. E., Juujärvi, P., Kooistra, L., & Pulkkinen, L. (2003). Dimensions of executive function-
ing: Evidence from children. British Journal of Developmental Psychology, 21(1), 59–80.
Leichtman, M. D., & Ceci, S. J. (1993). The problem of infantile amnesia: Lessons from fuzzy-
trace theory. In M. L. Howe & R. Pasnak (Eds.), Emerging themes in cognitive development
(Vol. 1, pp. 195–213)., Foundations New York: Springer.
Leichtman, M. D., & Ceci, S. J. (1995). The effects of stereotypes and suggestions on preschool-
ers’ reports. Developmental Psychology, 31, 568–578.
Leichtman, M., Wang, Q., & Pillemer, D. P. (2003). Cultural variation in interdependence and
autobiographical memory. In R. Fivush & C. Haden (Eds.), Autobiographical memory and
the construction of a narrative self: Developmental and cultural perspectives (pp. 73–98).
Mahwah: Erlbaum.
References 355

Leontjev, A. N. (1931). Development of memory. Moscow: Uchpedgiz.

Leslie, A. M., & Káldy, Z. (2007). Things to remember: Linits, codes, and the development of
object working memory in the first year. In L. M. Oakes & P. J. Bauer (Eds.), Short- and
long-term memory in infancy and early childhood (pp. 103–125). New York: Oxford
University Press.
Levin, J. R., & Pressley, M. (1978). A test of the developmental imagery hypothesis in children's
associative learning. Journal of Educational Psychology, 70(5), 691–694.
Lewis, M., & Brooks-Gunn, J. (1979). Social cognition and the acquisition of self. New York:
Plenum Press.
Liben, L. S., & Bowman, C. R. (2014). The development of memory from a Piagetian perspec-
tive. In P. J. Bauer & R. Fivush (Eds.), The Wiley handbook on the development of children’s
memory (Vol. I, pp. 65–86). Chichester: Wiley.
Lindberg, M. A. (1980). Is knowledge base development a necessary and sufficient condition for
memory development? Journal of Experimental Child Psychology, 30, 401–410.
Lindberg, M. A., Keiffer, J., & Thomas, S. W. (2000). Eyewitness testimony for physical abuse
as a function of personal experience, development, and focus of study. Journal of Applied
Developmental Psychology, 21, 555–591.
Lindsay, D. S., Johnson, M. K., & Kwon, P. (1991). Developmental changes in memory source
monitoring. Journal of Experimental Child Psychology, 52, 297–318.
Little, A. H. (1970). Eyelid conditioning in the human infant as a function of the interstimulus
interval. Unpublished master’s thesis, Brown University, Providence, RI.
Little, A. H., Lipsitt, L. P., & Rovee-Collier, C. (1984). Classical conditioning and reten-
tion of the infant’s eyelid response: Effects of age and interstimulus interval. Journal of
Experimental Child Psychology, 37, 512–524.
Lloyd, M. E., & Miller, J. K. (2014). Implicit memory. In P. J. Bauer & R. Fivush (Eds.), The
Wiley handbook on the development of children’s memory (Vol. I, pp. 336–366). Chichester:
Wiley.
Lloyd, M. E., & Newcombe, N. S. (2009). Implicit memory in childhood: reassessing devel-
opmental invariance. In M. L. Courage & N. Cowan (Eds.), The development of memory in
infancy and childhood (pp. 93–113). Hove: Psychology Press.
Lobsien, M. (1902). Experimentelle Untersuchungen über die Gedächtnisentwicklung bei
Schulkindern. Zeitschrift für Psychologie und Physiologie der Sinnesorgane, 27, 34–76.
Lockl, K., & Schneider, W. (2002a). Developmental trends in children’s feeling-of-knowing
judgements. International Journal of Behavioral Development, 26, 327–333.
Lockl, K., & Schneider, W. (2002b). Zur Entwicklung des selbstregulierten Lernens im
Grundschulalter: Zusammenhänge zwischen Aufgabenschwierigkeit und Lernzeiteinteilung
[The development of self-regulated learning in elementary school children: Associations
between task difficulty and allocation of study time]. Psychologie in Erziehung und
Unterricht, 49, 3–16.
Lockl, K., & Schneider, W. (2003). Metakognitive Überwachungs- und Kontrollprozesse bei der
Lernzeiteinteilung von Kindern [Monitoring and control processes in children’s allocation of
study time]. Zeitschrift für Pädagogische Psychologie, 17, 173–183.
Lockl, K., & Schneider, W. (2004). The effects of incentives and instructions on children‘s allo-
cation of study time. European Journal of Developmental Psychology, 1(2), 153–169.
Lockl, K., & Schneider, W. (2006). Precursors of metamemory in young children: The role of
theory of mind and metacognitive vocabulary. Metacognition and Learning, 1, 15–31.
Lockl, K., & Schneider, W. (2007). Knowledge about the mind: Links between theory of mind
and later metamemory. Child Development, 78(1), 148–167.
Lodico, M. G., Ghatala, E. S., Levin, J. R., Pressley, M., & Bell, J. A. (1983). The effects of
strategy-monitoring on children’s selection of effective memory strategies. Journal of
Experimental Child Psychology, 35, 263–277.
Loftus, E. F., & Pickrell, J. E. (1995). The formation of false memories. Psychiatric Annals, 25,
720–725.
356 References

Logie, R. H. (1995). Visuo-spatial working memory. Hillsdale: Erlbaum.

Logie, R. H., & Baddeley, A. D. (1990). Imagery and working memory. In P. J. Hampson, D. E.
Marks, & J. T. E. Richardson (Eds.), Imagery: Current developments (pp. 103–128). London:
Routledge.
Logie, R. H., & Pearson, D. G. (1997). The inner eye and the inner scribe of visuo-spatial work-
ing memory. Evidence from developmental fractionation. European Journal of Cognitive
Psychology, 9, 241–257.
Lorsbach, T. C., Katz, G. A., & Cupak, A. J. (1998). Developmental differences in the ability to
inhibit the initial misinterpretation of garden path passages. Journal of Experimental Child
Psychology, 71, 275–296.
Lorsbach, T. C., & Morris, A. K. (1991). Direct and indirect testing of picture memory in second
and sixth grade children. Contemporary Educational Psychology, 16, 18–27.
Lorsbach, T. C., & Reimer, J. F. (1997). Developmental changes in the inhibition of previously
relevant information. Journal of Experimental Child Psychology, 64, 317–342.
Lourenco, S. F., & Frick, A. (2014). Remembering where: The origins and early development of
spatial memory. In P.J. Bauer & R. Fivush (Eds.), The Wiley handbook on the development of
children’s memory (Vol. I., pp. 367–393). Chichester: Wiley.
Luciana, M., Conklin, H. M., Hooper, C. J., & Yarger, R. S. (2005). The development of non-
verbal working memory and executive control processes in adolescents. Child Development,
76(3), 697–712.
Luciana, M., & Nelson, C. (1998). The functional emergence of prefrontally-guided working
memory systems in four- to eight-year-old children. Neuropsychologia, 36, 273–293.
Luck, S. J., & Vogel, E. K. (1997). The capacity of visual working memory for features and con-
junctions. Nature, 390, 279–281.
Lukowski, A. F., & Bauer, P. J. (2014). Long-term memory in infancy and early childhood. In P.
J. Bauer & R. Fivush (Eds.), The Wiley handbook on the development of children’s memory
(Vol. I, pp. 230–254). Chichester: Wiley.
Luo, Y., Baillargeon, R., Brueckner, L., & Munakata, Y. (2003). Reasoning about a hidden object after
a delay: Evidence for robust representations in 5-month-old infants. Cognition, 88, B23–B32.
Lyons, W. (1986). The disappearance of introspection. Cambridge: MIT Press.
Malmberg, K. J. (2008). Recognition memory: A review of the critical findings and an integrated
theory for relating them. Cognitive Psychology, 57, 335–384.
Mandler, J. M. (1983). Representation. In J. H. Flavell & E. M. Markman (Eds.), Handbook of
child psychology (Vol. 3, pp. 420–493). New York: Wiley.
Mandler, J. M. (1990). Recall of events by preverbal children. In A. Diamond (Ed.), The develop-
ment and neural bases of higher cognitive functions (Vol. 608, pp. 485-503), Annals of the
New York Academy of Sciences. New York: New York Academy of Sciences.
Mandler, J. M. (2007). How do we remember? Let me count the ways. In L. M. Oakes & P. J.
Bauer (Eds.), Short- and long-term memory in infancy and early childhood: Taking the first
steps towards remembering (pp. 271–290). New York: Oxford University Press.
Mandler, J. M., Scribner, S., Cole, M., & DeForest, M. (1980). Cross-cultural invariance in story
recall. Child Development, 51, 19–26.
Marche, T. A. (1999). Memory strength affects reporting of misinformation. Journal of
Experimental Child Psychology, 73(1), 45–71.
Markman, E. M. (1973). Factors affecting the young child’s ability to monitor his memory.
Unpublished doctoral dissertation, University of Pennsylvania.
Martin, V. L., & Pressley, M. (1991). Elaborative-interrogation effects depend on the nature of
the question. Journal of Educational Psychology, 83, 113–119.
Masur, E. F., McIntyre, C. W., & Flavell, J. H. (1973). Developmental changes in apportion-
ment of study time among items in a multitrial free recall task. Journal of Experimental Child
Psychology, 15, 237–246.
Matthews, A., Ellis, A. E., & Nelson, C. A. (1996). Development of preterm and full-term
infant ability on AB, recall memory, transparent barrier detour, and means-end tasks. Child
Development, 67, 2658–2676.
References 357

Mazzoni, G. (1998). Memory suggestibility and metacognition in child eyewitness testi-

mony: The roles of source monitoring and self-efficacy. European Journal of Psychology of
Education, 13, 43–60.
Mazzoni, G., & Nelson, T. O. (1995). Judgments of learning are affected by the kind of encoding
in ways that cannot be attributed to the level of recall. Journal of Experimental Psychology:
Learning, Memory and Cognition, 21, 1263–1274.
McAuley, T., & White, D. A. (2011). A latent variables examination of processing speed,
response inhibition, and working memory during typical development. Journal of
Experimental Child Psychology, 108, 453–468.
McCabe, A., & Peterson, C. (1991). Getting the story: A longitudinal study of parental styles
in eliciting narratives and developing narrative skill. In A. McCabe & C. Peterson (Eds.),
Developing narrative structure (pp. 217–253). Hillsdale: Erlbaum.
McCall, R. B., & Carriger, M. S. (1993). A meta-analysis of infant habituation and recognition
memory performance as predictors of later IQ. Child Development, 64, 57–79.
McCauley, M. R., & Fisher, R. P. (1995). Facilitating children’s eyewitness recall with the
revised cognitive interview. Journal of Applied Psychology, 80, 510–516.
McCoon, G., & Ratcliff, R. (1992). Inference during reading. Psychological Review, 99, 440–466.
McDonough, L., & Mandler, J. M. (1994). Very long-term recall in infants: Infantile amnesia
reconsidered. Memory, 2, 339–352.
McDonough, L., Mandler, J., McKee, R., & Squire, L. (1995). The deferred imitation task as a
nonverbal measure of declarative memory. Proceedings of the National Academy of Science,
92, 7580–7584.
McGuigan, F., & Salmon, K. (2004). The time to talk: the influence of the timing of adult—child
talk on children’s event memory. Child Development, 75, 669–686.
McGuigan, F., & Salmon, K. (2005). Pre-event discussion and recall of a novel event: How are
children best prepared? Journal of Experimental Child Psychology, 91, 342–366.
Meacham, J. A. (1977). Soviet investigations of memory development. In R. V. Kail & J. W.
Hagen (Eds.), Perspectives on the development of memory and cognition (pp. 273–295).
Hillsdale: Erlbaum.
Means, M., & Voss, J. (1985). Star wars: A developmental study of expert and novice knowledge
structures. Journal of Memory and Language, 24, 746–757.
Mecklenbräuker, S., Hupbach, A., & Wippich, W. (2003). Age-related improvements in a concep-
tual implicit memory test. Memory & Cognition, 31, 1208–1217.
Mecklenbräuker, S., & Wippich, W. (1995). Implizites Gedächtnis bei Kindern: Bleiben
auch bei konzeptgesteuerten Aufgaben alterskorrelierte Differenzen aus? Zeitschrift für
Entwicklungspsychologie und Pädagogische Psychologie, 27, 29–46.
Meichenbaum, D. M., Burland, S., Gruson, L., & Cameron, R. (1985). Metacognitive assessment.
In S. Yussen (Ed.), The growth of reflection in children (pp. 3–30). New York: Academic Press.
Melot, A.-M. (1998). The relationship between metacognitive knowledge and metacognitive
experiences: Acquisition and re-elaboration. European Journal of Psychology of Education,
13, 75–89.
Meltzoff, A. N. (1985). Immediate and deferred imitation in fourteen and twenty-four-month-old
infants. Child Development, 56, 62–72.
Meltzoff, A. N. (1988). Imitation of televised models by infants. Child Development, 59,
1221–1229.
Meltzoff, A. N. (1995). What infant memory tells us about infantile amnesia: Long-term recall
and deferred imitation. Journal of Experimental Child Psychology, 59, 497–515.
Merritt, K. A., Ornstein, P. A., & Spicker, B. (1994). Children’s memory for a salient medical
procedure: Implications for testimony. Pediatrics, 94, 12–23.
Metcalfe, J. (2002). Is study time allocated selectively to a region of proximal learning? Journal
of Experimental Psychology: General, 131, 349–363.
Meumann, E. (1907). Sechste Vorlesung: Entwicklung der einzelnen geistigen Fähigkeiten beim
Kinde (Fortsetz.). Vorlesungen zur Einführung in die Experimentelle Pädagogik (Bd. 1).
Leipzig: Wilhelm Engelmann.
358 References

Mevarech, Z., & Fridkin, S. (2006). The effects of IMPROVE on mathematical knowledge, math-
ematical reasoning, and metacognition. Metacognition and Learning, 1, 85–97.
Mevarech, Z., & Kramarski, B. (1997). IMPROVE: A multidimensional method for teaching math-
ematics in heterogeneous classrooms. American Educational Research Journal, 34, 365–394.
Miller, G. A. (1956). The magical number seven, plus or minus two: Some limits on our capacity
for processing information. Psychological Review, 63, 81–97.
Miller, G. A., Galanter, E., & Pribram, K. A. (1960). Plans and the structure of behaviour. New
York: Holt, Rhinehart & Winston.
Miller, J. K., & Lloyd, M. E. (2011). More questions than answers: The development of the flu-
ency heuristic during childhood. In P. Higham & J. A. LeBoe (Eds.), Constructions of remem-
bering and metacognition: Essays in honor of Bruce Whittlesea (pp. 28–39). Basingstoke:
Palgrave Macmillan.
Miller, P. H. (1990). The development of strategies of selective attention. In D. F. Bjorklund
(Ed.), Children’s strategies: Contemporary views of cognitive development (pp. 157–184).
Hillsdale: Erlbaum.
Miller, P. H. (2014). The history of memory development research: Remembering our roots. In
P. J. Bauer & R. Fivush (Eds.), The Wiley handbook on the development of children’s memory
(Vol. I, pp. 19–40). Chichester: Wiley.
Miller, P. H., & Seier, W. L. (1994). Strategy utilization deficiencies in children: when, where,
and why. In H. W. Reese (Ed.), Advances in Child Development and Behavior (Vol. 25, pp.
107–156). San Diego: Academic Press.
Miller, P. H., Seier, W. L., Probert, J. S., & Aloise, P. A. (1991). Age differences in the capacity
demands of a strategy among spontaneously strategic children. Journal of Experimental Child
Psychology, 52, 149–165.
Miller, P. J., Wiley, A., Fung, H., & Liang, C. (1997). Personal storytelling as a medium of
socialization in Chinese and American families. Child Development, 68, 557–568.
Miscione, J. L., Marvin, R. S., O'Brien, R. G., & Greenberg, M. T. (1978). A developmen-
tal study of preschool children's understanding of the words "know" and "guess". Child
Development, 48, 1107–1113.
Mistry, J. (1997). The development of remembering in cultural context. In N. Cowan (Ed.), The
development of memory in childhood (pp. 343–368). Hove: Psychology Press.
Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A. H., Howerter, A., & Wagner, T. D.
(2000). The unity and diversity of executive functions and their contributions to complex
“frontal lobe” tasks: A latent variable analysis. Cognitive Psychology, 41, 49–100.
Moely, B. E., Hart, S. S., Leal, L., Santulli, K. A., Rao, N., Johnson, T., et al. (1992). The teach-
er’s role in facilitating memory and study strategy development in the elementary school
classroom. Child Development, 63, 653–672.
Moely, B. E., Hart, S. S., Santulli, K., Leal, L., Johnson, T., Rao, N., et al. (1986). How do teach-
ers teach memory skills? Educational Psychologist, 21, 55–72.
Moely, B. E., Olson, F. A., Halwes, T. G., & Flavell, J. H. (1969). Production deficiency in young
children’s clustered recall. Developmental Psychology, 1, 26–34.
Moely, B. E., Santulli, K. A., & Obach, M. S. (1995). Strategy instruction, metacognition, and
motivation in the elementary school classroom. In F. Weinert & W. Schneider (Eds.), Memory
performance and competencies: Issues in growth and development (pp. 301–321). Hillsdale:
Erlbaum.
Moore, C., Bryant, D., & Furrow, D. (1989). Mental terms and the development of certainty.
Child Development, 60(1), 167–171.
Morris, G., Baker-Ward, L., & Bauer, P. J. (2010). What remains of that day: The survival of chil-
dren’s autobiographical memories across time. Applied Cognitive Psychology, 24, 527–544.
Morrison, F. J., & Lord, C. (1982). Age differences in recall of categorized material:
Organization or retrieval? The Journal of Genetic Psychology, 141, 233–241.
Moynahan, E. D. (1978). Assessment and selection of paired associate strategies: A developmen-
tal study. Journal of Experimental Child Psychology, 26, 257–266.
References 359

Mullen, M. K., & Yi, S. (1995). The cultural context of talk about the past: Implications for the
development of autobiographical memory. Cognitive Development, 10, 407–419.
Murachver, T., Pipe, M.-E., Gordon, R., Owens, J. L., & Fivush, R. (1996). Do, show, and tell:
Children’s event memories acquired through direct experience, observation, and stories. Child
Development, 67, 3029–3044.
Murphy, K., McKone, E., & Slee, J. (2003). Dissociations between implicit and explicit memory
in children: The role of strategic processing and the knowledge base. Journal of Experimental
Child Psychology, 84, 124–165.
Murphy, M. D., Schmitt, F. A., Caruso, M. J., & Sanders, R. E. (1987). Metamemory in older
adults: the role of monitoring in serial recall. Psychology and Aging, 2, 331–339.
Myers, M., & Paris, S. G. (1978). Children’s metacognitive knowledge about reading. Journal of
Educational Psychology, 70, 680–690.
Myers, N., & Perlmutter, M. (1978). Memory in the years from two to five. In P. A. Ornstein
(Ed.), Memory development in children (pp. 191–218). Hillsdale: Erlbaum.
Nagy, L. (1930). Experimentelle Untersuchungen über die Entwicklung der Gedächtnis- und
Denkformen der Knaben und Mädchen zwischen 7-19 Jahren. In H. Volkelt (Ed.), Bericht
über den 11. Kongreß für Experimentelle Psychologie in Wien vom 9.-13. April 1929 (pp.
113–122). Jena: Gustav Fischer.
Naito, M. (1990). Repetition priming in children and adults: Age-related dissociation between
implicit and explicit memory. Journal of Experimental Child Psychology, 50, 462–476.
National Institute of Child Health and Human Development (NICHD). (2000). Report of the
National Reading Panel. Teaching children to read: An evidence-based assessment of the scien-
tific research literature on reading and its implications for reading instruction: Reports of the
subgroups (NIH Publication No. 00-4754). Washington, DC: U.S. Government Printing Office.
Naus, M. J., & Ornstein, P. A. (1983). Development of memory strategies: Analysis, questions,
and issues. In M. T. H. Chi (Ed.), Trends in memory development research (Vol. 9, pp. 1–30).
Basel: Karger.
Naus, M. J., Ornstein, P. A., & Aivano, S. (1977). Developmental changes in memory: The
effects of processing time and rehearsal instructions. Journal of Experimental Child
Psychology,23, 237–251.
Naus, M. J., Ornstein, P. A., & Kreshtool, K. (1977). Developmental differences in recall and
recognition: The relationship between rehearsal and memory as test expectation changes.
Journal of Experimental Child Psychology, 23(2), 252–265.
Neisser, U. (1962). Cultural and cognitive discontinuity. In T. E. Gladwin & W. Sturtevant (Eds.),
Anthropology and human behavior (pp. 54–71). Washington, DC: Anthropological Society of
Washington DC.
Neisser, U. (1967). Cognitive psychology. New York: Appleton-Century-Crofts.
Nelson, C. A. (1997). The neurobiological basis of early memory development. In N. Cowan
(Ed.), The development of memory in childhood (pp. 41–82). Hove: Psychology Press.
Nelson, C. A. (2000). Neural plasticity and human development: The role of early experience in
sculpting memory systems. Developmental Science, 3, 115–136.
Nelson, C. A., Monk, C. S., Lin, J., Carver, L. J., Thomas, K. M., & Truwit, C. L. (2000).
Functional neuroanatomy of spatial working memory in children. Developmental Psychology,
36, 109–116.
Nelson, K. (1986). Event knowledge: Structure and function in development. Hillsdale: Erlbaum.
Nelson, K. (1993). The psychological and social origins of autobiographical memory.
Psychological Science, 4, 7–14.
Nelson, K. (1994). Long-term retention of memory for preverbal experience: Evidence and
implications. Memory, 2, 467–475.
Nelson, K. (1996). Language in cognitive development: Emergence of the mediated mind. New
York: Cambridge University Press.
Nelson, K. (2014). Sociocultural theories of memory development. In P. J. Bauer & R. Fivush
(Eds.), The Wiley handbook on the development of children’s memory (Vol. I, pp. 87–108).
Chichester: Wiley.
360 References

Nelson, K., & Fivush, R. (2004). The emergence of autobiographical memory: A social cultural
developmental theory. Psychological Review, 111, 486–511.
Nelson, K., Fivush, R., Hudson, J., & Lucariello, J. (1983). Scripts and the development of mem-
ory. In M. T. H. Chi (Ed.), Trends in memory development research (Vol. 9, pp. 52–70). New
York: Karger.
Nelson, K., & Gruendel, J. (1979). Generalized event representations: Basic building blocks of
cognitive development. In A. L. Brown & M. E. Lamb (Eds.), Advances in developmental
psychology (Vol. 1, pp. 131–158). Hillsdale: Erlbaum.
Nelson, K., & Gruendel, J. (1981). Generalized event representations: Basic building blocks of
cognitive development. In M. E. Lamb & A. L. Brown (Eds.), Advances in developmental
psychology (Vol. 1, pp. 21–46). Hillsdale: Erlbaum.
Nelson, T. O. (1996). Consciousness and metacognition. American Psychologist, 51, 102–116.
Nelson, T. O., Dunlosky, J., Graf, A., & Narens, L. (1994). Utilization of metacognitive judg-
ments in the allocation of study during multitrial learning. Psychological Science, 5, 207–213.
Nelson, T. O., & Narens, L. (1980). Norms of 300 general-information questions: Accuracy of
recall, latency of recall, and feeling-of-knowing ratings. Journal of Verbal Learning and
Behavior, 19, 338–368.
Nelson, T. O., & Narens, L. (1990). Metamemory: A theoretical framework and new findings. In
G. Bower (Ed.), The psychology of learning and motivation: Advances in research and theory
(Vol. 26, pp. 125–173). New York: Academic Press.
Nelson, T. O., & Narens, L. (1994). Why investigate metacognition? In J. Metcalfe & A. P.
Shimamura (Eds.), Metacognition. Knowing about knowing (pp. 1–25). Cambridge: MIT Press.
Netschajeff, A. (1900). Experimentelle Untersuchungen über die Gedächtnisentwicklung bei
Schulkindern. Zeitschrift für Psychologie und Physiologie der Sinnesorgane, 24, 322–351.
Netschajeff, A. (1902). Über Memorieren. Sammlung von Abhandlungen aus dem Gebiete der
Pädagogischen Psychologie und Physiologie, 5, 293–329.
Newcombe, N. S., & Crawley, S. L. (2007). To have and have not: What do we mean when we
talk about long-term memory development? In L. M. Oakes & P. J. Bauer (Eds.), Short- and
long-term memory in infancy and early childhood: Taking the first steps towards remember-
ing (pp. 291–313). New York: Oxford University Press.
Newcombe, N., & Fox, N. A. (1994). Infantile amnesia: Through a glass darkly. Child
Development, 65, 31–40.
Newcombe, R., & Reese, E. (2004). Evaluations and orientations in mother child narratives
as a function of attachment security: A longitudinal investigation. International Journal of
Behavioral Development, 28(3), 230–245.
Nicolson, R. (1981). The relationship between memory span and processing speed. In M. J.
Friedman, J. P. Das, & N. O’Connor (Eds.), Intelligence and learning (pp. 179–183). New
York: Plenum.
Oakes, L. M., & Bauer, P. J. (Eds.). (2007). Short- and long-term memory in infancy and early
childhood: Taking the first steps toward remembering. New York: Oxford University Press.
Oakes, L. M., Hurley, K. B., Ross-Sheehy, S., & Luck, S. J. (2011). Developmental changes in
infants’ visual short-term memory for location. Cognition, 118, 293–305.
Oakes, L. M., & Luck, S. J. (2014). Short-term memory in infancy. In P. J. Bauer & R. Fivush
(Eds.), The Wiley handbook on the development of children‘s memory (Vol. I, pp. 157–180).
Chichester: Wiley.
Oakes, L. M., Ross-Sheehy, S., & Luck, S. J. (2006). Rapid development of binding in visual
short-term memory. Psychological Science, 17, 781–787.
Oakes, L. M., Ross-Sheehy, S., & Luck, S. J. (2007). The development of visual short-term
memory in infancy. In L. M. Oakes & P. J. Bauer (Eds.), Short- and long-term memory in
infancy and childhood: Taking the first steps toward remembering (pp. 75–102). New York:
Oxford University Press.
Ofen, N., Kao, Y.-C., Sokol-Hessner, P., Kim, H., Whitfield-Gabrieli, S., & Gabrieli, J. D.
E. (2007). Development of the declarative memory system in the human brain. Nature
Neuroscience, 10, 1198–1205.
References 361

Offner, M. (1924). Das Gedächtnis. Berlin: Reuther & Reichard.

Olson, I. R., & Newcombe, N. S. (2014). Binding together the elements of episodes: Relational
memory and the developmental trajectory of the hippocampus. In P. J. Bauer & R. Fivush
(Eds.), The Wiley handbook on the development of children’s memory (Vol. I, pp. 285–308).
Chichester: Wiley.
Ornstein, P. A. (Ed.). (1978). Memory development in children. Hillsdale, NJ: Erlbaum.
Ornstein, P. A. (1999). Comments: Toward an understanding of the development of memory. In
F. E. Weinert & W. Schneider (Eds.), Individual development from 3 to 12: Findings from the
Munich Longitudinal Study (pp. 94-105). New York: Cambridge University Press.
Ornstein, P. A., Baker-Ward, L., Gordon, B. N., & Merritt, K. A. (1997). Children’s memory for
medical experiences: Implications for testimony. Applied Cognitive Psychology, 11, 87–104.
Ornstein, P. A., Baker-Ward, L., Gordon, B. N., Pelphrey, K. A., Tyler, C. S., & Gramzow, E.
(2006). The influence of prior knowledge and repeated questioning on children’s long-term
retention on the details of a pediatric examination. Developmental Psychology, 42, 332–344.
Ornstein, P. A., Baker-Ward, L., & Naus, M. J. (1988). The development of mnemonic skill. In
M. Weinert & M. Perlmutter (Eds.), Memory development: Universal changes and individual
differences (pp. 31–50). Hillsdale: Erlbaum.
Ornstein, P. A., & Elischberger, H. B. (2004). Studies of suggestibility: Some observations and
suggestions. Applied Cognitive Psychology, 18, 1129–1141.
Ornstein, P. A., Gordon, B. N., & Larus, D. M. (1992). Children’s memory for a personally expe-
rienced event: Implications for testimony. Applied Developmental Psychology, 6, 49–60.
Ornstein, P. A., Grammer, J. K., & Coffman, J. L. (2010). Teachers’ “mnemonic style” and the
development of skilled memory. In H. S. Waters & W. Schneider (Eds.), Metacognition, strat-
egy use, and instruction (pp. 23–53). New York: Guilford Press.
Ornstein, P. A., & Haden, C. A. (2001). Memory development or the development of memory?
Current Directions in Psychological Science, 10, 202–209.
Ornstein, P. A., & Haden, C. A. (2009). Developments in the study of memory development. In
M. L. Courage & N. Cowan (Eds.), The development of memory in infancy and childhood
(pp. 367–385). Hove: Psychology Press.
Ornstein, P. A., Haden, C. A., & Coffman, J. L. (2011). Learning to remember: Mothers and
teachers talking with children. In N. L. Stein & S. W. Raudenbush (Eds.), Developmental
cognitive science goes to school (pp. 69–84). New York: Taylor & Francis.
Ornstein, P. A., Haden, C. A., & Elischberger, H. B. (2006). Children’s memory development:
Remembering the past and preparing for the future. In E. Bialystok & F. I. M. Craik (Eds.),
Lifespan cognition: Mechanisms of change (pp. 143–161). New York: Oxford University Press.
Ornstein, P. A., Haden, C. A., & San Souci, P. (2008). The development of skilled remembering
in children. In H. L. Roediger III & J. H. Byrne (Eds.), Cognitive psychology of memory (Vol.
2, pp. 715–744)., Learning and memory: A comprehensive reference Oxford: Elsevier.
Ornstein, P. A., & Light, L. L. (2010). Memory development across the lifespan. In W.F.
Overton (Vol. Ed.), R. M. Lerner (Ed.-in-Chief), Handbook of life-span development: Vol. 1.
Cognition, biology, and methods (pp. 259-305). Hoboken: Wiley.
Ornstein, P. A., Merritt, K. A., Baker-Ward, L., Furtado, E., Gordon, B. N., & Principe, G. F.
(1998). Children’s knowledge, expectation, and long-term retention. Applied Cognitive
Psychology, 12, 387–405.
Ornstein, P. A., & Naus, M. J. (1978). Rehearsal processes in children’s memory. In P. A.
Ornstein (Ed.), Memory development in children (pp. 69–99). Hillsdale: Erlbaum.
Ornstein, P. A., & Naus, M. J. (1985). Effects of the knowledge base on children‘s memory strat-
egies. In H. W. Reese (Ed.), Advances in child development and behavior (Vol. 19, pp. 113–
148). Orlando: Academic Press.
Ornstein, P. A., Naus, M. J., & Liberty, C. (1975). Rehearsal and organizational processes in chil-
dren’s memory. Child Development, 46, 818–830.
Ornstein, P. A., Naus, M. J., & Stone, B. P. (1977). Rehearsal training and developmental differ-
ences in memory. Developmental Psychology, 13, 15–24.
362 References

O’Sullivan, J. T. (1993). Preschoolers’ beliefs about effort, incentives, and recall. Journal of
Experimental Child Psychology, 55, 396–414.
Oura, Y., & Hatano, G. (1988). Memory for melodies among subjects differing in age and experi-
ence in music. Psychology of Music, 16, 91–109.
Overman, W. H., Bachevalier, J., Sewell, F., & Drew, J. (1993). A comparison of children’s per-
formance on two recognition memory tasks: delayed nonmatch-to-sample versus visual
paired-comparison. Developmental Psychobiology, 26, 345–357.
Palincsar, A. S. (1986). The role of dialogue in providing scaffolded instruction. Educational
Psychologist, 21, 73–98.
Palincsar, A. S., & Brown, A. L. (1984). Reciprocal teaching of comprehension-monitoring activ-
ities. Cognition and Instruction, 1, 117–175.
Paris, S. G. (1978). Coordination of means and goals in the development of mnemonic skills. In
P. Ornstein (Ed.), Memory development in children (pp. 259–273). Hillsdale: Erlbaum.
Paris, S. G., Cross, D. R., & Lipson, M. Y. (1984). Informed strategies for learning: A pro-
gram to improve children’s reading awareness and comprehension. Journal of Educational
Psychology, 76, 1239–1252.
Paris, S. G., & Lindauer, B. K. (1976). The role of inference in children’s comprehension and
memory for sentences. Cognitive Psychology, 8, 217–227.
Paris, S. G., & Lindauer, B. K. (1982). The development of cognitive skills during childhood.
In B. Wolman (Ed.), Handbook of developmental psychology (pp. 35–60). Englewood Cliffs:
Prentice-Hall.
Paris, S. G., Lindauer, B. K., & Cox, G. L. (1977). The development of inferential comprehen-
sion. Child Development, 48, 1728–1733.
Paris, S. G., Newman, R. S., & McVey, K. A. (1982). Learning the functional significance of
mnemonic actions: A microgenetic study of strategy acquisition. Journal of Experimental
Child Psychology, 34, 490–509.
Paris, S. G., & Oka, E. R. (1986). Children’s reading strategies, metacognition, and motivation.
Developmental Review, 6, 25–56.
Parker, J. F., & Carranza, L. E. (1989). Eyewitness testimony of children in target-present and
target-absent lineups. Law and Human Behavior, 13, 133–149.
Parkin, A. J. (1997). The development of procedural and declarative memory. In N. Cowan (Ed.),
The development of memory in childhood (pp. 113–137). Hove: Psychology Press.
Parkin, A. J., & Streete, S. (1988). Implicit and explicit memory in young children and adults.
British Journal of Psychology, 79, 361–369.
Pascalis, O., de Haan, M., Nelson, C. A., & de Schonen, S. (1998). Long-term recognition mem-
ory for faces assessed by visual paired comparison in 3- and 6-month-old infants. Journal of
Experimental Psychology: Learning, Memory, and Cognition, 24, 249–260.
Pascual-Leone, J. (1970). A mathematical model for the transition rule in Piaget’s developmental
stages. Acta Psychologica, 63, 301–345.
Pascual-Leone, J. (1978). Compounds, confounds, and models in developmental information process-
ing: A reply to Trabasso and Foellinger. Journal of Experimental Child Psychology, 26, 18–40.
Pascual-Leone, J. (1987). Organismic processes for neo-Piagetian theories: A dialectical causal
account for cognitive development. International Journal of Psychology, 22, 531–570.
Pascual-Leone, J. (2000). Reflections on working memory: Are the two models complementary?
Journal of Experimental Child Psychology, 77, 138–154.
Pascual-Leone, J., & Baillargeon, R. (1994). Developmental measurement of mental attention.
International Journal of Behavioral Development, 17, 161–200.
Pascual-Leone, J., & Johnson, J. (1999). A dialectical constructivist view of representation: Role
of mental attention, executives and symbols. In I. Sigel (Ed.), Development of mental repre-
sentation: Theories and applications (pp. 169–200). Mahwah: Erlbaum.
Pascual-Leone, J., & Johnson, J. (2011). A developmental theory of mental attention: Its appli-
cations to measurement and task analysis. In P. Barrouillet & V. Gaillard (Eds.), Cognitive
development and working memory: A dialogue between Neo-Piagetian and cognitive
approaches (pp. 13–46). New York: Psychology Press.
References 363

Patel, V. L., & Groen, G. J. (1991). The general and specific nature of medical expertise: A criti-
cal look. In K. A. Ericsson & J. Smith (Eds.), Toward a general theory of expertise: Prospects
and limits (pp. 93–125). Cambridge: Cambridge University Press.
Paz-Alonso, P. M., Larson, R. P., Castelli, P., Alley, D., & Goodman, G. S. (2009). Memory
development: emotion, stress, and trauma. In M. L. Courage & N. Cowan (Eds.), The devel-
opment of memory in infancy and childhood (pp. 197–239). Hove: Psychology Press.
Pearse, S. L., Powell, M. B., & Thomson, D. M. (2003). The effect of contextual cues on chil-
dren’s ability to remember an occurrence of a repeated event. Legal and Criminological
Psychology, 8, 39–50.
Pearson, P. D., Hansen, J., & Gordon, C. (1979). The effect of background knowledge on young
children’s comprehension of explicit and implicit information. Journal of Reading Behavior,
11, 201–209.
Pelphrey, K. A., & Reznick, J. S. (2003). Working memory in infancy. In R. Kail (Ed.), Advances
in child development and behavior (Vol. 31, pp. 173–227). New York: Academic Press.
Pelphrey, K. A., Reznick, J. S., Davis Goldman, B., Sasson, N., Morrow, J., Donahoe, A., et al.
(2004). Development of visuospatial short-term memory in the second half of the first year.
Developmental Psychology, 40, 836–851.
Perez, L. A., Peynircioglu, Z. F., & Blaxton, T. A. (1998). Developmental differences in implicit
and explicit memory performance. Journal of Experimental Child Psychology, 70, 167–185.
Perlmutter, J., & Myers, J. L. (1973). A comparison of two procedures for testing multiple con-
trasts. Psychological Bulletin, 79, 181–184.
Perlmutter, M., Schork, E. J., & Lewis, D. (1982). Effects of semantic and perceptual orienting
tasks on preschool children‘s memory. Bulletin of Psychonomic Society, 19, 65–68.
Perner, J. (1991). Understanding the representational mind. Cambridge: MIT Press.
Perner, J., & Ruffman, T. (1995). Episodic memory and autonoetic consciousness:
Developmental evidence and a theory of childhood amnesia. Journal of Experimental Child
Psychology, 59, 516–548.
Perrig, W., & Perrig, P. (1993). Life span aspects in explicit and implicit memory. The German
Journal of Psychology, 17, 307–309.
Perruchet, P., Frazier, N., & Lautrey, J. (1995). Conceptual implicit memory: A developmental
study. Psychological Research, 57, 220–228.
Peterson, C. (2002). Children’s long-term memory for autobiographical events. Developmental
Review, 22, 370–402.
Peterson, C., & Bell, M. (1996). Children’s memory for traumatic injury. Child Development, 67,
3045–3070.
Peterson, C., Jesso, B., & McCabe, A. (1999). Encouraging narratives in preschoolers: An inter-
vention study. Journal of Child Language, 26, 49–67.
Peterson, C., Morris, G., Baker-Ward, L., & Flynn, S. (2014). Predicting which childhood
memories persist: Contributions of memory characteristics. Developmental Psychology, 50,
439–448.
Peterson, C., & Parsons, B. (2005). Interviewing former 1- and 2-year-olds about medical emer-
gencies 5 years later. Law and Human Behavior, 29, 743–754.
Peterson, C., & Whalen, N. (2001). Five years later: Children’s memory for medical emergen-
cies. Applied Cognitive Psychology, 15, S7–S24.
Peterson, C., Warren, K. L., & Short, M. M. (2011). Infantile amnesia across the years: A 2-year
follow-up of children‘s earliest memories. Child Development, 82(4), 1092–1105.
Pezdek, K., Finger, K., & Hodge, D. (1997). Planting false childhood memories: The role of
event plausibility. Psychological Science, 8, 437–441.
Pezdek, K., & Hodge, D. (1999). Planting false childhood memories in children. The role of
event plausibility. Child Development, 70, 887–895.
Piaget, J. (1952). The origins of intelligence in children. New York: International Universities
Press.
Piaget, J. (1954). The construction of reality in the child. New York: Basic Books.
364 References

Piaget, J. (1962). Play, dreams, and imitation in childhood. New York: Norton.
Piaget, J., & Inhelder, B. (1973). Memory and intelligence. New York: Basic.
Pickering, S. J., Gathercole, S. E., Hall, M., & Lloyd, S. A. (2001). Development of memory for
pattern and path: Further evidence for the fractionation of visuo-spatial memory. Quarterly
Journal of Experimental Psychology, 54A, 397–420.
Pickering, S. J., Gathercole, S. E., & Peaker, S. M. (1998). Verbal and visual spatial short-term
memory in children: Evidence for common and distinct mechanisms. Memory & Cognition,
26, 1117–1130.
Pillemer, D. B., Picariello, M. L., & Pruett, J. C. (1994). Very long-term memories of a salient
preschool event. Applied Cognitive Psychology, 8, 95–106.
Pillemer, D. B., & White, S. H. (1989). Childhood events recalled by children and adults. In H.
W. Reese (Ed.), Advances in child development and behavior (Vol. 21, pp. 297–340). San
Diego: Academic Press.
Pipe, M.-E., Lamb, M. E., Orbach, Y., & Esplin, P. W. (2004). Recent research on children’s testi-
mony about experienced and witnessed events. Developmental Review, 24, 440–468.
Pipe, M.-E., & Salmon, K. (2009). Memory development and the forensic context. In M. L.
Courage & N. Cowan (Eds.), The development of memory in infancy and childhood (pp. 241–
282). Hove: Psychology Press.
Pohlmann, A. (1906). Einfluß der Lokalisation auf das Behalten. Experimentelle Beiträge zur
Lehre vom Gedächtnis. Berlin: Gerdes & Hödel.
Poole, D. A., & Lindsay, D. S. (1995). Interviewing preschoolers: Effects of nonsuggestive tech-
niques, parental coaching and leading questions on reports of nonexperienced events. Journal
of Experimental Child Psychology, 60, 129–154.
Poole, D. A., & White, L. T. (1991). Effects of question repetition on the eyewitness testimony of
children and adults. Developmental Psychology, 27, 975–986.
Poole, D. A., & White, L. T. (1993). Two years later. Effects of question repetition and retention
interval on the eyewitness testimony of children and adults. Developmental Psychology, 29,
844–853.
Poole, D. A., & White, L. T. (1995). Tell me again and again: Stability and change in the
repeated testimonies of children and adults. In M. S. Zaragoza, J. R. Graham, C. N. Gordon,
R. Hirschman, & Y. S. Ben Porath (Eds.), Memory and testimony in the child witness (pp.
24-43). Newbury Park: Sage.
Pressley, M. (1982). Elaboration and memory development. Child Development, 53, 296–309.
Pressley, M. (1995). What is intellectual development about in the 1990s? Good information pro-
cessing. In F. E. Weinert & W. Schneider (Eds.), Memory performance and competencies:
Issues in growth and development (pp. 375–404). Mahwah: Erlbaum.
Pressley, M. (2002). Reading instruction that works: The case for balanced teaching (2nd ed.).
New York: Guilford Press.
Pressley, M., & Afflerbach, P. A. (1995). Verbal protocols of reading. Hillsdale, NJ: Erlbaum.
Pressley, M., Borkowski, J. G., & O’Sullivan, J. T. (1985). Children’s metamemory and the
teaching of memory strategies. In D. L. Forrest-Pressley, G. E. MacKinnon, & T. G. Waller
(Eds.), Metacognition, cognition, and human performance (Vol. 1, pp. 111–153). Orlando:
Academic Press.
Pressley, M., Borkowski, J. G., & Schneider, W. (1987). Cognitive strategies: Good strategy users
coordinate metacognition and knowledge. In R. Vasta & G. Whitehurst (Eds.), Annals of child
development (Vol. 5, pp. 89–129). New York: JAI Press.
Pressley, M., Borkowski, J. G., & Schneider, W. (1989). Good information processing: What it is
and how education can promote it. Journal of Educational Research, 14, 857–867.
Pressley, M., & Brewster, M. E. (1990). Imaginal elaboration of illustrations to facilitate fact
learning: Creating memories of Prince Edward Island. Applied Cognitive Psychology, 4,
359–370.
Pressley, M., Cariglia-Bull, T., Deane, S., & Schneider, W. (1987). Short-term memory, ver-
bal competence, and age as predictors of imagery instructional effectiveness. Journal of
Experimental Child Psychology, 43, 194–211.
References 365

Pressley, M., & Dennis-Rounds, J. (1980). Transfer of a mnemonic keyword strategy at two age
levels. Journal of Educational Psychology, 72(4), 575.
Pressley, M., Forrest-Pressley, D., & Elliott-Faust, D. J. (1988). What is strategy instructional
enrichment and how to study it: Illustrations from research on children’s prose memory and
comprehension. In F. E. Weinert & M. Perlmutter (Eds.), Memory development: Universal
changes and individual differences (pp. 101–130). Hillsdale: Erlbaum.
Pressley, M., & Ghatala, E. S. (1990). Self-regulated learning: Monitoring from text learning.
Educational Psychologist, 25, 19–33.
Pressley, M., Goodchild, F., Fleet, J., Zajchowski, R., & Evans, E. D. (1989). The challenges of
classroom strategy instruction. Elementary School Journal, 89, 301–342.
Pressley, M., Graham, S., & Harris, K. (2006). The state of educational intervention research as
viewed through the lens of literacy intervention. British Journal of Educational Psychology,
76, 1–19.
Pressley, M., & Hilden, K. R. (2006). Memory strategies. In W. Damon & R. M. Lerner (Eds.-in-
chief), D. Kuhn & R. Siegler (Vol. Eds.), Handbook of child psychology: Vol. 2. Cognition,
perception, and language (6th ed., pp. 511–556). Hoboken: Wiley.
Pressley, M., & Levin, J. R. (1977). Developmental differences in subjects’ associative-learn-
ing strategies and performance: Assessing a hypothesis. Journal of Experimental Child
Psychology, 24, 431–439.
Pressley, M., & Levin, J. R. (1980). The development of mental imagery retrieval. Child
Development, 51, 558–560.
Pressley, M., Levin, J. R., & Delaney, H. D. (1982). The mnemonic keyword method. Review of
Educational Research, 52, 61–92.
Pressley, M., Levin, J. R., Ghatala, E. S., & Ahmad, M. (1987). Test monitoring in young grade
school children. Journal of Experimental Child Psychology, 43, 96–111.
Pressley, M., & McCormick, C. B. (1995). Advanced educational psychology of educators,
researchers, and policymakers. New York: HarperCollins.
Pressley, M., McDaniel, M. A., Turnure, J. E., Wood, E., & Ahmad, M. (1987). Generation
and precision of elaboration: Effects on intentional and incidental learning. Journal of
Experimental Psychology: Learning, Memory, and Cognition, 13, 291–300.
Pressley, M., & Van Meter, P. (1993). Memory strategies: Natural development and use following
instruction. In R. Pasnak & M. L. Howe (Eds.), Emerging Themes in Cognitive Development:
Vol. 2. Competencies (pp. 128–165). New York: Springer (10).
Pressley, M., & Woloshyn, V. (1995). Cognitive strategy instruction that really improves chil-
dren’s academic performance (2nd ed.). Cambridge: Brookline Books.
Pressley, M., Wood, E., & Woloshyn, V. (1990). Elaborative interrogation and facilitation of
fact learning: Why having a knowledge base is one thing and using it is quite another. In W.
Schneider & F. E. Weinert (Eds.), Interactions among aptitudes, strategies, and knowledge in
cognitive performance (pp. 200–221). New York: Springer.
Pressley, M., Wood, E., Woloshyn, V., Martin, V., King, A., & Menke, D. (1992). Encouraging
mindful use of prior knowledge: Attempting to construct explanatory answers facilitates
learning. Educational Psychologist, 27, 91–109.
Preyer, W. (1882). The mind of the child. Leipzig: Grieben.
Price, D. W. W., & Goodman, G. S. (1990). Visiting the wizard: Children’s memory for a recur-
ring event. Child Development, 61, 664–680.
Principe, G. F., & Ceci, S. J. (2002). “I saw it with my own ears”: The effects of peer conver-
sations on preschoolers’ reports of nonexperienced events. Journal of Experimental Child
Psychology, 83, 1–25.
Principe, G. F., Cherson, M., DiPuppo, J., & Schindewolf, E. (2012). Children’s natural con-
versations following exposure to a rumor: Linkages to later false reports. Journal of
Experimental Child Psychology, 113, 383–400.
Principe, G. F., Haines, B., Adkins, A., & Guiliano, S. (2010). False rumors and true belief:
Memory processes underlying children’s errant reports of rumored events. Journal of
Experimental Child Psychology, 107, 407–422.
366 References

Principe, G. F., Kanaya, T., Ceci, S. J., & Singh, M. (2006). Believing is seeing: How rumors can
engender false memories in preschoolers. Psychological Science, 17, 243–248.
Principe, G. F., & Schindewolf, E. (2012). Natural conversations as a source of false memories
in children: Implications for the testimony of young witnesses. Developmental Review, 32,
205–223.
Qin, J., Quas, J., Redlich, A., & Goodman, G. S. (1997). Children's eyewitness memory. In N.
Cowan (Ed.), Memory development in childhood (pp. 301–342). London: Psychology Press.
Quas, J. A., Goodman, G. S., Bidrose, S., Pipe, M.-E., Craw, S., & Ablin, D. S. (1999). Emotion
and memory: Children’s long-term remembering, forgetting, and suggestibility. Journal of
Experimental Child Psychology, 72, 235–270.
Quas, J. A., & Klemfuss, J. Z. (2014). Physiological stress reactivity and episodic memory in
children. In P. J. Bauer & R. Fivush (Eds.), The Wiley handbook on the development of chil-
dren’s memory (Vol. II, pp. 688–708). Chichester: Wiley.
Quas, J. A., Malloy, L. C., Melinder, A., Goodman, G. S., & D’Mello, M. (2007). Developmental
differences in the effects of repeated interviews and interviewer bias on young children’s
event memory and false reports. Developmental Psychology, 43, 823–837.
Rabinowitz, M. (1984). The use of categorical organization: Not an all-or-none situation. Journal
of Experimental Child Psychology, 38, 338–351.
Rabinowitz, M., & Chi, M. T. H. (1987). An interactive model of strategic processing. In S. J.
Ceci (Ed.), Handbook of the cognitive, social, and physiological characteristics of learning
disabilities (Vol. 2, pp. 83–102). Hillsdale: Erlbaum.
Rabinowitz, M., & McAuley, R. (1990). Conceptual knowledge processing: An oxymoron? In W.
Schneider & F. E. Weinert (Eds.), Interactions among aptitudes, strategies, and knowledge in
cognitive performance (pp. 117–133). New York: Springer.
Rabinowitz, M., Ornstein, P. A., Folds-Bennett, T. H., & Schneider, W. (1994). Age-related dif-
ferences in speed of processing: Unconfounding age and experience. Journal of Experimental
Child Psychology, 57, 449–459.
Radossawljewitsch, P. R. (1907). Das Behalten und Vergessen bei Kindern und Erwachsenen
nach experimentellen Untersuchungen. In E. Meumann (Ed.), Pädagogische Monographien
(Vol. 1, pp. 128–181). Leipzig: Otto Nemnich.
Ratner, H. H., Smith, B. S., & Dion, S. A. (1986). Development of memory for events. Journal of
Experimental Child Psychology, 41, 411–428.
Reber, A. S. (1993). Implicit learning and tacit knowledge. Journal of Experimental Psychology:
General, 118, 219–235.
Recht, D. R., & Leslie, L. (1988). Effect of prior knowledge on good and poor readers‘ memory
of text. Journal of Educational Psychology, 80, 16–20.
Reese, E. (2002). Social factors in the development of autobiographical memory: The state of the
art. Social Development, 11, 124–142.
Reese, E. (2009). The development of autobiographical memory: Origins and consequences.
In P. Bauer (Ed.), Advances in child development and behavior (Vol. 37, pp. 145–200).
Amsterdam: Elsevier.
Reese, E. (2014). Taking the long way: Longitudinal approaches to autobiographical memory
development. In P. J. Bauer & R. Fivush (Eds.), The Wiley handbook on the development of
children’s memory (Vol. II, pp. 972–995). Chichester: Wiley.
Reese, E., Haden, C., Baker-Ward, L., Bauer, P. J., Fivush, R., & Ornstein, P. A. (2011).
Coherence of personal narratives across the lifespan: A multidimensional model and coding
method. Journal of Cognition and Development, 12, 424–462.
Reese, E., Haden, C. A., & Fivush, R. (1993). Mother-child conversations about the past:
Relationships of style and memory over time. Cognitive Development, 8, 403–430.
Reese, E., Jack, F., & White, N. (2010). Origins of adolescents’ autobiographical memories.
Cognitive Development, 25, 352–367.
Reese, E., & Newcombe, R. (2007). Training mothers in elaborative reminiscing enhances chil-
dren’s autobiographical memory and narrative. Child Development, 78(4), 1153–1170.
References 367

Reese, H. W. (1962). Verbal mediation as a function of age level. Psychological Bulletin, 59,
502–509.
Reese, H. W. (1977). Imagery and associative memory. In R. V. Kail & W. Hagen (Eds.),
Perspectives on the development of memory and cognition (pp. 113–175). Hillsdale: Erlbaum.
Renninger, K. A., Hidi, S., & Krapp, A. (Eds.). (1992). The role of interest in learning and devel-
opment. Hillsdale: Erlbaum.
Reyna, V. F., & Brainerd, C. J. (1995). Fuzzy-trace theory: An interim synthesis. Learning and
Individual Differences, 7, 1–75.
Reyna, V. F., Holliday, R., & Marche, T. (2002). Explaining the development of false memories.
Developmental Review, 22, 436–489.
Reyna, V. F., & Kiernan, B. (1994). Development of gist versus verbatim memory in sentence
recognition: Effects of lexical familiarity, semantic content, encoding instructions, and reten-
tion interval. Developmental Psychology, 30, 178–191.
Reznick, J. S. (2007). Working memory in infants and toddlers. In L. M. Oakes & P. J. Bauer
(Eds.), Short- and long-term memory in infancy and early childhood: Taking the first steps
toward remembering (pp. 3–26). New York: Oxford University Press.
Reznick, J. S. (2009). Working memory in infants and toddlers. In M. L. Courage & N.
Cowan (Eds.), The development of memory in infancy and childhood (pp. 343–365). Hove:
Psychology Press.
Reznick, J. S. (2014). Methodological challenges in the study of short-term working memory in
infants. In P. J. Bauer & R. Fivush (Eds.), The Wiley handbook on the development of chil-
dren’s memory (Vol. I, pp. 181–201). Chichester: Wiley.
Reznick, J. S., Corley, R., & Robinson, J. (1997). A longitudinal twin study of intelligence in the
second year. Monographs of the Society for Research in Child Development, 62 (1, Serial No.
249).
Reznick, J. S., Morrow, J. D., Goldman, B. D., & Snyder, J. (2004). The onset of working mem-
ory in infants. Infancy, 6, 145–154.
Richmond, J., & Nelson, C. A. (2008). Mechanisms of change: A cognitive neuroscience
approach to declarative memory development. In C. A. Nelson & M. Luciana (Eds.),
Developmental cognitive neuroscience (2nd ed., pp. 541–552). Cambridge: MIT Press.
Richmond, J., & Nelson, C. A. (2009). Relational memory during infancy: Evidence of eye track-
ing. Developmental Science, 12, 549–566.
Ringel, B. A., & Springer, C. J. (1980). On knowing how well one is remembering: The persis-
tence of strategy use during transfer. Journal of Experimental Child Psychology, 29, 322–333.
Ritter, K. (1978). The development of knowledge of an external retrieval cue strategy. Child
Development, 49, 1227–1230.
Ritter, K., Kaprove, B. H., Fitch, J. P., & Flavell, J. H. (1973). The development of retrieval strat-
egies in young children. Cognitive Psychology, 5, 310–321.
Roberts, M. B., & Powell, K. P. (2006). The consistency of false suggestions moderates chil-
dren’s reports of a single instance of a repeated event: Predicting increases and decreases in
suggestibility. Journal of Experimental Child Psychology, 94, 68–89.
Roderer, T., & Roebers, C. M. (2010). Explicit and implicit confidence judgments and develop-
mental differences in metamemory: an eye-tracking approach. Metacognition and Learning,
5, 229–250.
Roebers, C. M. (2002). Confidence judgments in children’s and adults’ event recall and suggest-
ibility. Developmental Psychology, 38, 1052–1067.
Roebers, C. M. (2003). Remembering faces in social contexts. In G. Schwarzer & H. Leder
(Eds.), The development of face processing (pp. 135–148). Cambridge: Hogrefe.
Roebers, C. M. (2014). Children’s deliberate memory development: The contribution of strate-
gies and metacognitive processes. In P. J. Bauer & R. Fivush (Eds.), The Wiley handbook on
the development of children’s memory (Vol. II, pp. 865–894). Chichester: Wiley.
Roebers, C. M., Cimeli, P., Röthlisberger, M., & Neuenschwander, R. (2012). Executive func-
tioning, metacognition, and self-perceived competence in elementary school children: An
368 References

explorative study on their interrelations and their role for school achievement. Metacognition
and Learning, 7, 151–173.
Roebers, C. M., & Fernandez, O. (2002). The effects of accuracy motivation on children‘s and
adults’ event recall, suggestibility, and their answers to unanswerable question. Journal of
Cognition and Development, 3, 415–443.
Roebers, C. M., Gelhaar, T., & Schneider, W. (2004). “It’s magic!“ The effects of presentation
modality on children‘s event memory, suggestibility, and confidence judgments. Journal of
Experimental Child Psychology, 87, 320–335.
Roebers, C. M., & Howie, P. (2003). Confidence judgments in event recall: Developmental progres-
sion in the impact of question format. Journal of Experimental Child Psychology, 85, 352–371.
Roebers, C. M., Howie, P., & Beuscher, E. (2007). Can private reports enhance children‘s event
recall, lower their suggestibility, and foster their metacognitive monitoring compared to face-
to-face interviews? Computers in Human Behavior, 23, 749–769.
Roebers, C. M., Krebs, S., & Roderer, T. (2014). Metacognitive monitoring and control in ele-
mentary school children: Their interrelations and their roles for test performance. Learning
and Individual Differences,29, 141–149.
Roebers, C. M., Moga, N., & Schneider, W. (2001). The role of accuracy motivation on chil-
dren‘s and adults’event recall. Journal of Experimental Child Psychology, 78, 313–329.
Roebers, C. M., & Schneider, W. (2000). The impact of misleading questions on eyewitness
memory in children and adults. Applied Cognitive Psychology, 14, 509–526.
Roebers, C. M., & Schneider, W. (2001). Individual differences in children’s eyewitness recall:
The influence of intelligence and shyness. Applied Developmental Science, 5, 9–20.
Roebers, C. M., & Schneider, W. (2002). Stability and consistency of children‘s event recall.
Cognitive Development, 17, 1085–1103.
Roebers, C. M., & Schneider, W. (2005). The strategic regulation of children’s memory perfor-
mance and suggestibility. Journal of Experimental Child Psychology,91, 24–44.
Roebers, C. M., von der Linden, N., Schneider, W., & Howie, P. (2007). Children’s metame-
morial judgments in an event recall task. Journal of Experimental Child Psychology,97,
117–137.
Roediger, H. L. (1990). Implicit memory: Retention without remembering. American
Psychologist, 45, 1043–1045.
Roenker, D. L., Thompson, C. P., & Brown, S. C. (1971). Comparisons of measures for the esti-
mation of clustering in free recall. Psychological Bulletin, 76, 45–48.
Rogoff, B. (1981). Schooling and development of cognitive skills. In H. C. Triandis & A. Heron
(Eds.), Handbook of cross-cultural psychology (Vol. 4, pp. 233–294). Boston: Allyn & Bacon.
Rogoff, B., & Mistry, J. (1985). Memory development in cultural context. In M. Pressley & C. J.
Brainerd (Eds.), Cognitive learning and memory in children (pp. 117–142). New York: Springer.
Rogoff, B., & Mistry, J. (1990). The social and functional context of children’s remembering.
In R. Fivush & J. A. Hudson (Eds.), Knowing and remembering in young children (pp. 197–
222). Cambridge: Cambridge University Press.
Rogoff, B., & Waddell, K. J. (1982). Memory for information organized in a scene by children
from two cultures. Child Development, 53, 1224–1228.
Rohwer, W. D. (1973). Elaboration and learning in childhood and adolescence. In H. W. Reese (Ed.),
Advances in child development and behavior (Vol. 8, pp. 1–57). New York: Academic Press.
Rohwer, W. D. (1980). An elaborative conception of learner differences. In R. E. Snow, P. A.
Federico, & W. E. Montague (Eds.), Aptitude, learning, and instruction (Vol. 2, pp. 23–46).,
Cognitive process analyses of learning and problem Hillsdale: Erlbaum.
Rohwer, W. D., & Bean, J. P. (1973). Sentence effects and noun-pair learning: a developmental
interaction during adolescence. Journal of Experimental Child Psychology, 15, 521–533.
Rohwer, W. D., Raines, J. M., Eoff, J., & Wagner, M. (1977). The development of elaborative
propensity during adolescence. Journal of Experimental Child Psychology, 23, 472–492.
Roodenrys, S., Hulme, C., & Brown, G. (1993). The development of short-term memory span:
Separable effects of speech rate and long-term memory. Journal of Experimental Child
Psychology, 56, 431–442.
References 369

Rose, S. A., Feldman, J. F., & Jankowski, J. J. (2001). Visual short-term memory in the first year
of life: Capacity and recency effects. Developmental Psychology, 37, 539–549.
Rose, S. A., Feldman, J. F., & Jankowski, J. J. (2007). Developmental aspects of visual recogni-
tion memory in infancy. In L. M. Oakes & P. J. Bauer (Eds.), Short- and long-term memory in
infancy and early childhood: Taking the first steps towards remembering (pp. 153–178). New
York: Oxford University Press.
Rose, S. A., Feldman, J. F., & Wallace, I. F. (1992). Infant information processing in relation to
six-year cognitive outcomes. Child Development, 63, 1126–1141.
Rose, S. A., Gottfried, A. W., Melloy-Carminar, P., & Bridger, W. H. (1982). Familiarity and
novelty preferences in infant recognition memory: Implications for information processing.
Developmental Psychology, 18, 704–713.
Rosenshine, B., & Meister, C. (1994). Reciprocal teaching: A review of the research. Review of
Educational Research, 64, 479–530.
Rosenshine, B., Meister, C., & Chapman, S. (1996). Teaching students to generate questions: A
review of the intervention studies. Review of Educational Research, 66(2), 181–221.
Ross-Sheehy, S., Oakes, L. M., & Luck, S. J. (2003). The development of visual short-term
memory capacity in infants. Child Development, 74, 1807–1822.
Ross-Sheehy, S., Oakes, L. M., & Luck, S. J. (2011). Exogenous attention influences visual
short-term memory in infants. Developmental Science, 14, 490–501.
Rovee-Collier, C. (1995). Time windows in cognitive development. Developmental Psychology,
51, 1–23.
Rovee-Collier, C. (1996). Shifting the focus from what to why. Infant Behavior and
Development, 19, 385–400.
Rovee-Collier, C. (1997). Dissociations in infant memory: Rethinking the development of
implicit and explicit memory. Psychological Review, 104, 467–498.
Rovee-Collier, C., & Cuevas, K. (2009). The development of infant memory. In M. L. Courage
& N. Cowan (Eds.), The development of memory in infancy and childhood (pp. 11–41). Hove:
Psychology Press.
Rovee-Collier, C., & Gerhardstein, P. (1997). The development of infant memory. In N. Cowan
(Ed.), The development of memory in childhood (pp. 5–40). Hove: Psychology Press.
Rovee-Collier, C., & Hayne, H. (2000). Memory in infancy and early childhood. In E. Tulving
& F. I. M. Craik (Eds.), The Oxford handbook of memory (pp. 267–282). New York: Oxford
University Press.
Rovee-Collier, C., Hayne, H., & Colombo, M. (2001). The development of implicit and explicit
memory. Amsterdam: John Benjamins.
Rovee-Collier, C., Schechter, A., Shyi, G. C.-W., & Shields, P. (1992). Perceptual identification
of contextual attributes and infant memory retrieval. Development Psychology, 28, 307–138.
Rovee-Collier, C., Sullivan, M. W., Enright, M. K., Lucas, D., & Fagen, J. W. (1980).
Reactivation of infant memory. Science, 208, 1159–1161.
Rubin, D. C. (1986). Autobiographical memory. Cambridge: Cambridge University Press.
Rubinstein, S. L. (1973). Das Gedächtnis. In S. L. Rubinstein (Ed.), Grundlagen der Allgemeinen
Psychologie (pp. 359-409). Berlin: Volk & Wissen (originally published in 1940).
Rudy, L., & Goodman, G. S. (1991). Effects of participation on children’s reports: Implications
for children’s testimony. Developmental Psychology, 27, 527–538.
Rugg, M., Mark, R., Walla, P., Schloerscheidt, A., Birch, C., & Allan, K. (1998). Dissociation
of the neural correlates of implicit and explicit memory. Nature, 392(6676), 595–598.
doi:10.1038/33396.
Rundus, D. (1971). Analysis of rehearsal processes in free recall. Journal of Experimental Child
Psychology, 89, 63–77.
Russo, R., Nichelli, P., Gibertoni, M., & Cornia, C. (1995). Developmental trends in implicit and
explicit memory: A picture completion study. Journal of Experimental Child Psychology, 59,
566–578. doi:10.1006/jecp.1995.1026.
Salatas, H., & Flavell, J. H. (1976a). Behavioral and metamnemonic indicators of strategic
behaviors under remember instructions in first grade. Child Development, 47, 81–89.
370 References

Salatas, H., & Flavell, J. H. (1976b). Retrieval of recently learned information: Development of
strategies and control skills. Child Development, 47, 941–948.
Sales, J. M., & Fivush, R. (2005). Social and emotional functions of mother-child reminiscing
about stressful events. Social Cognition, 23, 70–90.
Salmon, K., & Pipe, M.-E. (1997). Pros and children’s event reports: The impact of a 1-year
delay. Journal of Experimental Child Psychology, 65, 261–292.
Samuel, A. G. (1978). Organizational vs. retrieval factors in the development of digit span.
Journal of Experimental Child Psychology, 26, 308–319.
Saywitz, K. J., Geiselman, R. E., & Bornstein, G. K. (1992). Effects of cognitive interviewing
and practice on children's recall performance. Journal of Applied Psychology, 77, 744–756.
Saywitz, K. J., Goodman, G. S., Nicholas, E., & Moan, S. F. (1991). Children’s memories of a
physical examination involving genital touch: Implications for reports of child sexual abuse.
Journal of Consulting and Clinical Psychology, 59, 682–691.
Saywitz, K. J., & Snyder, L. (1993). Improving children’s testimony with preparation. In G. S.
Goodman & B. L. Bottoms (Eds.), Child victims, child witnesses: Understanding and improv-
ing testimony (pp. 117–146). New York: Guilford Press.
Scarf, D., Gross, J., Colombo, M., & Hayne, H. (2013). To have and to hold: Episodic memory in
3- and 4-year-old children. Developmental Psychobiology, 55, 125–132.
Schaaf, J. M., Alexander, K. W., & Goodman, G. S. (2008). Children’s false memory and true disclo-
sure in the face of repeated questions. Journal of Experimental Child Psychology, 100, 157–185.
Schacter, D. L. (1987). Implicit memory: History and current status. Journal of Experimental
Psychology: Learning, Memory and Cognition, 13, 501–518.
Schacter, D. L. (1992). Understanding implicit memory: A cognitive neuroscience perspective.
American Psychologist, 47, 559–569.
Schacter, D. L., & Moscovitch, M. (1984). Infants, amnesics, and dissociable memory systems.
In M. Moscovitch (Ed.), Advances in the study of communication and affect (Vol. 9, pp. 173–
216)., Infant memory New York: Plenum.
Schank, R. C., & Abelson, R. P. (1977). Scripts, plans, goals and understanding. Hillsdale:
Erlbaum.
Scherf, K. S., Sweeney, J. A., & Luna, B. (2006). Brain basis of developmental change in visuo-
spatial working memory. Journal of Cognitive Neuroscience, 18(7), 1045–1058.
Schlagmüller, M., & Schneider, W. (2002). The development of organizational strategies in
children: Evidence from a microgenetic longitudinal study. Journal of Experimental Child
Psychology, 81, 298–319.
Schlagmüller, M., & Schneider, W. (2007). WLST 7-12: Würzburger Lesestrategie-Wissenstest
für die Klassen 7-12. Göttingen: Hogrefe.
Schlagmüller, M., Visé, M., & Schneider, W. (2001). Zur Erfassung des Gedächtniswissens bei
Grundschulkindern: Konstruktionsprinzipien und empirische Bewährung der Würzburger
Testbatterie zum deklarativen Gedächtnis. Zeitschrift für Entwicklungspsychologie und
Pädagogische Psychologie, 33, 91–102.
Schneider, W. (1985). Developmental trends in the metamemory-memory behavior relation-
ship: An integrated review. In D.-L. Forrest-Pressley, G. E. MacKinnon, & T. G. Waller
(Eds.), Cognition, metacognition, and human performance (Vol. 1, pp. 57–109). New York:
Academic Press.
Schneider, W. (1986). The role of conceptual knowledge and metamemory in the development of
organizational processes in memory. Journal of Experimental Child Psychology, 42, 218–236.
Schneider, W. (1989). Problems of longitudinal studies with children: Practical, conceptual, and
methodological issues. In M. Brambring, F. Lösel, & H. Skowronek (Eds.), Children at risk:
Assessment, longitudinal research, and intervention (pp. 313–335). New York: De Gruyter.
Schneider, W. (1993). Domain-specific knowledge and memory performance in children.
Educational Psychology Review, 5, 257–273.
Schneider, W. (1997). The impact of expertise on performance: Illustrations from developmental
research on memory and sports. High Ability Studies, 8(1), 7–18.
References 371

Schneider, W. (1998). Performance prediction in young children: Effects of skill, metacognition

and wishful thinking. Developmental Science, 1, 291–297.
Schneider, W. (1999). The development of metamemory in children. In D. Gopher & A. Koriat
(Eds.), Attention and performance XVII: Cognitive regulation of performance - Interaction of
theory and application (pp. 487–514). Cambridge: MIT Press.
Schneider, W. (2000a). Research on memory development: Historical trends and current themes.
International Journal of Behavioral Development, 24, 407–420.
Schneider, W. (2000b). Giftedness, expertise, and (exceptional) performance: A develop-
mental perspective. In K. A. Heller, F. J. Mönks, R. J. Sternberg, & R. F. Subotnik (Eds.),
International handbook of research and development of giftedness and talent (2nd ed., pp.
165–178). Oxford: Elsevier.
Schneider, W. (2008). The development of metacognitive knowledge in children and adolescents:
Major trends and implications for education. Mind, Brain, and Education, 2, 114–121.
Schneider, W. (2010). Metacognition and memory development in childhood and adolescence. In
H. S. Waters & W. Schneider (Eds.), Metacognition, strategy use, and instruction (pp. 54–81).
New York: Guilford Press.
Schneider, W. (2011). Memory development in childhood. In U. Goswami (Ed.), The Blackwell-
Wiley handbook of childhood cognitive development (2nd ed., pp. 347–376). London: Wiley.
Schneider, W. (2014). Individual differences in memory development and educational implica-
tions: Cross-sectional and longitudinal evidence. In P. J. Bauer & R. Fivush (Eds.), The Wiley
handbook on the development of children’s memory (Vol. II, pp. 947–971). Chichester: Wiley.
Schneider, W., & Artelt, C. (2010). Metacognition and mathematics education. ZDM: The
International Journal on Mathematics Education, 42, 149-161.
Schneider, W., & Bjorklund, D. F. (1992). Expertise, aptitude, and strategic remembering. Child
Development, 63, 461–473.
Schneider, W., & Bjorklund, D. F. (1998). Memory. In W. Damon (Editor-in-Chief), & D. Kuhn
& R. S. Siegler (Vol. Eds.), Handbook of Child Psychology: Vol. 2. Cognition, perception,
and language (5th ed., pp. 467-521). New York: Wiley.
Schneider, W., & Bjorklund, D. F. (2003). Memory and knowledge development. In J. Valsiner
& K. Connolly (Eds.), Handbook of developmental psychology (pp. 370–403). London: Sage.
Schneider, W., Bjorklund, D. F., & Maier-Brückner, W. (1996). The effects of expertise and IQ
on children‘s memory: When knowledge is, and when it is not enough. International Journal
of Behavioral Development, 19, 773–796.
Schneider, W., Borkowski, J. G., Kurtz, B. E., & Kerwin, K. (1986). Metamemory and motiva-
tion: A comparison of strategy use and performance in German and American children.
Journal of Cross-Cultural Psychology, 17, 315–336.
Schneider, W., Bös, K., & Rieder, H. (1993). Leistungsprognose bei jugendlichen Spitzensportlern
[Performance prediction in adolescent top tennis players]. In J. Beckmann, H. Strang, & E.
Hahn (Eds.), Aufmerksamkeit und Energetisierung (pp. 277–299). Göttingen: Hogrefe.
Schneider, W., & Bullock, M. (Eds.). (2009). Human development from early childhood to early
adulthood: Findings from a 20 year longitudinal study. New York: Psychology Press.
Schneider, W., Gruber, H., Gold, A., & Opwis, K. (1993). Chess expertise and memory for chess
positions in children and adults. Journal of Experimental Child Psychology, 56, 328–349.
Schneider, W., Knopf, M., & Sodian, B. (2009). Verbal memory development from early child-
hood to early adulthood. In W. Schneider & M. Bullock (Eds.), Human development from
early childhood to early adulthood: Findings from a 20 year longitudinal study (pp. 63–90).
New York: Psychology Press.
Schneider, W., & Körkel, J. (1989). The knowledge base and text recall: Evidence from a short-
term longitudinal study. Contemporary Educational Psychology, 14, 382–393.
Schneider, W., Körkel, J., & Vogel, K. (1987). Zusammenhänge zwischen Metagedächtnis,
strategischem Verhalten und Gedächtnisleistungen im Grundschulalter: Eine entwicklung-
spychologische Studie [Interrelationships among metamemory, strategic behavior, and
memory performance in elementary school children: A developmental study]. Zeitschrift für
Entwicklungspsychologie und Pädagogische Psychologie, 19, 99–115.
372 References

Schneider, W., Körkel, J., & Weinert, F. E. (1987). The effects of intelligence, self-concept,
and attributional style on metamemory and memory behaviour. International Journal of
Behavioral Development, 10(3), 281–299.
Schneider, W., Körkel, J., & Weinert, F. E. (1989). Domain-specific knowledge and memory
performance: A comparison of high- and low-aptitude children. Journal of Educational
Psychology, 81, 306–312.
Schneider, W., Körkel, J., & Weinert, F. E. (1990). Expert knowledge, general abilities, and text
processing. In W. Schneider & F. E. Weinert (Eds.), Interactions among aptitudes, strategies,
and knowledge in cognitive performance (pp. 235–251). New York: Springer.
Schneider, W., Kron, V., Hünnerkopf, M., & Krajewski, K. (2004). The development of young
children’s memory strategies: Findings from the Würzburg Longitudinal Memory Study.
Journal of Experimental Child Psychology, 88, 193–209.
Schneider, W., Kron-Sperl, V., & Hünnerkopf, M. (2009). The development of young children’s
memory strategies: Evidence from the Würzburg Longitudinal Memory Study. European
Journal of Developmental Psychology, 6, 70–99.
Schneider, W., & Lockl, K. (2002). The development of metacognitive knowledge in children
and adolescents. In T. Perfect & B. Schwartz (Eds.), Applied metacognition (pp. 224–257).
Cambridge: Cambridge University Press.
Schneider, W., & Lockl, K. (2008). Procedural metacognition in children: Evidence for develop-
mental trends. In J. Dunlosky & B. Bjork (Eds.), A handbook of memory and metacognition
(pp. 391–409). Mahwah: Erlbaum.
Schneider, W., Lockl, K., & Fernandez, O. (2005). Interrelationships among theory of mind,
executive control, language development, and working memory in young children: A longitu-
dinal study. In W. Schneider, R. Schumann-Hengsteler, & B. Sodian (Eds.), Young children's
cognitive development: Interrelations among executive functioning, working memory, verbal
ability, and theory of mind (pp. 259–284). Hillsdale: Erlbaum.
Schneider, W., & Pressley, M. (1989). Memory development between 2 and 20. New York: Springer.
Schneider, W., & Pressley, M. (1997). Memory development between two and twenty (2nd ed.).
Mahwah: Erlbaum.
Schneider, W., Schlagmüller, M., & Visé, M. (1998). The impact of metamemory and domain-
specific knowledge on memory performance. European Journal of Psychology of Education,
13, 91–103.
Schneider, W., & Sodian, B. (1988). Metamemory-memory behavior relationships in young chil-
dren: Evidence from a memory-for-location task. Journal of Experimental Child Psychology,
45, 209–233.
Schneider, W., & Sodian, B. (1991). A longitudinal study of young children’s memory behavior
and performance in a sort-recall task. Journal of Experimental Child Psychology, 51, 14–29.
Schneider, W., & Sodian, B. (1997). Memory strategy development: Lessons from longitudinal
research. Developmental Review, 17, 442–461.
Schneider, W., & Uhl, C. (1990). Metagedächtnis, Strategienutzung und Gedächtnisleistung:
Vergleichende Analysen bei Kindern, jüngeren Erwachsenen und alten Menschen
[Metamemory, strategy use, and memory performance: A comparison among children,
young adults, and the elderly]. Zeitschrift für Entwicklungspsychologie und Pädagogische
Psychologie, 22, 22–41.
Schneider, W., Visé, M., Lockl, K., & Nelson, T. O. (2000). Developmental trends in children’s mem-
ory monitoring: Evidence from a judgment-of-learning task. Cognitive Development, 15, 115–134.
Schneider, W., & Weinert, F. E. (1990). The role of knowledge, strategies, and aptitudes in
cognitive performance: Concluding comments. In W. Schneider & F. E. Weinert (Eds.),
Interactions among aptitudes, strategies, and knowledge in cognitive performance (pp. 286–
302). New York: Springer.
Schneider, W., & Weinert, F. E. (1995). Memory development during early and middle child-
hood: Findings from the Munich Longitudinal Study (LOGIC). In W. Schneider & F. E.
Weinert (Eds.), Memory performance and competencies. Issues in growth and development
(pp. 263–279). Mahwah: Erlbaum.
References 373

Schraw, G., & Moshman, D. (1995). Metacognitive theories. Educational Psychology Review, 7,
351–371.
Schröder, L., Keller, H., Kärtner, J., Kleis, A., Abels, M., Yovsi, R. D., et al. (2013). Early remi-
niscing in cultural contexts: Cultural models, maternal reminiscing styles, and children‘s
memories. Journal of Cognition and Development, 14, 10–34.
Schuchardt, K., Mähler, C., & Hasselhorn, M. (2008). Working memory deficits in children with
specific learning disorders. Journal of Learning Disabilities, 41, 514–523.
Schumann-Hengsteler, R. (1995). Die Entwicklung des visuo-räumlichen Gedächtnisses
[Development of visuo-spatial memory]. Göttingen: Hogrefe.
Schumann-Hengsteler, R. (1996). Children’s and adults’ visuospatial memory: The game concen-
tration. Journal of Genetic Psychology, 157, 77–92.
Schunk, D. H., & Zimmerman, B. J. (Eds.). (1998). Self-regulated learning: From teaching to
self-reflective practice. New York: Guilford Press.
Schwanenflugel, P. J., Fabricius, W. V., & Alexander, J. (1994). Developing theories of mind:
Understanding concepts and relations between mental activities. Child Development, 65,
1546–1563.
Schwarz, S., & Roebers, C. M. (2006). Age differences in the effects of social influence on chil-
dren’s eyewitness performance and their metacognitive monitoring. Journal of Experimental
Child Psychology, 94, 229–248.
Schwenck, C., Bjorklund, D. F., & Schneider, W. (2007). Factors influencing the incidence of
utilization deficiences and other patterns of recall/strategy-use relations in a strategic memory
task. Child Development, 78, 1771–1787.
Schwenck, C., Bjorklund, D. F., & Schneider, W. (2009). Developmental and individual
differences in young children‘s use and maintenance of a selective memory strategy.
Developmental Psychology, 45, 1034–1050.
Scribner, S., & Cole, M. (1972). Effects of constrained recall training on children’s performance
in a verbal memory task. Child Development, 43, 845–857.
Scribner, S., & Cole, M. (1981). The psychology of literacy. Cambridge: Harvard University
Press.
Selz, O. (1913). Über die Gesetze des geordneten Denkverlaufs [On laws of systematic thought].
Stuttgart: Spemann.
Sharp, D., Cole, M., & Lave, C. (1979). Education and cognitive development: The evidence
from experimental research. Monographs of the Society for Research in Child Development,
44 (Serial No. 178).
Shaughnessy, J. (1981). Memory monitoring accuracy and modification of rehearsal strategies.
Journal of Verbal Learning and Verbal Behavior, 20, 216–230.
Shimamura, A. P. (2000). Toward a cognitive neuroscience of metacognition. Consciousness and
Cognition, 9, 313–323.
Shin, H.-E., Bjorklund, D. F., & Beck, E. F. (2007). The adaptive nature of children’s overestima-
tion in a strategic memory task. Cognitive Development, 22, 197–212.
Shing, Y. L., & Lindenberger, U. (2011). The development of episodic memory: Lifespan les-
sons. Child Development Perspectives, 5(2), 148–155.
Shing, Y. L., Lindenberger, U., Diamond, A., Li, S. C., & Davidson, M. C. (2010). Memory
maintenance and inhibitory control differentiate from early childhood to adolescence.
Developmental Neuropsychology, 35, 679–697.
Shing, Y. L., Werkle-Bergner, M., Brehmer, Y., Müller, V., Li, S.-C., & Lindenberger, U. (2010).
Episodic memory across the lifespan: The contributions of associative and strategic compo-
nents. Neuroscience and Biobehavioral Reviews, 34, 1080–1091.
Shing, Y. L., Werkle-Bergner, M., Li, S.-C., & Lindenberger, U. (2008). Associative and stra-
tegic components of episodic memory: A life-span dissociation. Journal of Experimental
Psychology: General, 137, 495–513.
Short, E. J., & Ryan, E. B. (1984). Metacognitive differences between skilled and less skilled
readers: Remediating deficits through story grammar and attribution training. Journal of
Educational Psychology, 76, 225–235.
374 References

Shrimpton, S., Oates, R., & Hayes, S. C. (1998). Children’s memory of events: Effects of stress,
age, time delay and location of interview. Journal of Applied Cognitive Psychology, 12,
133–143.
Siaw, S. N., & Kee, D. W. (1987). Development of elaboration and organization in differ-
ent socioeconomic-status and ethnic populations. In M. A. McDaniel & M. Pressley (Eds.),
Imagery and related mnemonic processes. Theories, individual differences, and applications
(pp. 239–256). New York: Springer.
Siegel, L. S., & Ryan, E. B. (1989). The development of working memory in normally achieving
and subtypes of learning disabled children. Child Development, 60, 973–980.
Siegler, R. S. (1996). Emerging minds: The process of change in children’s thinking. New York:
Oxford University Press.
Siegler, R. S. (1998). Children’s thinking (3rd ed.). Upper Saddle River: Prentice Hall.
Siegler, R. S. (2006). Microgenetic analyses of learning. In D. Kuhn & R. S. Siegler (Vol.
Eds.), W. Damon & R. M. Lerner (Eds.-in-Chief), Handbook of Child Psychology: Vol. 2.
Cognition, perception, and language (pp. 464–510). New York: Wiley.
Simcock, G., & Hayne, H. (2002). Breaking the barrier: Children fail to translate their preverbal
memories into language. Psychological Science, 13, 225–231.
Simcock, G., & Hayne, H. (2003). Age-related changes in verbal and nonverbal memory during
early childhood. Developmental Psychology, 39(5), 805–814.
Slackman, E. A., & Nelson, K. (1984). Acquisition of an unfamiliar script in story form by young
children. Child Development, 55, 329–340.
Sluzenski, J., Newcombe, N. S., & Kovacs, S. L. (2006). Binding, relational memory, and recall
of naturalistic events: A developmental perspective. Journal of Experimental Psychology:
Learning, Memory, and Cognition, 32(1), 89–100.
Sluzenski, J., Newcombe, N. S., & Ottinger, W. (2004). Changes in reality monitoring and epi-
sodic memory in early childhood. Developmental Science, 7, 225–245.
Sluzenski, J., Newcombe, N. S., & Satlow, E. (2004). Knowing where things are in the second
year of life: Implications for hippocampal development. Journal of Cognitive Neuroscience,
16, 1443–1451.
Smirnov, A. A. (1973). Problems of the psychology of memory. New York: Plenum Press.
Smirnov, A. A., & Zinchenko, P. I. (1969). Problems in the psychology of memory. In M. Cole
& J. Maltzman (Eds.), A handbook of contemporary Soviet psychology (pp. 452–502). New
York: Basic Books.
Smith, B. S., Ratner, H. H., & Hobart, C. J. (1987). The role of cuing and organization in chil-
dren’s memory for events. Journal of Experimental Child Psychology, 44, 1–24.
Smith, D. A., & Graesser, A. C. (1981). Memory for actions in scripted activities as a function of
typicality, retention interval, and retrieval task. Memory and Cognition, 9, 550–559.
Smith, E. E., & Jonides, J. (1997). Working memory: A view from neuroimaging. Cognitive
Psychology, 33, 5–42.
Snyder, K. A. (2007). Neural mechanisms of attention and memory in preferential looking tasks.
In L. M. Oakes & P. J. Bauer (Eds.), Short- and long-term memory in infancy and early
childhood: Taking the first steps towards remembering (pp. 179–208). New York: Oxford
University Press.
Snyder, K. A., Stolariva, M., & Nelson, C. A. (2006). Neural correlated of novelty preferences.
Unpublished manuscript.
Sodian, B. (2005). Theory of mind. The case for conceptual development. In W. Schneider,
R. Schumann-Hengsteler, & B. Sodian (Eds.), Young children’s cognitive development:
Interrelationships among executive functioning, working memory, verbal ability, and theory
of mind (pp. 95–130). Mahwah: Erlbaum.
Sodian, B., & Schneider, W. (1990). Children’s understanding of cognitive cuing: How to manip-
ulate cues to fool a competitor. Child Development, 61, 697–704.
Sodian, B., & Schneider, W. (1999). Memory strategy development: Gradual increase, sudden
insight or roller coaster? In F. E. Weinert & W. Schneider (Eds.), Individual development from
References 375

3 to 12: Findings from the Munich Longitudinal Study (pp. 61–77). Cambridge: Cambridge
University Press.
Sodian, B., Schneider, W., & Perlmutter, M. (1986). Recall, clustering, and metamemory in
young children. Journal of Experimental Child Psychology, 41, 395–410.
Son, L. K., & Metcalfe, J. (2000). Metacognitive and control strategies in study-time allocation.
Journal of Experimental Psychology: Learning, Memory, and Cognition, 26(1), 204–221.
Sophian, C., & Wellman, H. M. (1983). Selective information use and perseveration in the search
behavior of infants and young children. Journal of Experimental Child Psychology, 35,
369–390.
Speer, J. R., & Flavell, J. H. (1979). Young children’s knowledge of the relative difficulty of rec-
ognition and recall memory tasks. Developmental Psychology, 15, 214–217.
Spence, M.-J. (1996). Young infants’ long-term auditory memory: Evidence for changes in pref-
erence as a function of delay. Developmental Psychobiology, 29, 685–695.
Squire, L. R. (1992). Declarative and nondeclarative memory: Multiple brain systems supporting
learning and memory. Journal of Cognitive Neuroscience, 4, 232–243.
Squire, L. R. (1999). Memory, human neuropsychology. In R. A. Wilson & F. C. Keil (Eds.), The
MIT encyclopedia of the cognitive sciences (pp. 520–522). Cambridge: MIT Press.
Squire, L. R., Knowlton, B., & Musen, G. (1993). The structure and organization of memory.
Annual Review of Psychology, 44, 453–495.
Squire, L. R., & Zola, S. M. (1996). Structure and function of declarative and non-declarative
memory systems. Proceedings of the National Academy of Sciences, 93, 13515–13522.
St Clair-Thompson, H. L., & Gathercole, S. E. (2006). Executive functions and achieve-
ments in school: Shifting, updating, inhibition, and working memory. Quarterly Journal of
Experimental Psychology, 59, 745–759.
Stein, B. S., Bransford, J. D., Franks, J. J., Owings, R. A., Vye, N. J., & McGraw, W. (1982).
Differences in the precision of self-generated elaborations. Journal of Experimental
Psychology: General, 111(4), 399–405.
Stern, C., & Stern, W. (1909). Erinnerung, Aussage und Lüge in der frühen Kindheit. Leipzig:
Barth.
Stern, W. (1910). Abstracts of lectures on the psychology of testimony and on the study of indi-
viduality. American Journal of Psychology, 21, 270–282.
Stern, W. (1914). Psychologie der frühen Kindheit. Leipzig: Quelle & Meyer.
Stern, W., & Stern, C. (1913). Die Kindersprache. Leipzig: Barth.
Sternberg, R. J. (1996). Costs of expertise. In K. A. Ericsson (Ed.), The road to excellence: the
acquisition of expert performance in the arts and sciences, sports and games (pp. 347–354).
Mahwah: Erlbaum.
Stipek, D. J. (1984). Young children's performance expectations: Logical analysis or wishful
thinking? In J. G. Nicholls (Ed.), The development of achievement motivation (pp. 121–142).
Greenwich: JAI Press.
Stormark, K. M. (2004). Skin conductance and heart-rate responses as indices of covert face rec-
ognition in preschool children. Infant and Child Development, 13, 422–433.
Tessler, M., & Nelson, K. (1994). Making memories: The influence of joint encoding on later
recall by young children. Consciousness and Cognition, 3, 307–326.
Thierry, K. L., & Spence, M. J. (2002). Source-monitoring training facilitates preschoolers’ eye-
witness memory performance. Developmental Psychology, 38, 428–437.
Thompson, W. C., Clarke-Stewart, K. A., & Lepore, S. J. (1997). What did the janitor do?
Suggestive interviewing and the accuracy of children's accounts, Law and Human Behavior,
21(4), 405–426.
Thompson, L. A., Fagan, J. F., & Fulker, D. W. (1991). Longitudinal prediction of specific cogni-
tive abilities from infant novelty preference. Child Development, 62, 530–538.
Tizzard-Drover, T., & Peterson, C. (2004). The influence of an early interview on long-term
recall: A comparative analysis. Applied Cognitive Psychology, 18, 727–743.
Towse, J., & Cowan, N. (2005). Working memory and its relevance for cognitive development.
In W. Schneider, R. Schumann-Hengsteler, & B. Sodian (Eds.), Young children’s cognitive
376 References

development: Interrelationships among executive functioning, working memory, verbal abil-

ity, and theory of mind (pp. 9–38). Mahwah: Erlbaum.
Towse, J. N., & Hitch, G. J. (1995). Is there a relationship between task demand and storage
space in tests of working memory capacity? Quarterly Journal of Experimental Psychology,
48A, 108–124.
Towse, J. N., Hitch, G. J., & Hutton, U. (1998). A re-evaluation of working memory capacity in
children. Journal of Memory and Language, 39, 195–217.
Trabasso, T., & Foellinger, D. B. (1978). Information processing capacity in children: A test of
Pasucal-Leone’s model. Journal of Experimental Child Psychology, 26, 1–17.
Trabasso, T., & Nicholas, D. W. (1980). Memory and inferences in the comprehension of narra-
tives. In F. Wilkening, J. Becker, & T. Trabasso (Eds.), Information integration by children
(pp. 215–242). Hillsdale: Erlbaum.
Triandis, H. C. (1989). The self and social behavior in differing cultural contexts. Psychological
Review, 96, 506–520.
Tulving, E. (1985). How many memory systems are there? American Psychologist, 40, 385–398.
doi:10.1037/0003-066X.40.4.385.
Tulving, E. (1972). Episodic and semantic memory. In E. Tulving & W. Donaldson (Eds.),
Organization of memory (pp. 381–403). New York: Academic Press.
Tulving, E. (2002). Episodic memory: From mind to brain. Annual Review of Psychology, 53,
1–25.
Tulving, E., & Madigan, S. (1970). Memory and verbal learning. Annual Review of Psychology,
21, 437–484.
Tulving, E. (2005). Episodic memory and autonoesis: Uniquely human? In H. S. Terrace & J.
Metcalfe (Eds.), The missing link in cognition: Origins of self-reflective consciousness (pp.
3–56). New York: Oxford University Press.
Turner, J. E., Henry, L. A., Smith, P. T., & Brown, P. A. (2004). Redintegration and lexical-
ity effects in children: Do they depend upon the demands of the memory task? Memory &
Cognition, 32(3), 501–510.
Turner, J. E., Henry, L. A., & Smith, P. T. (2000). The development of the use of long-term
knowledge to assist short-term recall. Quarterly Journal of Experimental Psychology, 53A,
457–478.
Tustin, K., & Hayne, H. (2010). Defining the boundary: Age-related changes in childhood amne-
sia. Developmental Psychology, 46, 1049–1061.
Ullman, H., Almeida, R., & Klingberg, T. (2014). Structural maturation and brain activity
predict future working memory capacity during childhood development. The Journal of
Neuroscience, 34, 1592–1598.
Underwood, B. J. (1957). Interference and forgetting. Psychological Review, 64, 49–60.
Usher, J. A., & Neisser, U. (1993). Childhood amnesia and the beginnings of memory for four
early life events. Journal of Experimental Psychology: General, 122, 155–165.
Van Bergen, P., Salmon, K., Dadds, M. R., & Allen, J. (2009). The effects of mother training
in emotion-rich, elaborative reminiscing on children’s shared recall and emotion knowledge.
Journal of Cognition and Development, 10, 162–187.
Van der Stel, M., & Veenman, M. V. J. (2008). Relation between intellectual ability and metacog-
nitive skillfulness as predictors of learning performance of young students performing tasks
in different domains. Learning and Individual Differences, 18, 128–134.
Van der Stel, M., & Veenman, M. V. J. (2010). Development of metacognitive skillfulness: A lon-
gitudinal study. Learning and Individual Differences, 20, 220–224.
Van Kraayenoord, C. E., & Schneider, W. (1999). Reading achievement, metacognition, reading
self-concept and interest: A study of German students in grades 3 and 4. European Journal of
the Psychology of Education, 14, 305–324.
Van Meter, P., & Pressley, M. (1994). The encoding of instruments when 10- to 14-year-olds
process isolated instrument-implicit sentences: More evidence of improved encoding during
childhood from elaborative instructions. Journal of Educational Psychology, 86, 402–412.
References 377

Van der Sluis, S., de Jong, P. F., & van der Leij, A. (2007). Executive functioning in children, and
its relations with reasoning, reading, and arithmetic. Intelligence, 35, 427–449.
Veenman, M. V. J. (2006). The role of intellectual and metacognitive skills in math problem solv-
ing. In A. Desoete & M. V. J. Veenman (Eds.), Metacognition in mathematics education (pp.
35–50). New York: Nova Science Publishers.
Veenman, M. V. J. (2011). Learning to self-monitor and self-regulate. In R. Mayer & P.
Alexander (Eds.), Handbook of research on learning and instruction (pp. 197–218). New
York: Routledge.
Veenman, M. V. J., Kok, R., & Blöte, A. (2005). The relation between intellectual and metacogni-
tive skills in early adolescence. Instructional Science, 33, 193–211.
Vertes, J. O. (1913). Das Wortgedächtnis im Schulkindesalter. Zeitschrift für Psychologie, 63,
19–128.
Vertes, J. O. (1931). Behalten und Vergessen des Kindes. Zeitschrift für Psychologie und
Physiologie der Sinnesorgane, 121, 241–354.
Vicari, S. (2001). Implicit versus explicit memory function in children with Down and Williams
syndrome. Down Syndrome Research and Practice, 7, 35–40.
Visé, M., & Schneider, W. (2000). Determinanten der Leistungsvorhersage bei Kindergarten-
und Grundschulkindern: Zur Bedeutung metakognitiver und motivationaler Einflussfaktoren
[Determinants of performance prediction in kindergarten and school children: The impor-
tance of metacognitive and motivational factors]. Zeitschrift für Entwicklungs-psychologie
und Pädagogische Psychologie, 32, 51–58.
Von der Linden, N., & Roebers, C. M. (2006). Developmental changes in uncertainty monitoring
during an event recall task. Metacognition and Learning, 1, 201–261.
Wagenaar, W. A., & Groeneweg, J. (1990). The memory of concentration camp survivors.
Applied Cognitive Psychology, 4, 77–87.
Wagner, D. A. (1974). The development of short-term and incidental memory: A cross-cultural
study. Child Development, 45, 389–396.
Wagner, D. A. (1978). Memories of Morocco: The influence of age, schooling, and environment
on memory. Cognitive Psychology, 10, 1–28.
Wagner, D. A. (1981). Culture and memory development. In H. Triandis & A. Heron (Eds.),
Handbook of cross-cultural psychology (Vol. 4, pp. 187–232). Boston: Allyn & Bacon.
Wagner, D. A., & Spratt, J. E. (1987). Cognitive consequences of contrasting pedagogies: The
effects of Quranic preschooling in Morocco. Child Development, 58, 1207–1219.
Wagoner, S. A. (1983). Comprehension monitoring: What it is and what we know about it.
Reading Research Quarterly, 18, 328–346.
Walker, C. H. (1987). Relative importance of domain knowledge and overall aptitude on acquisi-
tion of domain-related information. Cognition and Instruction, 4, 25–42.
Wang, Q. (2001). Culture effects on adults’ earliest childhood recollection and self-description:
Implications for the relation between memory and the self. Journal of Personality and Social
Psychology, 81, 220–223.
Wang, Q. (2007). “Remember when you got the big, big bulldozer?” Mother-child reminiscing
over time and across cultures. Social Cognition, 25, 455–471.
Wang, Q. (2008). Emotion knowledge and autobiographical memory across the preschool years:
A cross-cultural longitudinal investigation. Cognition, 108, 117–135.
Wang, Q. (2014). The cultured self and remembering. In P. J. Bauer & R. Fivush (Eds.), The
Wiley handbook on the development of children’s memory (Vol. II, pp. 605–625). Chichester:
Wiley.
Wang, Q., Doan, S. N., & Song, Q. (2010). Talking about internal states in mother-child
reminiscing influences children’s self-representations: A cross-cultural study. Cognitive
Development, 25, 380–393.
Wang, Q., Leichtman, M. D., & Davies, K. I. (2000). Sharing memories and telling stories:
American and Chinese mothers and their 3-year-olds. Memory, 8, 159–177.
Warrington, E. K., & Weiskrantz, L. (1968). A new method of testing long-term retention with
special reference to amnesic patients. Nature, 217, 972–974.
378 References

Waters, H. S. (2000). Memory strategy development: Do we need yet another deficiency? Child
Development, 71, 1004–1012.
Waters, H. S., & Andreassen, C. (1983). Children’s use of memory strategies under instruction.
In M. Pressley & J. R. Levin (Eds.), Cognitive strategy research: Psychological foundations
(pp. 3–24). New York: Springer.
Waters, H. S., & Kunnmann, T. W. (2010). Metacognition and strategy discovery in early child-
hood. In H. S. Waters & W. Schneider (Eds.), Metacognition, strategy use, and instruction
(pp. 3–22). New York: Guilford Press.
Waters, H. S., & McAlaster, R. (1983). Encoding variability and organization in free recall.
Journal of Experimental Child Psychology, 36, 380–395.
Waters, H. S., & Schneider, W. (2010). Common themes and future challenges. In H. S. Waters &
W. Schneider (Eds.), Metacognition, strategy use, and instruction (pp. 281–288). New York:
Guilford Press.
Waters, H. S., & Waters, E. (1979). Semantic processing in children’s free recall: The effects of
context and meaningfulness on encoding variability. Child Development, 45, 735–746.
Webb, S. J., & Nelson, C. A. (2001). Perceptual priming of upright and inverted faces in children
and adults. Journal of Experimental Child Psychology, 79, 1–22.
Weigle, T. W., & Bauer, P. J. (2000). Deaf and hearing adults‘ recollections of childhood and
beyond. Memory, 8, 293–309.
Weinert, F. E. (1962). Untersuchungen über einige Bedingungen des sprachlichen Lernens bei
Kindern und Jugendlichen. Vita humana, 5, 185–194.
Weinert, F. E. (1986). Developmental variations of memory performance and memory-related
knowledge across the life-span. In A. Sörensen, F. E. Weinert, & L. R. Sherrod (Eds.), Human
development: Multidisciplinary perspectives (pp. 535–554). Hillsdale: Erlbaum.
Weinert, F.E., & Kluwe, R.H. (Eds.). (1987). Metacognition, motivation and understanding.
Hillsdale, NJ: Erlbaum.
Weinert, F. E., & Perlmutter, M. (Eds.). (1988). Memory development: Universal changes and
individual differences. Hillsdale: Erlbaum.
Weinert, F. E., & Schneider, W. (Eds.). (1986). First report on the Munich Longitudinal Study
on the Genesis of Individual Competencies (LOGIC). Munich: Max Planck Institute for
Psychological Research.
Weinert, F. E., & Schneider, W. (Eds.). (1995). Memory performance and competencies. Issues in
growth and development. Mahwah, NJ: Erlbaum.
Weinert, F. E., & Schneider, W. (Eds.). (1999). Individual development from 3 to 12. Findings
from the Munich Longitudinal Study. Cambridge, UK: Cambridge University Press.
Weinert, F. E., Schneider, W., & Knopf, M. (1988). Individual differences in memory develop-
ment across the life-span. In P. B. Baltes, D. L. Featherman, & R. M. Lerner (Eds.), Life-span
development and behavior (Vol. 9, pp. 39–85). Hillsdale: Erlbaum.
Weinert, F. E., & Weinert, S. (1998). History and systems of developmental psychology. In A.
Demetriou, W. Doise, & C. van Lieshout (Eds.), Life-span developmental psychology (pp.
1–33). New York: Wiley.
Weizmann, F., Cohen, L. B., & Pratt, R. J. (1971). Novelty, familiarity and the development of
infant attention. Developmental Psychology, 4, 149–154.
Welch-Ross, M. K. (1995). An integrative model of the development of autobiographical mem-
ory. Developmental Review, 15, 338–365.
Welch-Ross, M. K., Fasig, L., & Farrar, M. J. (1999). Predictors of preschoolers’ self-knowl-
edge: Reference to emotion and mental states in mother-child conversation about past events.
Cognitive Development, 14, 401–422.
Wellman, H. M. (1977). Preschoolers’ understanding of memory-relevant variables. Child
Development, 48, 1720–1723.
Wellman, H. M. (1978). Knowledge of the interaction of memory variables: A developmental
study of metamemory. Developmental Psychology, 14, 24–29.
Wellman, H. M. (1983). Metamemory revisited. In M. T. H. Chi (Ed.), Trends in memory devel-
opment research (pp. 31–51). Basel: Karger.
References 379

Wellman, H. M. (1985). The origins of metacognition. In D.-L. Forrest-Pressley, G. E.

MacKinnon, & T. G. Waller (Eds.), Metacognition, cognition, and human performance (Vol.
1, pp. 1–31). Orlando, FL: Academic Press.
Wellman, H. M. (1988). The early development of memory strategies. In F. Weinert & M.
Perlmutter (Eds.), Memory development: Universal changes and individual differences (pp.
3–30). Hillsdale: Erlbaum.
Wellman, H. M. (1990). The child’s theory of mind. Cambridge: MIT Press.
Wellman, H. M., Collins, J., & Glieberman, J. (1981). Understanding the combination of
memory variables: Developing conceptions of memory limitations. Child Development, 52,
1313–1317.
Wellman, H. M., Cross, D., & Watson, J. (2001). Meta-analysis of theory-of-mind development:
the truth about false belief. Child Development, 72, 655–684.
Wellman, H. M., Ritter, K., & Flavell, J. H. (1975). Deliberate memory behavior in the delayed
reactions of very young children. Developmental Psychology, 11, 780–787.
Welsh, J. A., Nix, R. L., Blair, C., Bierman, K. L., & Nelson, K. E. (2010). The development of
cognitive skills and gains in academic school readiness for children from low-income fami-
lies. Journal of Educational Psychology, 102(1), 43–53.
Welsh, M. C., Pennington, B. F., & Groisser, D. B. (1991). A normative-developmental study
of executive function: A window on prefrontal function in children. Developmental
Neuropsychology, 7, 131–149.
Werner, H. (1926). Einführung in die Entwicklungspsychologie. München: Barth.
Werner, H. (1953). Einführung in die Entwicklungspsychologie (4th ed.). München: Barth.
Werner, J. S., & Perlmutter, M. (1979). Development of visual memory in infants. In H. W. Reese
& L. P. Lipsitt (Eds.), Advances in child development and behavior (Vol. 14, pp. 1–56). New
York: Academic Press.
West, T. A., & Bauer, P. J. (1999). Assumptions of infantile amnesia: Are there differences
between early and later memories? Memory, 7, 257–278.
Wetzler, S. E., & Sweeney, J. A. (1986). Childhood amnesia: An empirical demonstration. In D. C.
Rubin (Ed.), Autobiographical memory (pp. 191–201). New York: Cambridge University Press.
Whipple, G. M. (1909). The observer as reporter: A survey of the “psychology of testimony”.
Psychological Bulletin, 6, 153–170.
Whipple, G. M. (1911). The psychology of testimony. Psychological Bulletin, 8, 307–309.
White, S. H., & Pillemer, D. B. (1979). Childhood amnesia and the development of a socially
accessible memory system. In J. F. Kihlstrohm & F. J. Evans (Eds.), Functional disorders of
memory (pp. 29–73). Hillsdale: Erlbaum.
Whittaker, S. J. (1986). Eliminating alternatives: Preschool children’s use of indirect memory
cues. British Journal of Developmental Psychology, 4, 199–207.
Whittaker, S., McShane, J., & Dunn, D. (1985). The development of cueing strategies in young
children. British Journal of Developmental Psychology, 3, 153–161.
Wiebe, S. A., Espy, K. A., & Charak, D. (2008). Using confirmatory factor analysis to understand exec-
utive control in preschool children: I. Latent structure. Developmental Psychology, 44, 575–587.
Wiebe, S. A., Sheffield, T., Nelson, J. M., Clark, C. A. C., Chevalier, N., & Espy, K. A. (2011).
The structure of executive function in 3-year-olds. Journal of Experimental Child Psychology,
108, 436–452.
Willoughby, M. T., Blair, C. B., Wirth, R. J., & Greenberg, M. (2010). The measurement of exec-
utive function at age 3 years: Psychometric properties and criterion validity of a new battery
of tasks. Psychological Assessment, 22, 306–317.
Wimmer, H. (1976). Aspekte der Gedächtnisentwicklung [Aspects of memory development].
Zeitschrift für Entwicklungspsychologie und Pädagogische Psychologie, 8, 62–78.
Wimmer, H., & Perner, J. (1983). Beliefs about beliefs: Representation and constraining function
of wrong beliefs in young children's understanding of deception. Cognition, 13, 103–128.
Wimmer, H., & Tornquist, K. (1980). The role of metamemory and metamemory activation in the
development of mnemonic performance. International Journal of Behavioral Development, 3,
71–81.
380 References

Winner, E. (1996). Gifted children: Myths and realities. New York: Basic Books.
Woloshyn, V. E., Paivio, A., & Pressley, M. (1994). Use of elaborative interrogation to help stu-
dents acquire information consistent with prior knowledge and information inconsistent with
prior knowledge. Journal of Educational Psychology, 1, 79–89.
Woloshyn, V. E., Pressley, M., & Schneider, W. (1992). Elaborative-interrogation and prior-
knowledge effects on learning of facts. Journal of Educational Psychology, 84, 115–124.
Wood, E. J., Winne, P., & Pressley, M. (1988). Elaborative interrogation, imagery, and provided
precise elaborations as facilitators of children’s learning of arbitrary prose. Presented at the
annual meeting of the American Educational Research Association. New Orleans, LA.
Woody-Dorning, J., & Miller, P. H. (2001). Children’s individual differences in capacity: Effects
on strategy production and utilization. British Journal of Developmental Psychology, 19(4),
543–557.
Worden, P. E. (1974). The development of the category-recall function under three retrieval con-
ditions. Child Development, 45, 1054–1059.
Worden, P. E., & Sladewski-Awig, L. J. (1982). Children's awareness of memorability. Journal of
Educational Psychology, 74, 341–350.
Yendovitskaya, T. V. (1971). Development of memory. In A. V. Zaporozhets & D. B. Elkonin
(Eds.), The psychology of preschool children (pp. 89–110). Cambridge: MIT Press.
Yonelinas, A. P. (2002). The nature of recollection and familiarity: A review of 30 years of
research. Journal of Memory and Language, 46, 441–517.
Yussen, S. R., & Bird, J. E. (1979). The development of metacognitive awareness in memory,
communication, and attention. Journal of Experimental Child Psychology, 28, 300–313.
Yussen, S. R., & Levy, V. M. (1975). Developmental changes in predicting one's own memory
span of short-term memory. Journal of Experimental Child Psychology, 19, 502–508.
Yussen, S. R., & Levy, V. M. (1977). Developmental changes in knowledge about different
retrieval problems. Developmental Psychology, 13, 114–120.
Zabrucky, K., & Ratner, H. H. (1986). Children's comprehension monitoring and recall of incon-
sistent stories. Child Development, 57, 1401–1418.
Zajac, R., & Karageorge, A. (2009). The wildcard: A simple technique for improving children’s
target-absent lineup performance. Applied Cognitive Psychology, 23, 358–368.
Zelazo, P.D., Müller, Frye, D., & Marcovitch, S. (2003). The development of executive function
in early childhood. Monographs of the Society for Research in Child Development, 68 (3,
Serial No. 274).
Zola, S. M., & Squire, L. R. (2000). The medial-temporal lobe and the hippocampus. In E.
Tulving & F. I. M. Craik (Eds.), The Oxford handbook of memory (pp. 485–500). New York:
Oxford University Press.
Index

A comparison of infant performance, 45f

Absolute mental capacity, 148 developmental progression in, 44f
Accuracy-oriented memory behavior, 292 Arbitrary-invariant events, 82, 83t
Accurate memory, 128, 129, 284, 314 Asian American parents, 99
Activation field, 147 Associative learning, 18, 25, 204, 206, 207
Active rehearsal, 136, 195, 196, 197 Attentional refreshing, 153, 154
Age-memory association, 44, 90, 105, 110, Attention-deficit/hyperactivity disorders
270, 275, 286, 324. See also Age- (ADHD), 34
related changes/differences Auditory attention, 168
in declarative memory, 313 Auditory stimuli, 49
developmental studies, 155–156, 157, 220 Autobiographical memory, 22, 23, 72, 75,
in false memory, 127 79–80, 92, 115, 313–315. See also
knowledge base in, 221–224 Infantile amnesia
in performance, 47, 135, 142, 149, 150, adult-child conversation, 92–93
153, 160 conversations during events, 95–97
source memory in, 121 cross-cultural comparisons, 98, 99
in STS, 138 during development, 76f
Age-related changes/differences elaborative talk, 96–97
in cognition, 149 event recall, 79–80
in conceptual priming, 32, 36 event recall format, 80–82
in declarative memory, 73, 80 lasting time, 99–101
in explicit memory, 30, 31 maternal elaborations and, 94f, 106
in identification lineup task, 114 memory stabilization, 88
in implicit memory, 29 organization of, 100
in long-term memory, 63, 64, 65f, 66 reminiscing training, 96
in occipital power values, 48 sources of, 76f, 77
in remembering (see Age-memory Autonoetic consciousness, 105
association) Autosuggestion techniques, 14
in sequential ordering skill, 139
in synaptic efficiency, 55
Alphabetic retrieval strategy, 216 B
Amnesia test, 70 Background knowledge, 113–115
Analysis of covariance (ANCOVA) results, confidence-accuracy (CA) relation, 114
246 Baddeley’s model, 157, 179, 180, 181, 315,
Anoetic, 25 316
A-not-B task, 43, 44, 45, 46, 48 differentiation hypothesis, 167

© Springer International Publishing Switzerland 2015 381

W. Schneider, Memory Development from Early Childhood Through
Emerging Adulthood, DOI 10.1007/978-3-319-09611-7
382 Index

EF components, 168 memory experiments with adults, 10

EF-WM model, basic disagreement, 170 mind of child, 9–10
empirical tests of, 160–171 Russian research, 18–20
episodic buffer, 175, 176 testimonial competence, 10
memory for static and dynamic informa- Childhood amnesia. See Infantile amnesia
tion, 165, 166f Children
original WM model, 150, 151, 172 animal game activities, 84
and Pascual-Leone’s model, 155 eyewitness memory, 14–15
revised model, 153 fixed and variable items memory, 83–84
Basic memory capacities, 3, 4, 6, 131–133, memory cues, 86
180, 181, 316 novel events memory, 87–88
information processing speed and, salient events memory, 88–90
139–144 Chunking, 137, 138
item-ordering ability on, 138–139 Cognitive development, 144, 283, 303, 319,
short-term capacity, 133–136 320, 325
strategic variables on, 136–138 future of, 174
Behavioral re(-)enactments, 62, 80 neo-Piagetian theories, 145–148
Behavioral theories, 2, 9 working memory models, 148–154
Behaviourism, 6, 17 Cognitive neuroscience, 36, 73, 127, 129, 261,
age-related changes in, 73–74 311, 313, 315, 321
organizational behavior, 200, 202, 210 Cognitive performance, 149, 154, 167, 253,
strategic behavior, 186–193, 296, 301 319
Brain imaging studies, 68 Cognitive primitive, 149, 180, 315
Brain structures, 35, 36, 73, 311, 313 Cognitive revolution, 1, 2, 5, 20, 22, 310
Broca’s area, 171, 172 Collective reconstruction measure, 138
Compound visual stimulus information task
(CVSI), 145
C Concentration, 111, 187, 249, 252
Capacity, 144. See also Basic memory Conceptual knowledge, 32, 222, 223, 231,
capacities 232, 237
non-Piagetian models, 145–148 Conceptual priming
working memory models, 148–154 age-related effects, 31, 36
Case’s resource-sharing model, 157–160 categorical relations, 32
cognitive space hypothesis, 158 development of, 30–32
memory decay hypothesis, 158 Conditional metacognitive knowledge, 257
reactivation process, 159 Conditioned response (CR), 55
Category typicality, 201, 222, 237 Conditioned stimulus (CS), 55
Central computing space, 22, 145 Conditioning paradigms, 42, 54–57
Central executive (CE), 22, 149, 275 Confidence judgments (CJs), 265, 283–285
control over attention, 151f Constructive memory, 238, 239
functions of, 170, 171, 175 Content knowledge, 231, 237
-loaded higher-order processing skills, 166 Control-monitoring (CM) model, 292, 294
in revised model of working memory, 146f, Control processes, 135, 151, 183, 193, 258,
147 266, 285–294
Cerebellum, 36, 41, 67, 71 Control skills, structural equation model, 306f
Change-detection paradigm, 50–52 Corsi blocks task, 152f, 155, 156, 157, 164,
schematic representation of, 51f 165
Child psychology, 10 Cowan’s working memory model, 151f, 152f,
age-graded behavioral inventory, 10 176–179
American research, 17–18 central global processing rate, 177
cognitive revolution, 20 developmental change, 179f
foundation in Germany, 9 picture-name-rhyming computer game,
German research, 17 178
Index 383

Cracker-choice task, 53 Domain knowledge, 132, 253

Cue-utilization model, 275, 279 limited effects on subsequent learning,
Cued-recall tasks, 42, 215, 235, 297 240–241
Cumulative rehearsal, 140, 194, 195, 196, 197, Domain-general mechanism, 140
199, 218t, 219, 223, 303 Domain-related tasks, 249
Domain-specific knowledge, 4, 6, 229, 231,
234, 246, 252, 316
D and age difference, 246
Declarative memory, 2, 21, 40, 68, 73, 132, on memory development, 22
310, 311, 313, 314, 321. See also processing mechanisms, 149
Explicit memory Dorsolateral prefrontal cortex (DLPFC), 133,
and nondeclarative memory, 67 173
and VPC task, 69 Down syndrome, 34
Declarative metacognitive knowledge, 257, Dual-process retrieval models, 125
259, 261, 263, 307, 319 fuzzy-trace theory in, 126, 129
Declarative metamemory, 256, 261f
development, 269–275
measures of, 262–263 E
problems, 263–264 Early childhood, 311–313
Declarative visual recognition memory, 69 long-term memory, 312
Deese-Roediger-McDermott (DRM) para- short-term memory, 312
digm, 33, 124 Early encoding strategies, 189–190, 191f
Deferred imitation tasks, 57–59, 71 category cues, 189
Deferred-elicited imitation, 59, 66, 69, 70, 71, consistency, 190
72, 73, 312 effectiveness, 190
Delayed-JOL effect, 278, 279, 280f intentionality of memory behavior, 190
Delayed-nonmatching-to-sample (DNMS) organization, 192
task, 46–47 play-and-remember condition, 189
age-related performance differences, 47–48 sort-and-remember condition, 189
individual differences, 48 Early retrieval strategies, 186–188
infant frontal lobe development and, 48 concentration, 187
Delayed-reaction procedure, 39. See also context-specific, 193
Delayed-response task mother-child-memory talks, 192–193
Delayed-response task, 43, 44, 46 nonverbal memory, 187
developmental progression in, 44f retrieval cues, 188
Developmental differences Early-verbal children, 64
in dual-task performance, 219, 220, 221 Ease-of-learning (EOL) judgments, 265, 266,
in event recall, 79–80, 196, 197, 202, 204, 276–277
212, 235 Effort attribution bias, 283
in event recall format, 80–81 Elaborated knowledge base, 222
in JOLs, 280 Elaboration, 204–210
in learning and retention, 66, 131, 133, associative learning, 207
134f, 135, 136 elaborative transfer, 207
in memory, 20, 44, 137, 159, 160, 169, -generated condition, 205
184, 274, 316 imagery strategy instruction, 206–209
in source monitoring, 121 from life span perspective, 209–210
in style of rehearsal, 194 maintenance effects, 207
Developmental invariance hypothesis, 28, 29, in noun-pair learning, 205
35, 37, 311 -provided condition, 205
Developmental theory, 15, 76, 149 repetition-control condition, 205, 206
Developmental-representational theory, 262 transfer task, 208, 209f
Diencephalon, 68 Elaborative interrogation, 242, 244, 253
Digit span forward task, 171 Canadian facts, 245t
384 Index

versus precise elaboration, 243–244 Experiential factors, 181, 316

West German facts, 245t Expert-novice paradigm, 246, 247, 249, 251,
Elaborative talk, 96–97, 98 253
Elicited imitation, 58 Explicit memory, 2, 21, 27, 40, 131. See also
deferred imitation and, 59 Declarative memory
paradigm, 60–61 explicit declarative memory, 25, 41, 68,
Emotion, 97, 107, 321 310
-memory interactions, 322 deferred/elicited imitation tasks, 70
Empty narration, 63, 64 development course of, 321
Encoding strategies, during school years, 193 and implicit memory, 32–33, 68
elaboration, 204–210 in infants, 67–72
organizational strategies and, 197–204 External retrieval cues, 212–214
rehearsal, 193–197 Eyewitness memory, 108–111
for remembering complex materials, cognitive capacity, 110
210–212 correct and incorrect responses by age,
Episodic buffer, 150, 151, 153, 157, 175–176, 118f
177, 181 field or naturalistic studies, 109
Episodic knowledge, 231 interviewer characteristics, 117
Episodic memory, 2, 6, 10, 21, 75, 77, 106, laboratory-style studies, 109
108, 176. See also Autobiographical lasting time, 111–113
memory; Event memory stereotype condition, 119
ontogeny of, 209, 210 suggestibility, age differences in, 116–119
and semantic memory, 67, 72 Eyewitness research, 129, 314
tasks, 321, 323 Eyewitness situations
European American mothers, 98–99 facilitating young children’s performance
Event memory, 75, 313–315. See also in, 127–128
Autobiographical memory; interview techniques, 129
Sociocultural contexts mental context reinstatement, 128
event variability, 82, 83 nonverbal techniques, 128
during events conversations, 95–97 Eyewitness testimonies, 6, 14, 15, 116
and eyewitness memory, 109
novel events, 87–88
past events conversations, 91–94 F
prior knowledge on, 90–91 Face-recognition tasks, 115
restaurant event, 79f Facial recognition memory, 114
of routine events, 78–82 False memory creation, 121–125
salient events, 88–90 age-related increases in, 127
Event-related potentials (ERPs), 35, 59, 69, external suggestions, 127
70, 121, 321 fuzzy-trace theory’s prediction in, 124
Events, salience of, 115–116 implanted memories, 122
Executive functioning (EF), 166, 168, 170, memory errors, 122
305, 306 memory-sharing situations, 123
structural equation model, 306f percentage of reports, 122f
Executive-loaded working memory (ELWM), rumors, 122
170, 173, 174 studies of, 33
Experiences, 295 Familiarity
-based judgments, 276 process dissociation analyses, 126
effects on memory, 37, 90–91, 311. See and recollection, 125–127
also Event memory Feeling-of-knowing (FOK) judgments, 265,
in memory reinstatement and reactivation, 267, 281–283
61–63 age difference in, 281
past experience, 3, 37, 39, 42, 92, 97, 98, Field of mental attention, 147
106, 193, 311 Fixed items, memory for, 83–84
Index 385

Flavell, John, 3, 6, 21, 183, 184f Imagery strategy instruction, 206–208, 209f
Fluency, 33, 168, 170, 292 Immediate JOLs, 278, 280f
Forgetting, 9, 13, 90, 100, 309 Immediate memory, 13, 16
curve, 103 constructive aspects of, 12
as function of age, 66 Implicit memory, 2, 5, 21. See also Procedural
of habituated stimulus, 50 memory
mental verb, 268, 273 in clinical populations, 33–34
Freud’s model of repression, 103 development, 25–37, 310–311, 321
Full narration, 63, 64 and explicit memory, 32–33, 68
Functional magnetic resonance imaging in infants, 67–72
(fMRI), 35, 171, 174 Inaccurate memory, 284
Fuzzy-trace theory (FTT), 104, 112 Incidental memory, 25, 234
in dual-process retrieval models, 126 Independent locomotion, 59, 60
gist-connecting mechanism, 125 Individual-item JOLs, 278
in rumored event, 124, 125 Infancy, 311–313
susceptibility to suggestion and, 119 hide-and-seek tasks, 43–48
novelty-preference paradigms, 48–51
short-term/working memory in, 42
G Infantile amnesia, 11, 101–103
Gamma correlations, 281, 283, 288 autonoetic consciousness, 105
General intelligence, 249 boundary of, 106
cultural differences, 141, 142 explanations and evaluations, 103–108
General memory, 309, 325 forgetting curve, 103
development, 15–16 hippocampal neurogenesis, 107–108
General strategy knowledge, 228 information-processing system, 104
Global processing speed, 140, 149 social interaction, 105–106
Good strategy user model, 4 Western versus Eastern cultures, 102
Good-information processing model, 228, Infants
259, 304 behavior range, 61
Grouping, 135 development of long-term memory in,
active grouping, 137 54–57
Growth-chart-type functions, 65 duration of retention, 57f
implicit and explicit memory, 67–72
pre-explicit declarative memory, 68
H train task, 56
Habituation/dishabituation procedure, 40 transition period, 59–61
Hardware, 21, 131, 309 Inferential memory, 238
neural hardware, 73, 101, 313 Information processing, 1, 2, 9, 309, 310
Hide-and-seek tasks, 43–48. See also Delayed- and brain development, 320
response task limited capacity, 132
High-elaborative mothers, 93 in WM concept, 42
European American mothers, 98–99 and working memory models, 148–154
in rumored events, 123–124 Information processing speed, 135, 136, 181,
High-mnemonic teachers, 303 234, 246, 315, 316
Hippocampus, 60, 68, 321 and age difference, 246
primitive recognition, 70 developmental changes in, 139–141
visual attention and memory, 70 and memory span, 141–144
and VPC task, 69 Inhibition, 18, 147, 149, 166, 167, 168, 170
active suppression process, 169
Inner scribe, 165
I Instrument-explicit sentences, 238
Identification lineup task, 114 Instrument-implicit sentences, 238, 239
Imagery, 11, 221 Instrument-reference sentences, 238
386 Index

Intelligence, 3, 141, 299 L

knowledge and memory performance, Language developments, 92, 106, 129, 267,
249–252 314
nonverbal, 252, 269 Language processing system, 175
Intentional learning, 242 Letter span, developmental differences in, 134f
as function of learning procedure, 243f Limited capacity theories, 132, 145
Intelligence quotient (IQ), 50, 119, 246, 249, Lindberg’s hypothesis, 235
301, 302 List-learning paradigms, 265, 276
in metamemory illustration, 299, 300f Logical memory, 16, 163
recall tasks, 250, 251f Logical-invariant events, 82, 83t
Intentional lies, 14 Logical-variant events, 82, 83t
Intentional memory behaviors, 20, 187 Longitudinal studies, 202
Interitem associativity, 201, 222, 232, 237 drawback of, 204
Internal retrieval cues, 212–214 need for, 324
International conflict schema, 247 Long-term knowledge, 108, 162, 164, 244
Interstimulus interval (ISI), 55 Long-term memory (LTM), 11, 13, 22, 41,
Involuntary memory, 19, 25 131. See also Toddlers’ long-term
Item identification process, 139 memory
conditioning paradigms, 54–57
deferred imitation tasks, 57–59
J impact of language acquisition on, 63–64
Judgments of learning (JOLs), 265, 278–281 Long-term store (LTS), 1, 2, 131, 310
individual-item versus aggregate JOLs, 278 Low-elaborative mothers, 93
in rumored events, 123–124
Low-mnemonic teachers, 303
K
Knowledge. See also Domain knowledge
of animals, 233f M
versus aptitude effects, 251 Maturational factors, 181, 226, 316
conceptualizations of, 231–234 Mechanical learning, 16, 19
intelligence and memory performance, Medial temporal lobe (MTL), 60, 68, 321
249–250 amnesia, 70
in memory and learning, 238–240 Mediation
and memory performance, 231 deficiency, 185
of mental verbs, 268 theories, 183
Knowledge acquisition, 228, 241, 323, 324 Memorizing-effort heuristics, 292, 293
Knowledge base, 231, 249 Memory capacity development of, 315–316
definition, 232 Memory cues, 62, 86, 213, 216
on memory development, 318–319 Memory decay hypothesis, 158, 159
Knowledge base effects, 221–224, 234 Memory deceptions, 14
Battig and Montague (B&M) categories, Memory development, 2, 326
236f brain basis of, 7
category typicality, 222 development of memory, 324, 325, 326
effortless activation of semantic memory encoding and retrieval processes, 22
relations, 236–237 general metacognitive knowledge, 22
interitem associativity, 222 hardware, 21
item-specific effects, 234–236 intelligence hypothesis of, 3
nonstrategic effects of, 223 knowledge base on, 318–319
reflective abstraction, 223 memory strategies, 21
and strategies, 237, 241–245 modern era of, 183
on subsequent learning and memory, research spurt on, 3
246–249 software, 21
Index 387

strategic behavior in young children, Metacognitive knowledge, 113–115, 132, 256,

186–193 260–262
Memory errors, 33, 108, 122, 124 conceptualizations of, 256–257
Memory models. See Theoretical working monitoring, 114
memory models, empirical tests of structural equation model, 306f
Memory monitoring deficiency hypothesis, 277 target-absent lineups, 115
Memory performance Metamemory, 4, 6–7, 22, 224, 255
changes in, 12 and age differences, 246
and child’s strategy, 184 concept, 258
focusing attention, 185 development of (see Metamemory
intelligence and knowledge, 249–252 development)
and knowledge, 231 measures of declarative metamemory,
in strategy use deficiencies, 184–185 262–263
Memory plasticity, 107, 209, 321 on strategy use, 224–225
Memory processes, 11, 47, 67, 70, 150, 177, Metamemory and memory relations, 294–296
202, 232, 255, 262, 307, 319 causal modeling procedures, 300f
Memory reactivation, 61–63 correlational studies, 296–297
upper limit (UL) of, 63f evidence from multivatiate analyses,
Memory reinstatement, 61–63, 73, 128, 313 299–302
Memory rich teaching, 304 evidence from training studies, 298–299
Memory skills, 10, 39, 92, 97, 132, 193, 202, Metamemory development, 319–320
229, 229, 309, 322, 323, 324 declarative metamemory development,
Memory span development 269–275
impact of item-ordering ability on, development of monitoring, 275–276
138–139 ease-of-learning judgments (see Ease-of-
information processing speed, 141–144 learning (EOL) judgments)
strategic variables on, 136–138 feeling-of-knowing (FOK) judgments (see
Memory span tasks, 133 Feeling-of-knowing (FOK) judgments)
developmental differences in, 134f judgments of learning (see Judgments of
Memory strategies, 4, 132 learning (JOLs))
development of, 229, 316–317 monitoring and control processes, 285–294
memory talk, 230 precursors of metamemory, 268–269
teacher-child conversations, 230 retrieval monitoring, 283–285
Memory strength, 16, 284, 285 self-control, 275–276
Memory structures. See Brain structures Misremembering, 273
Memory system Mobile conjugate reinforcement, 40, 55, 57,
impact of advances in neuroscience, 34–35 69, 71, 73, 312
proposed taxonomy of, 26f Modal memory study model, 186
Memory talk, 192, 227, 230, 303, 317, 323 Model of good-information processing, 253, 319
Memory variable interaction, 274 Monitoring, 275–276
Memory vividness, 285 and control processes, 285–294
Memory-for-location tasks, 271 retrieval monitoring, 283–285
Memory-knowledge interactions, 323 study time monitoring, 293f
Mental space, 145. See also M-space Monitoring control (MC) model, 292
Metacognition, 249, 256 M-space, 145, 146, 147, 154, 155, 157
acquisition procedures, 228 Multimodal temporary store, 150
experiences, 256, 259 Multiple strategy use, 216–218
importance for education, 302–307 factors affecting, 218–229
main components of, 258f Multistore models of memory, 1
skills, 256, 259 Munich Longitudinal Study on the Genesis
taxonomy of, 261f of Individual Competencies (LOGIC),
vocabulary, 269 134, 139, 202, 203, 270, 272
388 Index

N organizational abilities, 199

National Institute of Child Health and Human ratio of repetition (RR) measure, 198
Development (NICHD), 306 results, 201t
Negative salient events, 100 sort-recall tasks, 200, 201
Neocortex, 35, 41, 67 Overestimation, 277, 283, 284
Neo-Piagetian models, 145–148
Neural hardware, 73, 101, 313
Neural network, 2, 41, 68 P
Neuropsychological evidence, 171–174 Paired-associate tasks, 237, 279, 286
brain activity role, 174 development differences, 286
development of ELWM, 173–174 Parent-child interactions, 323
differences in WM capacity, 172–173 Partial compensation models, 252
neural basis of developmental trends, 172 Pascual-Leone’s model, 154–157
Nine-month olds, 59–61 and Baddeley’s model, 155
memory reactivation and reinstatement, Pascual-Leone’s theory of constructive opera-
61–63 tors, 145
Nondeclarative memory, 2, 41 Passive rehearsal, 195
and declarative memory, 67 Perceptual priming, 33, 35, 36, 37
in infants, 68 and conceptual priming, 31, 311
Nonverbal memory, 16, 187 development of, 26–30
Novel events, memory for, 87–88 Performance prediction, 265, 276
Novelty-preference paradigms, 48–51 Perirhinal cortex, 68
Person variables, 269
Phonological (articulatory) loop (PL), 150,
O 160, 161, 164
Object search task, 43. See also A-not-B task of verbal memory span, 163
OECD PISA study, 264 Photorecognition, age-related changes, 65f
Offline methods, 266 Piaget’s theory, 39
Online methods, 266 Picture-sequencing tasks, 80
Open-ended questions, 85, 90, 92–93, 94, 95, Planning, 169
96, 100, 113, 128 Positron emission tomography (PET), 171
Operant conditioning, 25, 40, 42, 54, 55, 71, Postdictions, 265, 277
73, 312 Preference-for-novelty procedure, 40
Operating space, 148, 158 Prefrontal cortex (PFC), 68, 121, 133, 173, 321
Ordered recall, 58, 59, 61, 70 Premotor area, 172
Organization, 183, 184, 219 Priming, 37. See also Conceptual priming;
conceptual organization, 272 Perceptual priming
of items, 199 neurocortex areas, 35
of memory, 100, 190, 323 Prior knowledge, 231, 238, 239. See also
of nonverbal working memory, 172 Experiences
semantic organization, 296 Procedural memory, 2, 21. See also Implicit
subjective organization, 197, 198 memory
Organizational strategies, 197–204 Procedural metamemory, 256, 257, 261f, 262
adjusted ratio clustering (ARC) measure, measures of, 264–266
198, 200 problems, 266–267
development of memory, 204 Production deficiencies, 184, 185, 294
developmental differences and changes, 202 PsychINFO database, 23, 40
developmental trends, 199 infant memory publications, 41f
free-recall tasks, 200, 201
individual patterns of strategy acquisition,
203f R
longitudinal evidence, 202 Readily functional associative ability, 176
microgenetic study, 204 Reading-to-understand condition, 244
Index 389

Canadian facts, 245t fuzzy-trace theory’s prediction in, 124

West German facts, 245t in memory, 123
Recall, 73
as function of age, 236f
future research of, 74 S
patterns, 247, 248f Salient events, memory for, 88–90
readiness assessments, 266, 267 age-related differences, 89f
Recognition, 11, 25, 26, 27, 28, 33, 48 Schema-confirmation-deployment model,
auditory, 47, 49 84–85
visual, 49, 50, 69 confirmation phase, 85
Recognition test, 62, 239, 273, 281 deployment phase, 85
of animals, 46 Script acquisition process, 82
skin conductance responses, 35 Script-based memory, 78–79
Recollection, 21, 25, 27, 70, 129, 310, 315 versus unfamiliar novel information, 81
and familiarity, 125–127 Self-concept, 77, 92, 96, 92, 106, 256, 269,
Redintegration, 162 305
Re(-)enactment, 62, 65 structural equation model, 306f
age-related changes, 65f Self-control, 261, 275, 292, 305
nonverbal, 80 in study time allocation, 287f
Reflective abstraction, 223 Self-monitoring, 265, 266, 275, 282
Regrouping, 137 Self-recognition, 98, 105
Rehearsal, 135, 183, 193–197 Self-regulation processes, 255, 257, 265, 266,
contiguity effect, 197 275, 276, 286, 288, 289, 292, 294, 308,
control processes, 193 319, 320
individual stability, 195 Semantic knowledge, 30, 35, 231, 232, 235,
long-term memory (LTM), 196, 197 236, 272
style on recall, 194–195 Semantic memory, 2, 32, 236–237
Reinforcement paradigms, 71 and episodic memory, 67, 72
mobile conjugate reinforcement task, 73 Semantic networks, 232, 233, 234
Relational knowledge, 228 Semantic-organizational strategies, 272
Relational memory, 60 Sensitivity category, 256
Relational processing theory, 60 Sensory registers (SR), 1, 21, 131
Remembering complex materials, 210–212 Sentence span, development of, 135f
comprehension monitoring, 210 Sequence learning, 37
informed strategy training approach, 211 Set maintenance, 169
memorability-based strategies, 211–212 Short-term memory (STM), 11, 12, 22, 41,
reciprocal teaching, 211 131
Repetition suppression, 69–70 capacity and age, 133–136
Response latency time (IRT), 138 in infancy, 42
Retrieval cues, 212–214, 271 theoretical approaches, 144–147
Retrieval deficits, 213, 214 Short-term store (STS), 1, 6, 131
Retrieval failure hypothesis, 104 and age, 132
Retrieval monitoring, 283–285 Short-term working memory (STWM), 42, 43,
JOL and CJs comparison, 284 44, 45, 52
Retrieval strategies, 212 Single-item rehearsal, 195, 196
age differences and flexibility, 214–216, Skate questions, 271
215f Skilled memory, 248
external and internal retrieval cues, Social cultural developmental theory, 75. See
212–214 also Sociocultural contexts
Rote memory, 16 Society for Research in Child Development
Routine events memory, 78–82 (SRCD), 3, 21
for single instances of, 82–86 Sociocultural contexts
Rumors, 122 during events conversations, 95–97
390 Index

past events conversations, 91–94 of children, 260–262

socioemotional development, 97–99 development, 319
Software, 21, 131, 309 problem-centered approach, 260
Sort-recall task, 216 self-regulation, 261
performance, 251f Theory-based judgments, 276
Source memory problem, 120–121 Thinking, 1, 3, 5, 20, 99, 253
Source monitoring, 120 -aloud techniques, 267
developmental differences in, 121 area of brain, 149
and prefrontal functioning, 120 efficient thinking, 228
Span of apprehension, 151 generalized, 229
Spatial working memory, 172 good thinking, 242
Specific strategy knowledge, 228 as mental verb, 268
Speech rate wishful thinking, 267, 277, 283
Korean sample, 142, 143f Threshold models, 252
in short-term memory capacity, 144 Time window, 65
U.S. sample, 142, 143f Time-based resource-sharing (TBRS) model,
Speeded responding, 169 153, 157, 159, 160
Speed-related tasks, cultural differences, 141 Toddlers’ long-term memory
Spontaneous recovery, 50 development of, 54–57
Stimulus-action pairings, 146 impact of language acquisition on, 63–64
Stimulus-bound recognition memory, 11 preverbal experiences, 64
Storage space, 148 train task, 56
Strategy use, and age difference, 246 Total processing space, 148, 180
Stress, 322 Trace integrity framework, 66
Stressful events, 115 Trace-accessibility view, 282
Striatum, 36, 41, 67 Trace-based view, 282
Trauma/traumatic, 322
brain injury (TBI), 30
T childhood memories, 111
Task memory, 270 events, 115
Task-related metamemory, 273f
Task-relevant knowledge, 241
Task-switching U
hypothesis, 158 Unconditioned stimulus (UCS), 55
model, 153 Underestimation, 283
Teacher talk, 323 Utilization deficiencies, 185
Temporal understanding, 92
Temporally ordered recall, 61
Tests of recognition, 64 V
Text learning tasks, 276 Variable items, memory for, 83–84
Text processing tasks, 250 Variable strategy use, 216–218, 218t
Text-recall tasks, performance, 250, 251f cumulative-rehearsal task, 219
Theoretical working memory models, empiri- dual-task procedures, 219
cal tests of, 154 educational effects on strategy use,
Baddeley’s model, 160–171 226–228
Case’s resource-sharing model, 157–160 efficiency of processing, 219
Cowan’s working memory model, 176–179 environmental factors on strategy use,
episodic buffer evidence, 175–176 225–226
neuropsychological evidence, 171–174 good information processing, 228
Pascual-Leone’s model, 154–157 good strategy use, 228
Theory of mind, 105, 120, 258–259 imagery instructions, 221, 221f
acquisition, 268 knowledge base effects, 221–224
beliefs, 261 memory capacity demands, 219
Index 391

metamemory on strategy use, 224–225 W

mnemonic orientation, 227 Werner’s theory, 18
self-report questionnaires, 227 Why questions, 242, 243, 244, 245
short-term memory capacity, 220 Williams syndrome, 34
Variables category, 256. See also Declarative Wishful thinking hypothesis, 267, 277, 283
metamemory Word length effect, 162
Verbal and visuospatial working memory, 164 Word span, developmental differences in, 134f
Verbal memory Word-pair method, 12, 13
concept of, 325 Working memory (WM), 22, 39, 42, 132
span, 161 development of, 315–316
Verbal protocol analysis, 266 dorsolateral prefrontal cortex, 133
Verbal recall, age-related changes, 65f, 66 hypothesis, 161
Verbatim traces, 112 in infancy, 42
Violation-of-expectations procedure, 46, 47 model, 149
Visual attention, 168 revised model of, 146
Visual cache, 165 theoretical approaches, 144–147
Visual habituation, 49 World knowledge, 231, 232, 234
Visual memory functions, 11 by children, 239
Visual paired-comparison (VPC) task, 47 Würzburg-Göttingen longitudinal study, 217
Visual recognition memory, 49
Visual stimuli, 49
Visuospatial processing tasks Y
cultural differences, 141 Young children, strategic behavior in, 186
information processing, 42 early encoding strategies, 189–190
working memory task, 172 early retrieval strategies, 186–188
Visuospatial sketch pad (VSSP), 150, 155, encoding versus retrieval strategies,
164, 165, 166, 170, 174, 180 190–193
Visuospatial storage system. See Visuospatial
sketch pad (VSSP)
Voluntary memory, 19