MSM – Multi-Store Memory Model

 Created by Atkinson and Shiffrin 1968

 Linear Model – information thrown through the system in one
direction
 Passive stores – stores hold on to information before being passed
on or lost

Features of MSM Stores

 ENCODING – different information types/formats that the brain uses

to store memory. This is the process of converting information into
memory traces (code) for storage and can be visual (images),
acoustic (sounds), or semantic (meaning).

 CAPACITY – This refers to how much information can be held in the

memory store.

 DURATION – how long information can be held in that store before it

is lost. This is the amount of time the information is held in the
memory store. The maximaum amount of time that a memory trace
can stay in a store.

After hearing a trigram, participants were asked to count

backward in threes or fours from a specified random digit
number until they saw a red light appear (then they
recalled the trigram). This is known as the Brown
Peterson technique, which aimed to prevent rehearsal.

MSM STORES

1. Sensory Register (SR)

Sensory information from the senses is detected and recorded

automatically

All information found in the short term or long-term memory

Coding: store depends on the sense organ it comes from

2. Short-Term memory

Passes information to the long-term memory LTM through rehearsal, either

maintenance rehearsal or elaborative rehearsal

Coding – information in short term memory is stored acoustically, in form

of sound/spoken words
Capacity – Miller suggested this is small, 5-9 items. Can be improved by
chunking (making small groups/sets)

Duration – short, 18-30secs, can be extended through verbal rehearsal

(rehearsal loops)

Information is lost due to displacement or decay

3. Long-Term memory

Information stored may last permanently

Coding: Information stored semantically, this is in the form of “meaning”

Capacity: No limit has been found, information can be lost not because it
is “out of the room” but because it is not accessible

Duration: no limit as people as old as 100 years can still remember

specific details about their childhood

If a person is not able to rehearse information, it will not

transfer to their long-term memory store.
Introduction

Memory is a fundamental cognitive process that enables

individuals to store, retain, and retrieve information. One of the
earliest cognitive models proposed to explain memory is the
Multi-Store Model (MSM), developed by Atkinson and Shiffrin
(1968). The MSM suggests that memory is made up of three
distinct components—sensory memory, short-term memory
(STM), and long-term memory (LTM)—and that information
passes through these stages in a linear sequence. This model
has been influential in shaping our understanding of memory,
though it has also received criticism for its oversimplified
structure. This essay will describe, explain, and evaluate the
MSM, using empirical studies by Peterson & Peterson (1959)
and Glanzer & Cunitz (1966) to support the model’s claims,
while also considering its limitations.

Paragraph 1: The structure and assumptions of the MSM

The Multi-Store Model assumes that memory consists of three

separate stores that differ in capacity, duration, and encoding.
Sensory memory briefly holds information from the
environment for a fraction of a second and encodes it based on
the sensory modality (e.g., iconic for visual, echoic for
auditory). If attention is paid to this information, it moves into
short-term memory, which has a limited capacity (around 7±2
items, according to Miller) and duration of about 15–30
seconds. Information is primarily encoded acoustically in STM.
Through rehearsal, information can be transferred to long-term
memory, which is assumed to have an unlimited capacity and
duration and encodes information semantically. Retrieval
occurs when information is brought back from LTM into STM for
conscious use. This linear process underpins the MSM’s
structure and explains how information is processed in the
mind.
Paragraph 2: Peterson & Peterson (1959) supports the
limited duration of STM

Support for the MSM comes from Peterson and Peterson (1959),
who investigated the duration of STM. In their study,
participants were shown trigrams (three-letter combinations
like “XQZ”) and asked to recall them after various delays (3, 6,
9, 12, 15, or 18 seconds). During the delay, participants had to
count backwards in threes from a random number to prevent
rehearsal. The results showed that as the delay increased,
recall accuracy declined significantly, with only about 10% of
participants accurately recalling the trigrams after 18 seconds.
This finding supports the MSM's claim that STM has a limited
duration and highlights the importance of rehearsal in
maintaining information. Without rehearsal, information quickly
decays from STM, consistent with the assumptions of the
model.

Paragraph 3: Glanzer & Cunitz (1966) supports separate

STM and LTM stores

Further evidence for the MSM is provided by Glanzer and Cunitz

(1966), who demonstrated the serial position effect—a
phenomenon in which people tend to recall the first and last
items on a list better than those in the middle. In their study,
participants heard a list of words and recalled them either
immediately or after a 30-second distractor task. The
immediate recall group exhibited both primacy (better recall of
first items) and recency effects (better recall of last items),
while the delayed recall group showed only the primacy effect.
This suggests that the last items were lost from STM due to the
distractor task, while the first items were already encoded into
LTM. These findings support the MSM's claim that STM and LTM
are distinct and functionally separate memory systems.

Paragraph 4: Evaluation – Strengths and empirical

support
The MSM has several strengths. Firstly, it is supported by
robust experimental research that shows a clear distinction
between STM and LTM in terms of duration, capacity, and
encoding. Studies like those by Peterson & Peterson and
Glanzer & Cunitz provide replicable, controlled evidence for the
model’s structure. Secondly, the MSM was one of the first
cognitive models to offer a clear framework for understanding
memory processes, and it provided a foundation for subsequent
models, such as the Working Memory Model and levels-of-
processing theory. Its simplicity also makes it a useful starting
point for understanding how memory works in everyday life.

Paragraph 5: Evaluation – Limitations and criticisms of

the MSM

Despite its contributions, the MSM has been criticized for being
overly simplistic and linear. It assumes that each memory store
is unitary, but case studies of brain-damaged patients
challenge this assumption. For example, patient KF (Shallice &
Warrington, 1970) had impaired verbal STM but intact visual
memory, suggesting that STM is not a single store. Additionally,
the MSM overemphasizes the role of rehearsal in memory
transfer. In real life, people often remember emotionally
significant or meaningful events without rehearsal. Moreover,
the model does not account for semantic processing, which
is better explained by alternative models such as Craik and
Lockhart’s Levels of Processing model. These criticisms
highlight that human memory is more dynamic and complex
than the MSM suggests.

Conclusion

In conclusion, the Multi-Store Model of memory was a

groundbreaking contribution to cognitive psychology, offering a
systematic framework for understanding how information is
processed, stored, and retrieved. It is well-supported by
empirical research, particularly the studies by Peterson &
Peterson (1959) and Glanzer & Cunitz (1966), which validate
the distinctions between STM and LTM. However, the model’s
simplicity is also its main limitation. It fails to account for the
complexity of different types of memory and the influence of
meaning and emotion on memory retention. Therefore, while
the MSM provides a useful foundational theory, it should be
considered alongside more recent models that incorporate the
multidimensional nature of memory.

Glanzer & Cunitz (1966)

Explanation and Evaluation

1. Aim:

To investigate the serial position effect and examine

whether there is evidence for separate short-term memory
(STM) and long-term memory (LTM) stores, as predicted
by the Multi-Store Model (MSM).

2. Procedure:

 Participants were read a list of 20 words, one at a time.

 They were then asked to recall the words in any
order (free recall).
 Two conditions:
o Immediate recall condition: Participants recalled
words straight after hearing them.
o Delayed recall condition: A 30-second distractor
task (counting backwards) was added before recall
 to prevent rehearsal and remove information from
STM.

3. Results:

 Immediate recall group showed a serial position

curve:
o Primacy effect: Better recall of words at the
beginning of the list.
o Recency effect: Better recall of words at the end of
the list.
 Delayed recall group:
o Recency effect disappeared, but primacy effect
remained.

4. Conclusion:

 The primacy effect reflects LTM encoding — early

words had time to be rehearsed and transferred to long-
term memory.
 The recency effect reflects STM storage — last words
were still in short-term memory unless disrupted.
 These findings support the Multi-Store Model by
showing that STM and LTM are separate systems with
different characteristics.

5. Evaluation:

Strengths:

 Controlled laboratory setting → High internal validity;

cause-effect relationships can be inferred.
 Supports empirical evidence for the Multi-Store Model.
 Clear operationalisation of STM and LTM effects (via
primacy and recency).

Limitations:
  Low ecological validity → Word list recall is artificial and
may not reflect everyday memory processes.
 Assumes rehearsal is the only way to transfer
information to LTM — real-life memory often involves
meaning, emotion, or distinctiveness.
 Sample bias → Typically college students; results may
not generalize to wider populations.

Balanced Insight:

Although the study strongly supports the Multi-Store Model's

idea of separate STM and LTM, it also reveals the
model’s oversimplification. It doesn't explain why we
remember some items better or account for different types
of long-term memory (e.g., semantic vs. procedural).

Summary 2

Glanzer & Cunitz (1966) provides key evidence for the Multi-
Store Model by demonstrating the primacy and recency effects,
which indicate the presence of distinct short- and long-term
memory systems, although its artificial method limits ecological
validity.