This article was downloaded by: [University of Sussex Library]

On: 29 March 2013, At: 14:35

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Research Quarterly for Exercise and Sport

Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/urqe20

Schema Theory (1975): Retrospectives and

Prospectives
a
Karl M. Newell
a
Department of Kinesiology, Pennsylvania State University, USA
Version of record first published: 26 Feb 2013.

To cite this article: Karl M. Newell (2003): Schema Theory (1975): Retrospectives and Prospectives, Research Quarterly for
Exercise and Sport, 74:4, 383-388

To link to this article: http://dx.doi.org/10.1080/02701367.2003.10609108

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to
anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contents
will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses should
be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims,
proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly in
connection with or arising out of the use of this material.
 Special Topics

Research Quarterly for Exercise and Sport

©2003 by the American Alliance for Health,
Physical Education, Recreation and Dance
Vol. 74, No.4, pp. 383-388

Schema Theory (1975): Retrospectives and Prospectives

Karl M. Newell

A briefcommentaryis providedon the theoretical assumptions, scholarly impact and continuing influenceoftheschematheory of
motorlearning (Schmidt, 1975). The traditional contrasts ofschematheory to the coordinative structureor dynamical systems
framework arereemphasized, and limitations ofthevariabilityofpractice experiments noted. A centralproblemfor theories of
Downloaded by [University of Sussex Library] at 14:35 29 March 2013

motorlearning is changeovertime, thebasison which learning is typically defined. Most theories including schemahave,
however, undervalued the importanceofthe time-dependent nature ofchangein deference to the almostexclusivestudy ofthe
amount ofsomeaveraged change in behavioraloutcome. Thepersistent and transitorychangers) in movementand outcome that
areobserved in action arereflections ofmultiple timescales ofchangein a dynamical system.

Key words: coordination, dynamical systems, skill acqui- learning. In light of the 25 years of theoretical and em-
sition pirical activities stimulated by the 1975 paper, I will iter-
ate some of the traditional criticisms of schema theory,

I t is now over 25 years since Richard A. Schmidt (1975)

published the schema theory of motor learning. This
landmark paper has had a huge impact on motor learn-
particularly those that emerged through the theoreti-
cal framework of coordinative structures (Kugler, Kelso,
& Turvey, 1980, 1982). The introduction of the self-or-
ing and control at large, a perspective reflected in the ganization metaphor to motor learning and control pro-
fact that the paper is an Information Sciences Institute vided a paradigm shift challenge to the symbol-based
"citation classic." It is a timely anniversary to reflect on representational account of schema theory. The tradi-
the ideas advanced in the schema paper, its impact to tionallimitations or criticisms of schema theory are all
date, and its potential for continued influence within as relevant now as they were 25 years ago, and they de-
motor learning. Motor learning scholars within kinesi- serve reemphasis here as a vehicle for understanding
ology and exercise and sport science domains have been the basis of the challenges that continue today in advanc-
active participants in creating the impact of this theory ing beyond the tenets of motor learning schema theory.
in that they have contributed a range of experimental In addition, this paper will dwell on one aspect of
tests and elaborations of the hypotheses outlined in the learning that needs to be considered more directly in
original schema paper. future motor learning theories. I am referring to the
The focus of this paper is to provide a brief com- issue of changein movement for action, a phenomenon
mentary on the contributions and limitations ofschema that is by definition central to motor learning and de-
theory and, perhaps more importantly, its continuing velopment but is handled indirectly by schema theory
potential for impact on future developments in motor and its antecedent influences. Throughout the 20th
century, learning has been studied almost exclusively
by a focus on the amount of change in the task outcome
Submitted: May 7, 2002
dependent variable (as witnessed by the almost exclu-
Accepted: March 3, 2003
sive use of analysis ofvariance models), without sufficient
Karl M. Newell is with the Department of Kinesiology at The consideration of the time-dependent (dynamic) prop-
Pennsylvania State University. erties of change in movement and outcome.

ROES: December 2003 383

 Newell

Theoretical Lineage ofSchema Theory ordination modes), being restricted by definition to the
novelty of a given action class. Second, the storage limits
The notion of schema has a long although not of the system are rarely seen as pressing in contempo-
strongly influential history as a construct in psychology, rary accounts of cognition and action, perhaps because
particularly in developing ideas about human learning the need for representational parsimony can be ap-
and memory. Various interpretations of the schema con- proached by dynamics and other theoretical orientations
struct over the years largely failed to gain any foothold to representation.
in the motor learning domain until the Schmidt (1975)
paper (however, see Bartlett, 1932). Schmidt's schema
theory of motor learning was a reflection of and stimu-
lated by the then current theorizing of Pew (1974) about Inadequacies of the Symbol-Based Program
schema and motor performance. Metaphor
The ideas that Schmidt (1975) outlined in the
schema learning theory paper were also developed The introduction ofthe coordinative structure theory
about the same time as those of the motor program ofmotor control and its subsequent elaboration into what
(Keele, 1968; Keele & Summers, 1975). The motor pro- is now more often called a dynamical systems framework
Downloaded by [University of Sussex Library] at 14:35 29 March 2013

gram concept was an application of the computer meta- to motor learning and control (see Kelso, 1995; Kugler &
phor to human performance in general. In a sense, the Turvey, 1987) posed a number of challenges not only to
link of motor programs and schema theory provides one Schmidt's (1975) schema theory but also to representa-
of the more important theoretical instances ofintegrat- tional accounts of motor control and learning in general.
ing motor control and motor learning, respectively. Kugler and colleagues emphasized the problems that
Another important theoretical influence on the motor programs and schema have with a discrete mode of
development ofschema theory was the closed-loop theory representation rather than a dynamic mode. They out-
of Adams (1971). Indeed, Schmidt's schema (1975) pa- lined the difficulties for specifying the representations
per can be interpreted as, in essence, a "schernatized" that are created and how much detail a representation
version of the Adams closed-loop theory but with greater should include. In short, the coordinative structure theory
emphasis on the recall mechanism. The reason for this emphasized the lawsofdynamics rather than the rules of
view is that the memory traces of recall and recognition, prescriptions for movement in action.
as outlined in Adams's theory were retained in schema The original coordinative structure theory papers
theory but given a generalized representational status. and their subsequent developments led to a broad rang-
In essence, schema is a generalized rule for motor out- ing set of papers that, when taken as a collective, cap-
put that, in particular, led to hypotheses about the role ture a theory of motor learning and control, even if it is
of practice variability in schema memory formation. not labeled as such in some unified sense. There are
A significant theoretical claim of the schema theory now numerous contributions about the dynamics of
(1975) paper was that it accommodated two major chal- motor control from informational (e.g., Lee, 1980) and
lenges for motor learning. The first was the novelty prob- movement perspectives (e.g., Beek & Beek, 1988;
lem oflearning, namely how an individual learns a motor Haken, Kelso, & Bunz, 1985), together with the change
task he or she has not performed previously. The sec- of those dynamics in motor learning (e.g., Mitra,
ond was the so-called storage problem for the memory Amazeen & Turvey, 1998; Newell, Mayer-Kress, & Liu,
of movement in action, which held that the memory ca- 2001; Schoner, 1989) and development (e.g., Thelen
pacity of the human central nervous system is insuffi- & Smith, 1994) and, moreover, the role of instructional
cient to capture the large number of movement strategies in motor learning (Newell, 1996; Newell &
representational details necessary to cover the activity Valvano, 1998). This theoretical framework is not an all-
range in which an individual engages over a lifetime. encompassing view of the motor learning phenomena,
Thus, learning in the schema theory is predicated on but by any consideration it provides as much if not more
the development of "more," and "better" motor pro- theoretical coherence and experimental prediction as
grams in that with learning an individual develops a any traditional theories oflearning the embraced by the
wider repertoire of task-relevant motor skills to handle movement domain in the last century.
the demands of everyday activities. Given these developments, it is not surprising that
The novelty and storage issues are still central prob- motor learning and control is often said to have two
lems for theories of motor learning and control and are camps or schools of thought. One is still a cognitive "sym-
likely to remain so. In retrospect, though, schema theory bol based camp" generally related to the traditions of
does not seem to be either essential or sufficient to schema theory and motor programs, and a second is a
handle the issues at stake. First, schema did and does "dynamical camp" that has arisen as a competing, theo-
not accommodate the novelty of new action classes (co- retical and empirical enterprise. Twenty years ago, when

384 ROES: December 2003

 Newell

the ideas of the coordinative structure theory were in- not, however, filtered through to influence the motor
troduced, there was considerable discussion and argu- learning and control domain, even with those who seek
ment about the merits of the competing theoretical more common ground between cognitive accounts, such
propositions (e.g., Kelso, 1981; Kugler et aI., 1980). as schema, and the dynamical systems framework.
These arguments and debates have waned over the years
to the point that now researchers either follow or work
within one particular point ofviewor the other, and theo-
retical debate on the issues (that still prevail) is con- Schema Theory and the Variability ofPractice
spicuous by its absence. It appears that scholars simply
see the constructs as reflections of different paradigms The major experimental work that has been con-
and are pursuing empirical work in a "normal science" ducted as a test of schema theory is in regard to its pre-
mode, consistent with whichever framework they are dictions about the variability of practice. Schema theory
following implicitly or explicitly. In Kuhnian terms, holds that a generic memory rule is more appropriately
there was a period of "extraordinary science" about 20 developed if there is a broad range of instances of the
years ago, and now the two paradigms have agreed to action (initial conditions, response specifications, past
disagree and entered back into normal science. sensory consequences, desired outcome) the learner
Downloaded by [University of Sussex Library] at 14:35 29 March 2013

The coordinative structure perspective outlined a experiences and can associate. This proposal leads to
number of challenges to schema theory (Kugler et aI., the idea that variability of the practice conditions will
1980, 1982), and I will list a few here. First, it raised the formulate a more task relevant and persistent general
logical point that order and structure in movement ob- rule for learning and performance over a given class of
served in motor skills does not necessitate the need for activity. Schema theory did not, however, provide spe-
a motor program to prescribe that order in all levels of cific constructs and hypotheses about the structure of
detail. To put it another way in a phrase used many times, the practice variability.
order and timing in rhythmic movement organization Understanding the role of practice is a fundamen-
does not require that a program or "central timer" ex- tal problem for motor learning. It is a concept, never-
ists, Second, the argument against the symbol-based theless, that has been studied to date with less rigor and
computer metaphor does not necessarily imply no emphasis than one might imagine given its centrality to
memory or representations for action, rather it chal- learning. Nevertheless, Schmidt iterated in the target
lenges (at least) a particular form of representation- paper (1975; see also Schmidt & Shapiro, 1982) that
the symbol representation for movement. A final point there is strong experimental support for the schema
arising from the work of Kugler and Turvey (1987) is proposition that variable practice is more effective than
that the motor program metaphor fails to exploit the nonvariable, single task, or single parameter practice in
intrinsic dynamics of structural and functional con- developing the schema rule. My reading of the relevant
straints that channel dynamical organization to move- studies is, however, that the experimental support for
ment output. This omission relates, among other issues, the variable practice hypothesis is not as strong as gen-
to the logical experimental difficulty schema theory has erally claimed (see also van Rossum, 1990). A central
in handling how the first movement pattern within a problem is that the poor experimental design of most
schema movement class is produced. variability of practice studies leaves them open to a tra-
The coordination structure theory and its develop- ditional "similarity" interpretation for transfer (Hold-
ments through dynamical systems (Kelso, 1995; Thelen ing, 1976; Osgood, 1949).
& Smith, 1994) raised directly the contrasting roles and Another issue that has made the tests of schema
theoretical challenges of symbols and dynamics, to- predictions less than satisfactory and, indeed, has un-
gether with their place in a general theory of movement dermined the examination of learning theories for a
and action. These are deep philosophical and theoreti- century, is that the study of practice effects has been
cal issues that have remained largely outside the day-to- trivialized in most cases to 100 trials or so in a single
day experimentation of motor learning and control, no session (Newell, 1985). There are, of course, exceptions
matter the theoretical perspective. Thus, for example, to this situation, and more studies are examining per-
variability of practice experiments can still be run as an formance beyond a single practice session; but much of
empirical question without reference to this important the study of motor learning is confined to limited prac-
theoretical issue. There have been a number of new ap- tice durations (see Magill, 1985; Schmidt, 1982). This
proaches to representation and action in the past 20 is important to note, because learning is fundamentally
years, including connectionism and neural networks, about the relatively persistent changes of behavior rather
and more recently attempts to provide the basis of a than the transitory ones always present in repetitive at-
unified cognitive neuroscience framework for mind, tempts at particular activities (e.g., warm-up decrement,
body, and action (Clark, 1997,2(01). These ideas have noise-like fluctuations in performance, etc). A chal-

ROES: December 2003 385

 Newell

lenge in motor learning is to find ways to understand One example of this is the study of new coordination
the relatively persistent from the relatively transient patterns, a key movement phenomenon that was also
changes and link those relative time scales of change to outside the realm of schema theory.
mechanisms or processes assumed to reflect the coor- The schema theory of motor learning emphasized
dination solution. the interesting question of the role of practice variabil-
Many of the experimental tests of schema theory, ity, and, as implied earlier, the experimental or practi-
including the more recent studies of knowledge of re- cal aspects of this are still not well understood. But even
sults and other practice variables, are dominated by rela- within the concept of practice, many other important
tively transient or temporary effects that do not learning phenomena need to be investigated. One, for
necessarily reflect the persistent kinds of change nec- example, is the nature of the variability for task-appro-
essary to infer learning. It is not that transitory change is priate practice, an issue being addressed under the
unimportant (it can hold positive features for learning), construct of contextual interference. Indeed, all the
but rather it needs to be distinguished in a theoretical traditional strategic aspects of practice and learning are
and practical sense from the persistent changes charac- open to further consideration, as many have not been
terizing motor learning. Thus, a challenge is to conduct investigated systematically for some time or have been
an agenda of experimentation and theoretical develop- limited to particular classes of tasks.
Downloaded by [University of Sussex Library] at 14:35 29 March 2013

ment consistent with the fundamental phenomena of In the debates about the relative merits of schema
learning skills in daily activities. The traditions of experi- theory and dynamical systems theory, it is often claimed
mental psychology, of which schema theory is in part a that one theory or the other cannot handle "this" find-
product, have contributed many principles and insights ing or "that" situation. Theories by design inherently
to motor skill learning, but they have also created "red narrow the domain that they are considering to avoid
herrings" to discovery of the principles of learning and the challenge of being "all things to all people, and
the coordination and control of movement in practice. hence nothing at all." At this stage of science in motor
learning and control, it is worth remembering that theo-
ries are primarily heuristic devices tending to empha-
size experimentally the strengths (that is, domain) of
Learning Phenomena and Schema Theory their particular tenets but ignore the weaknesses or ar-
eas eliminated by design. Thus, a consequence is that
Schmidt (1975) emphasized the view that only a some contrasts that have been made between schema
schema account with representational properties could theory and dynamical approaches to motor learning have
handle particular kinds of phenomena in motor learn- missed the central challenges oflearning and have not
ing and control. Obviously, there is a wide range of basic led to advancing the problems at hand.
phenomena of change that any motor learning theory
needs to accommodate. And there is no question the
examples raised by Schmidt are real phenomena of
motor control that are interesting in their own right. Practice and the Time Scales of Learning
Whether or not these phenomena provide the basis for
challenges to a dynamical account of motor learning, as A central phenomenon that needs to be understood
Schmidt outlined, is another question. The interpreta- in motor learning is the question of change and the role
tion seems to fall into the trap noted by Kugler and col- of practice in changing behavior. Behavior change is
leagues 20 years ago (1980; and iterated above) that relative in the sense of its degree of persistence or its
determining order in movement organization does not relatively temporary nature. This can be considered in
necessitate a symbolic program to prescribe that order. a cohesive fashion by thinking about the multiple time
A more central challenge for a new motor learning scales of change in motor learning and development
theory is to identify and accommodate the key phenom- (Newell et al., 2001), as a vehicle to describe and un-
ena of motor learning. Indeed, it is useful to raise the derstand the time-dependent nature of change and,
question: what are the fundamental phenomena that hence, learning.
reflect persistent and transitory change and need to be There is a long history of studying learning curves
handled in a motor learning theory. Much of what has (or performance curves as some still prefer to call them),
transpired in the motor learning domain has followed but their importance dropped off in discussions of
the paradigms of experimental psychology that over the motor learning after A Newell and Rosenblum (1981)
years largely narrowed the range of tasks examined. As championed the power law function for both cognitive
a consequence, the breadth oflearning phenomena that and motor learning. Indeed, the issue oflearning curves
need to be accommodated in a motor learning theory has largely been ignored for the past 20 years in spite of
has been underestimated by default (Newell, 1985). its central importance to the field. Recently, however,

386 ROES: December 2003

 Newell

there have been challenges to the robustness of a power potentially have direct relevance to issues in the theory
law interpretation of learning, both in cognitive of representation and action (d. Clark, 1997,2001), have
(Heathcote, Brown, & Mewhort, 2000) and motor not significantly impacted the experimental study ofhu-
(Newell et al., 2001) tasks. The question ofwhat changes man motor learning. As the adage goes, it takes the in-
in the outcome and movement dynamics with practice troduction ofa new theory to replace an extant theory as
and learning is still a relevant issue for study. the guiding framework for experimental science. Dem-
An important theoretical claim from our recent onstrating experimental limitations or principled prob-
paper on time scales in motor learning is that one can lems to a theory is often not sufficient for its demise.
accommodate the qualitative types of change (functions The dynamical systems approach to motor learning,
of learning) in movement and action by a small group as the label implies, directly relates to the problem of
of principles from dynamical systems theory (K. Newell change in behavior over time. Schema theory, like other
et al., 2001). In other words, we do not need a different theories oflearning before it, may appearto provide the
theory to handle the different kinds oflearning curves same level of directness to the change issue because of
or patterns of change that arise from learning and adap- the general claim to consider learning, but schema was
tive behavior over time. Rather, a range of time scales of and is both metaphorically and in reality more of a static
change reflect a small set of dynamical principles that than dynamic theory oflearning. A consideration of the
Downloaded by [University of Sussex Library] at 14:35 29 March 2013

can, in effect, form the basis for a unified dynamical ac- time scales of change may provide a formal basis for
count to the change in behavior. In this dynamical ac- developing a theory of practice in motor learning.
count of motor learning, the issue of time scales of New theories of motor learning, though, will most
change is a central construct for considering the time- usefully be predicated on an interdisciplinary focus to
dependent nature of practice and learning, although movement in action rather than an isolated disciplinary
more cognitive-level accounts are not precluded by this view,such as represented by schema theory through the
approach. traditions of psychology. In the past 25 years, there has
Instructional strategies for motor learning can build been an explosion of an interdisciplinary systems ap-
on these concepts to understand the role interventions proach to movement science manifest today in so many
by teachers and coaches have in facilitating learning. exciting ways at many levels of analysis. These changing
This is a traditional province of cognitive approaches to times lead naturally to the suggestion that Schmidt's
skill learning, and schema theory still provides an intui- (1975) schema theory will have the additional distinc-
tive and general backdrop to current considerations of tion of representing the last occasion in which an ex-
instructional strategies and practice. Nevertheless, there clusively behavioral psychological theory for studying
are principled ways in which one could build a dynami- motor learning and control will receive as much schol-
cal account of interventions in motor learning, and work arly recognition.
in this regard has begun in recent years (Newell, 1996;
Newell & Valvano, 1998).

References
Adarns.]. A (1971). A closed-loop theory of motor learning.
Summary Comments Journal ofMotor Behavior, 3, Ill-ISO.
Bartlett, F. C. (1932). Remembering. Cambridge, UK: Cam-
Schmidt's (1975) schema theory in motor learning bridge University Press.
and control is still with us today, but it is more a back- Beek, P.]., & Beek, W. (1988). Tools for constructing dynami-
ground general influence to ongoing experimental cal models of rhythmic movement. Human Movement
work than a foreground set of theoretical propositions Science, 7,301-342.
driving an active empirical agenda. The background Clark, A (1997). Being there: Putting brain, body, and world to-
influence has been retained, because schema theory is gether again. Cambridge, MA: MIT Press.
still viewed as the predominant "cognitive" account of Clark, A. (2001). Mindware. New York: Oxford University Press.
human experimental motor learning, and, moreover, it Haken, H., Kelso,]. A. S., & Bunz, H. (1985). A theoretical
model of phase transitions in human hand movements.
provides hypotheses about the role of practice in motor
Biological Cybernetics, 51, 347-356.
learning. This general theoretical influence of schema
Heathcote, A, Brown, S., & Mewhort, D.]. K. (2000). The
theory has prevailed, I suspect, because it has not been power law repealed: The case for an exponential law of
supplanted by other cognitive theories that have rel- practice. P~ychonomic Bulletin & Reoieui, 7, 185-207.
evance for the experimental study of motor learning. Holding, D. H. (1976). An approximate transfer surface.Jour-
The theoretical developments in connectionism, artifi- nal ofMotor Behavior, 8, 1-9.
cial intelligence, neural networks, and general compu- Keele, S. W. (1968). Movement control in skilled motor per-
tational approaches to brain and behavior, that formance, Psychological Bulletin, 70,387-403.

ROES: December 2003 387

 Newell

Keele, S. W., & Summers,j.J. (1975). The structure of motor Newell, K. M., Mayer-Kress, G., & Liu, V-To (200 I). Time scales
programs. In G. E. Stelmach (Ed.), Motor control: Issues in motor learning and development. P5ychologiral Reoieui;
and trends (pp. 109-142). New York: Academic Press. 108,57-82.
Kelso.]. A. S. (1981). Contrasting perspectives on order and Newell, K. M., & Valvano,J. (1998). Therapeutic intervention
regulation in movement.lnj. Long&A. Baddley (Eds.), as a constraint in learning and relearning movement skills.
Attention and performance IX (pp. 437-457), Hillsdale, NJ: SrandinavianJoumal oJOrrupational Therapy, 5,51-57.
Lawrence Erlbaum and Associates. Osgood, C. E. (1949). The similarity paradox in human learn-
Kelso, J. A. S. (1995). Dynamic patterns: The self-organiuuion oj ing: A resolution. Psychological Review, 56, 132-143.
brain and behavior. Cambridge, MA: MIT Press. Pew, R. W. (1974). Human perceptual motor performance.
Kugler, P.N., Kelso,j. A. S., & Turvey, M. T. (1980). On the In B. H. Kantowitz (Ed.), Human information processing:
concept of coordinative structures as dissipative struc- Tutorials in performance and cognition(pp. 1-39). New York:
tures: 1. Theoretical lines of convergence. In G. E. Lawrence Erlbaum and Associates.
Stelmach &j. Reqiun (Eds.), Tutorials in motor behavior Schmidt, R. A. (1975). A schema theory of discrete motor skill
(pp. 1-49). New York: North-Holland. learning. Psychological Review, 82, 225-260.
Kugler, P. N., Kelso,j. A. S., & Turvey, M. T. (1982). On the Schmidt, R. A. (1982). Motor control and learning. Champaign,
control and coordination of naturally developing systems. IL: Human Kinetics.
In J. A. S. Kelso & J. E. Clark (Eds.), The deoelopment oj Schmidt, R. A., & Shapiro, D. C. (1982). The schema theory:
Downloaded by [University of Sussex Library] at 14:35 29 March 2013

movement rontrol and co-ordination (pp. 5-78). New York: Recent evidence and developmental implications. InJ.
Wiley. A. S. Kelso &j. E. Clark (Eds.), The deoelopment oJ move-
Kugler, P. N., & Turvey, M. T. (1987). Information, natural law, ment control and co-ordination (pp. 113-150). New York:
and the self-assembly 0/ rhythmic movement: Theoretical and Wiley.
experimental investigations. Hillsdale, NJ: Lawrence Schoner, G. (1989). Learning and recall in a dynamic theory
Erlbaum and Associates. of coordination patterns. Biological Cybernetio, 62,39-54.
Lee, D. N. (1980). Visuo-rnotor coordination in space-time. Thelen, E. & Smith, L. B. (1994). A dsnamic systemwptIToach to
In G. E. Stelmach &J. Requin (Eds.), Tutorials in motor the deoelopment oj cognition and action. Cambridge, Mass:
behavior (pp. 281-296). Amsterdam: North-Holland. MIT Press.
Magill, R. A. (1985). Motor learning: Concepts and applications. Van Rossum, J. H. A. (1990). Schmidt's schema theory: The
Dubuque, IA: Brown. empirical base of the variability of practice hypothesis:
Mitra, A., Arnazeen, P. G., & Turvey, M. T. (1998). Intermedi- A critical analysis. Human Movement Science, 9, 387-436.
ate motor learning as decreasing active (dynamical)
degrees offreedom. Human Movement Science, l Z, 17-66.
Newell, A., & Rosenbloom, P. S. (1981). Mechanisms of skill
acquisition and the law of practice. In j. R. Anderson Author's Notes
(Ed), Cognitive skills and their acquisition (pp. I-55).
Hillsdale: NJ: Lawrence Erlbaum and Associates. I would like thank Dagmar Sternad and Danny Russell
Newell, K. M. (1985). Coordination, control and skill. In D.
for helpful comments on an earlier version of the pa-
Goodman, 1. Franks, & R. B. Wilberg (Eds.), Differingper-
per. Please address all correspondence concerning this
spertiues in motor learning, memory, and control (pp. 295-
318). Amsterdam: North-Holland.
article to Karl M. Newell, College of Health and Human
Newell, K. M. (1996). Change in movement and skill: Learn- Development, The Pennsylvania State University, 201
ing, retention, and transfer. In M. Latash & M. Turvey Henderson Building, University Park, PA 16802.
(Eds.), Dexterity and its development (pp. 393-429).
Hillsdale, NJ: Lawrence Erlbaum and Associates. E-mail: KMN I@psu.edu

388 RDES: December 2003