BJMB
Brazilian Journal of Motor Behavior
Special issue:
The role of practice in motor learning
Profeta,
Ugrinowitsch
2022
VOL.16
N.2
134 of 142
Extensive practice in motor learning: an overview and future directions
VITOR L. S. PROFETA
1
| HERBERT UGRINOWITSCH
1
1
Universidade Federal de Minas Gerais, Escola de Educação Física, Fisioterapia e Terapia Ocupacional, Grupo de Estudos em Desenvolvimento e Aprendizagem
Motora (GEDAM).
Correspondence to: Vitor Leandro da Silva Profeta, Ph.D., Universidade Federal de Minas Gerais, Av. Pres. AntĂ´nio Carlos, 6627 - Pampulha, Belo Horizonte - MG,
31270901.
Telefone: +55 31 3409-2394
email: vitorprofeta@hotmail.com
https://doi.org/10.20338/bjmb.v16i2.279
ABBREVIATIONS
LT Limb and target location
T Target location only
LT-200 Limb and target location -
Moderate practice (200 trials)
LT-2000 Limb and target location -
Extensive practice (2000 trials)
T-200 Target location only - Moderate
practice (200 trials)
LT-2000 Target location only - Extensive
practice (2000 trials)
PUBLICATION DATA
Received 15 12 2021
Accepted 05 03 2021
Published 01 06 2022
ABSTRACT
Extensive practice is observed when performance reaches a plateau despite practice continuation. Although
extensive practice promotes better results in retention tests, its status regarding transfer tests is not clear. The
Adaptive Process Approach states that transfer will benefit from extensive practice. In turn, the Specificity of
Practice Hypothesis states that transfer may be harmed after extensive practice. Each perspective has been
around for 30 years, accumulating enough information to support its claims. However, they have not been
directly contrasted, and each perspective has independently developed its methodological strategy to
understand the role of extensive practice in motor learning. Therefore, it is possible that results supporting the
perspectives are by-products of their methodological strategies, providing narrow validation limits to both
perspectives. In order to better understand the effects of extensive practice on transfer, it is necessary to
contrast the Adaptive Process Approach with the Specificity of Practice Hypothesis in a single and unifying
methodological framework.
KEYWORDS: Extensive practice | Adaptive Process approach | Specificity of Practice | Hypothesis | Transfer
INTRODUCTION
Practice is essential for motor learning. In the beginning, both performance error
and variability are high. After a period of practice, errors decrease to levels that allow one
to consistently achieve the task goal, leading to performance stabilization. Performance
stabilization is characterized by either a small variability in performance or an asymptote
curve indicating that a plateau has been reached. Extensive practice refers to the
continuation of practice after performance stabilization, and it has been shown that
extensive practice affects how one responds to new task demands
1
. On the one hand,
studies conducted in the scope of the Adaptive Process Approach have found that
extensive practice may facilitate transfer. The Adaptive Process Approach predicts that
changes in the organization of a skill promoted by extensive practice yield flexibility to
handle new situations
2,3
. On the other hand, studies conducted in the scope of the
Specificity of Practice Hypothesis have found that extensive practice impairs transfer
4,5
.
We believe that the discrepancy in the results between the Adaptive Process
Approach and the Specificity of Practice Hypothesis offers an opportunity for theoretical
advances. Platt
6
proposed the method of strong inference, which emphasizes the need of
focusing on the exclusion of a hypothesis as an important step to foster scientific growth.
More specifically, contrasting predictions in a way that at least one of them may be
excluded as a reasonable explanation of a phenomenon of interest is an important step in
BJMB
Brazilian Journal of Motor Behavior
Profeta,
Ugrinowitsch
2022
VOL.16
N.2
Special issue:
The role of practice in motor learning
135 of 142
any scientific inquiry
6
. The lack of methods for the exclusion of hypotheses in studies of
motor learning has been highlighted elsewhere
7
. The present article aims to offer an
argument about how adopting a strong inference stance may advance our understanding
of the effects of extensive practice on motor learning.
This paper is organized as follows: we first review studies on extensive practice in
the scope of the Adaptive Process Approach. Then, we review studies on extensive
practice in the scope of the Specificity of Practice Hypothesis. Finally, in the last section,
we suggest a methodological way of contrasting the two perspectives in a single
experiment to gain more insights into the role of extensive practice in motor learning.
THE ADAPTIVE PROCESS APPROACH
The importance of the organization of structures of control for learning has been
highlighted by the proponents of the Adaptive Process Approach
2,3
. It is suggested that an
action program controls a motor skill. An important assumption in the Adaptive Process
Approach about how an action program controls a motor skill is that human beings are
open systems hierarchically organized
8
; consequently, an action program should also be
hierarchically organized in macro- and microstructures
9
.
The macrostructure is responsible for the components’ sequence and
organization, such as relative timing and relative force
10
. In this sense, the macrostructure
controls the important aspects for movement patterns or consistency in motor behavior,
which is a consequence of components’ organization based on the individual’s intention
9,10
. It is important because behavior consistency allows for reaching the goals with
reliability.
The microstructure is responsible for the movement variability, such as total time
and total force. Consequently, the microstructure controls the important aspects for
adaptability in motor behavior. Behavior adaptability is necessary since the environment
constantly changes, making it impossible to plan all the details in advance. Thus, the
microstructure is generated in every trial based on the constraints imposed by the
macrostructure
9,10
. Combining a selected macrostructure and a generated microstructure
in a single two-level control structure diminishes the central demand for controlling all the
motor skills aspects, addressing two of the main motor behavior features, i.e., consistency
and adaptability.
The agenda of the Adaptive Process Approach involves the investigation of how
the macro- and microstructures of action programs modify when facing different types of
perturbations. The general strategy of investigation involves assigning individuals to
different groups defined by their amounts of practice or levels of performance stabilization
of a given task. Stabilization refers to a state in which an individual has enough practice to
consistently reach the goal of a task, named as the performance criterion of the task
11,12
.
Of importance, although levels of stabilization are defined concerning results of movement
execution, levels of performance stabilization are suggested to reflect levels of
organization of systems’ internal constraints
12
.
Often, investigators are interested in comparing two levels of stabilization: practice
until the stabilization of performance (stabilization group) and practice beyond the
BJMB
Brazilian Journal of Motor Behavior
Profeta,
Ugrinowitsch
2022
VOL.16
N.2
Special issue:
The role of practice in motor learning
136 of 142
stabilization of performance (extensive practice or superstabilization
a
group). In this
context, extensive practice refers to the ability to reach the performance criterion
repeatedly. Figure 1 simulates the operationalization of the concept of stabilization.
Although extensive practice does not promote significant gains in performance, it is
expected that extensive practice promotes gains in the capacity to adapt and transfer.
Figure 1. Operationalization of both stabilization and extensive practice in the context of the Adaptive
Process Approach. In a hypothetical task, the goal is achieved when the error is smaller or equal to 2.0 cm
(dashed line). Stabilization is assumed when the performance is achieved repeatedly under a small
bandwidth (e.g., three times in a row, reached in trial number 115), which is named the performance criterion.
Extensive practice ends after repeatedly achieving the performance criterion (e.g., six blocks of three trials in
a row, reached in trial number 166). The performance criterion is characteristic of each motor task.
There is evidence that extensive practice plays an important role in adapting to
both predictable
12
and unpredictable
13
perturbations. For instance, Fonseca et al.
13
compared the adaptation of participants of stabilization and extensive practice groups to
unpredictable changes (i.e., perturbations) in a visual stimulus during a timing coincident
sequential motor task. Participants of each group were asked to perform a sequential
movement pattern touching five components, in which the last component of the sequence
was timed to coincide with a moving visual stimulus. In the first phase of the study, the
speed of the visual stimulus was predictable, and groups differed regarding their amount of
practice. The amount of variability in movement execution and the average result of
movement sequences (i.e., error) were similar between groups. In the second phase, the
sequence of movement was kept constant, and unpredictable variations in the speed of
the visual stimulus were introduced. Results indicated that the extensive practice group
adapted better to the perturbation than the stabilization group.
a
In the Adaptive Process Approach, extensive practice is often referred to as superstabilization. The term
superstabilization also encompasses the movement control structure level of organization after an amount of practice
that goes beyond what is necessary for an individual to stabilize her/his performance. Since our focus is on the amount
of practice, we will stick with the term extensive practice.
BJMB
Brazilian Journal of Motor Behavior
Profeta,
Ugrinowitsch
2022
VOL.16
N.2
Special issue:
The role of practice in motor learning
137 of 142
In general, the extensive practice has led to higher performance variability during
practice but better adaptation in tasks with different demands such as timing coincident
12,13
, isometric force control
15
, and interception of a virtual moving target
11
. Therefore, the
effects of the extensive practice should not be ascribed to particularities of task
constraints. A possible explanation for the results favoring the extensive practice is that an
increase in performance variability after stabilization would represent further exploitation of
task solutions, providing the action program with more resources to deal with new
situations. Support for that hypothesis more directly was conducted by Ugrinowitsch et al.
(2014)
3
.
The study design of Ugrinowitsch et al. (2014)
3
was similar to others in the
Adaptive Process Approach. In the first phase of the study, three groups were defined
based on their stabilization level: pre-stabilization, stabilization, and superstabilization (i.e.,
extensive practice). As expected, performance variability was higher in the extensive
practice group than in the stabilization group. In turn, there was no difference in
performance variability between the extensive practice group and the pre-stabilization
group. When facing a perturbation in the second phase, the pre-stabilization group
maintained low-performance accuracy and high variability. However, the extensive practice
group maintained its performance accuracy but diminished variability. The latter was
related to modifications in the macrostructure, indicating that variability after performance
stabilization was functional, promoting a higher capacity for adaptation.
Overall, the findings of the extensive practice in the Adaptive Process Approach
have been interpreted as evidence of the continuity of the motor learning process beyond
performance stabilization. It seems that extending practice beyond performance
stabilization compared to practice until performance stabilization allows for getting more
information related to the task and the environment. However, the role of perceptual
information in the control and adaptation of motor skills has not been addressed in the
Adaptive Process Approach. This is an issue that has been discussed in the scope of the
Specificity of Practice Hypothesis.
THE SPECIFICITY OF PRACTICE HYPOTHESIS
The Specificity of Practice Hypothesis arose to challenge the common statement
in the literature on motor learning that individuals become less dependent on
environmental information in more advanced stages of learning.16 Notice that similar to
the expected effect of the extensive practice predicted in the scope of the Adaptive
Process Approach, this idea assumes that learning advances towards the elaboration of
an internal representation such as an action program or an internal model, which becomes
more sophisticated with practice. Therefore, a better internal representation is seen as
relatively protected against environmental influences, and movement would need fewer
corrections. The Specificity of Practice Hypothesis states that individuals’ dependence on
environmental, situational information is higher in the extensive practice.
For instance, Proteau et al.