First, some disclaimers:
My intention is to provide this to all new BDG hires, as part of their onboarding. Understanding motor learning/ control, and skill acquisition is a must for the modern coach. I want them to be able to ask questions.
I did not write the majority of this article. Credit instead goes to Muratori et al., (2014) for their aptly named manuscript: “Applying Principles of Motor Learning and Control to Upper Extremity Rehabilitation”. If you want to read the original article, please do so here. It’s a great read for all interested in skill acquisition during the rehabilitative process.
After reaching the end of their paper, I realized there may be some value in rewriting it from a motor learning and control in baseball perspective. So, I’ve done just that. Again, the majority of this blog post is a copy of the 2014 paper - and anything in black ink has been directly transcribed from the original version. My small contribution has been to replace all examples of upper limb rehabilitation from the original paper with representative baseball pitching examples - found in red ink.
This article was published in 2014. Some of the claims and statements made are slightly outdated and more vague than they necessarily are as of 2022.
Outline (click section to skip to it)
Without further adieu, let’s dive in!
~
The challenge of creating meaningful change in baseball pitching mechanics and/or performance may be met by weaving key concepts of motor learning and control into coaching protocols.
However, in order to effectively integrate these concepts into baseball pitching programs, motor learning and motor control strategies need to be better understood. The purpose of this review is to outline key principles of motor learning and motor control that can be used to foster skill acquisition in baseball pitchers. To illustrate the application of these principles for baseball pitchers, we will consider the case of “Jack”, a collegiate male baseball pitcher.
1.0 MOTOR LEARNING AND CONTROL
The attainment of motor skills involves a process of motor learning whose principles
integrate information from psychology, neurology, physical education, and rehabilitation
research. Together these disciplines shape our understanding of how individuals progress
from novice to skilled motor performance. Infants learning to reach and grasp use the perceptions they have of their own body and abilities to secure objects of various shapes and sizes. Adults must often accommodate to the strength and sensory changes that occur with aging/training/injuries/etc., to modify how they perform tasks. In the same sense, pitchers may need to un/relearn previously acquired motor skills to help develop “better technique” and improve outcomes.
1.1 Systems Model
Motor control theories provide a framework to guide the interpretation of how learning or
re-learning movement occurs. Perspectives in motor control are based on evolving models of the nervous system and represent the paradigm shifts that have taken place throughout
history. Historically, when the concepts of an existing paradigm begin to limit the way
movement and behavior are interpreted, new paradigms are developed. For example, in the
early 1900's voluntary movement was thought to occur through reflex linkages. This
paradigm led to numerous theories of motor control that have been replaced as knowledge of the nervous system expands. Although the assumptions associated with varied motor control theories differ, most current theories have incorporated a Systems view of distributed control of the nervous system. A Systems model suggests that movement results from the
interaction of multiple systems working in synchrony to solve a motor problem. The
advantage of the Systems model is that it can account for the flexibility and adaptability of
motor behavior in a variety of environmental conditions. Functional goals as well as
environmental and task constraints play a major role in determining movement. This frame
of reference provides a foundation for developing intervention strategies based on task goals that are aimed at improving motor skills.
To exemplify how movement problems are solved, consider our case, Jack, as he attempts to learn how to throw a baseball harder or more efficiently than he did previously. To be successful, he must learn how to solve this motor problem with the constraints imposed by the coaching staff (in this case, a drop-step constraint drill). A Systems model of control suggests multiple factors, both internal and external to Jack, need consideration when he performs this functional movement. Internal factors may include strength, mobility, coordination, pain level, motivation, cognition, autonomic function, and readiness. External factors may include the drill selected, slope of the mound, material of the mound/ shoes/ plyo balls, and outside distractions. In order to complete the task of throwing harder, all available systems must work together to produce a successful strategy.
1.2 Degrees of Freedom Problem
Producing a single optimal strategy for movement presents a significant problem to the
nervous system. Nikolai Bernstein, a nineteenth century Russian neurophysiologist who
challenged the contemporary reflex theories of movement that dominated his field,
pioneered the concept of multiple systems working together to create movement. He argued that to perform smooth and efficient voluntary movement one must overcome the degrees of freedom problem. Bernstein recognized that when multiple systems interact, there are vast movement options (degrees of freedom) available to perform the same action. For example, Jack could achieve pelvis rotation by flexing and internally rotating and extending his hip, or he could achieve it by abducting and then extending his hip to drive pelvic rotation. This redundancy takes place at multiple levels within the CNS. For example, muscles can fire in different ways to control particular movement patterns or joint motions.
In addition, many different kinematic/movement patterns can be executed to accomplish one specific outcome or action. A functional pitcher would be able to employ multiple strategies (throwing from a slightly different release height, throwing a different pitch, effectively moving through a different constraint drill, etc.) to achieve the same goal - efficiently throwing a hard strike to the exact same location. Bernstein has suggested that a key function of the CNS is to control redundancy by minimizing the degrees of freedom or the number of independent movement elements employed. The resolution to the degrees of freedom problem will vary depending on the characteristics of the learner as well as the components of the task and environment. For example, oblique pain may increase Jack’s likelihood of co-contraction to stabilize his body against undesired movement as he attempts to throw the ball. Similarly, during the early stages of learning, Jack may produce very simple movements and limit the amount of joint motion and speed of the drill by increasing joint co-contraction. As the task is learned, Jack’s co-contraction is likely to decrease, the drop-step and throwing motion will appear more fluid, and the CNS will demonstrate multiple motor resources to achieve the tasks.
1.3 Dominant Theories
The question of how specific movement patterns are selected out of the vast number of
options available has a major influence on how coaches intervene. Many theories have
developed describing how multiple systems might come together to produce a functional
movement. However, two distinct classes of theory have dominated the discussion for more
than forty years. The first focuses on central control of movement instructions (e.g., Motor
Program Theory [MPT]) and the second on dynamic self-organization of multiple sub-
systems around a meaningful goal (e.g., Dynamic Pattern or Dynamical Systems Theory
[DST]).
Motor Program Theory initially suggested that some form of neural storage of motor plans
took place and that these motor plans were retrieved as needed to achieve motor goals.
Three major issues arose around the ability of MPT to adequately explain voluntary
movement; a storage problem, a novelty problem, and the problem of motor equivalence.
The storage problem is the result of the huge repertoire of human movements. Where are the motor plans for the movements stored? It would seem there would need to be an infinite
storage capacity in the nervous system to contain all the plans necessary for the variety of
movement available. The second issue, the novelty problem, addresses the ability to plan
new actions. How is there a program for a movement that has never been performed before?
Finally, there is the issue of motor equivalence --the same action can be accomplished using
different patterns of coordination. How is this possible if the action is the result of a program?
Some of the issues outlined above in MPT have been addressed with the generalized motor
program (GMP) theory proposed by Schmidt. In his work, Schmidt argues that motor
programs do not have to be specified for every action. Rather, there are generalized
programs that contain rules for a large class of similar actions. This minimizes the storage
needs, accounts for novelty (new actions are merely versions of other actions previously
performed and, therefore, part of an existing class of movements), and explains motor
equivalence by arguing that rules of a GMP are not muscle specific; rather there are
invariant features that the program specifies, including timing and force coordination. These
invariants help define classes of movement and minimize the absolute amount of
information that must be stored.
Conversely, DST proposes that rather than a sequence of motor steps that are “stored”;
movement is an emergent property10 occurring as the neuromuscular system interacts with the environment; an online adaptation specific to the task at hand. In DST, physical
movement is constrained by characteristics of the individual (size, cognition, motivation,
etc.), environment (light, gravity, etc.), and task (goals, rules, etc.). Although the CNS is
still necessary for initiating movement and monitoring ongoing movement for error, it is just
one subsystem responsible for the eventual motor output. An assumption of DST is that
while certain movement patterns are preferred they are not obligatory and, therefore, new
patterns of movement can emerge when there is a shift somewhere in the system. This is
an attractive idea when working with pitchers who demonstrate ‘inefficient’ mechanics - this would represent a “shift in a sub-system” allowing for a new adaptive motor pattern to emerge. The downside is that even small injuries may themselves act as constraints, pushing movement strategies in an inefficient direction.
It is not clear whether one theory will prevail or a compromise of these two theories will
evolve that better answers how movement occurs. Bernstein suggested that the outcome of
a movement is represented in a motor plan (e.g., aiming a ball toward a target), and
distributed at different levels of the CNS. This is a concept that many theories have adopted.
Although the specific organization of motor plans is not known, flexible neural
representations of the dynamic and distributed processes through which the nervous system
can solve motor problems seem to exist. A motor program has evolved into an abstract
representation of a movement that centrally organizes and controls the degrees of freedom.
Learning comes from an interaction and strengthening among multiple systems and there
may be strong neural connections between related systems that can be crudely viewed as
representations. This internal representation needs to be matched to the external
environment and functional movement likely emerges as a result of this interaction.
2.0 SKILL ACQUISITION
Skilled actions are those that demonstrate consistency, flexibility and efficiency.
Consistency refers to the repeatability of performance – is the individual able to perform the
task consistently over a period of trials conducted over a number of sessions? For example, can Jack repeat performance across multiple training sessions? Flexibility (transferability) refers to the ability to adapt and modify task performance based on changing environments or conditions. For instance, can Jack transfer his performance from a drop-step to an in-game situation? Efficiency usually pertains to the capabilities of the cardiovascular and musculoskeletal systems. Can Jack perform across multiple innings of work, with elevated pitch counts? It is important to realize that performance of an activity indicates that one has attained that skill; however, within any motor task people can possess various levels of skill.
2.1 Stages of Skill Acquisition
In the early part of Jack’s training, his movements may be poorly controlled and his movement goals may be simple and limited. For example, Jack may make errant throws, spike the ball into the ground, or look clumsy moving through drills. He may need to focus a lot on the movements. As he keeps improving, Jack may exhibit a larger repertoire of movements and move with less effort or greater efficiency. These attributes exemplify Jack’s progression through stages of skill acquisition.
As coaches, we must determine where individuals are struggling along the learning continuum so that we can target our interventions appropriately. Although various stages of learning have been proposed, a two-stage model proposed by Gentile introduces key components for clinicians to consider when designing intervention strategies.
In Gentile's model, there are two objectives for the initial stage of learning: (1) to learn the
basic movement pattern needed to achieve the goal; and (2) to identify components of the
environment important to the task. Gentile further classifies environmental characteristics
into regulatory and non-regulatory features. Regulatory features of the environment include
all aspects necessary for successful performance of the task. Thus, when learning how to create hip-shoulder separation using the drop-step constraint drill from a mound, Jack must consider the type of ball being used (regulation vs plyo ball vs weighted ball), the target he is aiming for, the slope of the mound, the surface friction of his shoes and the mound, etc. Non-regulatory features are those aspects of the environment that are present - and may even be distracting - but are not integral to performance of the task. In our example, the colour of the ball, presence of other members in the facility, or whether or not Metallica or Bad Bunny are playing are all non-regulatory features. Even though the features may alter the way the movement is ultimately produced, Jack does not need to attend to these characteristics to perform the drop-step.
When they are in the initial stage of learning individuals should be encouraged to actively
explore the environment through trial and error. This stage is a period in which the basic
dynamics of movements are experienced and new strategies are tested within the limits of
player safety. It is often considered a cognitive stage as performers must solve a series of
problems experienced as they try various movements. Coaches can aid learning by structuring the environment to maximize regulatory features and minimize non-regulatory features as individuals actively search for appropriate movement strategies.
Once a coordinative pattern develops that allows for some degree of success, and the
performer is able to distinguish between regulatory and non-regulatory features of the
environment, the later stage of learning begins. During this phase of refinement the focus
switches from “what to do” to “how to do” the movement better. Thus, this later stage of
learning is characterized by a less cognitive process of consolidation in order to improve
motor efficiency and movement flexibility (e.g. the ability to perform the task under
different conditions).
It is important to remember that learning is not linear. Instead motor performance follows
the `power law of learning' with large improvements noted during early practice and smaller
rates of improvement displayed as practice continues. We often see periods of great
improvement followed by plateaus or even regression in our players. During these periods it
it is possible that, while performance appears worse, learning is still occurring. During the off
periods individuals may be fatigued or have decreased attention or they may be attempting
new strategies to perform the task. However, evidence suggests that memory consolidation
for long-term storage continues during performance plateaus and plateaus are followed by
new periods of observable improvement.
2.2 Explicit vs. Implicit Learning Processes
Gentile has suggested that motor skill learning involves two parallel yet distinct learning
processes, explicit and implicit, complementing the stages of learning discussed above.
Although these two processes change at different rates, and appear to take place in different stages, they overlap during skill learning. During explicit learning the performer's focus is on attainment of the goal as in the initial stage of learning. In an attempt at early success, the performer develops a “map” between their body structure and the conditions within the environment.
Initially, Jack must understand the movements he can make in order to perform the drop-step and effectively create hip-shoulder separation. He must attend to changes in his movements shape/structure and its relationship to external conditions and demands as he attempts to problem solve through tasks. We as coaches need to provide simple, relevant cues (creating easily affordable environments) to assist with problem solving.
Whenever movement patterns can be consciously adapted by the performer they are considered to be regulated by explicit processes. However, success at achieving a task goal does not necessarily imply that movement performance is efficient.
During extended practice in the later states of learning, Jack’s motor control strategies should be refined, indicating the predominance of implicit processes. Implicit learning will occur over a gradual period of time as he learns to unconsciously merge successive movements, couple simultaneous components and regulate intersegmental force dynamics inherent in specific tasks. Intersegmental force dynamics incorporate active forces produced through muscle contraction and passive forces such as motion-dependent torques (joint movement obtained without muscle contraction) that occur naturally in the environment or as a result of movement. The variability typically observed in young children and novel performers as they learn particular motor skills allows them to develop a range of force production patterns. Although explicit processes dominate the early stages and implicit processes in the later stages, it is important to understand that both processes are present throughout (re)learning. Jack may not be able to perform the most basic version of a drop-step without changes in force dynamics and/ or gradation. Likewise, as his skill improves, new conditions will be confronted and conscious attention will be allocated as necessary.
2.3 Measurement of Motor Learning
Motor learning is measured by analyzing performance in three distinct ways: acquisition,
retention and transfer of skills. Acquisition is the initial practice or performance of a new
skill (or new control aspect of a previously learned motor skill). For Jack, this means the practice of efficiently throwing a ball towards a target by creating greater hip-shoulder separation via the constraints of the drop-step drill, as he incorporates the components of sequencing, balance control, strength, and movement efficiency.
Retention is the ability to demonstrate attainment of the goal or improvement in
some aspect, following a short or long time delay in which the task is not practiced. This means that Jack would be able to effectively achieve better hip-shoulder separation in a drop-step at the end of one training session and again at the beginning of a new session on a different day without further practice or cueing. If he is successful, he demonstrates that he has retained the ability to effectively create hip-shoulder separation in a drop-step.
Transfer requires the performance of a task similar in movement yet different from the original task practiced in the acquisition phase (e.g., altered force or timing). For example, Jack would display transfer if he demonstrated kinematic features of the drop-step during a janitor constraint drill, or a regular pitch to an in-game scenario.
Acceptable performance of a motor skill within a single session (or series of sessions) does not demonstrate that the skill has been learned. A skill is not considered truly learned until retention and/or transfer of that particular skill is demonstrated. It is imperative when determining a client's level of independence that we consider whether we have measured performance at a point in time or learning of the skill so that it can be performed in the environments and under the conditions necessary for the client to be successful.
2.4 Classification of Motor Skills
Three useful classification systems for motor skills include defining the:
size of the movement – gross or fine motor skills;
beginning and end points – discrete or continuous; and
characterizing the stability of the environment in which the task is being performed - open or closed.
These three classification schemes can be used to organize and plan task practice.
Fine and gross motor skills are familiar to baseball coaches. Fine motor skills are those that use the small muscles of the hands, such as the fingers for manipulation of grips in throwing different pitches. Gross motor skills use the larger muscles of the trunk and extremities. Both small and large muscles may be included in various drills (including the drop-step) for Jack, such as stabilizing the pelvis while allowing the trunk to rotate about it, delivering energy into the throwing arm.
Movements classified as discrete or continuous may be controlled by different
mechanisms. Discrete movements have a defined beginning and end point. Common
examples are turning on a light, pushing a button, raising your hand in class, slipping on a
shoe, and goal-directed reaching. Continuous or rhythmic motions are those with no clear
start or end. Walking, playing the drums, swimming, and driving all represent continuous
tasks. The distinction between these classes of movements is often difficult to discern. Is
continuous motion a series of discrete movements or, conversely, are rhythmic movements
functional units while discrete are merely abbreviated rhythmic motions? As movements
fall along a continuum between these classes, the term serial movements has been proposed
to account for movements that are continuous but with clear discrete components. In my opinion, baseball pitching could be considered a serial movement, since throwing an individual pitch could be considered discrete, but “pitching” in a game has much less distinct endings…
Finally, skills can be classified as open or closed based on the temporal and spatial features
of the environment where a task is performed. A closed skill is one in which the performer
can start and stop at any time because the regulatory features of the environment remain
constant, such as throwing a plyo ball against a wall, or lifting weights post-game by oneself.
Open skills require the performer to conform to changes in the environment for success. Predictive abilities are essential. For example, catching a ball requires you to move your hands in time with the movement of the ball to make contact. Similarly, when trying to throw to a catcher, you must conform to their expectation of pitch type and setup location to be successful. As coaches, we frequently start with closed skills because we are working with pitchers in a training facility or gym, but we must move to environments that are progressively more open to provide a challenge and optimize an individual's independence.
3.0 PROMOTING SKILL ACQUISITION
3.1 Practice
We know that to gain expertise a skill must be rehearsed repeatedly. However, there are
many variables to consider when structuring the way practice should ensue including the
amount, the type, and the schedule. As presented above, the best practice design should not simply promote immediate performance effects, but ensure long-term learning by promoting retention and transfer of skill.
3.2 Amount of Practice
It is well supported that the best way to improve at any skill is to practice, practice and do
more practice. The more time devoted on task the more opportunity an individual has to
improve their capabilities. This is readily observed by thinking about the development of throwing abilities in younger athletes. It takes a number of years for a child to be able to perform a throw that is similar to how it is performed in adults, let alone a drop-step constraint drill. During those years many movements are made as the younger “pitcher” initially tries to hold a regulation baseball or throw it to a target, and eventually attempts to get hitters out. During practice not all of the attempts and combinations of movements are successful but each attempt provides the young athlete with information about both what to do, and what not to do in order to achieve the goal.
In training and rehabilitation, the underlying principle, more practice is better, is readily observed when interpreting the literature on constraint-induced movement therapy (CIMT). In this paradigm, initially designed for adults post-stroke, the unaffected arm is restrained, requiring the individual to use the affected arm to complete numerous repetitions of various tasks that challenge the system. Results from this type of program have been promising showing that intense structured practice leads to improvements in function, quality of movement, timing, and even changes in the neural substrates of the brain. which correspond to improved movement capabilities. Though not directly the same, similar interpretations can be made with regard to baseball training capabilities and practice times.
3.3 Whole vs. Part Practice
Should a motor task be practiced in its entirety (whole) or should it be broken into separate
parts? The answer is not an easy one and is multifactorial involving an in-depth understanding of the movement in question. To decide if part practice may be beneficial the task must be analyzed based on the number of segments as well as the degree that those segments are interdependent on one another. In continuous motor tasks, the current portion of a movement is dependent on the movement just completed and, therefore, these tasks are best practiced in their entirety. Sequences of movements that will be coupled, such as the different phases of the throwing motion, may lend themselves to be divided, practiced, and then combined for whole practice since there are segments that are clearly separated from one another. Pitchers are commonly seen working through “dry reps” of different parts of their throw separately, then attempt to incorporate that change into their full delivery.
This part practice can be beneficial if used properly since a learner can perform pieces of the movement and have some degree of success providing increased motivation to learn the skill. However, deciding how to divide a task is often a difficult decision for a coach. For example, it may be easy to see how Jack can practice the stride phase as different from the follow through phase; however, it is actually more difficult to separate such tasks since there is a temporal relationship between them. Artificially breaking apart a task that does not lend itself to part practice may not benefit motor learning and may even hinder the process.
While it appears clear that more practice is better, how practice sessions should be structured to ensure optimal learning is less clear. Should a lot of practice be performed at once (massed) or should rest breaks be sprinkled throughout (distributed)? Should only one task be practiced (constant) or should different tasks or variations of the singular task be
introduced (variable)?
The concepts of massed and distributed practice define different ways that practice can be
undertaken. For example, if you wanted to work with Jack on his hip-shoulder separation using a drop-stop, you may decide to perform 30 trials during a training session, but should he practice all 30 trials at once? Massed practice requires all the trials to be performed in a manner that minimizes the amount of rest between trials so there is more time on task than there is spent during rest. Distributed practice divides repetitions into smaller chunks to allow for rest between trials (e.g., 5 trials now, 5 in 10 minutes, etc.).
There is no empirical evidence to support that either of these schedules lead to superior motor skill learning, however, depending on the goal of the practice session and the individual's capabilities (strength, endurance, cognition, ability to focus on task), incorporating either a massed or distributed schedule may lead to better learning and should be taken into account when designing an intervention.
3.4 Constant vs. Variable Practice
Performance of only one task exactly the same way time after time is termed constant
practice. Using our earlier example, this would involve Jack performing the same drop-step drill using the same mound, and the same plyo ball for every attempt. While this may improve Jack’s ability to perform this particular task, the literature to date has found there may be a reduced ability to retain and transfer a skill to an in-game situation following constant practice.
For transfer of skill to occur, study results have suggested that variable practice may be more effective. Variable practice involves performing variations of the task or completely different tasks throughout a training session. For Jack this would mean using different drill variations (step-back, janitor, etc.) in order to train. Variability could also be introduced by placing accuracy targets in different spots, or by changing the mounds or plyo balls being thrown. What is not clearly understood is how much variation should be present to encourage optimal learning?
Three categories are used to describe variability and practice order - blocked, random and
serial. Training hip-shoulder separation with constraint drills as our example, the current intervention is designed so that Jack will engage in 30 practice trials, performing three different constraint drills. A blocked schedule would require that all practice be completed under one condition before moving on to the next. For example, 10 drop-steps, 10 janitors, 10 step-backs. A random schedule maximizes variability. As with blocked, Jack will still perform 30 trials, but each drill is practiced in random order. If random and blocked practice schedules are considered as the anchors on a continuum, a serial schedule is somewhere in between. A serial schedule requires a sequence to be followed until all 30 trials are completed (e.g., drop-step, janitor, step-back… 10 times).
The ability to predict what task will be performed from trial to trial differs depending upon
the type of practice schedule, with blocked having the highest and random the lowest degree of predictability. The effect of different practice schedules on learning is termed contextual interference. Briefly, during blocked practice there is low interference or disruption in memory as a person practices multiple trials repeatedly. However, in random practice there is high interference because trials are interrupted by other tasks. Study results have shown that while higher contextual interference (random practice) may lead to poor performance it frequently leads to better learning (as measured with retention and transfer tests) compared to blocked practice. This may occur because in random practice the skill must be reconstructed on each attempt, allowing an individual to practice a variety of strategies. This benefit appears to be lost when learning very complex tasks or in individuals with significant neurological impairments. For example, Jack’s difficulty with multi-step commands and trouble focusing on tasks may make random practice ineffective. Furthermore, blocked practice may be best because of the increased chance for success during practice providing motivation to continue with practice.
As players become more comfortable solving motor problems, it is considered good practice to continually interview the players and understand what their current short and long term goals and perceptions are. Considering the roles motivation, randomization, effective transfer, etc. play in designing effective training protocols, the difficulty of practice should likely be paired to the intentions of the player.
3.5 The Role of Mental Practice
Mental practice involves cognitive rehearsal and imagining of a motor action with the goal
of improving performance but without the production of overt physical movement. Research has demonstrated that, depending on the task, improvements in motor skill can occur with mental practice alone, however, when mental practice is combined with physical practice the improvement in skill is magnified. It is hypothesized that mental practice is successful in helping improve skill because, when performed, the neural processes involved in imagining the movement are very similar to those required for physical performance. Further, although there is no overt movement with mental practice, EMG studies have demonstrated that low level muscle activity (submovement activity) occurs.
Thus, if Jack was to use mental practice to work on his hip-shoulder separation, he should be instructed that it is not a relaxing meditation. Instead, Jack should visualize himself going through the steps of throwing, concentrating on the movements required and “feeling” how his imagined body is moving.
3.6 Specificity and Location of Practice
Task-specific or task-oriented practice is an approach to coaching that focuses on
performance of functional tasks that are meaningful to the individual. In order for this type
of practice to be successful, a coach must be able to accurately assess their player and
identify their limitations and deficits. It becomes the job of the coach to accurately
arrange the environment to provide the proper affordances so that the task, or a modified
version of the task, can successfully be completed by the player. An affordance is the
reciprocal relationship or “fit” between a player and the environment and can drive the
composition of the movement. For example, when working on hip-shoulder separation with Jack, if the goal for the session is to improve maximum hip-shoulder separation, then following the principles of task-oriented practice, Jack might be asked to perform split stance thoracic spine rotations, or heavy medicine ball hook-em’s.
While task oriented training may appear simple, setting the environment to target the
movements you desire your player to practice is quite difficult and time consuming. If your
environment is not set properly the performer may not be successful and thus may become
frustrated and not have the drive to want to continue to practice. Further, it is important to
ensure the activity performed is challenging and meaningful to the individual as to not be
perceived as boring or useless.
Practice which leads to optimal learning depends on the task being learned and the
characteristics of the learner (e.g., age, stage of learning). The best practice design will not
simply promote immediate performance effects, but more importantly will promote long-
term learning.
4.0 THE ROLE OF FEEDBACK
Feedback refers to information an individual receives pertaining to the performance of a
task. It is generated from two sources. The first is referred to as internal or task-intrinsic,
which is information about the movement gained through interpretation of sensory, visual
and auditory experiences. The second is external feedback, commonly referred to as
augmented feedback since this information is primarily used to enhance task-intrinsic
feedback. However, sometimes external feedback may have a more important role and
serve as a replacement if impairment is present in a sensory system. Augmented task-related
feedback can come in many forms such as verbal, visual (demonstration), or physical
(manual guidance). Use of augmented feedback can greatly enhance a person's ability to
learn a task, but when and how should it be provided? Here we will explore the types of
augmented feedback available and discuss the content, timing and frequency of this
feedback and its role in enhancing learning.
4.1 Types of Augmented Feedback
Verbal augmented feedback is provided as either knowledge of results or knowledge of
performance. Knowledge of results is information related to the outcome and in most
cases is redundant information because by the time a task is completed the performer is
usually aware if they were successful. Knowledge of performance pertains to information
regarding execution of the task and typically relates to the type or quality of the movement.
Although it is usually redundant information, knowledge of results may be useful at any
time during the learning process but is particularly useful in the earlier stages because it can
serve as a motivator (e.g., great job, you opened your hand). Knowledge of performance is
best used in the later stages after the goal of the movement is realized and now the
information provided about varying aspects of the movement can be more readily
understood by the performer.
4.2 Composition of the Augmented Feedback
Feedback can be delivered many ways. With regards to knowledge of performance the coach can provide descriptive information regarding the past movement (e.g., you rotated too soon in the drill) or prescriptive information offering a possible solution to be used for the next attempt (e.g., the next time you move, keep your torso facing the back wall for longer).
If used in the traditional sense, these are both passive methods of providing information (the learner is being told what happened or what to do next). A better way to use these tools may be to engage the learner in a dialogue by asking leading questions so that they become an
active participant in trying to solve the movement problem (e.g. what do you think happened on the last attempt? What will you do differently on the next attempt?). Active learning is thought to be crucial because it emphasizes the process of continuing to solve new and different motor problems as they arise, rather than just being provided with the solution. For example, when working with Jack to improve his hip-shoulder separation through a drop-step constraint drill, we may ask him how he intends to create separation. This approach would require Jack to assess the situation and his ability to ultimately formulate and execute a motor plan, rather than just being provided with a solution.
In addition, depending on the learner and the stage of learning they are in, the coach
must decide what type of feedback to provide such as:
whether the information given should be general (about the movement itself) or specific (about a particular portion of a movement);
qualitative (that was better than last time) or quantitative (that was 10 degrees greater than last time);
comprised of information related to an internal focus of attention (rotate your chest to your rear glute) or external focus of attention (reach chest to back wall as long as possible);
and whether information should be provided regarding the correctness of the movement or on the errors.
Overall, it appears that individuals in the early stage of learning, such as Jack in the early
phase of learning how to throw, would do better with prescriptive or active-engaging feedback that is more general and qualitative in nature with an external focus on what was done correctly.
4.3 Frequency of Augmented Feedback
Feedback can be provided concurrently during a task or it can be provided after a task is
performed (terminal). While the goal of concurrent feedback is to have an immediate effect
on the movement being performed, studies have suggested that concurrent feedback may actually hinder retention and transfer because the performer or the individual performing the
task, becomes reliant on that feedback to complete the task successfully. This is an
important consideration for hand coaches. So, how much feedback should be provided?
Studies in healthy populations are suggesting that less feedback is best. This does not suggest that no feedback should be provided. Augmented feedback can enhance learning but it has also been found that too much can hinder learning. Our players naturally become dependent on feedback when it is provided to them. Feedback that is more frequent encourages passive rather than active participation and can reduce one's ability to perform those skills. Thus, the amount of feedback should be reduced as training progresses.
This can be done by (1) using an intermittent schedule of when feedback is given that
consists of summary or average information regarding a selection of previous attempts; (2)
using a faded schedule that initially provides a lot and then is reduced as practice continues;
(3) setting up boundaries (bandwidth) where feedback will be given only if an error is too
large or too small; or (4) allowing the performer to have some control and decision-making
over when and what type of assistance is provided during task practice. The frequency of
feedback is dependent on the stage of the learner and the learner themselves.
4.4 Before Task Performance – Modeling
Modeling “refers to the process of reproducing actions that have been executed by another
individual”. These demonstrations are powerful sources of information given prior to task
performance that provide the observer with the general movement pattern and the goal of the movement as well as information about the way submovements are coordinated and related to one another that are otherwise difficult to put into words. Studies indicate that the learner readily adopts the demonstrated strategies (whether effective or ineffective) for use to perform a motor skill. Thus, the model that will demonstrate the movement must be
carefully chosen.
The type of model used has been found to influence motivation, and therefore subsequent
learning. While mastery or expert models provide important strategies for the learner these
strategies may be so advanced that they are unusable for the learner. In these cases, a less
skilled model, or even better, a less-skilled peer model (someone with similar physical
characteristics) maybe more beneficial because the learner can notice correct and incorrect
aspects of the movement and strategies the model used to correct for errors. Thus, in the case of Jack, the coach serving as the model for how to perform a drop-step may not provide Jack with the most useful information, it may be best to have a peer (another player who just learned the drop-step) perform this activity for Jack to watch and problem solve through.
4.5 Manual Guidance
Manual guidance is when a coach passively moves a player in an effort to
provide more appropriate proprioceptive feedback, but is this the best way to teach an
individual how to perform a task? It goes without argument that when safety is a concern
then it is important that a coach is close by or has their “hand-on” the learner but what
about when learning or relearning a skill? It has been suggested that manual guidance is, in a way, concurrent feedback. Thus, the guidance that a coach provides becomes part of the
regulatory features of the task and the participant becomes dependent on that feedback to
complete the task successfully. Thus, while performance may improve, learning (as
measured through retention and transfer) may be slowed since during practice the learner did not need to actively solve the motor problem over and over. So, when working with Jack as he tries to improve his hip-shoulder separation, should the coach practice in a manner that is always hands-off? Well, not exactly.
First, as mentioned above, safety is of the utmost importance. Barring that, there are other times when manual guidance may be needed. For example, we may take our hands or attach a core velocity belt to Jack’s hips in order to take him through the movement he is being asked to perform to give him an idea of the movement we would like him to produce. Yet, if manual guidance is used, the coach should attempt to remove the physical assistance as soon as possible so the player does not become dependent on the guidance.