A Critical Evaluation of How Theories of Motor Learning
Louise Warren Final Assignment Critically evaluate how theories of motor learning and motor control place varying degrees of importance upon feedback and emphasise different types of feedback. Research on motor learning and control has been debated for many years. According to Schmidt & Wrisberg (2007, 5), Franklin M. Henry (1904-1993) was one of the pioneers of motor skill research in the laboratory at the University of California at Berkeley. Some of his students, including Richard Schmidt, whose work is discussed in this essay, went on to develop their own theories and lead enquiry into the motor skill field (Schmidt & Wrisberg 2007).
How we acquire motor skills, for example voluntary movements of the limbs, body and head, is the study of motor learning and motor control. This is achieved by the state of our “neuromuscular system” (Magill 2007, 3). Motor skills are produced through practice and become part of the long-term memory through repetition of skills (Wilmerding & Krasnow 2009). Edward Thorndike (1874-1949), was considered to be one of the early researchers into the effects of stimulus-response in motor skill learning (Salmoni 1989, 197).
Thorndike’s Connectionism is one of the models of motor skill learning which this essay will discuss in terms of how feedback impacts learning and what type of feedback can be considered in each case. An example of learning a motor skill A dance teacher working with a student wants to improve the execution of an assemble dessus. First she decides to take the student back to the barre to focus on the action of the leg as it quickly slides to the side from the demi-plie, while simultaneously pushing off the supporting leg, then joining (assembling) the feet in the air in 5th position (Royal Academy of Dance 997, 83). The use of the barre provides an aid in learning, to help the student control the speed of the step, assisting with stability in the upper body. The student may improve her technical understanding of the step in several ways: through attention, focus and observing the teacher demonstrate the step (Krasnow & Chatfield 1996). Other ways include execution, feedback from the instructor and further repetition (Krasnow & Chatfield 1996); although repetition in itself does not always improve performance and may depend on whether feedback is given (Wilmerding & Krasnow 2009, 3).
What is feedback? Feedback aids the student by giving them information on the state of their performance or actions (Schmidt & Wrisberg 2007). There are two types of feedback: intrinsic and extrinsic, which is also known as augmented feedback (Schmidt & Wrisberg 2007, 285). Intrinsic feedback is the “sensory-perceptual information that is a natural part of a performing skill” (Magill 2007, 332). This is information received through the senses.
In the above teaching example, intrinsic feedback could be received visually, as the student observes the instructor demonstrate the correct execution; auditory, as she hears the sound of the foot swish against the floor before the jump and through proprioception, by feeling the forefoot push into the floor. Augmented or external feedback provides additional information to the student which they are unable to detect intrinsically, usually through an instructor or another external source such as video (Schmidt & Wrisberg 2007, 286).
Instructors will need to decide if augmented feedback is warranted depending on how complex the skill is and think about the experience of the learner by questioning are they beginners, or more advanced in their skill-set (Schmidt & Wrisberg 2007, 295)? The response would then determine the type of training and feedback utilised, however the instructor also needs to make a comparison between knowledge and actual skill for example: A novice given complete information on how to ski, a set of decision rules, and then launched from a mountain top would most likely end up in an orthopaedic ward .
Bandura cited in Salmoni 1989, 207 Augmented feedback can be further divided into two areas: knowledge of results (KR), which refers to the outcome of the performing skill (Magill 2007, 333) and knowledge of performance (KP), giving the student information about the “movement characteristics which [lead] to the performance outcome” (Magill 2007, 334). Using the teaching example previously, KR could be used to inform the student if she had reached appropriate learning outcomes for jumps: for example; “Secure, stretched leg extensions” (Royal Academy of Dance 2010) on the assembling of the legs in the air.
KP can be utilised in dance teaching to provide feedback about the quality of the movement. In our example of the student practicing assemble dessus at the barre, KP may be used to help improve the movement; for example a teacher may give a verbal correction such as “ensure your knees are over your toes during the demi plie before the jump”. Thorndike’s Connectionism Thorndike believed that the learning process was evidenced as the strengthening of bonds between stimulus and response. Additionally, he believed that learning was a trial and error process (Salmoni 1989, 197).
If the connection resulted in a “satisfying state of affairs” (Salmoni 1989, 197), and the outcome was encouraging to the subject, then the connection was strong. If, on the other hand, the outcome was not satisfying to the subject and was an “annoying state” (Salmoni 1989, 197); then the connection was weak (Salmoni 1989). This he described as the Law of Effect (Schmidt & Wrisberg 2007, 312). He thought that external feedback as KR should be given after every movement task to aid learning in the absence of intrinsic feedback or the connection would not be strong (Schmidt & Wrisberg 2007, 312).
With regard to the frequency of KR feedback that Thorndike originally proposed in order for the strengthening of stimulus-response bonds, other theorist’s have since questioned this. More recent research has “concluded that decreased relative frequency of KR increased learning” (Salmoni, Schmidt and Walter cited in Salmoni 1989, 45). However, further experimental evidence did not fully support this hypothesis; but as a result, further enquiry was developed in this area (Salmoni 1989, 45). One of the negative aspects of frequent feedback has been discussed by Magill.
He expresses a possible side-effect being “an attention-capacity overload” (Magill 2007, 359). Adams closed-loop theory Adams theory contrasted with Thorndike’s theory of stimulus response and its role in reinforcement of learning. He believed that during the initial stages of learning, knowledge of results was not responsible for strengthening the bonds between stimulus and response, but that it delivered error information to the subject, and influence subsequent performance trials (Salmoni 1989). This would allow them to make gradual adjustments to the movement to improve execution (Salmoni 1989).
Adams construct utilised a “reference mechanism” (Adams 1976, 90) which referred to a perceptual trace and memory trace (Salmoni 1989, 201). Perceptual trace is explained as stored movement information in which the KR compares the desired movement with the current movement to inform and improve the acquisition of the skill (Adams 1976) and the “extent of the movement” (Salmoni 1989, 201). Memory trace is the initiation and choice of the movement (Brian Mac Sports Coach 2011[online]). In contrast to Thorndike’s theory, Adams believed that “delay of knowledge has no effect on movement acquisition” (Salmoni 1989, 201).
However, he felt that during the initial stages of learning a movement, that knowledge of results was essential for acquisition of the skill and that withdrawal of feedback at this stage would adversely affect the learning outcome (Salmoni 1989). Adam’s theorised two stages of learning: the verbal – motor stage, which was the initial acquisition of the skill through verbal-cognitive control and motor-stage, in which the skill had become controlled internally and mostly “error free” (Salmoni 1989, 201).
During the initial stages of acquiring a motor skill, feedback is given to inform the student of the status of the movement which will create a response. This could be either to carry on the movement as per initial instructions if it was correct, or give augmented feedback to improve or correct the movement error (Magill 2007). As the student constantly refers back to the reference model, he improves the motor skill to the point where feedback can be reduced and he has acquired the skill. This is the closed loop theory of learning.
An example of closed loop learning could be viewed as a dance student learning to develop correct posture in a ballet class. The student first receives a visual aid feedback; for example, the teacher’s correct stance, or a picture or verbal description. The teacher gives the student feedback indicating whether she has achieved correct posture with the use of verbal and kinaesthetic (hands on) cues. Eventually, the student has the image internalised and reaches the motor skill stage where she no longer needs to be reminded of the correct stance for ballet.
One of the limitations of this theory is that it is a time consuming process due to the system of constantly “comparing the actual and expected feedback” in order to control movements (Schmidt & Wrisberg 2007, 74). Schmidt also expresses another limitation. He describes that closed-loop control can be “inadequate for performing skills that are brief in duration” (Schmidt & Wrisberg 2007, 75); for example where a batter in a baseball game has to quickly asses a change in the direction of the ball in flight (Schmidt & Wrisberg 2007, 75). Therefore, Adam’s closed loop theory seemed more suited to less complex skills (Rosenbaum 2009).
However, as Salmoni expresses “the theory was responsible for stimulating a great deal of research in this area” (Salmoni 1989, 200) including Schmidt’s motor schema theory. Motor Schema theory In 1975, Richard Schmidt sought to continue research from Adams closed loop theory. He believed that the relationship formed between the acquired movement information and the subsequent evaluation of the motor responses would lead to the subject forming schemata which would “determine specific actions” (Jacko 2009, 398), in order to achieve learning outcomes.
Recall schema, encompasses the initial motor learning experience, including the response specifications and outcomes; this was “response production” (Salmoni 1989, 201). Recognition schema also included internal and external feedback through sensory conditions (Salmoni 1989, 201) and considered the accuracy of the response (Salmoni 1989). Unlike Adams, he believed that actions are not stored in the memory as singular moves that are connected, but that we refer to abstract relationships and rules about movement (Brian Mac Sports Coach 2011[online]).
His theory described an abstract class of movements known as a “generalized motor program” (Magill 2007, 89) or (GMP) in which the subject recalls a class of movements which is retrieved when a skill involving that class of movements is utilised (Magill 2007). This theory was designed to improve discrete movement skills and use practice variables to ensure the class of movements could occur in a variety of situations (Leitch 2011). Examples of these movements could be the actions of: a tennis serve, bowling a ball and a golf swing. .
Schmidt’s theory placed an emphasis on knowledge of results to strengthen both recall and recognition schemas (Salmoni 1989). According to Sherwood & Lee (2003, 378) both KP and KR were used frequently after every movement trial to aid in the evaluation of the ideal movement, therefore; “updating the memory representation” (Sherwood & Lee 2003, 378). Schmidt believed that withdrawal of KR would weaken the “recognition schema” (Sherwood & Lee 2003, 381). However, subsequent research has questioned the effectiveness of immediate feedback in retention and transfer of learning tests (Sherwood & Lee 2003, 378).
This is evidenced by Swinnen (1996, 52), who explains that the learner may develop an overdependence on continuous feedback which then becomes part of the process, thereby affecting internal processing. An example of over-use of feedback could be viewed where a dance teacher constantly dances with the students, cues them before every exercise and corrects many aspects of the step. The students may not internalise the organisation of the steps by copying the movements of the teacher, or develop a reliance on the cues and corrections to remind them of the composition.
However, observational learning, not copying without internalising the movement, can lead to motor skill learning as discussed in the next example. Bandura’s Model of Observation learning Albert Bandura regards motor skill learning as the act of the learner acquiring information of movement patterns from verbal instruction and modelling (demonstrating). Augmented and intrinsic feedback further strengthens the learning process. He believes that observational learning occurs due to the subject viewing the “effects of their actions during enactive learning” (Salmoni 1989, 204).
Bandura’s theory is also known as “cognitive mediation theory” (Magill 2007, 318) and describes four processes. Firstly, attention; how the learner focuses attention on the task and factors which affect this (Learning Theories Knowledgebase 2011[online]). He believed that it was important for the learner to pay full attention and not just observe (watching passively) to affect optimal learning (Magill 2007, 318). Secondly, in the retention process of the observed action, memory of the task occurs including organising mental images and motor rehearsal (Magill 2007).
Thirdly, during the reproduction process, the mental image of the action is turned into a physical action. At this stage successful production of the action depends on the capability of the learner (Magill 2007). The final step in Bandura’s theory is motivation; there must be a good reason (an incentive) for the learner to imitate the task (Learning Theories Knowledgebase 2011[online]). Motivation can influence a students desire to continue with the task. This is often delivered through encouraging (positive) verbal feedback to indicate how well a student is performing.
Research has shown this can be effective in helping to develop a student’s self-efficacy (Magill 2007, 335). According to Bandura, without all four processes completed, the action will not be performed (Magill 2007, 318). According to Salmoni, Bandura’s theory was similar to Schmidt’s in that he “believes that people learn generative rules of action rather than specific action patterns” (Salmoni 1989, 204). Progress is monitored through feedback and movement corrections which correspond with the desired action (Salmoni 1989).
Through observing the model and also receiving verbal feedback, some researches have found an increase in self-efficacy leading the learner to have belief in their ability and result in greater motivation (Zetou et al. 2008). This can be achieved through observing an expert perform the skill to a high level or watching themselves via video and receiving feedback. A study by Tzetzis, et al. cited in Zetou et al. (2008) compared self-modelling, (the subject observing themselves); observing experts with verbal feedback and the use of traditional teaching methods in novice skiers.
The results indicated that the self-modelling and observing experts approach produced superior outcomes to the traditional teaching methods (Zetou et al. 2008). However it was noted that care must be taken to ensure the students are not focusing on observing the negative aspects of their own performance which can lead to a decrease in motivation (Alkire et al in Zetou et al. 2008). Observational learning can be incorporated into a dance classroom by using different teaching strategies. Here is an example discussed by Richard Glasstone (1999).
Dancers learn through visual observation …Teaching them to watch each other carefully is one of the best ways of helping them to correct their own faults. ” Richard Glasstone 1999, 61 However, as with other motor learning theories, more research is needed in order to better understand exactly how movement cues are learnt; in this case, through observation and that “a comprehensive theoretical model” (Scully & Newell cited in Salmoni 1989, 208) needs to be developed. Conclusion