Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Augmented concurrent error information and the acquisition of the continuous gross motor skill of forward.. Russell, Susan Leilani 1981-03-26

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata


831-UBC_1981_A7_5 R97.pdf [ 3.48MB ]
JSON: 831-1.0077341.json
JSON-LD: 831-1.0077341-ld.json
RDF/XML (Pretty): 831-1.0077341-rdf.xml
RDF/JSON: 831-1.0077341-rdf.json
Turtle: 831-1.0077341-turtle.txt
N-Triples: 831-1.0077341-rdf-ntriples.txt
Original Record: 831-1.0077341-source.json
Full Text

Full Text

AUGMENTED CONCURRENT ERROR INFORMATION AND THE ACQUISITION OF THE CONTINUOUS GROSS MOTOR SKILL OF FORWARD OUTSIDE EDGES IN FIGURE SKATING by SUSAN LEILANI RUSSELL B.P.E., University of British Columbia, 1978 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF PHYSICAL EDUCATION in THE FACULTY OF GRADUATE STUDIES School of Physical Education and Recreation We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA March 1981 (c)Susan Leilani Russell, 1981 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Pk^<, i CL.t± V ^uc^-Uo The University of British Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 ABSTRACT An experiment was designed to test the effects of in creased availability of FB cues through the use of augmented concurrent error information on the acquisition of the contin uous, gross motor skill of consecutive forward outside edges in figure skating. Each of 2 groups of 10 Ss per group skated 20 trials, jco.nsisting of 6 consecutive outside edges, on each of 2 days. Percent time spent on edge was measured as the dependent variable. In addition to verbal KR*and instruction al information, received from an instructor working in a double blind condition, the experimental group also received immediate, concurrent, response generated error information from a telemetric monitoring device. On day 2, all subjects from both groups performed an additional 10 trials, without the aid of KR, instructional information, or information from the monitoring device. Two hypotheses were tested. Hypothesis 1, which stated that subjects having increased accessibility to relevant FB cues would show a faster rate of skill acquisition, was sup ported by the results of this study and is in keeping with closed-loop motor control theory. Hypothesis 2, which stated that artificially enhancing error information does not inhibit progress to the motor stage of skill acquisition,as^refleeted in performance maintenance when KR is withdrawn, was also sup ported by the results. •• There was no significant change in performance when KR and the telemetric monitoring device were withdrawn on Day 2 of the experiment. iii TABLE OF CONTENTS Page LIST OF TABLES v LIST OF FIGURES vi CHAPTER I STATEMENT OF THE PROBLEM 1 Introduction ....Statement of the Problem 5 Hypotheses 6 Definitions 7 Delimitations 9 AssumptionsLimitations 11 II REVIEW OF THE LITERATURE 12 Adams' Closed-Loop Motor Control Theory 12 Selective Attention Implications ... 21 Allocation of Attention 22 Some Considerations 26 III METHODS AND PROCEDURES 9 Subjects 2Apparatus 9 Experimental Design 31 Procedures 33 IV RESULTS AND DISCUSSION 37 Results from Experiment One 37 iv Page Discussion of Results from Experiment One kl Results from Experiment Two k5 Discussion of Results from Experiment two k6 Summary of Results 51 V SUMMARY AND CONCLUSIONS 3 c Conclusions 55 DiscussionSuggestions for Further Research .... 57 BIBLIOGRAPHY 59 APPENDICES. A. Apparatus 6k B. Instructions to Subjects 69 C. Trial Scores By Subjects 72 V LIST OF TABLES TABLE Page 1 Mean Percent On-Edge Times for Groups in Experiment One 38 2 ANOVA Table for Experiment One 39 3 ANOVA Table for Experiment Two 47 4 Mean Performance Scores for Groups la in Experiment Two 48 A-l Percent On-Edge by Subject for the Experimental Group in Experiment One 73 A-2 Percent On-Edge by Subject for the Control Group in Experiment One 75 A-3 Percent On-Edge by Experimental Subjects in Last Five Trials of Experiment Two . 77 A-4 Percent On-Edge by Control Subjects in Last Five Trials of Experiment Two . 78 vi LIST OF FIGURES FIGURE Page 1 Edge pattern during test situations 10 2 Comparison of flat and edge tracings .... 10 3 Strip chart recording of edge/flat times . . 32 k Pattern for consecutive forward outside edges along a coincidental axis for an experimental trial 35 5 Significant trials effect showing general improvement of performance scores over time ^-0 6 Slopes of learning curves for experiment one k2 7 Mean performance scores for quarter segments of trial one kk 8 Average performance scores by groups and conditions k9 9 Mean performance scores for last five trials of experiment one and last five trials of experiment two 50 A-l Device architecture 65 1 CHAPTER I STATEMENT OF THE PROBLEM Introduction The importance of edge control to the sport of figure skating cannot be overemphasized. The controlled, curving edge is the basis of virtually all skills required and must be mastered before proficiency in any area of the sport can be attained. To this point, therefore, it is of great im portance to the figure skating coach that beginners in the sport acquire edge control with good technique as early in the skater's career as possible, in order that progression to advanced work can occur. Control and perfection of the edge is attained through training on compulsory figures, to which skaters dedicate many hours of practice. In motor learning terminology, the skill of skating a compulsory figure is defined in terms of a continuous skill, requiring regulation by the skater during his/her perfor mance (Fitts and Posner, 1967)• The rules of the sport of figure skating require that only one edge of the hollow ground blade profile be in contact with the ice at one time during the execution of a compulsory figure. Violation of this procedure results in the occurrence of a flat, where 2 "both edges of the blade are in contact with the ice at the same time. A flat is defined by the rules of compulsory figure skating as an error, and can be readily identified by looking at the marks or tracing left by the skater on the ice as a result of the performance. A flat tracing is characterized by a double track as opposed to a single track left by a cleanly skated edge (Figure 2 ). Thus the angle of the blade to the ice is critical throughout the performance of an edge. It is important that the skater makes use of kinesthetic information so that he can learn to "feel" the difference between an edge and a flat, and correct his performance when a flat occurs. This view is consistent with closed loop control theories of motor skill acquisition, where feedback arising from an ongoing response is compared to a central reference of correctness for error detection. The major function of a closed loop system is to minimize the extent of error in terms of the deviation of a system's output from a central reference of correctness or desired goal (Schmidt, 1976). In skating edges, the difference between skating an edge and a flat usually only requires a 'very fine adjust ment to the blade-ice angle, and thus for this refinement to occur, there must be a very accurate internal reference of correctness to detect the error. In Adams' Closed Loop 3 Motor Control Theory (1971). the internal reference of correctness is referred to as the perceptual trace (PT), and is said to be weak and ill-defined during early stages of learning. The PT consists of a 'pool' of traces or images from past performances, making it weak, unreliable, and subject to forgetting during early stages of learning where few past experiences with the task have occurred. Thus the detection of a flat based on the interaction of feedback and the PT is not necessarily dependable. Sub jects at these early stages must rely heavily on peripheral rather than central mechanisms for error detection. Adams claims that during these early stages of skill acquisition, learners use feedback (FB) in relation to knowledge of results (KR) to adjust their performances towards a criterion. This early dependence on KR is charac teristic of what Adams has termed the verbal-motor stage of skill acquisition. KR is used to draw the performer's at tention to appropriate FB cues which would otherwise go unattended. This allows the subject to form a strong per ceptual trace which will act as a reference of the criterion response when KR is no longer available. When the per former can use the PT for the control of his performance in the absence of KR, and performance can be maintained, he is said to be functioning in the motor stage. I* Traditionally, KR is given after a performance is completed. Studies "by Bilodeau and Bilodeau (1958) show that KR delay does not affect learning except in cases of continuous tasks. In a continuous task, such as skating edges, performers must rely on some form of short term memory (STM) to retain the motor information of a perform ance so that it may later be used in relation to KR to be compared to a criterion response. During the period of time where a skater is holding information about a pre viously skated edge, hist-performance is continuing, thus adding more and more information to be retained and proc essed. Fitts and Posner (1967) define STM as a system which loses information rapidly in the absence of sustained attention. Due to the limited processing capacity of the central nervous system (CNS), individuals have to select pertinent items from this temporary store depending on the demands and goals of the current activity. Unattended in formation fades away so that it cannot be recalled, conse quently, all environmental and sensory information face the same limitations due to the process of selective at tention. Before a strong PT can be laid down, the learner has first to perceive the appropriate error information by attending to it, and then hold this attended information in memory so that at the end of the performance, it can be used in relation to KR to adjust subsequent responding. The learner will not attend to, hence not retain, informa-5 tion about an error if in fact he is not aware that an error has occurred. Information that is not attended to stands subject to fading and forgetting, and thus may not be available for use in relation to KR. Based on these theoretical considerations the ef ficiency of learning a continuous task under traditional KR delay methods can be questioned. It would seem plau sible that in a situation where appropriate error informa tion could be made available for attention immediately at the onset of the error, learning would occur at a faster rate . This study was designed to investigate the effects of increased availability of FB information to a per former' s attention on the acquisition of the continuous gross motor skill of skating consecutive forward outside edges. Statement of the Problem This study is a practical application of the the oretical rationale of closed-loop learning, as it-'pertains to the acquisition of the continuous gross motor skill of skating forward outside edges. The purpose of the in-6 vestigation is two-fold. The first concern is to examine the effects of increased availability of relevant feedback cues to a performer's.attention on the rate of acquisition of the self-paced continuous gross motor skill of skating forward outside edges. The second concern is to determine whether or not aid in recognizing the important feedback cues through a FB augmenting device will affect the pro gression of performers from the verbal motor stage of skill acquisition, to the motor stage, that is, will the learner develop a dependency which will debilitate per formance once the device is withdrawn. Hypotheses The hypotheses are: 1. Subjects having increased accessibility to relevant FB cues will show a faster rate of skill acquisi tion, as reflected via a reduction in error scores. 2. Artificially enhancing error information will not inhibit progress to the motor stage, as reflected in maintenance of performance when KR and the monitoring device are withdrawn. A strong PT will be inferred from performance maintenance when KR is withheld. 7 Definitions Closed-loop motor control. Behaviour which is self-regulated and accordingly adjusted for correctness "by comparing current FB to a centrally established reference mechanism is considered closed-loop in nature (Adams, 1968,1971) • Knowledge of results (KR). Information which is ex ternally provided relating discrepancies between desired and achieved responding is considered knowledge of results. Feedback (FB). Sensory in nature, feedback refers to the internal after effects of responding. Perceptual trace (PT). In Adams' closed-loop motor control theory (1968), the referential memory system which is made up of past movement consequences and is used to compare current iB^againsMfor the detection and adjust ment of errors, is called the perceptual trace. The perceptual trace consists of a complex distribution of traces from a learning situation that consists of a series of trials (Adams, 1971)• 8 Rate of learning. In this study, rate of learning will he determined by change in performance over trials, as reflected in error scores. Edges. When a skate blade is sharpened the bottom of the blade is hollowed the length of the blade, producing two 'edges', one on either side of the hollow. The skater may tilt his blade slightly to lift one of these edges off the ice. When a skater leans his body so that his weight is on the lateral side of the foot, he has lifted the in side edge of the blade off the ice and is skating on the outside edge of the blade. Conversely, if a skater leans to that the weight is transferred to the medial side of the foot, he is skating on the inside edge. During the execution of an 'edge', the skater must be on one foot skating on a controlled curve. In a test situation skaters are asked to skate consecutive inside or consecutive outside edges on alternate sides of a coincidental axis, thereby necessitating the changing of feet at the end of each completed half circle (Figure 1 ). Flats. A flat occurs when both edges of the blade are in contact with the ice at the same time. In the execution of an edge, a flat is considered a fault and 9 can be readily identified by the double track mark it leaves on the ice, as opposed to a single track mark left by an edge (Figure 2> ). Novice skater. In this study, novice skater will refer to skaters who are capable of passing at least the first two National Skating Tests, but no higher than the fourth test. Delimitations 1. The study is delimited to skaters ages 17 years and over who are not capable of passing any higher than the fourth National Skating Test. 2. Learning will be assessed over 40 trials, spread evenly over two days. 3« The study is delimited to the effects of wearing the telemetric monitoring device. Assumptions 1. This study is based upon the assumption that Adams' closed-loop theory of motor control is valid and in this situation: the task of skating forward outside edges is subject to closed-loop control. 10 Figure 2. Comparison of flat and edge tracings. 11 2. An internal reference of correctness such as the perceptual trace does exist, and development of the perceptual trace results in learning. Limitations 1. This study is limited by the sample size of twenty subjects. 2. This study is limited to novice skaters age seventeen years and over. 12 CHAPTER II REVIEW OF THE LITERATURE It is the intention of this review of literature to outline Adams' closed-loop motor control theory as a theoretical framework upon which this study is based, and to review related theoretical areas. Adams' Closed-Loop Motor Control Theory In 1971, Adams proposed a closed-loop motor control theory for self-paced motor behaviour which differed from previous attempts to explain this process in that it afforded feedback (FB) a joint learning and performance role. Until this theory was introduced, views on the role of FB in controlling motor "behaviour were clearly divided. One group of researchers, lead by James (1890) with his response chaining hypotheses, gave FB a pure performance regulation role, with no lasting effects on behaviour. As an alternative to the response chaining hypotheses, Lashley and his associates (Lashley, 1917, 1951; Lashley and Ball, 1929; Lashley and McCarthy, 1926) proposed a motor programming theory, following the concept of some 13 central internalized control plan being laid down for each movement sequence. Clearly, Lashley and his associates saw FB as a pure learning variable. It is beyond the scope of this study to discuss all the theories and studies on FB and learning. (For a complete review of FB theories, see Adams, 1968). In an attempt to overcome the inadequacies and prob lems of either a continuous response produced stimulus controlled theory (such as James, 1890), or an open-loop motor program theory (such as Lashley, 1917), Adams in troduced his theory of motor control where feedback serves two functions (Adams, 1968, 1971). In its first role, feedback provides input for the formation of an error detection mechanism called the perceptual trace, the function being a referential memory system against which future responding can be evaluated. This referential memory system is made up of past movement consequences. It is not a single trace as its name implies, but rather a complex distribution of traces in a learning situation that has a series of trials (Adams, 1971)- In a discussion of Adams' theory, Marteniuk (1976) describes the perceptual trace as , ... the image of environmental and response produced stimuli, a large part of which is kinasthetic in origin, which acts as an indi vidual's reference level when determining the appropriateness of a response. Evidence for a centrally represented image formed "by feedback that influences a motor response has been found in studies by Greenwald and Albert (1968), and Zeloznik and Spring (1976), who in similar experiments found that passive subjects could learn a given response by watching active subjects performing the response. The results from these studies suggest that subjects form an image of the required response from the visual cues of the active subjects. Studies by Adams and Goetz (1973) and Marshall (1972) show that this image is used after a response to detect error, fitting a closed-loop schema. Once this error detection mechanism is formed, the second function of feedback is to provide input to the above mechanism for comparison and evaluation so that errors can be detected. Within his closed loop framework, Adams (1971) proposes two memory states that are individually respon sible for the initiation of an appropriate response when a stimulus occurs, and for the regulation and evaluation of the response in terms of correctness. The concept of 15 two memory states was derived from verbal learning research where differences between recognition and recall states have been demonstrated (Luh, 1922; Bahrick, 1965; Postman, J.enkins and Postman, 1948; Kintch, 1968). Evidence of the existence of separate recall and recognition states in motor memory can be found in Adams and Goetz (1973), Christina and Merriman (1977), Marshall (1972), Newell (1974), Newell and Chew (1974), Schmidt and White (1972), and Schmidt and Wrisberg (1973a, b). The medium of motor recall in Adams' theory is the memory trace which is responsible for the selection and initia tion of the appropriate response to a given stimulus. The memory trace is a limited, open-loop motor program which determines the initial direction of a response. The recognition medium of a movement in Adams' theory is the perceptual trace and, as previously discussed, operates in a closed-loop manner as the reference mechanism against which current feedback is compared for response evaluation. The strength of the recognition mechanism, henceforth called the PT is a function of the amount of relevant feedback stimuli and the amount of exposure to them (Adams, Goetz, and Marshall;. 1972; Adams, Gopher and Lintern, 1977; Marshall, 1972; Newell, 1974; Wallace, De Oreo and Roberts, 1976; Zelaznik and Spring, 1976). On any given trial, feedback stimuli from an ongoing response are compared 16 against perceptual traces from previous responses. Thus, "both stimuli, which persist from past trials through learn ing, as well as momentary stimuli from a current trial, determine behaviour giving feedback a dual learning and performance role. Before the PT can act as the primary response control mechanism in a motor skill situation, it must be developed. In early stages of learning where the perceptual trace is weak and ill-defined, knowledge of results (KR) is used in addition to feedback to direct subjects towards the cri terion response to be learned (Adams, 1971). Bilodeau and Bilodeau (1961) consider KR to be the strongest variable involved in learning and performance. A well established principle in motor learning is that KR is necessary for learning, hence improvement in performance, to occur (Adams, 1971). As early as 1927, Thorndike, using a line drawing task, found no improvement in performance scores for subjects who were not given KR. Trowbridge and Cason (1932) asked blindfolded subjects to draw a three inch line., and found that in situations where no augmented KR was given, subjects failed to im prove . Similar findings have been reported by Baker and Young (I960), McGuigan, Hutchens, Eason and Reynolds (1964), Elwell and Grindley,(1938). 17 Bilodeau and Bilodeau (1958) used a lever positioning task to show that withdrawing KR in early trials of learning led to a substantial decrement in performance. However, after sufficient practice, performance was maintained on KR withdrawal trials. Newell (197^) indicated the need for KR in early practice trials, using a linear, ballistic displacement task. Adams, Goetz and Marshall (1972) conducted a self-paced linear positioning task experiment, varying amounts of KR and practice, and found that skill acquisition was best in situations where KR was highly augmented. A second principle of KR is that the rate of improve ment depends on the preciseness of KR (Adams, 1971; Magill, 1980). Trowbridge and Cason (1932) varied the amount and precision of KR in a line drawing task and found improve ment was best in conditions of more precise KR. A study by Smoll (1972) using a bowling skill showed better results were achieved by groups receiving more precise KR. Howe-ver, the amount of time needed to make use of KR information has been found to be dependent on the preciseness of the KR. Too precise KR can actually be detrimental to learn ing (Rogers, 1974) unless sufficient time is allotted for information processing. A third principle of KR concerns the frequency of its occurrence. Findings in this area show the more frequently 18 KR is given, the better learning and performance become. (Bilodeau and Bilodeau, 1958; Magill, 1980; Marteniuk, 1976). The fourth principle of KR is of particular importance to this study and regards the timing of KR. In a learning situation there are two time periods which are of concern. The first period is the time between performance and KR and is called the KR delay interval. Most studies in this area have been conducted using discrete motor tasks and have found that KR delays of up to one hour do not affect learn ing. However, where performance involves repetitions of the movement, learning is adversely affected if a repetition occurs between the original movement and delivery of its related KR (I. Bilodeau, 1956; Lavery and Suddon, 1962). Thus, the timing of KR in regards to the KR delay interval seems of particular importance to continuous motor skills where acquisition is adversely affected by KR delay. How ever, there are few studies documenting this effect, since the bulk of studies dealing with KR delay have used discrete motor tasks. The second time period of importance is called the post KR delay interval and refers to the time which elapses between KR and the start of the next performance. The 19 findings from studies on post-KR delay indicate that there must he sufficient time for information processing to occur. The time needed for information processing will vary with the preciseness of KR, and the stage of learning of the performer. In summary, KR as information to direct error correc tion is necessary for learning to occur in early stages of skill acquisition (Elwell and Grindley, 1938; Bilodeau and Bilodeau, 1961; Newell 197^; Magill, 1980; Stelmach, 1970; Baker and Young,I960; McGuigan, Hutchens, Eason and Reynolds, 1964; Thorndike, 1927). There is no tendancy for respond ing to move towards a precise criterion when KR is absent. KR helps to ensure the proper development of a representative model of criterion responding. This is achieved through matching sensory FB from a performance with KR. This pro cess enhances the acquisition of a skill by facilitating the learning process, when used over a relatively large number of KR trials According to Adams' theory, subjects use FB in relation to KR to adjust subsequent responding towards a criterion performance. Sufficient practice with KR allows feedback to lay down a strong PT that will act as a reference of the criterion when KR is no longer available. 20 In situations where KR is not forthcoming, the PT "becomes the dominant response control mechanism, acting as an in ternal, subjective form of KR. The shift in response control from KR to PT after sufficient criterion level practice in Adams' theory is defined in terms of two stages. The first, or the verbal motor, stage is characterized by a weakly defined PT due to lack of exposure and practice with appropriate and rel evant stimuli. The subject is forced to use feedback cues in relation to KR from outside sources in order to form verbal strategies pursuant to future performances. During the verbal motor stage, response accuracy deteriorates if KR is withheld (Adams, Goetz and Marshall, 1972; Bilodeau, Bilodeau, and Schumsky, 1959; Boulter, 1964; Newell, 1974; Bilodeau and Bilodeau, 1958).because of the weak perceptual trace and its susceptibility to forgetting or progressive ambiguity in the absence of verbal intervention. In the motor stage, a strong PT has been formed over a relatively large number of KR trials and takes over as the dominant control mechanism of motor performance. Subjects can rely on matching current feedback to the PT for error detection without external directives. In the motor stage, response accuracy will not deteriorate if KR is withdrawn. This 21 effect has been shown by Bilodeau and Bilodeau (1958), Newell (1974) and Bilodeau, Bilodeau and Schumsky (1959). In summary, the key variables for strengthening the PT are the amount of relevant feedback stimuli (quantity) and the amount of practice (frequency) with them. The learner must perceive the relevant stimuli from amongst the many irrelevant stimuli which bombard him from the environment before strengthening of a representative PT can occur. Beginners experience difficulty in learning a new skill due to their Inability to determine for them selves what their errors are and how to correct them. In early stages of skill acquisition, the learner uses FB in relation to KR to attend to relevant stimuli, deter mine errors and develop an internal reference of correct ness. The process by which this identification and selec tion of pertinent information eventually occurs is called selective attention. Selective Attention Implications In a skill situation relevant cues from the environ ment must be attended to before information processing can proceed. Gentile (1972) refers to these relevant cues as the regulatory stimulus subset, as opposed to non-regula tory information within the learning environment. These 22 cues are referred to as regulatory since the movement pattern must conform to them if the predetermined goal is to be accomplished. The process whereby an individual attends to the regulatory stimulus subset at the exclusion of other environmental information is called selective attention with perception being the end result. Selective attention is necessary due to the limited information pro cessing capacity of our central nervous system. Allocation of Attention Early studies on selective attention (Broadbent, 1958; Moray, 1959; Cherry, 1953) show consistently that an unattended signal has little chance of reaching memory or conscious awareness. Moray (1959) presented two dif ferent messages simultaneously (one to the right ear and one to the. left) to subjects, asking them to repeat the message of one ear. The results from this study indicate that there was no memory for the unattended message. Similarly, Cherry (1953) presented two simultaneous mes sages to subjects.asking them to pay attention to the message in one ear by repeating what they heard. At the end of the messages subjects were asked questions about what they heard on the unattended ear. Subjects noticed only very general characteristics of the messages such as 23 changes in language spoken, voice, etc., with almost no recall of its verbal content. Mowbray (1953) showed that even when subjects were asked to attend to two messages, they were unable to divide their attention between them. Consistently, recall on one of the two messages was poor. In a perceptual motor skill situation the performer is bombarded with simultaneously presented information from a variety of sources. Due to the limits imposed by selective attention, an inexperienced motor performer finds himself severely handicapped in his ability to process relevant information and, therefore, perform a specific skill. Studies by Cherry (1953) and Mowbray (1953) show dramatic limitations in our abilities to deal with simul taneously presented information. Results of these studies show that when two messages arrive simultaneously and only one is attended to, the unattended message is poorly analyzed with almost no specific characteristics available for recall. In a complex motor skill, a novice performer must deal with information from his environment, plus many appropriate feedback cues from his response. Results from selective attention studies suggest that the novice performer may be unable to attend to more than one source of information and thus the many mistakes characteristic of a novice performance .'.may., be the result of perceptual limita tions imposed by poor utilization of selective attention 24 (Marteniuk, 1976). Further, the fact that a performer can detect, recognize and compare information does not guarantee that he will attend to the information that is relevant for a successful performance. The performer must learn from past experience and/or external information sources to attend to and select the relevant cues while disregarding or attenuating the others. Not yet capable of distinguishing what is relevant from what is not, the early learner begins, (Adams' 'verbal-motor' stage) and he is forced to rely on KR from outside sources to draw his attention to the rele vant feedback cues. In turn, the individual uses the in formation gained from KR, in relation to FB cues, to eval uate his performance in order to modify future attempts. If the performer is unable to retain feedback information from his previous attempt, evaluation and subsequent mod ification of performance is not possible. Thus, we are concerned with the performer's ability to retain informa tion over short periods of time so that the information can be used to evaluate and modify his motor performance, when used in relation to KR. Studies by Bilodeau (1956") and Lavery and Suddon (1962) have shown that in motor tasks where repetitions of the movement take place between the original movement and its KR, learning is adversely affected. This effect 25 is presumably due to the limits imposed by the capacity and duration of short term memory (STM) and subsequent information processing. In the case of skating edges, where a skater is holding information in memory about a previously skated edge, his performance is continuing, adding more and more information to be retained in STM for post-performance analyzing and processing. Short term memory is a storage system that is limited in its operation, capacity and duration. Fitts and Posner (1967) define it as a system which loses information rapidly in the absence of sustained attention. A classic experi ment by Sperling (I960) demonstrated that much more in formation is accessible to STM that can be recalled. He demonstrated that while subjects were identifying items in memory, other items were being lost. For items where attention had not been focussed, recall was very poor. Peterson and Peterson (1959) and Broadbent (195^), in similar experiments, showed that forgetting in STM is due to lack of attention. Thus, if an item is not attended to in STM it will fade within a very short period of time so that it is no longer available for recall. If an early learner doesn't attend to the proper feedback cues, he will find it difficult to use KR in comparison to them in his attempt to create the proper PT. 26 Some Considerations The limits imposed by selective attention and STM severely handicap the early learner in the development of the PT which is necessary for the eventual progression to the motor stage of skill acquisition. In a situation where augmented error information could be generated to draw attention to the appropriate and relevant cues, one would expect a more positive performance and an accelerated learning curve. In an attempt to show this affect, Goldstein and Rittenhouse (195^) produced disappointing results. Using a target shooting task, subjects in the experimental group received a buzzer when their guns were on target. As would be expected, the performance of the experimental group was much better than the control group, who received no error correction. However, performance dropped off considerably when the buzzer produced KR was removed. This suggested that the buzzer had acted as a performance regulating factor rather than as a learning aid. The major drawback of this experiment was that the researchers did not provide subjects with true error in formation. Rather than receiving critical error informa tion, the subjects had only to wait to hear the buzzer and fire. In"early stages of learning, performers do not try to repeat the previous behaviour, but instead they attempt to modify their behaviour to make it more correct. 27 When learners make errors early in learning, and KR values are large, they are not responding on the basis of move ments that they recognize as having made before, because this would cause them to repeat past errors. For learning to occur, performers must use KR to make the next response different from the previous one; he must use the perceptual trace in relation to KR from an outside source (the cri terion reference standard) and adjust the response accord ingly on the next trial (Adams, 1971)- The information provided in the Goldstein and Rittenhouse experiment let performers repeat rather than correct past performances. The subjects became reliant on a cue which was not a part of the desired task and that was only relevant to the on going performance. It seems logical that had the buzzer sounded when subjects were in error, the subjects would become more sensitive to the right response in order to eliminate the sound of the buzzer, i.e., error. In a similar study by Annett (1959). utilizing a target pressure task, augmented, concurrent FB resulted in much better performance scores, but when withdrawn, performance de teriorated rapidly and dramatically indicating no learning effect. It is hypothesized that had these studies employed the augmented error information in such a way as to allow 28 subjects to modify previous responses, i.e., learning FB versus performance or action FB, the withdrawal trials would not have produced the performance decrements observed. The following experiment examined the effects of increased availability of FB cues through the use of im mediate, concurrent, augmented KR on the acquisition of the continuous gross motor skill of skating forward out side edges. 29 CHAPTER III METHODS AND PROCEDURES The continuous, self-paced motor skill used in this study was that of skating consecutive forward outside edges along a coincidental axis. Subjects Twenty female volunteer novice skaters, ages 17 and over, were recruited from the University of British Columbia's student and staff population. Subjects were randomly divided into two groups of ten per groups Apparatus A telemetric monitoring device was designed to monitor a skater's blade position and activity on the ice, in terms of blade/ice angle, during the attempted execution of forward outside edges along a coincidental axis. Error information based on the measurement of the skater's performance was provided to the skater during 30 performances in the form of a non-irritating, electronic "beeping noise . The apparatus consisted of an outrigger styled, ice angle sensor which was fastened to the skater's "blade prior to performance and, via a small "battery pack attached to the skater's "belt, transmitted signals both to an earphone worn by the skater and to a telemetry re ceiver with its associated recording equipment (see Appendix A for a detailed description). In particular, an audible tone was generated through the earphone whenever the blade-ice angle exceeded a predetermined value. This predetermined value was the angle at which both edges of the blade contacted the ice at the same time resulting in a flat, or beyond this point where only the inside edge of the blade was in contact with the ice. The error tone generated through the earphone persisted until the blade-ice angle was corrected and brought back into the critical range of the desired movement, thus producing the execution of an acceptable outside edge. The ice-angle sensor was constructed of aluminum for its durable and light weight characteristics, and was of an outrigger design that clipped onto the top of the skate blades, where it would not impede performance, and would run along the ice surface parallel to the control blade. 31 During each trial the telemetry signal was trans mitted to a data collection station where blade activity in terms of edge/flat times was recorded on a strip chart for purposes of the monitoring and future analysis of per formance (see Figure 3 )• The total system consisted of a pair of sensors to convert the blade-ice angle to an electrical signal, a detector to analyze the angle signal, an earphone, and assorted standard devices to transmit and record the detector output (see AppendixAA ) . Experimental Design The experiment consisted of two phases. In the first phase, subjects were required to skate twenty trials on each of two days, for a total of forty trials. Each trial consisted of six consecutive forward outside edges, per formed across the width of the ice surface (Figure 4). An average experimental session from phase one of this experiment required approximately 25 minutes to complete. In the second phase of the experiment, which occurred on the second day of testing after trials 21 - 40 were completed, subjects were required to perform an additional 10 trials while wearing, but without receipt of any in-1. ia lb 1. A sample strip chart produced by the performance of forward outside edges by a novice skater. Points a and b indicate a change of skating foot. 2. A sample strip chart produced by .the performance of forward outside edges by a skilled skater. Points a and b indicate a change of skating foot. Figure 3- Strip chart recordings for edge/flat times. 33 formation from, the telemetric monitoring device or the instructor. An experimental session from phase two re quired approximately 35 minutes to perform. Each subject's total trial time varied as did his time per arc. The error information recorded on the strip chart permitted the precise measurement of each attempt, thus each trial. From this chart recording all calculations were made possible and the percentage trial time spent on edge was recorded as the dependent measure of this study. The data was analyzed as a 2 X 2 X 20 randomized groups design using analysis of variance, with repeated measures on the last two factors, e.g., 2 groups, 2 days, 20 trials per day. In addition, the last five trials of phase two of the experiment were analyzed with the last five trials of phase one in a 2 X 2 X 5 randomized.groups design with repeated measures on the last factor for the pur pose of detecting any changes in performance when the monitoring device was removed. Procedure s The experimental group in this study wore the tele-metric monitoring device which provided Ss with instan-3^ taneous error information every time response execution was imperfect, i.e., they went 'off-edge'. In addition, at the end of each trial each subject received verbal "KR about the success of their performance, as well as teacher generated instructional information on how to improve the next performance. The instructor for the experiment was unaware of the group (experimental or control) to which the subject belonged. This experiment was conducted in this blind manner in order to control experimenter bias. The control group also wore the telemetric monitoring device for the performance recording purposes but did not receive the error generated tone through the earphone from it. Prior to the start of an experimental session the angle sensors were attached to the skate blades and the critical angles were set by manually manipulating the skates on a flat surface and adjusting the angle threshold con trols. The Ss then put their skates on and the critical angles were rechecked. Prior to each session each skater was given the same verbal instructions which were read to them by the experi menter. The skater's task was to learn to skate six consec utive forward outside edges (three right, three left) along a coincidental axis as cleanly as possible (see Figure 4). 35 i.' Figure 4 . Skaters were asked to skate consecutive forward outside edges along a coincidental axis. 36 After receiving their instructions, the skaters observed a demonstration of a sample trial and then at tempted the appointed task for themselves. Subjects were instructed to commence the next trial immediately after receiving information from the instructor regarding the previous performance, i.e., an effort was made to control the duration of the intertrial interval. . Only one subject was present during an experimental session to prevent Ss from overhearing FB being given to other Ss. All subjects were given the same directions prior to the experiment (see Appendix B). 37 CHAPTER IV RESULTS AND DISCUSSION Results From Experiment 1 Results from Experiment 1, as shown in Table 1, clearly indicate performance of the experimental group was better than that of the control group, as indicated by higher percentage pn-edge scores. A 2 X 2 X 20 analysis of variance was employed and confirmed a signficant groups effect (p<.001) (Table 2). While the mean performance over both days for the experimental group was approximately 77%, it was only 39% for the control group, averaged over the same period of time. In fact the experimental group means for each of the performance days greatly exceeded those of the control group. The ANOVA further confirmed a significant trials effect showing general improvement of performance scores over time (Figure :5)• The nature of the trials effect within days was different between Day 1 and Day 2, and this difference was not constant between groups, resulting in a significant Groups X Days X Trials interaction effect 38 Table 1 Mean Percentage On-Edge Time for Groups on Day 1 and Day 2 in Experiment 1. Group 1 Group 2 (Control) (Experimental) x=70.09fo x=39.89f* x=8k.37fo X=39.02% X=77-23fo Table 2 ANOVA Table for Experiment One Source Sum of Degrees of Mean &,a\t F p< squares freedom square G Error 357299.85 173843.12 1 181 357299.85 9657-95 37 -00 0.0001 D D X G Error 3107 .08 979.92 6797^.77 1 1 18 3107.08 979.92 3776.38 0 .82 0 .26 0.3764 0.6167 T T X G Error 20889-29 198<7? 75 67150.42 19 19 342 1099.44 104.62 196.35 5 .60 0-53 0.0001 0.9472 D X T D X T X G Error 4622.30 6125.82 55916.48 19 19 342 243.28 322.41 163.50 1.49 1 -97 0.0867 0.0094 100 * * Experimental Control • • 90 J 80 J * * * * * * * * * * * * * * j# # * * # * * * c+ H.7o CD §60 CD ^ cn CD 50 40 30 20 10 Day 1 8 10 i2 1'4 1'6 18 Trials 2b 22 Day 2 IE 26 28 3'o 3'2 3^ 36 IB *Jo Figure 5- Significant trials effect showing general improvement of performance scores over time. 41 (p<.01). This interaction effect was due primarily to a G X D X T linear component. On Day 1 the performance of the experimental group improved by approximately 26%> from trial 1 to trial 20, while the control groups' performance only improved by approximately 12%. A linear regression analysis revealed the slope of the experimental group's improvement scores on Day 1 to be 1.3, as compared to the control group's slope of only .63. By Day 2, the experi mental group's performance had stabilized (slope =.12), while the- control group showed erratic but steady improve ment (slope =1.09) (Figure 6) . Discussion of Results from Experiment 1 The purpose of Experiment 1 was to determine the validity of hypothesis one which stated subjects having increased accessibility to relevant FB cues through im mediate, concurrent KR would show a greater rate of skill acquisition as reflected in error scores. The mean per formance score of the control group for each of the trials at no time matched or even approached the mean performance of the experimental group.. Initial inspection of the data caused concern over the large difference (approximately 26%) which existed between the experimental group and the control group 100 A * * Experimental Control 90 J 80 CD §60 CD CD 50 * * * *^¥T* t^oj-ope 1.3 * * * * # * I" * * * * * * » Slope = .12 40 30 20 -10 Slope = .63 Day 1 Slope = 1.09 Day 2 1 Jl I FT<5 12 1% l'6 18 3)2l 31 26 28 3»0 3'2 34 36 3&" <Jo Trials Figure 6. Slopes of learning curves for experiment one. 43 after the first trial and the effectiveness of the random ization procedure for assignment of subjects to groups was questioned. In light of the fact that one trial con sisted of 6 consecutive edges, it was conceivable that much information was lost about the performance of the in itial attempts at skating an edge. When the information recorded on the strip chart (blade activity in terms of on-edge/off-edge times per trial) for the first trial was broken down into segments of one quarter and the on-edge time measured for each segment the resulting analysis saw the initially reported difference between the two groups reduced to approximately 7% after the first quarter (Fig ure 7)• In addition, the experimental group demonstrated a very steady improvement rate, while the control group displayed erratic changes that were not always positive. Due to the considerable improvement in performance scores demonstrated by the experimental group throughout trial one, it is certainly reasonable to assume that the large difference observed between the two groups after trial 1 was in fact attributable to considerable learning already having taken place throughout trial 1 by the experimental group. Clearly, the experimental groupss performance was significantly better than that of the control group, as reflected in error scores; the result of a faster rate of * * Control ' ' Experimental 100 . 90 . 20 10 . 1 2 3 4 Figure £. Quarter segments of trial 1, more accurately presenting •lnt^rgroup differences at the commencement of the experiment. 4$ learning. This finding is in keeping with Adams closed-loop motor control theory, and with the findings of Thorndike (1927), Bilodeau., Bilodeau and Schumsky (1959), Bilodeau (1956), Goldstein and Rittenhouse (195^), and Annett (1959), where KR, used in relation to FB, produced accelerated learning. While the analysis showed no significant Days, Days X Groups, or Days X Trials effects, these effects must be interpreted in light of the Days X Groups X Trials effect, where the nature of the trials effect within days was dif ferent between Day 1 and Day 2 and this difference was not constant between groups. Results from Experiment 2 Experiment 2 was performed after all trials from experiment 1 had been completed by the performer. Ex periment 2 consisted of ten trials on Day 2 performed without the access to the error information produced by the device. Results from the last five trials of Exper iment 1 and the last five trials of Experiment 2 were analyzed in a 2 X 2 X 5 ANOVA to investigate performance maintenance. 46 Analysis of the results failed to reveal significant Trials, Groups X Trials, Conditions X Trials, or Groups X Trials X Conditions effects (Table 3)• Thus there was no change in performance over the five trials of either con dition. Further, the data analyzed over groups fails to show any significant change in performance from condition 1 to condition 2, and whatever small changes did occur, both groups changed in the same way in their mean performance over the two conditions (Figures 8 and 9)• Results from the analysis indicate a significant Groups effect, with the mean for the experimental group being approximately 87% on edge, and the mean for the control group being approximately 48% on edge (Table 4). Discussion of Results from Experiment Two The purpose of Experiment 2 was to determine the validity of the hypothesis that artificially enhancing error information will::mot inhibit progress to the motor stage of skill acquisition, as reflected by performance maintenance when KR and the added FB are withdrawn. The absence of significant Groups X Trials, or Groups X Condi tions effects demonstrates performance maintainance by both groups and can be interpreted as demonstrating suc cessful progression of the experimental group from the Table 3 ANOVA Table for Experiment Two Source Sum of Degrees of Mean F p<£ squares freedom square y..;:e V -^i G 76061.70 1 76061.70 19.06 0.0004 Error 71849.21 18 3991.62 Cond 258.78 1 258.78 0.54 0.4710 CG 3.15 1 3-15 0.01 0.9361 Error 8588.98 18 477 .17 T 499.30 4 124.83 1.29 0.2824 TG 949.16 4 237-29 2.45 0.0537 Error 6972.41 72 96.84 CT 722.50 4 180.63 2.45 0.0535 CTG 123.49 4 30.87 0.42 0.7942 Error 5301.13 72 73-63 48 Table 4 Means for Experiment 2, Showing Performance Stability by both Groups from Experiment 1 to Experiment 2. Group 1 Group 2 Condition 1 Condition 2 x=46.51% x=85-76% x=49.03% x=87.?8% X=47.77% X=86.77% 49 100 -• 90 -80 -70 -% 60 -Time on 50 -Edge 40 -30 -20 -10 -Experimental Control Condition 1 I Condition 2 Figure 8. Average performance score by groups and conditions 50 loo •90 -I 80 -I Experimental 70 J % 60 -I Time on Edge 50 -I Control ' 40 -I 30 -I 20 4 10 12 3 4 5 Last 5 trials of Experiment 1 5 Last 5 trials of Experiment 2 Figure 9. Sesuljs f-roriioExpe rime nt 2. Comparison of the performance of both groups over the last five trials of Experiment 1 and the last five trials of Experiment 2. 51 verbal motor stage to the motor stage of skill acquisition, where withdrawal of KR does not affect performance. The findings from Experiment 2 are in keeping with studies by Newell (1974), Bilodeau, Bilodeau and Schumsky (1959) and Bilodeau and Bilodeau (1958). These findings are also in agreement with Adams' closed-loop theory. Presumably, a strong representative PT has been formed by the experimental group over the relatively large number of KR trials and thus became a reliable performance control mechanism when KR was withdrawn. Summary of Results Analysis of the results obtained from Experiments one and two are in agreement with closed-loop motor control theory for self-paced motor tasks. Fundamental to Adams' closed-loop motor control theory is the development of a strong and representative PT. Consistent and appropriate responding without the aid of KR infers the existence of a strong and representative PT. The development of a PT is said to be affected by the amount and practice with relevant feedback cues. Thus, exposure to greater amounts of relevant FB cues permits accelerated development of a strong PT, resulting in a faster rate of skill acquisition. A faster rate of skill acquisition is reflected in a 52 steeper, positive learning curve which plateaus with high and stable performance scores. This investigation provides support for the effects of additional amounts of relevant FB cues on the development of a representative PT. The existence of a strong PT is inferred by accelerated learning curves by the experimental group in experiment one. Maintenance of consistently high performance scores by the experimental group in experiment two can be inter preted as performers functioning in the motor stage of Adams' closed-loop control theory, where control shifts from peripheral to central and KR withdrawal does not affect performance. In this stage, performers rely on a strong perceptual trace as their performance regulation device. 53. CHAPTER V SUMMARY AND CONCLUSIONS The main objective of this investigation was to study the effects of enhancing the accessibility of re levant FB information on the rate of skill acquisition of the continuous self-paced motor skill of skating forward outside edges. Based on the theoretical framework of Adams' closed-loop motor control theory experimentation was conducted in accordance with the hypothesis that sub jects having increased accessibility to relevant performance scores (via FB) in a motor skill situation would show a fast er rate of skill acquisition. This experiment involved 20 volunteer novice skaters, performing the self-paced motor skill of skating consecu tive forward outside edges along a coincidental axis. Skaters were required to skate twenty trials of six con secutive outside edges on each of two days (6 X 20 X 2 = 240). IThe experimental group received error information instantaneously from a specially designed telemetric mon itoring device (worn by both groups) every time their per-, formance varied from the criterion reference, i.e., they 5^ went 'off edge'. Skaters of both groups received appro priate KR from an instructor, who was teaching/coaching/ evaluating in a double blind condition. There was initial concern that the artificial.Ven-hancement of the accessibility of relevant information by use of a FB device might cause performers to become de pendent on the device, rather than on FB cues inherent to the task, i.e., a regulating rather than a learning effect. Such an effect would not allow a representative PT to be established and, therefore, hinder progression from the verbal motor to the motor stage of skill acquisition. Thus, a subproblem related to this study examined the ef fects of withdrawal of information provided by such a FB device. To investigate this subproblem, a second experi ment in this study was performed in accordance with the hypothesis that artificially enhancing error information would not inhibit progress to the motor stage of skill acquisition, as reflected in the maintenance of performance levels when KR and the information from the FB device were withdrawn. In this experiment, subjects were required to perform an additional ten trials of six consecutive out side edges (10 X 6 = 60) on day two, where the augmented FB and KR were withheld. Performance was analyzed over the last five of these ten trials, and compared with the last five trials of experiment one. 55 Conclusions The conclusions formulated from this investigation are as follows: 1. The provision of concurrent, response generated, relevant FB information results in a faster rate of skill acquisition for the task of skating consecutive forward outside edges. This conclusion is in keeping with closed-loop motor control theory for self-paced motor tasks. 2. Artificially enhancing error information through the use of a telemetric monitoring device does not inhibit progress to the motor stage of skill acquisition. Discussion The results and findings of this study are of particular interest to the figure skating coach. Expedient acquisition of sound manoevers and technique are necessary in order that progression to more advanced work can proceed. The device designed for this study has allowed acquisition of the fundamental skill of skating forward outside edges in sig nificantly less time than is normally experienced. In its present form, the device used in this study would not be practical as a learning aid in a beginner's 56 figure skating program. The time required to fasten the equipment to the skater and set the electronic controls defeats its practicality. Likewise, the presence of bulky wires, although taped out of the way in this experiment, are too cumbersome for the young, beginning figure skater. Sophistication of the device to a point where its operation was completely telemetric, eliminating the bulky wires, and reducing set-up time would make it a valuable aid for the novice learning and mastering compulsory figures. The rewards reaped from use of such a device could be substan tial. Although impossible to put an accurate time value on the acquisition of clean, controlled edges, it is safe to say that under normal learning conditions, we are deal ing with at least several months of practice time to de velop a 'feel' for the difference between flats and edges. This experiment produced this 'feel' after only one and one half hours of practice for most experimental subjects. A further application of this device would be for skaters who are experiencing difficulty in the performance of a compulsory figure due to flats. Error information can be gained from the device without an instructor having to be present. This saves valuable time for both the skater and coach, as well as many dollars in lesson fees. Ap plication of the device in this manner extends its use beyond the beginning skater to those learning new figures 57 as a result of progression to more advanced work and those experiencing difficulty with particular aspects of more advanced levels of the figure eight in regards to edge control, e.g., flats to center or changes of edge during an inside push-off. Rapid acquisition of controlled edges (as can be produced with this device) would have tremendous beneficial effects on all aspects of figure skating (freeskate, dance and compulsory figures) as the curving edge is the basis of virtually all skills in figure skating. Suggestions for Further Research The literature available for testing closed-loop learning for motor performance deals only with performance of discrete motor tasks. There is no literature available testing the theoretical concerns of closed-loop learning for continuous self-paced motor skills. Research in this area would help to extend the theoretical base of our knowledge of skill learning and be of value to practition ers dealing with instructional methods for continuous motor tasks. Further, it is suggested that research the present study, but involving more subjects, groups and 58 varying KR and FB enhancement withdrawal periods "be employed to determine where progress to the motor stage from the verbal-motor stage occurs during the accelerated rate of learning. It is further suggested that the device used in this experiment might prove useful in remedial skill learning situations, and further experimentation in the remedial area be conducted. Further experimentation should also occur using different age groups to determine the applicability and practicality of the device in dif ferent learning situations based on age. 59 BIBLIOGRAPHY Adams, J.A. Response feedback and learning. Psychologi cal Bulletin, 1968, 70, 486-504. Adams, J.A. A closed-loop theory of motor learning. Journal of Motor Behavior, 1971, 3, 111-149. Adams, J.A., & Goetz, E.T. Feedback and practice as variables in error detection and correction. Journal of Motor Behavior, 1973, 5. 217-224. Adams, J.A., Goetz, E.T., & Marshall, P.M. Response feedback and motor learning. Journal of Experimental  Psychology, 1972, 92, 391-397-Annett, J. Learning a pressure under conditions of im mediate and delayed knowledge of results. Quarterly  Journal of Experimental Psychology, 1959, 11, 3-15' Bahrick, H.P. The ebb of retention. Psychological Re view, 1965, 72, 60-73-Baker, C.H., & Young, P. Feedback during training and retention of motor skills. Canadian Journal of  Psychology, I960, 14, 257-264T" Bilodeau, E.A., & Bilodeau, I.M. "Variable frequency of knowledge of results and the learning of a simple skill. Journal of Experimental Psychology, 1958, 55, 379-383^ CaT~ Bilodeau, E.A., Bilodeau, I.M., & Schumsky, D.A. Some effects of introducing and withdrawing knowledge of results early and late in practice. Journal of  Experimental Psychology, 1959. 58, 142-144. Boulter, L.R. Evaluation of mechanisms in delay of knowledge of results. Canadian Journal of Psych ology, 1964, 18, 281-291. Broadbent, D.E. Perception and Communication. London: Pergamon Press, 1958. Cherry, E.C. Some experiments on the recognition of speech with one and two ears. Journal of the  Acoustical Society of America, 1953, 25, 975-979-60 Christina, R.A., & Merriman, W.J. Learning the direc tion and extent of a movement: A test of Adams' closed-loop theory. Journal of Motor Behavior, 1977, 9, 1-9-Deutsch, J.A., & Deutsch, D. Attention: Some theoretical considerations. Psychological Review, 1963, 70, 80-90. Elwell, J.L., & Grindley, G.C. The effects of knowledge of results on learning and performance. British  Journal of Psychology, 1938, 29, 39-54. Fitts, P.M., & Posner, M.J. Human performance: Basic concepts in psychology series. Belmont, Ca.: Brooks/ Cole, 1967. Gentile, A.M. A working model of skill acquisition with application to teaching. Quest XVIII, January 1972, 3-23-Gentile, A.M., & Nacson, J. Organizational processes in motor control. Exercise and Sport Reviews, 1976, 4, 1-29-Goldstein, M.P., & Rittenhouse, C.H. Knowledge of re sults in the acquisition and transfer of a gunnery skill. Journal of Experimental Psychology, 1954, 48, 187-196"; Greenwald, A.G., & Albert, S.M. Observational learning: A technique for elucidating S-R mediating processes. Journal of Experimental Psychology, 1968, 76, 267-272. Holding, D.H. Principles of Training. London: Pergamon Press, 1965. James, W. Principles of Psychology (Vol. 1). New York: Holt, 1890. (a) Kintsch, W. Recognition and free recall of organized • lists. Journal of Experimental Psychology, 1968, . 78, 481-4^ Lashley, K.S. The accuracy of movement in the absence of excitation from the moving organ. American Journal  of Physiology, 1917, 43, 169-194. 61 Lashley, K.S. The problem of serial order in behavior. In L.A. Jeffress (Ed.), Cerebral mechanisms in  behavior. New York: Wiley, 1951, 112-136. Lashley, K.S., & Ball, J. Spinal conduction and kinaes-thetic sensitivity in the maze habit. Journal of  Comparative Psychology, 1929, 9, 71-156. Legge, D., & Barber, P. Information and Skill. Great Britain: Methuen, 1976. Luh„ C.W. The conditions of retention. Psychological  Monographs, 1922, 142, 31. McQuigan, F.J., Hutchens, C., Eason, N., & Reynolds, T. The retroactive interference of motor activity with knowledge of results. Journal of General Psychology. 1964, 70, 279-281. Magill, R.A. Concepts and applications. Dubuque, Iowa: W.C. Brown, 1980. Marshall, P.H. Recognition and recall in short-term motor memory. Journal of Experimental Psychology, 1972, 95, 147-153-Marteniuk, R.G. Information processing in motor skills. Toronto: Holt, Rinehart & Winston, 1976. Moray, N. Attention in Dichotic listening: Affective cues and the influence of instructions. Quarterly  Journal of Experimental Psychology, 1959, 11, 56-60. Mowbray, G.H. Simultaneous vision and audition: The comprehension of prose passages with varying levels of difficulty. Journal of Experimental Psychology, 1953, 46, 365-372. Newell, K.M. Knowledge of results and motor learning. Exercise and Sport^Sciences Reviews, 1976, 4, 196-228~. Newell, K.M., & Chew, R.A. :Recall and recognition in motor learning. Journal of Motor Behavior, 1974, 6, 245-253. Norman, D.A. Memory and attention. New York: John Wiley and Sons, 1969 • Norman, D.A. Memory and attention (2nd ed.) >"~-New York: John Wiley and Sons, 1976, 62 Oswald, I., Taylor, A., & Treisman, M. Discrimination responses to stimulation during human sleep. Brain, I960, 3, 440-453. Peterson, L.R., & Peterson, M.J. Short term retention of individual verbal items. Journal of Experimental  Psychology;,1959, 58, 193-198. Postman, L., Jenkins, W.O., & Postman, D.L. An exper imental comparison of active recall and recognition. American Journal of Psychology, 1948, 6l, 511-519-Reynolds, A.G., & Flagg, P.W. Cognitive psychology. Cambridge, Mass.: Winthrop, 1977-Robb, M.D. The dynamics of motor skill acquisition. New Jersey: Prentice-Hall, 1972. Schmidt, R.A. Control processes in motor skills. Exercise and Sport Sciences Reviews, 1976, 4, 229-26T! Schmidt, R.A., & White, J.L. Evidence for an error detection mechanism in motor skills: A test of Adams' closed-loop theory. Journal of Motor Behavior, 1972, 4, 143-153-Schmidt, R.A., & Wrisberg, C.A. Further tests of Adams' closed-loop theory: ...Response produced feedback and the error detection mechanism-. In I.D. Williams and L.M. Wankel (Eds.), Proceedings of the Fourth Canadian  Psycho-Motor Learning and Sport Psychology Symposium. Fitness and Amateur Sport Directorate, Department of National Health and Welfare, Ottawa, Canada, 1973-Schmidt, R.A., & Wrisberg, C.A. Further tests of Adams' closed-loop theory: Response-produced feedback and the error detection mechanism. Journal of Motor  Behavior, 1973, 5, 155-164. Singer, R.N. Motor learning and human performance (3rd ed.). New York: MacMillan, 1980. Smoll, F.L. Effects of precision of information feedback upon acquisition of a motor skill. Research Quarterly, 1972, 43, 489-493-Sperling, G. Information available in brief visual presentation. Psychological Monographs, I960, 74, 1-498. 63 Stafford, E.M. Knowledge of results and the perceptual  trace. Unpublished Master's Thesis, University of British Columbia, 1978. Stelmach, G. Learning and response consistency with augmented feedback. Ergonomics, 1970, 13, 421-425-Stelmach, G. (Ed.). Motor control: Issues and trends. New York: Academic, 1976. Stelmach, G. Learning and response consistency with augmented feedback. Ergonomics, 1970, 13, 421-425. Stelmach, G., Kelso, J., & McCullagh, P. Preselection and response biasing in short term motor memory. Memory and Cognition, 1976, 4, 62-66. Treisman, A.M. Monitoring and storage of irrelevant messages in selective attention. Journal of Verbal Learning and Verbal Behavior, 1964c, 3, 449-459-Treisman, A.M. Selective attention in man. British  Medical Bulletin, 1964d, 20, 12-16. Wallace, S.A., De Oreo, K.L., & Roberts, G.C. Memory and perceptual trace development in ballistic timing. Journal of Motor Behavior, 1976, 8, 133-137-Welford, A.T. Skilled performance: Perceptual and motor  skillsl London: Scott & Foresman, 1976 Whiting, H.T.A. Input and perceptual processes in sports skill, lln E.J. Glencross (Ed.), Psychology  and Sport. Australia: Hill, 1978. Zelaznik, H., & Spring, J. Feedback in response recog nition and production. Journal of Motor Behavior, 1976, 8, 309-312. 64 APPENDIX A Apparatus 65 Right |-Sensor comparing circuit Right Set-Point Knob earphone earphone jack radio transmitter Experimenter's Station radio waves Clock (Runs on 'no "beep' = on edge time for trial) Chart Recorder Interface Radio [_J. (Radio jack) Figure A-l. Device Architecture 66 The apparatus used consists of a pair of sensors to convert blade-ice angle to an electrical signal, a detector to analyze the angle signal, a headphone and assorted standard devices to transmit and record the de tector output. Input Generator Blade-ice angle is measured by a mechanical sensor that is attached to the skate blade. The sensor is equipped with a pair of outrigger-styled ice-tracking arms that are arranged to rotate the shaft of a potentiometer which is connected so as to output a voltage that is a monotonic function of blade-ice angle. Since the ice manouevers executed by the subject uses left and right feet alternately, one blade-ice angle sensor is fitted to each skate. The right skate sensor is equipped with a switch that determines whether the left or the right sen sor signal drives the apparatus. When the right skate blade is on the ice, the right skate sensor signal is used, when the right skate blade is lifted off the ice the left skate sensor is used. 6? Tone Generator The sensor signals are processed by a portable, adjustable threshold detector. When the active sensor signal exceeds the set point value, the detector emits an audio-frequency tone signal that simultaneously trans mitted to the subject via an earphone and to the experi menter's station via frequency modulated VHF radio tele metry link. Separate threshold adjustments exist for each skate. As a functioning unit the sensor-detector set provides signals to the subject and experimenter whenever blade-ice angle of the currently active skate exceeds the critical value. Recording Equipment Equipment at the experimenter's station consists of a telemetry receiver that drives the measuring equip ment. Current measuring equipment configuration includes a chart recorder with simple manual event marker, used for general session history recording and for measurement of session duration, and a high resolution event timer, used for measuring total time spent over the blade-ice angle threshold in each session. 68 Procedure At the start of an experimental session the angle sensors are attached to the skate blades and the critical angles are set by manually manipulating the skates on a flat surface and adjusting the threshold controls. Value of the critical angle is set before the skates are put on. 69 APPENDIX B Instructions to Subjects 70 Prior to the commencement of each subject's per formance the experimenter read the following instructions "You will be asked to perform the skating skill of forward outside edges along the red line on the ice. An outside edge consists of a one foot glide on a curve, where the foot that is off the ice will be on the outside of that curve. This will be demonstrated for you. You will be asked to perform a total of 50 trials, 20 today, and the rest on one other day within the week. Each trial consists of 6 consecutive outside edges across the ice, alternating from right to left foot, and startin on the right foot. Your starting position will be given to you, and an example of a trail will be demonstrated. You will receive instructions about how to improve your performance throughout the trials. The object of this exercise is to stay on the outside of the blade for the entire curve that you are skating (minus your pushing time). This requires that you keep your foot tilted to the outside. If you receive a tone from the earphone, this means you are not on edge. To shut the tone off, you must correct your foot position by holding your ankle 71 up over the edge. This will he demonstrated. As soon as you go "back on edge, the tone will cease. It is im portant to your final score that you try to stay on edge as much as possible. Only the experimental group of this study will receive the beep from the apparatus. The instructor is not aware of which group you are in. It is important to the results of this experiment that this information is kept from her at all times." 72 APPENDIX C Trial Scores by Subjects Table A-l Percent On-Edge T;ime by Subject for the Experimental group in Experiment 1 . Day 1 SI S2 S3 S4 S5 S6 S7 S8 S9J S10 Tl 51 .8 91 .5 55-5 72.2. 25.6 62.6 50.0 10.3 71.5 92.4 T2 47.6 74.6 54.9 89-2 24.7 76.2 94.3 31.8 63.I 93-5 T3 38.1 79.3 36.9 64.2 40 .6 69.5 92.4 38.4 78.4 47.8 T4 45-4 36.8 90.5 75-2 83.6 68.8 66.0 10.2 84.2 91 .9 T5 20.9 47.2 82.2 65-6 56.3 75-6 97-6 86.9 92.4 95.7 T6 38.6 57-9 81.6 69-0 61.0 88.5 95.8 69-9 93-5 92 .8 T7 66.7 39-8 95.8 81.2 84.5 74.4. 96.2 69-8 86.0 69-9 T8 74.2 71 -3 88.5 74.5 77.3 77.2 93.4 75-7 90.9 94.2 T9 81.8 84.2 83-5 78.0 84.5 94.0 96.8 73-9 86.2 96.1 T10 87.8 94.9 85.4 64.1 96.3 77.1 94.9 77.9 73-6 . 96.3 Til 70.7 97.5 91.0 77.8 76.9 75.9 98.9 90.9 69.6 95.1 T12 69.1 90.0 75-2 68.3 76.1 66.8 97.5 68.3 75-1 93-4 T13 70.4 97-7 88.9 74.6 68.0 73-5 97-1 95-0 77-5 87.6 Tl4 77.5 94.7 95.1 75.9 64.6 69.0 94.6 91.1 79.7 94.9 T15 74.1 94.6 92.9 64.6 87.O 66.5 99.1 95.8 70.7 92.9 T16 74.5 92.3 94.4 82.7 89-1 74.6 98.9 97.1 72.2 85.2 T17 75-5 97.4 95.7 69.3 87.O 70.7 92.9 96.3 67.9 91.6 T18 72.9 97.6 76.6 76.6 93-1 82.0 93.^ 91.1 97-5 87.7 T19 84.1 98.0 88.9 62.0 96.7 83.3 74.5 90.8 97-4 87.2 T20 73.4 94.8 87-9 71 -9 90.2 77-5 74.4 95.9 97". 0 84.6 Table A-l (Continued) Percent fin-Edge Time by Subject for the Experimental Group in Experiment 1. Day 2 SI S2 S3 S4 S5 S6 S7 S8 S9 S10 T21 44.8 97-8 87.1 78.5 90.4 82.9 57-9 87-2 87-9 94.6 T22 64.3 95.1 89.6 85.2 95-1 83.9 67-3 75.3 96.2 92.4 T23 76.9 91.2 83.9 88.4 96.0 78.1 70.8 90.5 89-9 92.5 T24 68.4 83-3 91 -3 86.6 94.4. 84.2 62.1 91.1 92.5 91.4 T25 67.4 63.6 92.0 80.5 85-3 85.0 70.0 85-9 84.2 93-8 T26 70.4 93-4 92.2 93-8 98.9 84.2 54.2 83.2 93.3 94.7 T27 66.1 92.3 80 .0 90.2 95.8 89-5 51.6 88.9 90.6 92.8 •T28 74.9 91.1 89-4 89.6 95-7 93-3 55-5 93-0 87.8 93.5 T29 59.3 84.7 89-0 94.6 93.0 93-2 62.1 88.7 97-4 94.3 T30 69.6 94.3 86.0 98 .0 92.1 94.0 56.1 93-8 92.2 95.5 T31 64.5 91.9 81 .9 85.4 88.8 91.1 60.2 79.7 82.0 86.7 T32 62.3 89.7 81 .7 89-1 92.3 92.1 42.9 91.4 79.3 89-3 T33 91-9 88.1 87.1 90.4 91.4 92.6 28 .2 88.4 68.5 92.3 T34 93-4 92.1 83-3 90.9 89-9 78.1 21 .0 92.9 83-3 91.6 T35 94.6 88.7 91.8 90.9 95-6 90.0 52.6 97.7 85-5 87-7 T36 95-9 96.5 91.1 88.2 84.9 93-7 21 .4 97-9 71.1 91.6 T37 98.2 93-9 86.0 92 .4 90.5 93-7 29.2 95.8 86.0 89-4 T38 95.6 95-0 84.8 92.5 91.5 91.7 23.4 95-8 92.0 91.3 T39 90.9 95-1 90.3 93-2 98.2 95-4 64.9 96.5 68.0 94.6 T40 90.7 94.5 78.8 82.4 95.1 93.8 53.1 94.8 83.7 93-2 Table A-2 Percent dn-Edge Time by Subject for the Control group in Experiment 1. Day 1 SI S2 SS3 S4 S5 S6 S7 S8 S9 S10 TI 44.1 02.7 69.3 oo;-o 54.1 43.6 00.6 09.7 29.0 71.5 T2 34.5 05-0 64.9 00.8 58.4 56.7 00.2 13.3 60.9 33-1 T3 13-6 04.9 65-3 02.3 37.0 61.1 00 .6 07-3 56.8 39.3 T4 17.5 01.7 69.3 06.4 51.8 48.4 01 .0 07.2 32.1 37.6 T5 14.6 02 .6 58.9 17.1 68.6 58.5 00.8 06.6 79.0 34.6 T6 18,6 02 .4 84.1 09.5 60.9 43.9 00.8 26.2 82.5 40 .4 T7 35-6 01.9 78.6 04.2 62.1 59-9 03.0 26.7 89-7 29.1 T8 35.3 07 .1 64.2 05.3 65-1 50.8 00.1 20.3 94.6 51.1 T9 53.3 01.3 79.6 03.9 41.5 50.0 00.0 23.2 92.9 42.0 T10 44.3 00.5 90.3 01.7 68 .0 31.0 00.5 26.2 88.5 35-9 Til 28.3 48.5 61.2 09.1 67.7 52.3 02.6 26.4 87.8 34.6 T12 28.4 46.9 62.4 06.9 43.8 46.3 01 .0 21.9 91.6 42 .2 T13 24.5 37.0 48.1 06.4 65.6 57.7 04.2 23.9 92.0 51 .1 Tl4 26.7 41 .5 68.3 07.6 64.2 55-1 02.9 38.4 86.4 63.4 T15 22.6 41.3 65-9 02.7 72.6 60.4 01 .8 38.0 42.3 71 .4 T16 27.1 29.1 64.8 01.6 76.9 55.8 00.5 40.1 59-9 52.4 T17 35-5 28.8 60.2 00.3 58.3 63.O 04.7 45.1 72.1 48.7 T18 36.6 34.1 69-4 15.6 50.8 48.7 05.2 46.5 56.8 52.1 T19 27.2 32 .2 56.8 04.1 49.5 50.1 13-5 51.3 70.3 52.9 T20 35-1 30 .4 49.8 07.9 66.1 52.6 13.3 31.7 73.7 45.3 Table A-2 (Continued) Percent Qn-Edge Time by Subject for the Control group in Experiment 1. Day 2 SI S2 S3 S4 S5 S6 S7 S8 S9 S10 T21 00.1 03.4 15.0 07.6 75-3 00.9 22.9 31.8 77-6 25.3 T22 00.2 06.2 23.3 00.8 63.1 00.6 17.5 50.6 79.1 44.0 T23 00 .0 01.8 63.6 07.5 70.3 01.5 16.7 50.0 83-4 36.1 T24 00.7 01.5 51.8 08 .0 67.5 05.3 18.4 48.4 88.1 40 .5 T25 00.0 06.0 42.0 01.7 71.0 03.5 09.8 51-5 85-9 48.1 T26 12.8 19-4 08 .0 05.1 56.6 01.2 10.4 66.2 94.2 40.5 T27 14.5 26.6 29.0 13.5 23.9 00.8 15-9 80.3 88.9 43.6 T28 22.7 18.5 17.9 13-1 65.1 53.7 24.7 84.7 84.9 40 .8 T29 02.5 08.8 06.7 15.2 52.5 54.1 40.5 85-5 84.9 48.8 T30 12.8 20 .3 29-6 15.8 20.3 44.9 22 .8 64.2 87.0 28 .0 T31 08.5 39.0 21 .5 20.8 54.8 54.2 04.9 65-9 90.6 17-8 T32 05.1 50.8 20 .1 44.7 61.1 53.1 08 .9 50.6 85.0 19.3 T33 11 .5 47.5 37-0 66.3 76,7 50.2 25.5 84.4 83.O 20.9 T34 08.8 36.0 57.8 63.4 23.5 50.3 12.2 61.8 82.7 44.4 T35 12.7 54.2 60.0 37-5 53-9 52.7 21 .4 82.4 84.2 49.8 T36 12.4 18 .3 26.0 55.2 19.3 75-8 05.6 67-5 84.1 28.5 T37 56.1 32.3 23.1 65-8 60.7 78.6 23.4 68.1 87.1 37-2 T38 33-8 55.9 17-6 63-6 68.7 66.1 16.0 51.9 87-5 17.2 T39 00.4 46.6 35-9 45.5 37-2 72.0 10.7 66.9 86.9 33-5 T40 02.0 51.5 29.0 58.6 60.0 77.6 13.5 73.8 87-5 33.0 Table A-3 Percent On-Edge Time by Experimental Subjects in last Eive Trials of Experiment 2. Trials SI S2 S3 S4 S5 S6 S7 S8 S9 S10 T6 82.5 96.2 T7 82.5 96.8 T8 76.8 97.6 T9 82.1 94.4 T10 86.2 96.6 96.3 95.2 94.4 92.6 72.6 94.9 90.5 92.8 68.7 88.1 94.9 94.9 76.3 91.9 92.6 84.5 96.9 94.1 91-6 94.0 67.8 91-6 86.2 89-5 69-0 95-8 76.6 90.4 78.1 89-4 86.7 94.3 82.0 94.2 88.6 91.1 77-0 71-8 85.9 93.6 Table A-4 Percent th-Edge Time by Control Subjects in Last Five Trials of Experiment 2. Trials SI S2 S3 S4 S5 S6 S7 S8 S9 S10 T6 02.3 52.9 37-0 26.9 62.7 94.5 23.8 74.4 98.4 40.7 T7 02.5 18.0 50.7 52.6 62.9 92.2 36.2 79.0 81.5 40.6 T8 14.2 33.8 33-3 23.5 67.9 96.4 25.4 73.8 98.6 43.4 T9 04.7 33.4 33-4 14.1 18.2 92.3 10.3 71.0 94.5 31-3 T10 01.7 52.4 34.6 33.4 49.2 96.7 34.9 83.4 79-7 42.0 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items