Psychological Modeling, Cognitive Representations, and Learning to Squat
by Peter Daniel Catina, Ph.D
How many times have you heard a coach at the contest screaming a long list of instructions to their lifter immediately prior to the execution of the squat? Guess what? It doesn’t work! In fact, it’s often too complicated, the athlete cannot readily process it, and it’s more of a distraction than anything else. Let’s face it, if your lifter doesn’t know what to do 30 seconds before his or her opener…. it’s too late! Thought precedes language. Unfortunately, some coaches seem to have very big mouths and very small brains. This is always a bad combination. More coaches should realize that they are not the center of attention, the lifter is. Instead of screaming a paragraph of useless information, coaches should limit their instructional cues to a few positively charged syllables of proper technique and encouragement.
Introduction
The purpose of this article is to present a multifaceted understanding of the processes underlying psychological modeling and their relationships to learning the free weight squat exercise. In order to benefit from the favorable training adaptations afforded by the barbell squat, performing the exercise with proper technique is absolutely crucial. Proper visual demonstration or modeling is vital for learning a complex motor skill such as the squat. Visual demonstration conveys a vast array of informational cues that are far more relevant to facilitate an observer's motor skill acquisition than information conveyed through verbal instruction. Successful performance of the squat exercise depends on the ability to hold visual symbols in memory for a short interval of time and maintain internal descriptions of the relevant biomechanical factors necessary to execute the motor task efficiently. In the following study, subjects who performed the squat after exposure to a video-taped demonstration of proper technique exhibited significantly higher scores in both the accuracy of their cognitive representation of the modeled action, as well as in performance technique than subjects who performed the squat without exposure to video-taped demonstration. These data suggest that visual demonstration of the squat is a positive factor in enhancing the performance of novice lifters. There are, of course, many factors that influence motor skill acquisition.
Upon visiting almost any fitness center, at least two things will be evident. Either people do not perform the squat at all, or many of those who do, perform it incorrectly. This is due to false information and improper instruction. Another problem is that people learn to squat while looking at their refection in a mirror. Although mirrors do provide a modicum of necessary feedback as to how one is progressing in terms of appearance, they are an inappropriate orientation for observing the execution of a motor task, especially one as complex as the squat. You wouldn't teach someone how to approach a bowling lane while facing a mirror, or demonstrate a tennis serve while facing the learner. Learning to squat in front of a mirror is not consistent with the notion of acquiring a cognitive representation through observational learning.
These concepts are easily recognized by standing in front of a mirror and raising your right hand. The image in the mirror seems to raise its left hand. Write a word on a piece of paper and hold the paper so that the word can be seen in the mirror. The word seems to be written backwards. Mirrors change what you see. When light strikes the reflection of a lifter in a mirror, only a small part of that light travels in the correct direction to reach that lifter’s eyes and the paths to each eye cross over to the other side before they reach their destinations. To get to the lifter’s eyes, light from every point on the reflection takes a different path. Light from the highest point ends up lowest after being reflected by the mirror. Likewise, light from the lowest point ends up highest after being reflected.
Squatting in front of a mirror is one of the most counterproductive practices I have ever witnessed, especially for a beginning powerlifter. What a shock it must be for a novice lifter to arrive at the contest and find that there are no mirrors! Few things are more important than simulating contest environment during training sessions. I’ve even gone so far as to bring newspaper and tape to the gym and cover up the mirror where I squat. Whatever works for you, comic strips, a poster of Shania Twain, or a picture of the meanest, ugliest, strictest head-referee on the planet (which is the most likely thing you’ll see at the actual contest). Think of it as a surefire way to keep those pesky bodybuilders away from the rack when you’re squatting. Try it sometime, just cover up the mirrors, and watch all the body builders disappear along with their reflections. The intent of this article is not necessarily to condemn mirrors (they’re great for shaving), but rather to show that proper visual demonstration is an effective method for motor skill acquisition.
Since the squat is a full-body exercise, it elicits one of the highest hormonal responses provided by all resistance exercises. Hormonal actions that influence the adaptations to the squat exercise include, but are not limited to, improved force production, stimulation of cartilage growth, and enhanced size (Fleck & Kraemer, 1987).
Performing the squat exercise with proper technique is crucial in order to benefit from these favorable adaptations. Teaching proper form to a student requires visual demonstration by the instructor. The literature in the area of motor skill performance as it relates to or is affected by modeling primarily considers visual perception of the modeled information as a mediating variable on behavior, but what ensues between perception and behavior does not appear to be fully addressed. Apparently, a myriad of intermediary components is essential in transforming the patterns of movement demonstrated by the model into appropriate actions to be accomplished by the learner.
Psychological Modeling
Modeling is an effective means of conveying relevant information to facilitate an observer's motor skill acquisition (Gould & Roberts, 1981). The literature in the area of modeling primarily considers visual perception as a mediating variable on behavior. Visual demonstrations of motor tasks are retained by the learner in the form of internal messages, which are recorded and saved in the pre-frontal cortex of the brain for future retrieval (Goldman-Rakic, 1987). It is evident that the observer somehow retains the modeled action and can later replicate what was seen in the absence of verbal instruction (Williams, 1994).
The significance of the modeling process lies in its effect on the behavior of the observer. The observer does not merely watch the action passively. But perceives the environmental information rather actively. According to Gibson (1968), the observer must differentiate the information within the optic array into its most useful dimensions. In other words, particular events and attributes are singled out for observing and describing what exactly ensues between perception and action. So, it is crucial to demonstrate proper technique to the observer in the most effective manner, especially when one considers the many people in various weight rooms that are using bad technique, thereby setting a bad example to those observers. When a coach gives too many verbal cues, it causes the lifter to think about too many things. This may result in confusion and attenuate performance. Too many instructions make it difficult for the student to totally isolate one strategy from another. In doing so, some of the information is lost. It may be that the student is using various combinations of strategies and cannot be focused into using the most effective one. Therefore, it is important that verbal instruction be as clear and parsimonious as possible. The coach should first demonstrate the squat with the learner standing behind him or her. This type of visual modeling facilitates motor skill reproduction. However, there is a multiplicity of variables and co-actors that are linked to observational learning which will be expounded upon within the remaining sections of this article.
Bandura (1986) suggests that behavior is mediated by exposure to the model and that repeated exposure to the model will improve the quality of the cognitive representation which will, in turn, facilitate performance. The concept of modeling is presumed to be controlled by four sub-components: "Attention", a conjecture that people cannot learn much by observation unless they attend to, and perceive accurately the significant features of the modeled behavior. "Retention", where it is submitted that people cannot be influenced by observation if they do not remember it. "Motor reproduction process", which is the conversion of symbolic representations into appropriate actions. "Motivation" which proposes that people are more likely to adopt a modeled behavior if it results in rewarding consequences. According to Martens, Burwitz, and Zuckerman (1976), the successfulness of the modeling process is limited by the difficulty of the motor task. This relationship between modeling and performance is predicated on two circumstances: an accurate perception of what is to be accomplished by way of strategy or technique, and the ability of the learner to reproduce the demonstrated action. The instructor should give immediate visual feedback by demonstrating the movement with an emphasis on correcting existing mistakes or replaying videotape of the student's performance, thereby providing visual assessment of the motor task in a timely manner. This will give the student a reference point from which he or she can improve performance.
According to Adams (1986), this knowledge of results enables the observer to correct errors in movement technique. The greater the accuracy of the cognitive representation of the modeled action, the greater the skill acquisition will be in the subsequent reproductions of it. This is consistent with the schema theory proposed by Schmidt (1975), which states that sensory consequences and actual outcomes, for a given set of initial conditions, could be related by the subject.
Cognitive Representations
Cognitive representations may be construed as "mental blueprints" comprising an essential link between perception and action. The brain not only categorizes these non-language representations; it also builds successive layers of categories such as shape, movement, and sequence. In this way, the learner organizes visual information, events, and their relationships.
The notion of a cognitive representation, although it has been called many different things by many different researchers, has been explored since the origin of psychology as a science. For example, James (1894) characterizes the notion of a cognitive representation when addressing the idea of movement images. More recently, researchers who have described, theorized, and attempted to measure changes in cognitive representations include (Cooper & Podgorny, 1976; Corballis, 1979; Broadbent, 1984; Matthews, 1988; Masson, 1990; Carroll & Bandura, 1990). A cognitive representation has two basic functions. One is to regulate movement production, and the other is to serve as a standard of corrections for the detection of error between the cognitive representation and response-produced feedback (Adams, 1986). While it may be possible to detect error in the movement of a reflection in a mirror, it is impossible to construct an accurate cognitive representation and correct the error through the immediate feedback from a mirror. The learner should be able to use his or her cognitive representation as a reference of correctness in order to form a hypothesis about how to perform the movement better. This is accomplished most effectively by modeling via visual demonstration from the instructor and/or implementing video-taped performance for assessing proper technique in the barbell squat.
Proper Squat Technique
Proper technique for squatting includes, but is not limited to, placing the bar approximately 1-3 inches below the anterior deltoid, which affords more efficient biomechanics by lowering the center of gravity, as long as the bar is not placed exceedingly low on the shoulders. Generally, the feet should be slightly wider than shoulder width. This will increase the availability and usage of the larger and more powerful muscles and enable the lifter to shorten the distance traveled. The lifter should start the descent by leading with the hips rather than with the knees so that the shins are perpendicular to the floor. The heels should be flat on the floor for the entire duration of the lift. Raising the heels up predisposes the knees to injury and shifts the center of gravity forward forcing the lower back to compensate for the displaced load. The lifter should have fully inhaled while starting the descent. The breath should be expelled when the "sticking point" is reached in the ascent, which is typically around thirty degrees of extension. This technique will increase interstitial leverage and aid in keeping the torso erect by forcing the chest out in front of the bar. As one can see, there is a complex network of movement underlying efficient squat performance.
Arousal Level and Increased Resistance in the Squat
It is bit idealistic to assume that gross motor activities, such as the squat, require high levels of arousal for optimal performance and conversely that gross motor activities are adversely affected by low levels of arousal. The idea should be that optimal levels of arousal are suitable for optimal levels of performance. Perhaps an appropriate increase in resistance must be used to elicit an optimal arousal level. This topic is of particular importance to beginning powerlifters because of the driving force of the human ego to place too much weight on the bar before proper technique is developed. Of course, the object of the sport is to lift more weight; after all, powerlifting is the essence of true strength expression, but not at the expense of proper form.
I am currently conducting a pilot study to further investigate the effect of increased resistance in subsequent trials of the squat. So far, the evidence suggests that gradually increasing the weight increases the accuracy of the cognitive representation and actually improves technique. One would think that making a novel and complex task even more difficult would increase inhibition, but carefully choosing the "correct" increment of resistance is responsible for heightened performance. This seems likely due to psychological factors of knowing that there is more weight on the bar, which may stimulate response behaviors such as increased heart rate, blood pressure, respiration, and most importantly the enhancement of reflexes, force production, and possibly reaction time. This psychological phenomenon seems to manifest itself physiologically by mediating by an increase in catecholamine secretion. According to Davis, Hitchcock, and Rosen (1991), fear or increased arousal may result from activation of a single area of the brain (the central nucleus of the amygdala). The projections from the central nucleus of the amygdala to the ventral tegmental area may mediate stress-induced changes in dopamine synapses located in the frontal cortex resulting in increased vigilance and attention. These alterations in autonomic activity may have a synergistic effect, thereby increasing the subject's ability to perform a gross motor task such as the squat when an appropriate increase in resistance is added to the bar.
Visualization and Imagery
Imagery is a pervasive form of experience and is clearly important for all individuals interested in the acquisition of free-weight lifting skills. The effectiveness of imagery techniques has been demonstrated regarding sport tasks of accuracy, concentration, and strength (Lee & White, 1990). The physiological and psychological benefits from practicing visualization improve athletic performance, and have a direct application to the increase of strength (Murphy, Woolfolk, & Budney, 1988). Imagery rehearsal of the desired sequence of sensory-motor behavior units involved in a good performance has been used both by itself and as a part of multiple models (e.g., visuo-motor behavior rehearsal, which combines imagery, relaxation, and actual performance (Suinn, 1980). An interesting point to consider is that individual differences in imagery ability can influence motor task performance. Subjects with greater imagery ability achieve higher scores in replicating movement patterns than do subjects of lesser imagery ability (Goss, 1986). Experimental research has indicated that "high imagers" exhibited significantly greater recall scores than "low imagers. Visual imagery facilitates the short-term retention of visually presented sequences of movement.
Clearly, it has been demonstrated that mental practice enhances performance (Grouios, 1992). However, there still exists the misconception that once mental practice is learned, it can be used in an appropriate situation with a reliable frequency. Teaching students mental practice techniques is one thing, teaching them to be able to initiate those techniques in specific situations such as learning to perform the squat exercise requires a close examination of technique and appropriate feedback in order to correct mistakes. Also, since it is possible for a lifter to mentally rehearse the execution of proper technique and retain the image of the correct motor pattern. Likewise, it is possible for a lifter to mentally rehearse the execution of poor technique and retain the image of the incorrect motor pattern. This is important when one considers that nearly all non-powerlifters perform the squat incorrectly, thus setting a bad example for gullible observers.
My research idea was to see if exposure to video-taped demonstration of the squat by an expert model would positively affect performance by increasing the accuracy of the learner's cognitive representation, as well as the level of performance technique in the squat. It was hypothesized that subjects who were exposed to a video-taped performance of the squat by an expert model would exhibit greater scores on the questionnaire analysis, the video analysis, and the three-dimensional figure analysis, than subjects who were not exposed to the video-taped demonstration. The following section is an abbreviated description of the experiment without most of the excruciatingly detailed statistical procedures.
Method
Twenty-four subjects were sampled, half of whom watched a video tape of the model performing a barbell squat. The subjects were randomly allocated into two groups according to individual time slots to undergo the experimental session. Time slots were numbered consecutively one through twenty-four. Odd numbered subjects were assigned to group A, while even numbered subjects were assigned to group B. Criteria for admission into the study were that the subject be male, matched for age, body weight, and height, with no history of chronic knee or lumbar spine maladies, have little or no experience in resistance training with free weights or formal instruction in proper squat technique. Subjects ranged in age (18 to 30 years), body weight (150 to 210 lbs.), and height (65 to 74 inches). This screening method normalized severe differences in anthropomorphic measurements.
Design
Three levels of analysis were established among the data with the following variables: the dependent variable was the cognitive representation of the model, and the independent variable was the exposure to the video-taped demonstration of the squat. Squat performance technique and cognitive representation were assessed at three levels of analysis. (1) The questionnaire analysis, which measured cognitive representation. (2) The video analysis, which measured squat performance technique. (3) The three-dimensional figure analysis, which measured the degree of similarity between the position of the model and the position of the subjects during the performance task. Multiple levels of analysis were used to clarify treatment effects on cognitive representation and motor performance.
The data analysis required a distribution-free, nonparametric test be used to measure differences between and within the two groups due to the relatively small sample size. The Wilcoxon Matched-Pairs Signed-Ranks Test and the Mann-Whitney U Test were selected. These statistical procedures were used to test the null hypotheses, which stated that there is no difference between scores in group A (exposure to video-taped demonstration) and group B (no exposure to video-taped demonstration). The experimental hypotheses are directional and state that scores will be greater in group A than those in group B. The alpha .05 level of significance was selected for all three categories of analysis.
Instrumentation
A video camera was used to film all performances. The camera was positioned 6.1 meters away and perpendicular to the primary sagittal plane of motion (lateral orientation). The optical axis of the camera was set at 1.5 meters from the floor so that each subject's range of motion could be readily observed, regardless of their differences in height. This format was also used to film the squat performance of an expert model who rendered a technically proper squat for the purpose of visual demonstration. Three orientation aspects were observed: rear, lateral, and 45 degrees (combination of rear and lateral). The following five critical areas of the kinematic pattern were emphasized by stopping the video at specific, pre-selected frames to illustrate the appropriate technique throughout the execution of the lift: the erect position phase, the beginning of the descending phase, the bottom position phase (thighs slightly below parallel), the middle of the ascending phase, and the recovery to the original starting position.
The decision to use an expert model was two-fold. The filming of the expert performance served as the modeling stimulus for the subjects, and subsequently became a template of proper execution of the motor task, thereby establishing performance criteria. The subject's performances were then compared to the template, and deviations in postural alignment and speed of movement during the lift were used as a mode of scoring.
Changes in cognitive representation were measured by a ten-item questionnaire which evaluated the clearness of the subject's cognitive representation of their performance. The questionnaire assessment was calibrated by using the Betts rating scale (Sheehan, 1969) ranging from (1) "perfectly clear and vivid" to (7) "no discernable image at all" The direction of magnitude in the Betts rating scale was reversed in order to maintain congruent data. The ten-item questionnaire measuring the imagery associated with performing the squat exercise follows:
1. You are near the barbell, the feel of the floor beneath your feet, the details of the bar and plates.
(each item was rated 1 2 3 4 5 6 7).
2. The sound of the buzzer, you grasp the bar, the feel of the knurls, the feel of the weight as you
cautiously place it across your shoulders.
3. The sensation of your lungs as you stand motionless, taking your final breath and holding it before the
commencement of the lift.
4. Your hips and knees bending as you lower yourself into the descending phase of the lift.
5. The precise moment that you felt you had achieved adequate depth.
6. The explosive force exerted on the bar as you reverse the direction of the movement and initiate the
ascending phase of the lift.
7. Your feet still pressing hard against the floor as your legs extend.
8. The sensation of your lungs as you forcefully exhale.
9. The precise moment that you regained the original starting position.
10. The feeling of relief as the weight is released off of your shoulders and you are relaxed.
Changes in cognitive representation were additionally measured by the use of a three-dimensional, adjustable, wooden figure. The subjects were required to assemble the wooden figure into a position which they determined to be the best three-dimensional representation of their performance.
Procedure
Each group was presented with the task of performing a barbell squat. Prior to the performance task, both groups thoroughly read the following written instructions:
1. Grasp the bar with your hands in front of you. Align your hands with the orange markings on the bar. This is approximately the correct hand placement.
2. Dip your head under the bar while keeping your hands firmly in place and position the bar across the upper portion of your shoulder blades.
3. Let the bar settle into the most comfortable and natural position on your shoulders.
4. Take a deep breath and hold it.
5. At the sound of the whistle, gradually bend your hips, then knees, and lower yourself as if you were preparing to sit down in a chair that is very low to the ground. Then reverse the direction of the bar as quickly as possible by pushing your feet very hard against the floor and thrusting your head backwards.
6. Keep your feet as flat as possible.
7. Once you have initiated the ascending phase of the lift, you must forcefully exhale and keep pushing with your feet until your legs fully extend and you have regained the original starting position.
8. Reposition the bar onto the rack, release the bar from your shoulders, and you may relax.
After the written instruction session was completed, half of the subjects were exposed to the video tape of the model performing a barbell squat while the remaining half were not. Both groups performed a squat with the bar loaded at 1/3 the lifter's body weight. Only one repetition was permitted, thereby reducing the occurrence of motor learning due to repetition. A whistle was sounded, thus marking the commencement of the performance task requirement. Immediately following each trial, all subjects completed a two-part assessment of their cognitive representation. Part 1 consisted of completing the questionnaire and part 2 consisted of assembling the wooden figure which was then photographed and later compared with the model's assembly of the wooden figure.
Results
As explained earlier, adaptation in cognitive representation was measured using three categories of analysis. The questionnaire scores, the video-taped performance of the squat, and subjects’ positioning of the three-dimensional wooden figure. Data were arranged in such a way that differences in paired observations within groups, as well as differences in observations between groups could be readily observed. The experimental hypotheses were directional, meaning that the scores of subjects who were exposed to the video tape were predicted to be significantly greater than the scores of those who were not exposed to the video tape.
Results from the questionnaire analysis:
It was hypothesized that the subjects who were exposed to the video-taped demonstration would have greater cognitive representation scores than subjects who were not exposed to the video tape. This hypothesis was supported between groups as The Mann-Whitney test yielded a U value of .50. The null hypothesis was rejected at the p < .005 level. Since the first analysis (the questionnaire) revealed a significant interaction for cognitive representation that was characterized by an increase in cognitive representation for trials involving exposure to the video-taped demonstration, a second and a third analysis was conducted to further clarify treatment effects on squat performance technique and cognitive representation.
Results from the video analysis:
A set of ten criteria was used to measure differences between the video-taped performance of the model and the video-taped performance of naive subjects. The examination of the variables was divided into the descending and ascending phases of the lift. The dependent variables observed were:
1. Vertical velocity of the bar during the descent phase.
2. Absolute trunk angle during the descent phase.
3. Head position during the descent phase.
4. Backward horizontal hip movement during the descent phase.
5. Forward horizontal knee movement during the descent phase.
6. Forward horizontal bar movement during the descent phase.
7. Vertical bar velocity during the ascent phase.
8. Trunk involvement during the ascent phase.
9. Forward horizontal hip movement during the ascent phase.
10. Knees bowed in or out during the ascent phase.
It was hypothesized that subjects exposed to the video-taped demonstration would have greater scores in squat performance technique than subjects who only received written instruction. The performance scores were derived through objective examination by the researcher and two other elite powerlifters and qualified referees. This hypothesis was supported between groups as the value of U was equal to zero. The null hypothesis was rejected at the p < .001 level.
Results from the wooden figure analysis:
A set of seven criteria was used to measure differences between the positioning of the wooden figure by the subjects and the positioning of the wooden figure by the researcher, thereby providing a three-dimensional interpretation of the subjects' performance. The dependent variables observed were:
1. Body-mass distribution.
2. Horizontal bar-ankle distance.
3. Absolute trunk angle.
4. Tibia angle.
5. Femur angle.
6. Foot angle (flat, represented by 0 degrees, or not flat represented by greater or less than 0 degrees).
7. Horizontal knee-bar distance.
It was hypothesized that the positioning of the wooden figure by subjects exposed to the video tape would demonstrate a higher degree of similarity to the model than the positioning of the wooden figure by the subjects who only received written instruction. This hypothesis was supported between groups as the value of U was 4.5 which is significant at the p < .05 level.
Conclusion
While it may be possible to detect error in the movement of a reflection in a mirror, it seems impossible to construct an accurate cognitive representation and correct the error through the immediate feedback from a mirror. The learner should be able to use his or her cognitive representation as a reference of correctness in order to form a hypothesis about how to perform the movement better. According to Adams (1986), this knowledge of results enables the observer to correct errors in movement technique. Although the present study does not directly allow its subjects to receive knowledge of results, the wooden figure presents a problem-solving task that allowed the subjects to construct a more accurate cognitive representation to enhance performance. Even though the subjects in the present study were not given the opportunity to practice the movement, the wooden figure provided an assessment of movement and produced a plan of action derived from information acquired in the trial. This is consistent with the schema theory proposed by Schmidt (1975), which states that sensory consequences and actual outcomes, for a given set of initial conditions, could be related by the subject.
The free-weight squat is an open-chain, multi-joint, full-body movement, with a higher level of complexity compared to resistance exercises on machines. Therefore, the additional demands on balance, control, and technique make the squat a very challenging skill to acquire and an even more challenging skill to teach. The four modeling processes "attention", "retention", "motor reproduction", and "motivation" outlined above, are necessary for the learner to acquire proper technique in the squat through observational learning. The learner must understand the purpose of a physical action such as squatting, practice organizing the sequential movements, and receive further demonstration of the same task in conjunction with mental rehearsal in order to enhance performance. The relevant aspects of proper technique are most effectively conveyed to the student by visual demonstration because it provides a complete symbolic representation of the kinematic information in the complex sequential components in the action pattern of the barbell squat.
I hope that my research will benefit both novice as well as elite powerlifters. I wish that I knew this stuff when I first started competing. And, even though I’m older, I still approach training and competition with fresh ideas. Always remember that a closed mind is usually empty. It is with great honor and enthusiasm that I share my research with the powerlifters who read this magazine. This is but one piece of convergent evidence, it does not provide conclusive evidence on it's own. This is a consequence of the large number of ambiguities that have plagued the understanding of cognitively represented images in the past.
Hopefully, this study has provided new aspects and through the distinctiveness of further research, new possibilities can be intuited. Knowledge of facts is certainly valuable. But, facts conceal ideas. In science, one must search for ideas; if there are no ideas, then there is no science.
Questions and comments may be sent to the author via email: Liftheavy@earthlink.net
References Cited
Adams, J.A., (1986). Use of the model's knowledge of results to increase the observer's performance. Journal of Human Movement Studies, 12, 89-98.
Bandura, A., (1986). Social Foundations of Thought and Action: A cognitive theory. New York: Prentice-Hall.
Brodbent, D.E., (1984). The Maltese cross: A new simplistic model for memory. Behavior and Brain Sciences, Mar., 7,1, 55-94.
Carroll, W.R., and Bandura, A., (1990). Representational guidance of action production in observational learning: A causal analysis. Journal of Motor Behavior, 22,1, 85-97.
Cooper, L.A., and Podgorny, D., (1976). Mental transformation and visual comparison processes: Effects of complexity and similarity. Journal of Experimental Psychology:Human Perception and Performance, 2, 503-514.
Corballis, M.C., (1979). Memory retrieval and the problem of canning. Psychological Review, Mar., 86, 2, 154-157.
Davis, M., Hitchcock, J.M., and Rosen, J.B., (1991). Neural mechanisms of fear conditioning measured with the acoustic startle reflex. In J. Madden (ed.) Neurobiology of Learning, Emotion, and Affect, Raven Press, New York.
Fleck, S.J. and Kraemer, W.J. (1987). Designing Resistance Training Programs.
Human Kinetics, Champaign, Il.
Gibson, J.J., (1968). What gives rise to the perception of motion? Psychological Review. 75, 335-346.
Goldman-Rakic, P.S., (1987). Circuitry of the prefrontal cortex and the regulation of behavior by representational memory. Handbook of Physiology. 5, 373-417.
Goss, A.E., (1986). Paired Associates Learning. Academic Press.
Gould, D., and Roberts, G.C., (1981). Modeling and motor skill acquisition. Quest, 33, 214-230.
Grouios, G. (1992). The effect of mental practice on diving performance. International Journal of Sport Psychology, 23, 60-69.
James, W., (1894). Thought before language: a deaf-mute's recollections. In Essays in Psychology. Harvard University Press, London, England, 278-291.
Lee, T.D., and White, M.A., (1990). Influence of an unskilled model's practice schedule on observational motor learning. Human Movement Science, 9, 349-367.
Martens, R., Burwitz, L., and Zuckerman, J., (1976). Modeling effects on motor performance. Research Quarterly, 47, 277-291.
Masson, M.E., (1990). Cognitive theories of skill acquisition. Special issue: The learning memory and perception of perceptual motor skill. Human Movement Science, 9, 3, 221-239.
Matthews, P.B., (1988). Proprioreceptors and their contribution to somatosensory mapping: complex messages require complex processing. Canadian Journal of Physiology and Pharmacology, 66, 430-438.
Murphy, S.M., Woolfolk, R.L., and Budney, A.J., (1988). The effects of emotive imagery on strength performance. Journal of Sport and Exercise Psychology, 10, 334-345.
Schmidt, R.A., (1975). A schema theory of discrete motor skill learning. Psychological Review, 82, 225-260.
Sheehan, P.W., and Neisser, V., (1969). Some variables effecting the vividness of imagery in recall. British Journal of Psychology, 60, 71-80.
Suinn, R.M., (1980). Seven Steps to Peak Performance: The mental training manual for athletes. Toronto: Lewiston, N.Y., H. Huber Publishing Co.
Williams, J.G., (1994). Motoric Modeling: Theory and Research. Journal of Human Movement Studies, 24, 6, 237-279.