|
Researched
and Composed by
Jacob Wilson, BSc. (Hons), MSc. CSCS and
Gabriel “Venom” Wilson, BSc. (Hons), CSCS
Abstract
Thorndike and
Woodworth (1901) proposed the Identical Elements Theory of Transfer,
which suggests that the amount of transfer or benefit training in one
situation would have on another would be determined by the number of
elements that the two situations had in common. In this context the
purpose of this paper was to review the current theoretical explanations
behind the Specificity Hypothesis in order to identify those elements.
Theoretical rationales cover the dominant explanations from the past 100
years of research.
Introduction
As discussed in
‘Historical Perspectives in Specificity’ Thorndike and Woodworth (1901)
proposed the Identical Elements Theory of Transfer, which suggests that
the amount of transfer or benefit training in one situation would have
on another would be determined by the number of elements that the two
situations had in common. Hergenton and Olson (2005) stated that ‘
Thordike believed that learning will transfer from the classroom to the
environment outside only insofar as the two situations are similar.’ In
1906 he summarized his thoughts by suggesting that “A man may be a
gifted poet, but an ignoramus in music; he may have a wonderful memory
for figures and only a mediocre memory for localities, poetry or human
faces; school children may reason admirably in certain sciences and be
below the average in grammar…’
Thorndike was
primarily concerned with the transfer of cognitive tasks. However, the
small amount of transfer seen in intelligence tasks, has also been found
in Motor tasks (Henry, 1958; Drowatzky and Zuccato,1967;
Schmidt and Lee, 1999, Sawyer, 2005, Sawyer,
Ostarello, Sues, and Demsey 2002, Elliott & Jaeger, 1988; Proteau,
Marteniuk, & Levesque, 1992; Proteau 1992, Soucy & Proteau, 2001,
Bennett & Davids, 1995, 1997, 1998; Elliott, Lyons, & Dyson, 1997;
Proteau, Tremblay, & DeJaeger, 1998; Robertson & Elliott, 1996). This
was summarized in a recent study by Sawyer, Ostarello,
Sues, and Demsey (2002) who state
‘research has shown that the transfer effect between movements at
different resistances, rates, and movement patterns is limited (13, 15).
So practicing one task to improve performance on a second task
(transfer) may not be as effective as practicing the desired task
itself. Improvement in squat press performance would not necessarily
result in improved 40-yd dash time (15).’
Thordike proposed
that any transfer seen was a function of the elements which underlie any
two situations. Nearly a century latter, two of the dominant scientists
in Motor Learning, Schmidt and Lee (1999) acknowledged the importance of
Thordike and Woodworth’s (1901) contribution, however they suggested
that ‘ the problem with this theory was that it never specified what the
‘elements’ could be…’ In this context the purpose of this paper was to
review the current theoretical explanations behind the Specificity
Hypothesis in order to identify those elements. Theoretical rationales
from the past century will be examined. A second purpose of this paper was
to identify differences between individuals in their ability to perform
a given task.
Foundational
Theories Which Address Specificity
The first theories
explaining Specificity and transfer were devised by Connectionists or
Behavior Learning Psychologists, who along with Cognitive Learning
Psychologists provided the precursor framework to the modern day field
of Motor Learning. According to Sawyer (2005) many of the underlying
processes of learning, whether from a Cognitive or Motor standpoint are
similar. Therefore these early theories have great bearing to motor
skill aquisition.
The connectionists
were so called because they felt that the acquisition of a skill
(learning) required the formation of a bond between a stimulus and a
response (Hull, 1943; 1952; Hergenhahn and Olson, 2005, Schmidt et
al., 1999) . The stimulus could be a pitch in baseball, with the
response being a perfectly executed swing, or a most muscular pose
during a bodybuilding contest. In this context performance was measured
as the proportion of correct responses to a given set of stimuli (Hergenhahn
and Olson, 2005) . They were known as behaviorists because they were
empiricists, meaning that they had to see the results and manipulations
of their experiments. This required clear observation of behavior or
performance (Skinner, 1974, 1978, 1988). It was thought that skill
acquisition could be inferred through behavior.
S-R
Figure 1
Figure 1 graphically demonstrates The bond between a Stimulus and a
Response.
Pavlov’s Theory
of Specificity
Pavlov was one of the earliest connectionists. Pavlov was responsible
for the discovery of classical conditioning. Essentially, he found that
pairing a neutral stimulus or a stimulus that would not normally cause a
response with an innate stimulus or a stimulus that causes a reflexive
response would eventually lead to the neutral stimulus being able to
cause the reflexive response (Pavlov, 1928, 1941, 1955). For example,
normally a bell (neutral stimulus) would not cause a dog to salivate.
However, if every time an experimenter rings a bell they feed the dog,
soon the dog associates the bell with the feeding stimuli (Innate or
Unconditioned Stimulus) to the point where they will drool upon hearing
the bell.
In this context he
proposed one of the first fundamental theories in learning. A
fundamental theory attempts to use principles that are as close as
possible to "reality.” However, his was quite exotic in its attempt
(Pavlov, 1955). Pavlov believed that an innate stimulus activated an
innate neural center in the brain. Activation would cause the neural
center to ‘radiate’ out electrical activity. The radiating wave would
travel out and activate the center responsible for causing the response.
The innate stimulus center may be a center for feeding stimuli while
the response center would initiate salivation. If the neutral stimulus,
such as a bell activated another neural center, specific to that stimuli
at the same time then the innate stimulus center (food center) and the
neutral stimulus center (bell center) would both radiate out at the same
time such that their fields would overlap. He called this overlap
‘Generalization.’ The overlapping of the two centers’ fields of
excitation formed a connection between the neutral stimuli center and
the innate stimuli center. With enough pairings (practice) the
connection becomes so strong that presenting the neutral stimulus alone
can excite the innate neural center and thus indirectly elicit the
innate response.
Pavlov believed
that each stimuli activated a specific neural center. He therefore
suggested that the greatest improvement in practice would come from
activating the specific neural center that the investigator would want
to cause the desired response in. Therefore if the experimenter wanted
a bell at a frequency of ‘X’ to cause an animal to drool, then they
needed to continually pair the innate stimuli such as food with a bell
rung at frequency X. He felt that generalization or transfer of
training occurred in so much as the neutral stimuli were similar to each
other. Similar stimuli centers were suggested to be close to each
other. In this context when they were activated, their wave of
excitation could activate other similar stimuli. Therefore if a bell
was rung at a frequency of Y, it would activate a distinct neural center
which, if similar to X, would activate neural center X, which could then
activate the response.
In summary, Pavlov suggested that the more similar the conditions of
practice were to the criterion task, the greater the transfer.
While Pavlov’s theory has solid predictive value, the physiological
aspects of his theory do not adequately fit with current knowledge of
the nervous system (Hergenhahn and Olson, 2005).
Hull’s
Explanation on The Specificity of Performance and Learning
Hull (1943, 1952)
was the first to quantitatively distinguish between performance and
learning or skill acquisition.
Hull
(1943, 1952) denoted performance as the reaction potential of an
organism. The reaction potential was defined as the probability, and
speed with which a behavior occurred to a given stimulus. In this
context he quantitatively defined performance through ‘Hull’s
equation.’ The equation is a complex set of variables which attempts to
compute the performance of an individual in a given situation
(see Wilson 2005
for a review). For example, what is the
probability of an athlete lifting X amount of weight on the bench press
on a given set? His work was critical because it predicted that
performance was determined by both permanent and temporary factors. For
example, a person may have learned how to be polite and smile when a non
amusing joke was told. However if they are tired, they may not express
this learned behavior. It is only recently however when the distinction
between performance and learning has become standard in motor behavior
research (Schmidt and Lee, 1999; Sawyer and Noel, 2001). The critical
nature of such distinction was notably supported by Sawyer and Noel
(2001). They had participants practice a tracking task with and without
an audience. In acquisition it was found that the group performing in
front of an audience had a lower performance than the no audience
group. However, when the audience was removed in retention there was no
difference in performance. This suggested that though they may have
both learned how to track, their distinct performances were due to the
differing environments.
Hull’s
(1943) equation in numerous situations has had great predictive value,
and can predict the average response in a wide variety of situations.
In this context,
Hull
does an excellent job of predicting the Specificity of Performance. His
theory would predict that the performance of a given individual in
situation A would transfer to situation B in so much as the situations
were similar. This is similar to the identical elements of transfer
theory, except that it acknowledges temporary variables, and identifies
those variables as clearly as possible.
A perfect example
is the Specificity of pre game warm ups. In motor control there is a
unique drop in performance classified as the Warm Up Decrement. In a
review Adams (1961) provided extensive evidence that after a break, the
first trials on a given task are normally below the performance of the
last trials. One rationale to explain this is the Set Hypothesis.
Schmidt and Lee (1999) define a Set as one or more temporary internal
states that underlie and support the skill in question. They further
suggest that during practice various supporting Sets are constantly
adjusted to maximize performance, while at rest these Sets fall to a
point that supports rest. Arousal is an excellent example of a Set. In
order to maximize performance, arousal will need to be heightened to a
specific level which supports that performance. This causes several
issues with bench players. Starters are able to warm up and raise their
Set to game standards; however bench player’s Sets are lowered to states
specific to rest. Therefore when they come off the bench they may
experience severe warm up decrements.
His theory also
predicts that performance will be specific to the environment,
suggesting that practice conditions should closely mimic such
conditions. This was again supported by the Sawyer and Noel (2001)
study above, which suggested that performance in practice without an
audience (fans) may not accurately predict how a team will perform in
front of an audience. A further example is Olympic lifting. Often high
level performers can only reach their top performances when in
competition (Landers, 1980). This may reflect such aspects as higher
arousal levels elicited by a loud crowd, while near fellow competitors.
In this context, to prepare for a contest, the Olympic lifter should
consider participating in smaller competitions before the major
competition to prepare themselves to operate in that environment.
Feedback conditions are also critical to take into consideration in
practice. Evidence suggests that feedback can have several positive and
negative effects. One of the negative effects is known as Dependency
(Schmidt and Lee, 1999; Sawyer 2005, Salmoni, Schmidt, & Walter, 1984;
Schmidt, 1991; Young & Schmidt, 1992). Dependency occurs when feedback
is given too frequently (see Swinnen, 1996, for a review on the
‘Guidance Hypothesis’). For example, if a coach continually stands near
their athletes and constantly talks them through their events in
practice, their performance will begin to depend on the coach’s
guidance. However, during competition when such interaction is not
allowed a severe decrement will be seen in performance. This suggests
that even coaching should factor in practice how to maintain a similar
environment to game time conditions. In this context Swinnen et al.
(1990) suggests that continual feedback not intrinsic to the movement
causes the participant to rely more heavily on that feedback rather than
intrinsic feedback.
In summary,
Hull’s
equation predicts that performance in one situation can predict
performance in another situation in a similar manner to the identical
elements of transfer theory. This is unique as it suggests that even if
a person performs the criterion task, if the environment or internal
state (arousal, hunger, glycogen stores) of an individual change, the
extent of that change will directly change performance. Therefore
practice conditions should mimic the criterion environment or
competition and the internal state that the athlete seeks to excel in.
Guthrie’s
Theory of Specificity
Guthrie (1950,
1952) provided perhaps the most parsimonious theory of all the
connectionists. He again defined performance as the probability of a
response occurring to a given stimulus set. In this context, Guthrie
formulated his ‘ One Law of Learning.” Guthrie suggested that ‘-
"A stimulus pattern gains its full associative strength on the occasion
of its first pairing with a response (Hergenhahn and Olson, 2005)." As
stated the acquisition of a skill in the connectionist framework occurs
as a bond is formed between a stimulus and a response, or if an already
existing bond is strengthened. The one Law of learning suggests that
when a response occurs to a given stimuli, not only is the bond formed,
but at full strength. It is important to understand that if the
learning process were to be graphed, it would create a negatively
accelerated curve (Balov, 1971; Fitts and Posner's 1967). This means
that at the start of learning, improvement occurs rapidly and slows as
the process continues. This seems at first to conflict with Guthrie’s
theory. However, Guthrie’s theory was molecular in nature. He
suggested that in the environment there were an endless number of
stimuli. For example the temperature, humidity, surrounding lighting,
and people serve as each separate stimuli. Due to processing
limitations, no organism can sample all of the stimuli in the
environment at once. However, when they make a response all the stimuli
sampled in the environment are attached to that response. The learning
curve accelerates rapidly at the beginning because more and more stimuli
could be attached to the given response. Each time the organism
attached new stimuli, a lower number of new stimuli could be attached to
the response, and therefore the possibility of improvement lowered.
Practice was therefore a function of attaching more and more stimuli to
a given response.
While Guthrie’s
theory seems extremely simplistic, it has a tremendous amount of
predictive value and perhaps explains home field advantage and practice
specificity better than any other (see Types of Specificity in article
4). A common example can be found in the student who studies all week
at home for an up and coming mid term. They feel extremely confident
entering the exam. However, when they sit down and put pen to paper
they completely blank. None of the answers come to them. However when
they again arrived at home they remembered the answers exactly. Guthrie
would predict that the answers or what he called responses did not occur
because the stimuli present in the student’s home, were not present in
the class room environment and therefore the probability of producing
the given response was severely lowered.
This concept can
also explain home field advantage in athletics. The performance of a
given individual is suggested to be dependent on the surrounding stimuli
(Wright and Shea,
1991; Courneya & Carron, 1992). The disadvantage of being on the road
is that the amount of stimuli which elicit a given athletic response are
not present.
The implications
are again to practice in a surrounding environment that is as close to
game time conditions as possible. For example if a game is at night,
teams should practice as much as possible that week underneath the
stadium lights.
Henry’s Specificity
Hypothesis
Henry’s background
was experimental psychology. Being familiar with studies involving the
specificity of intelligence, Henry proposed the Specificity Hypothesis
of Motor Behavior, which suggests that the underlying attributes of a
motor skill or task are specific to that skill or task and not
transferable (task-specific) (Sawyer, 2005). In this context an
attribute can be defined as the underlying capacity within an individual
which allows for the expression of skill. These are presently viewed as
genetically predisposed and not easily modifiable by practice (Sawyer,
2005, Sawyer, Ostarello, Sues, and Demsey 2002, Schmidt and Lee, 1999).
Attributes can be subdivided into body configuration attributes such as
height, intellectual attributes, and motor attributes which are
important for the expression of movement behavior such as various types
of movement speeds, and muscle fiber ratios. Henry’s (1958) Specificity
Hypothesis suggests that individuals have an extremely large compilation
of attributes, up to thousands. It was further suggested that each
attribute was independent and not effected by the capacity of any other
attribute. In this context, each individual contained a collection of
genetically predisposed attributes which varied in strength from poor to
extremely strong. The final aspect of the Specificity Hypothesis posits
that each task or skill is dependent on a great compilation of
attributes. By changing the task, the underlying collection of
attributes which support the task are suggested to change in order to
support the new task. The evidence supporting the Specificity Hypothesis
is overwhelming, and reviewed in part six of this series. It is so
extensive that Sawyer (2005) suggests that it is one step below being
classified as a Law.
The predictions made by the Specificity Hypothesis are that correlations
between any two tasks should be low. It is critical to understand that
Henry (1958) proposed that these attributes were task specific. A
classic example is a study by Drowatzky and Zuccato (1967) who
investigated participants’ performances on six different balance tasks.
The study was meant to investigate whether an underlying attribute such
as balance existed. It was found that specificity between balance tasks
ranged from 96 % to 90 %! This suggests that balance is not general in
nature but task-specific.
According to Henry
(1958) the greatest gains in any practice session will come from
practicing the criterion task itself. However, recently Ostarello
(2005) suggested that individuals should realize that Specificity is not
an all or none phenomenon. Going back to the concept of a negatively
accelerated learning curve, participants should realize that as they
progress in experience, they will receive less return with greater
effort. In this context, a small amount of transfer can provide a
significant benefit (Sawyer, Ostarello, Sues, and Demsey 2002). This
concept is investigated further in this series when conditions of
practice are discussed.
The Specificity
Hypothesis is critical and useful in that it identifies the elements
which were missing in Thorndike’s and Woodworth’s Identical Elements of
Transfer Theory. It also suggests that the similarity of the collection
of attributes supporting any two tasks will converge in so much as the
tasks themselves are similar.
Similarity of any
two tasks can be subdivided into numerous categories. However, five are
typically discussed. These factors include the Rate at which the task
is performed, the Resistance provided through the task, the Pattern of
the task, the Environment the task is performed in, as well as the
underlying Processes which occur during task execution.
Conclusion
The purpose of
this paper was to identify the dominant theories explaining the
Specificity phenomenon. Pavlov suggested that the specificity of any
two tasks was a reflection of the proximity between the neurological
structures controlling the two tasks. Hull posited that performance was
made of a number of sub elements, such as internal and external states.
The closer these states were from practice to the criterion task the
greater the transfer between any two tasks. Guthrie introduced a theory
which suggested that learning involved practicing a particular response
in the context of specific stimuli and that the transfer between any two
environments would be proportional to the number of shared stimuli
within those two environments. Finally, Henry (1958) proposed the
Specificity Hypothesis which suggests that the underlying attributes of
an act or task are specific to that act or task and not transferable
(task-specific). In the above theories, the commonality found is within
the concept that transfer of training is proportional to the similarity
between any two tasks.
References and Sources Cited
Robertson, S., &
Elliott, D. (1996). Specificity of learning and dynamic balance.
Research Quarterly for Exercise and Sport, 67,69-75.
Elliott, D., &
Jaeger, M. (1988). Practice and the visual control of manual aiming
movements. Journal of Human Movement Studies, 14, 279-291.
Proteau, L.,
Marteniuk, R. G., & Levesque, L. (1992). A sen-sorimotor basis for motor
learning: Evidence indicating specificity of practice. The Quarterly
Journal of Experimental Psychology, 44A, 557-575.
Proteau, L.
(1992). On the specificity of learning and the role of visual
information for movement control. In L. Proteau and D. Elliott (Eds.),
Vision and motor control. (pp. 67-103). Amsterdam: North-Holland.
Elliott, D.,
Lyons, J., & Dyson, K. (1997). Rescaling an acquired discrete aiming
movement: Specific or general motor learning? Human Movement Science,
16, 81-96.
Soucy, M. C., &
Proteau, L. (2001). Development of multiple movement representations
with practice: Specificity vs. flexibility. Journal of Motor Behavior,
33, 243-254.
Bennett, S.J., &
Davids, K. (1995). The manipulation of vision during a powerlift squat:
Exploring the boundaries of the specificity of learning hypothesis.
Research Quarterly for Exercise and Sport, 66,210-218.
Bennett, S.J., &
Davids, K. (1997). The effect of task constraints on the manipulation of
visual information and the implications for the specificity of learning
hypothesis. Human Movement Science, 16, 379-390.
Bennett, S.J., &
Davids, K (1998). Manipulating peripheral visual information in manual
aiming Exploring the notion of specificity of learning. Human Movement
Science, 17, 261-287.
Proteau, L.,
Tremblay, L., & DeJaeger, D. (1998). Practice does not diminish the role
of visual information in on-line control of a precision walking task:
Support for the specificity of practice hypothesis. Journal of Motor
Behavior, 30, 143-150.
Skinner, B. F.
(1974). About behaviorism. (New York: Knopf)
Skinner, B. F.
(1978). Why don't we use the behavioral sciences? Human Nature, 1,
86-92.
Skinner, B. F.
(1987b). Whatever happened to psychology as the science of behavior?
American Psychologist, 42, 780-786.
Hull.
C. L. Principles of Behavior. Appleton-Century-Crofts, New York, 1943.
Pavlov. I. P.
(1928). Lectures on conditioned reflexes: Twenty-five years of objective
study of the higher nervous system activity (behavior) of animals. Vol.
1. W. H. Gantt (Trans. & Ed.), N.Y.: International Publishers.
Pavlov, I. P.
(1930/1955). Selected works. Moscow: Foreign Languages Publishing House.
Pavlov, I. P.
(1932/1955). Selected works. Moscow: Foreign Languages Publishing House.
Adams, J.A.
(1961). The second facet of forgetting: A review of warm-up decrement.
Psychological Bullitin, 58, 257-273
D.M. Landers, The
arousal-performance relationship revisited. Research Quarterly for
Exercise and Sport, 1980, 51, 77-90
Swinnen, S. P.
(1996). Information feedback lot motor skill learning: A review. In H.
N. Zelaznik (Ed.), Advances in motor learning and control (pp. 37-65).
Champaign, IL: Human Kinetics.
Swinnen,
S. P., Schmidt, R. A., Nicholson, D. E., & Shapiro, D. C. (1990).
Information feedback for skill acquisition: Instantaneous knowledge of
results degrades learning. Journal of Experimental Psychology: Learning,
Memory, and Cognition, 16, 706-716.
Salmoni, A. W.,
Schmidt, R. A., & Walter, C. B. (1984). Knowledge of results and motor
learning: A review and critical reappraisal. Psychological Bulletin, 95,
355-386.
Schmidt, R. A.
(1991). Frequent augmented feedback can degrade learning: Evidence and
interpretations. In J. Requin & G. E. Stelmach (Eds.), Tutorials in
motor neuroscience (pp. 59-75).
Amsterdam:
Kluwer.
Young, D. E., &
Schmidt, R. A. (1992). Augmented feedback for enhanced skill
acquisition. In G. E. Stelmach & J. Requin (Eds.), Tutorials in motor
behavior II (pp. 677-693).
Amsterdam:
North-Holland.
Guthrie, E.R. The
Psychology of Learning: Revised Edition. Harper Bros: Massachusetts.
1952.
Guthrie, E.R. and
F. Powers. Educational Psychology. Ronald Press Co: New York. 1950.
Fitts,
P. M., & Posner, M. I. (1967). Human performance.
Monterey, CA:
Brooks/Cole.
Baloff, N.
"Extensions of the Learning Curve—Some Empirical Results." Operations
Research Quarterly, 22.44 (1971).
Wright DL, Shea
CH.
Contextual dependencies in motor skills. Mem Cognit. 1991
Jul;19(4):361-70.
Courneya, K. S., &
Carron, A. V. (1992). The home advantage in sport competitions: A
literature review. Journal of Sport & Exercise Psychology, 14, 13-27.
© ABC Bodybuilding Company. All rights reserved.
Disclaimer |
