Logan: Principles of Learning

             PRINCIPLE: STIMULUS-RESPONSE CONTIGUITY

          There is an increased likelihood that the organism will 
          repeat the last response made on the previous occasion
          of a stimulus event.

     This principle is, to S-R theory, what sensory conditioning is to
S-S theory.  Whereas the latter asserts that stimuli become associated
simply as a result of occurring contiguously in space and time, the
former asserts that a response becomes associated with a stimulus simply
as a result of practice.  Just as classical conditioning might be
interpreted as a particular case of sensory conditioning, so equally
might it be interpreted as a particular case of S-R contiguity.  That is
to say, the paradigm of classical conditioning could be viewed as one way
in which to insure that the response of interest does indeed occur in the
context of the antecedent stimulus.  The essence of the principle is that
the temporal contiguity of a response and a stimulus is sufficient to
produce learning.

     This is not an easy principle to demonstrate experimentally, but an
illustrative study would involve placing a rat in a running wheel that is
wired so that any time that the rat runs, a tone sounds.  In such a
study, there is no US that reflexively elicits running behavior; we have
only arranged for a tone to sound whenever the rat is "spontaneously"
running.  Then, after some period of exposure to these circumstances, the
tone is sounded when the rat is not running.  There is evidence that the
rat will start to run after the tone sounds.  The somewhat tenuous nature
of this type of support for the Principle of S-R Contiguity is that the
tone is not sounded until after the rat has started to run; a more ideal
but technically impossible procedure would be to have the tone come on
just before the rat spontaneously began to run.

     It is important to recognize that for all S-R theorists, S-R
contiguity is a necessary condition for learning.  At issue is whether
contiguity is sufficient.  Those who would deny the principle of S-R
contiguity contend that some other conditions must also be satisfied for
learning to occur.  Specifically, the most common alternative theory
contends that, in addition to S-R contiguity, reinforcement is also
necessary for learning.  The popularity of S-R reinforcement theory dates
back to the 1920's with Thorndike's statement of the Law of
Effect, that organisms tend to learn responses that are followed by
"satisfying" consequences.  Subsequently, however, almost all learning
theorists have accepted the sufficiency of the contiguity principle, be
it the S-R or the S-S version.  You learn what you practice, regardless
of whether or not you want to learn it and regardless of whether or not
you are reinforced for doing so.

     As might by now be anticipated there is an alternative explanation
of the S-R contiguity principle.  The reason for repeating this
controversial thread is not because one or the other approach is right or
better than the other, but because viewing the principle from
contrasting perspectives may help to increase one's understanding of the
principle itself.  In this case it is necessary to introduce the
important concept or response-produced feedback.  It is presumed that
whenever a response occurs, it is accompanied by some form of response
occurs, it is accompanied by some form of proprioceptive or kinesthetic
stimuli from the glands, muscles, and joints involved in that response.
The important function of this feedback is to inform the organism of the
behavior that is taking place.  We shall have occasion to refer to such
response-produced feedback throughout this book.

     As regards the S-R contiguity principle, it is only necessary to
recognize that the exteroceptive stimulus occurs contiguously with
the feedback stimulus.  Specifically, at the same time that the rat feels
himself running, he hears the tone.  Hence, what at first glance to be
S-R think about running.  This does not necessarily mean that the rat
will actually start to run, but there is a strong tendency to do whatever
we are thinking about doing.  It is known, for example, that when humans
are instructed to think about moving their fingers, just the
thought will lead to small muscular contractions in the fingers.

     A special case of the S-R contiguity principle occurs in response
chains.  We refer to a BEHAVIOR CHAIN when the feedback from each
response sets the occasion for the next response.  Almost all responses
can be viewed as a chain of more molecular units; for example, a rat's
bar-press entails approaching the bar, raising up on the hind legs,
placing the front paws on the bar, pressing down, and then releasing.
The entire sequence becomes integrated as a result of S-R contiguity,
where the S is response-produced feedback of the preceding response.  In
similar fashion, a great many human responses, from pronouncing a word to
executing a dance step, are behavior chains that run off as each
component action, in turn, leads on to the next action.

     In spite of its apparent simplicity the S-R contiguity principle has
wide ranging practical implications.  It says that mere repetition
will make any response habitual.  This applies not only to motor behavior
but also to mental activities.  For example, alertness, worrying,
daydreaming, and preoccupation with sexual, aggressive, or belittling
thoughts can become habitually associated with various contexts.  Every
time we engage in one or another form of behavior, we increase the
likelihood of continuing to engage in it.  As with all basic Principles
of Learning the S-R contiguity principle is indifferent as to whether the
behavior is desirable or undesirable, adaptive or maladaptive, stupid or
intelligent.  Every lie is toward the making of a liar; every crime is
toward the making of a criminal; every cigarette is toward the making of a
smoker.  But fortunately in such situations, the principle is reversible;
not engaging in previously habitual behavior can make not engaging in it
also habitual.

TERMS: Contiguity (s-r), sensory conditioning, response-produced
feedback, Law of Effect, response, stimulus, exteroceptive stimulus,
interoceptive stimulus, reinforcement theory.