Use of the Premack principle in controlling the behavior of nursery school children
Premack’s principle (Premack, 1959) can be
stated: if behavior B is of higher probability
than behavior A, then behavior A can be made
more probable by making behavior B contingent upon it.
In a preliminary exploration of nursery
school procedures, three 3-yr-old subjects (Ss)
were available three hours a day, five days a
week, for about one month. On the first day,
in the absence of any aversive control, verbal
instructions usually had little effect on the
Ss’ behavior. When they were instructed to sit
in their chairs, Ss would often continue what
they were doing-running around the room,
screaming, pushing chairs, or quietly working
jigsaw puzzles. Taking Premack seriously, such
behaviors were labeled as high probability behaviors and used in combination with the
signals for them as reinforcers. These high
probability behaviors were then made contingent on desired behaviors. For example,
sitting quietly in a chair and looking at the
blackboard would be intermittently followed
by the sound of the bell, with the instruction:
“Run and scream.” The Ss would then leap to
their feet and run around the room screaming.
At another signal they would stop. At this
time they would get another signal and an
instruction to engage in some other behavior
which, on a quasi-random schedule, might be
one of high or low probability. At a later
stage, Ss earned tokens for low probability behaviors which could later be used to “buy” the
opportunity for high probability activities.
With this kind of procedure, control was
virtually perfect after a few days. For example,
when Ss were requested to “sit and look at the
blackboard” (an activity which in the past
had intermittently been interrupted by the
signal for some higher probability behavior),
they were under such good control that an
observer, new on the scene, almost certainly
would have assumed extensive aversive control was being used.
An examination of high probability behaviors quickly showed that many, if not most
of them, were behaviors which ordinarily
would be suppressed through punishment. Extrapolating from this we were able to predict
the reinforcing properties of some behaviors
which had never been emitted. For example,
throwing a plastic cup across the room and
kicking a waste basket had never been observed but proved to be highly reinforcing
activities after they had once been evoked by
instructions. (Some unpredicted behaviors
proved to be highly reinforcing, e.g., pushing
the experimenter around the room in his
caster-equipped chair.)
In summary, even in this preliminary, unsystematic application, the Premack hypothesis
proved to be an exceptionally practical principle for controlling the behavior of nursery
school Ss.
Citation: Homme LE, Debaca PC, Devine JV, Steinhorst R, & Rickert EJ. 1963. Use of the Premack principle in controlling the behavior of nursery school children. Journal of the Experimental Analysis of Behavior, 6, 544.