Educational Psychology—Limitations and Possibilities

B. F. Skinner 203
There are primarily four types of reinforcement schedules: (a) fixed-interval (FI), (b) fixed-ratio
(FR), (c) variable-interval (VI), and (d) variable-ratio (VR). In FI reinforcement, organisms are
given or exposed to reinforcement stimulus on a fixed time schedule. When an organism becomes
conditioned to an FI schedule of reinforcement, its behavior becomes stable. The general rule
with FI reinforcement is that an organism’s rate of responding is inversely proportional to the
interval between reinforcements. In this type of reinforcement schedule, organisms learn that
responses early in the interval are never reinforced immediately and organisms will tend to pace
the responses and “pile up” its responses toward the end of the interval. In an FR schedule,
the reinforcement stimulus is provided after a fixed number of responses have been exhibited
by the organism. With this schedule, the organism learns that rapid responding is important.
There is a direct correlation between the rate of responding and the rate of reinforcement, that
is, the higher the rate of responding the higher the rate of reinforcement. In a VI reinforcement
schedule, time is a critical factor. After an organism has learned a particular response, the
amount of time it takes for the next reinforcement stimulus to be presented keeps changing. It
will not be possible for an organism to learn the time interval accurately. Organisms tend to
respond at an extremely stable rate under the VI schedule. In a VR reinforcement schedule, an
organism is given the reinforcement stimulus after a different number of responses have been
exhibited. In short, variable number of responses is required to produce successive reinforcers.
Reinforcing well-learned behaviors on a VR schedule generate extraordinarily high rates of
performance.
An overview of operant conditioning has been presented. Behavior, which is a series of
stimulus–response connections, is followed by a consequence, and the nature of the consequence
(e.g., presence or absence of reinforcing stimulus) modifies the organism’s tendency to exhibit or
inhibit the behavior in the future.
OPERANT CONDITIONING APPLIED TO EDUCATIONAL PSYCHOLOGY
Most biographical accounts of B.F. Skinner suggest that Skinner’s interest in educational
psychology began on that fateful day of November 11, 1953, Father’s Day, when Skinner visited
his daughter’s fourth-grade arithmetic class. While sitting at the back of his daughter’s classroom,
Skinner observed that the students were not receiving prompt feedback or reinforcement from
their teacher and were all moving at the same pace despite differences in ability and preparation.
Skinner had researched delay of reinforcement and knew how it hampered performance. If
mathematical-problem-solving behavior is perceived as a complex series of stimulus–response
connections that had to be effectively established, then the teacher in Skinner’s daughter’s fourthgrade
arithmetic class definitely needed help. It was simply impossible for the teacher with twenty
or thirty children to shape mathematical-problem-solving behavior in each student. In operant
conditioning theory, the concept of shaping requires that the best response of the organism be
immediately reinforced. In the math class, however, some of the students had no idea of how
to solve the problems, while other students breezed through the exercise and learned nothing
new. Furthermore, the children did not find out if one problem was correct before doing the next
problem. They had to answer a whole page before getting any feedback, and then probably not
until the next day.
That afternoon, Skinner constructed his first teaching machine. The first teaching machine that
was developed by Skinner was a device that presented problems to learners in random order. This
machine simply practiced and rehearsed skills or behaviors already learned. Learners did not
learn any new responses or new behaviors. A few years later Skinner developed and incorporated
programmed instruction into the learning machines. Learners would respond to content to be
learned that were broken down into small steps. The first responses of each content sequence