Premack Principle

Description of the Strategy

By the time the average undergraduate student has completed his or her college career, chances are that he or she has been exposed to the Premack principle virtually dozens of times in the formal classroom setting. Classes such as Learning, Child Development, and Behavior Modification are sure to have brought the student into contact with one of the most universally recognized principles in the field of behavior change. And yet, even prior to embarking on that college career, it is a certainty that most students had been exposed to the Premack principle in a less formal manner—through its application to them by parents, teachers, and perhaps other authority figures seeking to elicit a desired behavior. When a parent withholds a child's playtime with friends until the child has completed homework, the Premack principle is employed. When a parent withholds time on the computer until the child has straightened his or her room, the Premack principle is employed.

The Premack principle states that it is possible to use a high probability behavior, that is, a behavior one regularly chooses to engage in, in order to reinforce participation in a low probability behavior, i.e., a behavior one does not regularly choose to engage in.

David Premack first introduced this concept in the mid-1960s to account for results found in a new experimental procedure he had developed in his research with rats. In Premack's research, a rat was first placed in a box that allowed it to run on a wheel or drink water from a spout as freely as it wanted. Baseline data were collected on the amount of time the rat spent engaging in each behavior. In a second phase, the rat was required to engage in a specified amount of one activity (e.g., drinking) in order to gain the other activity (running). Then the contingencies were reversed, that is, the rat was required to engage in running in order to get drinking. Premack found that when a behavior was more probable (occurred more often than the other behavior) during the initial baseline phase, it could be used to increase (beyond baseline levels) the probability that the less frequent behavior would occur.

Based on this work, Premack suggested that reinforcement can involve contingency between behaviors, blurring the line between response (behavior) and reinforcer. Whereas response and reinforcer were historically thought of as two distinct, separate classes, Premack defined them by their function in a particular situation and their relative probabilities in that situation. A behavior can act as a reinforcer for other behaviors that are lower in probability and can be a reinforceable behavior when contingent on other more probable behaviors. In practice, basing contingencies on the probabilities of behaviors should increase the ability to predict the effectiveness of a reinforcer. If we know that a child freely chooses to draw pictures more often than read books, then we know that the former is more likely an effective reinforcer compared to the latter and that it may even be used to reinforce the latter. Furthermore, the strength of any reinforcer is based on the relative disparity in probability between it and the instrumental behavior. If reading books and playing on the computer are relatively equally enjoyable activities for a child, neither is likely to be a strong reinforcer for the other. However, if reading books is highly preferred to playing on the computer, then, under the Premack principle, its possible role as a reinforcer for computer time is more certain.

...