Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)




Ulrich and Flavell (1970) note that although aggression is more a problem today than it has ever been, and although there have been centuries of concern, little has been done to develop effective controls. One conceptual framework for explaining the occurrence and control of aggressive behavior is social learning theory. Bandura (1973) suggests that aggressive behaviors that are reinforced will be strengthened. Several studies with children support this notion (Cowan & Walters, 1963; Davitz, 1952, & Lovaas, 1961). Those aggressive behaviors that are punished or unrewarded will be weakened or eliminated. Hawkins, Peterson, Schweid, and Bijou (1966) suggested that extinction is one effective technique for eliminating aggressive behavior. Pendergrass (1971) suggested that timeout can be an effective technique for suppression of aggressive, verbal behavior. Support for the effectiveness of punishment as a technique for suppressing aggression was offered by Deur and Parke (1970). There has been no comprehensive study to compare the effectiveness of these three methods.

The present study compared the effectiveness of punishment, timeout, and extinction in suppressing an aggressive response in children. In addition, two levels of both timeout and punishment were compared in order to explore the effects of intensity of these two treatments. The subjects were 61 first, second and third grade males. There were three phases in the study. In the first phase, the acquisition phase, the subjects were trained in an aggressive response (punching a clown's face). After a two minute period in which the subjects acquainted themselves with the apparatus, they received 10 CRT reinforcements (M & M’s), and five FR3 reinforcements for punching. In the second phase of the experiment, the treatment phase, the subjects were exposed to one of five treatments: 30 second timeout, 60 second timeout, low punishment, high punishment, or extinction. None of the subjects received reinforcement during the treatment phase. In the timeout groups a door covered the punching apparatus contingent on each response for a period of either 30 or 60 seconds. The punishment groups received a tone of either low or high intensity contingent on each response. The subjects in the extinction group were allowed to continue to punch as in the previous phase, but they received no reinforcement. The subjects were continued in the treatment phase until they requested to stop (suppressed) or until 30 minutes had passed. All subjects were seen in a follow-up phase one week later in order to assess the persistence of the suppression. The subjects received no reinforcement during this phase and were allowed to punch until they suppressed or until 10 minutes had passed.

Several hypotheses were made. First, it was hypothesized that there would be a difference in the suppression of aggressive responding produced by the treatment methods in the treatment phase. More specifically, it was hypothesized that extinction would lead to less suppression than would either timeout or punishment during the treatment phase. Further, high levels of punishment and timeout were hypothesized to lead to more suppression than low levels. Secondly, it was hypothesized that timeout would lead to more persistence of suppression in the follow-up phase than would punishment. Third, extinction was hypothesized to lead to less persistence of suppression in the follow-up phase than would either timeout or punishment.

The primary hypotheses of this study were only partially supported. Timeout produced more efficient suppression than either punishment or suppression. However, punishment did not produce more suppression than extinction. Further, there were no differences in the suppression produced by high and low levels of either timeout or punishment.

Timeout required fewer trials before suppression than did the other treatment methods. There was a tendency for timeout to require more time in treatment before suppression than other treatments. However, more research is necessary to clarify this tendency. It was suggested that the lack of difference between the two levels of timeout may indicate that short durations of timeout are as effective as longer durations in producing suppression.

There were no significant differences between punishment and extinction on any of the measures of suppression during the treatment phase. The lack of effectiveness of punishment was attributed to the absence of cognitive or verbal structuring accompanying the buzzers. This finding stresses the importance of cognitive structuring in the effectiveness of treatments with humans. Also, it suggests that punishment may not be effective without verbal structuring with human subjects. Further research seems to be necessary to determine whether the impact of punishment comes from the physical act, the instructions accompanying that act, or from some interaction between these two.

Contrary to the third hypothesis there were no significant differences in the persistence of the suppression produced by any of the treatment methods. In spite of the efficiency of the suppression produced by the timeout during the treatment phase, it did not appear to produce more persistent suppression than the other treatments.