Developmental Psychology 1972, Vol. 7, No. 3, 266-269 Schedule of Punishment and Inhibition of Aggression in Children 1 ROSS D. PARKE2 AND JAN L. DEUR University of Wisconsin Ninety 8-10-year-old boys served in an experiment concerning the relative effectiveness of inconsistent and consistent punishment for suppressing a hitting response. Persistence was the main dependent index. The data showed that consistent punishment resulted in faster inhibition than incon- sistent punishment; subjects who were punished showed less persistence than subjects placed on an extinction schedule. Variations in the type of instructions accompanying the punishment did not affect the subjects' per- sistence on the task. The relationship between this finding and previous field reports of the ineffectiveness of punishment is briefly noted. The main purpose of this study was to provide information concerning the relative effectiveness of inconsistent and consistent punishment for controlling aggression in children. Although previous research has in- dicated that punishment is an effective means of producing inhibition of aggression (Deur & Parke, 1970; Hollenberg & Sperry, 1951), there is little information concerning the suppressive value of different schedules of punishment. While Deur and Parke found that inconsistent reward and punishment led to increased resistance to continuous punish- ment and enhanced resistance to extinction, there was no direct comparison of the effec- tiveness of different types of schedules of punishment in their study. In fact, nearly all previous investigations of punishment sched- ules have employed animal subjects (e.g., Azrin, 1956; Estes, 1944) and in general have confirmed the prediction that consistent punishment is more effective than intermit- 1 This study was supported in part by Research Grants GS-1847 and GS-31885x from the Na- tional Science Foundation to the first author and by United States Public Health Service Training Grant 144-8920 from the National Institute of Mental Health to the University of Wisconsin. Thanks are extended to Kenneth Stevens, Princi- pal, and the staff of DeForest Elementary School for their cooperation. 2 Requests for reprints should be sent to Ross D. Parke who is now at the Pels Research Insti- tute, Yellow Springs, Ohio 45387. tent punishment—at least while punishment is in effect. The single study comparing con- sistent and inconsistent punishment with children was recently reported by Leff (1969). Using a resistance to deviation par- adigm, this investigator found that consist- ently punished subjects made fewer errors during a discrimination-learning task than intermittently punished children; however, on a subsequent behavioral test for response inhibition, children in the two types of pun- ishment-training groups did not differ. To determine the suppressive value of consistent and inconsistent punishment on another be- havior, namely, aggression, and to relate the findings to earlier work on aggressive behav- ior, the effects of two punishment schedules on suppression of hitting a Bobo doll was examined. It was expected that inconsistent punishment would be a less effective inhibi- tor than consistent punishment. A second purpose was to examine the ef- fects of pairing different verbal instructions with the punishing stimulus—a loud buzzer. In a previous study (Deur & Parke, 1970), subjects were told that the buzzer meant "you are playing the game badly," which could be interpreted as a failure manipula- tion. A nonfailure instruction, namely, that the buzzer was "a bad noise," was included to assess the generalizability of the previous findings to other types of instructional sets accompanying the buzzer-punisher. 266