Person perception and the Berkowitz paradigm for the study of aggression

Three experiments in person perception were conducted to investigate the conditions under which naive observers label an actor as aggressive and to ascertain how this label affects the reactions of the observers to the actor. These studies were performed by exposing subjects to written descriptions of Berkowitz's much-used paradigm for the study of aggression in order to additionally demonstrate that naive observers do not label in the same way as do research psychologists. In Experiment 1 the hypothesis was confirmed that harm-doing behavior is labeled as aggression only when it is antinormative. When harm doing was justified by the instigation or attack of another person, its use did not lead to the actor being labeled as aggressive. In Experiment 2 it was found that violation of the limits imposed by the experimenter on the number of shocks that could be used by actors led observers to characterize them as aggressive even when their action was justified by the prior instigation of the other party. Violation of two norms led to a more aggressive rating of the actor than did the violation of only one norm. Experiment 3 showed that actors who are labeled as aggressive are disapproved of and avoided by observers. The results of these studies are discussed in terms of the ecological validity of Berkowitz's research paradigm and metatheoretical implications for the scientific study of harm-doing behavior.