Sources of variance in short term reactions to media violence
Research on the effects of media violence indicates that, at least under laboratory conditions, observation of violent stimuli leads to increased aggression. Numerous theories have been formulated to account for this effect, and most have received a degree of empirical support. This has led to theoretical fragmentation of the research. It is argued that a basic conceptual framework is required to guide future progress. It has further been shown that a number of factors affect the extent to which reactions are observed. It is argued that any comprehensive theory of the effects of media violence must take account of the role of individual, film and situational factors, and the possible importance of reactions other than aggression. Two similar theories separately advanced by Berkowitz  and Huesmann , based respectively on the priming of an associative network of aggression related responses, and the encoding and retrieval of aggression related scripts, may jointly have the potential to form such a comprehensive conceptual framework. Previous theories may be considered as special extensions of the Berkowitz/Huesmann model. A series of experiments was performed to test the predictions of this theory, examining the influence of multiple individual difference and film factors on groups of conceptually related affective, cognitive and physiological responses. Findings indicated that violent films prime responses associated with both aggression and anxiety. These results are better interpreted within Berkowitz's later  model of the formation of anger, and suggest that Huesmann's model should be expanded to include anxiety related scripts. It was established that both film and individual factors influence the nature and intensity of viewer reactions, but the way in which they do this is not entirely clear. Ways are suggested in which future research may provide information abut how this occurs.