Pergamon Aggression and Violent Behavior, VoL 1, No. 2, pp. 163-177, 1996 Copyright © 1996 Elsevier Science Ltd Printed in the USA. All rights reserved 1359-1789/96 $15.00 + .00 SSDI 1359-1789(95)00014-3 LIMITATIONS OF LABORATORY PARADIGMS FOR STUDYING AGGRESSION James T. Tedeschi University at Albany, State University of New York Brian M. Ouigley Research Institute on Addictions, Buffalo, New York ABSTRACT. The construct validity of four laboratory paradigms used in studying aggres- sion (the teacher~learner, essay evaluation, competitive reaction time game, and Bobo mod- eling paradigms) is examined. It is argued that the first three paradigms under-represent the construct of aggression because they deal only with situations of retaliation which have been sanctioned by a third party legitimate authority (the experimenter) and because research par- ticipants are given no choice other than physical forms of harm-doing as a means of respond- ing to attacks. Additionally, the teacher~learner and essay evaluation paradigms employ cover stories which make the research participants' intentions and motivations unclear or even counter to the proposed theory. The Bobo modeling paradigm may not examine aggres- sion at all, rather, imitative behavior of "rough and tumble play" in which there is no intent to harm. It is proposed that the focus of research on aggression should be the intentions and motivations of the actor rather than simple attack-retaliation situations. Future research needs to examine the motivations of subjects in the traditional paradigms to determine if they are situations in which participants intend to cause harm. Additionally, in order to examine the full range of phenomena which aggression theorists wish to explain, a multimethod approach combining both laboratory and non-laboratory studies must be utilized. THOUSANDS OF EXPERIMENTS have been performed to examine the causes of aggres- sive behavior since the publication of Frustration and Aggression (Dollard, Doob, Miller, Mowrer, & Sears, 1939). Laboratory research on human aggression has been largely con- fined to four sets of research procedures (or paradigms), which will be described in detail below. Given that so much of current aggression theory depends upon such a limited set of Correspondence should be addressed to James T. Tedeschi, State University of New York at Albany, Department of Psychology, 1400 Washington Avenue, Albany, NY 12222. 163