ABSTRACT
One of the classic issues in social science concerns whether decent people are prone to behave indecently when banded together in a group (G. Allport, 1985). Our approach to this issue involves a contrast between interindividual and intergroup behavior in the context of a matrix game (typically a prisoner’s dilemma game [PDG]), in which communication between players is allowed (Insko, Schopler, Hoyle, Dardis, & Graetz, 1990; Insko et al., 1987, 1988, 1992, 1993, 1994; McCallum et al., 1985; Schopler, Insko, Graetz, Drigotas, & Smith, 1991; Schopler et al., 1993, 1994, 1995). The majority of these studies were conducted in a suite in which individuals or groups were located in different “home” rooms that were connected to a central room. After examining a version of a PDG matrix provided for a given trial, individuals or group representatives (or in some instances entire groups) went to the central room to discuss possible action with their opponent (or opponents) and then returned to their home rooms, where a choice was made. In the groups condition, the choice was made by the group as a whole. The final decision was recorded on a form, which was carried back to the central room and given to the experimenter. The experimenter announced the decisions by the two individuals or the two groups, and distributed the money that had been earned. Although there were a few studies involving just one trial, typically the procedure was repeated for 10 trials.
