This article describes computer simulations in which pairs of ''individuals'' in large groups played a prisoners' dilemma game. The individual's choice to cooperate or not was determined by 1 of 3 simple heuristics: tit-for-tat; win-stay, lose-change; or win-cooperate, lose-defect. Wins and losses were determined through the comparison of a play's outcome with the average outcome of the individual's neighbors. The results revealed qualitative differences between small and large groups. Furthermore, the prevalence of cooperation in the population depended in predictable ways on the heuristic used, the values of the payoff matrix, and the details of the social comparison process that framed the outcomes as wins or losses.
- SOCIAL IMPACT