Biased face recognition in the Prisoner's Dilemma game

Students were asked to participate in a series of single-shot Prisoner's Dilemma games, Each subject chose a strategy before the experiment began and was required to play the same strategy with each opponent, These ''opponents'' were photographic reproductions of young women and men, accompanied by labels indicating whether they were cooperators or defectors, Subjects also rated the attractiveness of the opponents, One week later, subjects again rated the photographs without descriptions, Half of the photographs had been seen previously and half were new, Subjects were asked to report which of the photographs they remembered from the previous week, Effects of the four factors (strategy and sex of opponents, and strategy and sex of subjects) on the number of photographs recognized were investigated, The strategy and sex of subjects had no apparent effects, whereas strategy and sex of opponents had a significant interactive influence: Males who had been portrayed as defectors and both groups of females were recognized at higher rates than males who had been portrayed as cooperators, The present study indicates that the biased face recognition of potential cheaters is too robust to be affected by the strategy and sex of subjects, whereas a difference results from the sex of faces presented to subjects, (C) 1997 Elsevier Science Inc.