You can judge a book by its cover: the sequel.: A kernel of truth in predictive cheating detection

Abstract

In accordance with evolutionary models of social exchange, we suggest the possible existence of a limited predictive cheater detection module. This module enables humans, to a certain extent, to predict how willing another might be to cooperate or not. Using unknown target subjects who had played a one-shot prisoner's dilemma game earlier, we asked participants in two experiments to rate how cooperative these target subjects were. Pictures were taken of the target subjects at three different moments: a neutral-expression picture taken prior to the game, an event-related picture taken during the decision-making moment of a practice round, and an event-related picture taken during the decision-making moment of a proper round. We found that participants in the experiments could accurately discriminate noncooperative pictures from cooperative ones, but only in response to those taken during the proper round. In both neutral-expression pictures and practice-round pictures, identification rates did not exceed chance level. These findings leave room for the existence of a predictive cheater detection module that deduces someone's decision to cooperate from event-related facial expressions.