Group correspondence biases and the provision of public goods.

Two experiments examined whether trait inferences and cooperative decisions in a social dilemma are influenced by biased perceptions of the group's past success or failure in resolving the dilemma. In each experiment, Ss joined a group that had previously either succeeded or failed to provide a step-level public good. Ss were informed of (1) the number of members who had contributed to the good, and (2) the investment quota, defined as the number of contributors needed for the good to be provided. The number of group members who had contributed was held constant, and the past success or failure of the group was determined by varying the severity of the quota. Ss in both experiments displayed the correspondence bias, judging successful groups as more cooperative and competent than unsuccessful ones. Moreover, past group success tended to engender cooperation in subsequent public goods problems, particularly those featuring severe or unknown quotas.