Bad Apples, Worse Apples? the Effect of Framing Social Dilemmas on the Bad Apple Effect

Abstract

The present study aimed to clarify differences between participants’ reactions to the manipulation of feedback in two social dilemma types, the Public Good dilemma and the Resource dilemma. In both dilemmas, the group’s four-person composition was manipulated: either the three fictive group members with whom the participant was confronted exhibited fairly unanimous choices (Low variance), or each of the group members exhibited a unique choice pattern – with one group member furthering his or her own interests, one group member furthering the group’s interests and one group member in between (High variance). Participants’ contributions in the Public Good dilemma and harvests in the Resource dilemma were the dependent variables. In the High variance condition, participants were expected to reciprocate on the defective group member’s behavior – an effect that was expected to be stronger in the Resource dilemma than in the Public Good dilemma. The results did not indicate any difference in strength between the two dilemmas. However, a strong interaction between Variance and a stable trait measure, the participants’ Social Value Orientation, was observed. Participants with a Proself value orientation did defect in the High variance condition, while participants with a Prosocial value orientation did not. Moreover, in the High variance condition, Prosocials reported a stronger identification with the group, and a greater obligation to contribute than did Proselfs. The authors speculate that Prosocials perceive that defecting would harm the contributing group members – and as a consequence, increase their contributions to voice their disapproval of the defective group member’s behavior.