| For three decades, behavioral economists and social psychologists studying experimental public good dilemmas have consistently observed that contributions to public goods decline over time. Game theorists have simultaneously made the normative argument that, in social dilemmas with fixed end-points, cooperative behavior is "irrational." This dissertation presents the results of three studies that challenge both the generality of the previous empirical findings and the presumed efficacy of the normative prescription by identifying a key condition under which cooperation can be sustained and effective. Using archival data, Study 1 documented a "consistent contributor effect"---how a consistent cooperator influences others, resulting in higher contribution levels within their groups, and superior outcomes for consistent contributors across groups. Study 2 experimentally manipulated the presence of a consistent contributor (CC) and replicated the findings from Study 1, thereby strengthening causal inference. Study 3 tested the impact of consistent contributor characteristics on their efficacy in yielding the CC effect. Smart "high status" CCs were found to have a marginally greater impact on others' choices than "lower status" CCs did, but both high and low status CCs elicited the effect relative to control groups. Although both high and low status CCs elicited the effect, only high status CCs were given credit by their fellow group members for being influential. Taken together, the behavioral results and the results of supplementary questionnaire data suggest that the mechanisms driving the consistent contributor effect include the norm of reciprocity (experience of social obligation), the establishment and encouragement of cooperative social norms, and the material and psychological reduction of risk perceptions for CCs' fellow group members. Results have implications for how we think about what constitutes "rational choice," and how we implement and teach strategic analysis and action. |
