Variance, norms and cooperative behavior in public goods games

This study examines the relationship between the variance of others' contributions, social norms (empirical and normative expectations), and cooperative behavior using a classic linear public goods game. The following results are observed. First, the variance of a participant's group members' contributions had a negative impact on their contributions, empirical expectations, and normative expectations. Second, deviations from the mean, whether negative or positive, were deemed less socially appropriate. Third, while there was a strong relationship between variance, social norms, and cooperative behavior, the mediating effect of social norms was found to be insignificant. Finally, there were some notable findings regarding behavior type. Although free riders and cooperators exhibited distinct behavioral patterns, their normative expectations were similar. Free riders expected others to cooperate, but their empirical expectations were significantly lower than cooperators' expectations, which were aligned with their actual contributions. These findings contribute to research on the relationship between distribution heterogeneity, social norms and cooperative behavior. Furthermore, these findings provide valuable insights into management practices.

Cooperation in Networked Collective-Action Groups: Information Access and Norm Enforcement in Groups of Different Sizes

Norms, typically enforced via sanctions, are key to resolving collective-action problems. But it is often impossible to know what each individual member is contributing to group efforts and enforce cooperation accordingly. Especially as group size increases, people commonly have access to the behaviors of—and can sanction—only those to whom they are tied in a broader network. Here we integrate two streams of research: one conceptualizing ties in networked collective-action groups as access to information about what others are doing and a second where ties represent information plus opportunities to enforce cooperation via punishment. While both have pointed to the cooperation benefits of more ties in the network, we argue that these benefits will depend on group size and whether ties provide access to information about what others are doing or whether they also entail opportunities for norm enforcement. Our experiment demonstrates that densely tied information networks facilitate cooperation but only when the group size is small. When people can also enforce their ties’ cooperation, however, densely tied networks particularly benefit larger groups. The results demonstrate how network-level properties and individual-level tie patterns intersect to promote contributions in small and large collective-action groups.

Cognitive Blame Is Socially Shaped

Blame is not only a cognitive process but also a social act of moral criticism. Such acts of criticism often serve to correct a transgressor’s behavior but can be costly—to the moral critic, the transgressor, and the community. To limit these costs, blame is socially regulated: Communities set standards of evidence for blame and expect individuals to provide warrant, or justification, for their expressed judgments by pointing to appropriate evidence. We describe the path model of blame, which captures the cognitive processes that underlie blame judgments and that specify the kind of evidence that counts as warrant for blame. We then show how the varying costs of blaming put social pressure on the moral critic to be accurate and fair. We also identify conditions under which these pressures are weakened and standards of evidence decline: for example, when the transgressor has low status or is an out-group member, when the critic has high status or is anonymous, or when interactions are online. We close with numerous open questions that we hope will inspire further research into the hypothesis that cognitive blame is socially shaped.

Reputations in mixed-role markets: A theory and an experimental test

The traditional understanding of reputation systems is that they secure trust between strangers by publicly calling out cheaters. In modern, online markets, it is increasingly common for providers of a good to also act as consumers, and vice versa. We argue that in such mixed-role markets, reputation systems serve a second important function: They allow providers who lend out their possessions (such as their house, car or tools) to earn reputational credits that can be spent on future borrowing, especially when lending lacks monetary compensation. In an experiment that introduces a new game, "the Lending Game", we show that, consistent with our argument, information on past lending leads subjects to lend to those who have themselves lent before, increasing overall lending. However, when lending is financially compensated, this mechanism of reciprocal lending ceases to operate.

Group leaders establish cooperative norms that persist in subsequent interactions

The temptation to free-ride on others’ contributions to public goods makes enhancing cooperation a critical challenge. Solutions to the cooperation problem have centered on installing a sanctioning institution where all can punish all, i.e., peer punishment. But a new, growing literature considers whether and when the designation of a group leader—one group member, given the sole ability to administer punishment—is an effective and efficient alternative. What remains unknown is whether and to what extent these group leaders establish cooperative norms in their groups via their own contributions to the public good, their use of sanctions, or both. Nor has past work examined whether leaders’ behaviors have lasting effects on non-leaders’ cooperation in subsequent interactions, outside of the leader’s purview. Here I show that leaders’ contributions to the public good predict non-leaders’ subsequent cooperation. Importantly, the effect is not limited to cooperation within the institution: the effect of leaders’ contributions continue to predict non-leaders’ contributions in a later interaction, where sanctions are removed. This process is mediated by non-leaders’ increased contributions in the institution, suggesting that leaders have effects on followers that shape followers’ subsequent behaviors. These effects occur above and beyond a baseline tendency to be influenced by non-leader group members; they also occur above and beyond the influence of peers in groups under a peer punishment institution. Results underscore how critical it is that groups install cooperative leaders: followers model their leaders’ cooperation choices, even in decisions external to the original institution and outside of the leader’s watch.