Advancing Cooperative Behavior and Its Measurement

• Applicant: Peter Katuscak, Ph.D.
• Subject Area: Economic Policy, Applied Economics
• Term: since 2026
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 569777791

This project advances our understanding of cooperative behavior in social dilemmas through three experimental studies. These studies will inform the design of more effective fundraising mechanisms and organizational structures to improve social efficiency. Study 1 is motivated by the observation that material incentives for voluntary contributions to a public good (PG) might crowd out of pro-social motivations to contribute. In our previous work, we demonstrate this for the case of public good lotteries. The aim of this study is to examine whether a modified mechanism in which participants can upfront (before deciding on their contribution) commit to forgo the lottery prize and instead contribute it to the PG can increase public good provision. This could happen because the mechanism aims to preserve the pro-social motivations for those driven by such motivations while, at the same time, preserving material incentives for those driven by such incentives. Study 2 investigates whether peer punishment increases cooperation in multi-level public good games (MPGGs) like it does in single-PG games. In an MPGG, individuals can allocate resources to three uses: a private account, one of local PGs, each benefiting a small group, and a global PG benefiting everyone. Marginal social efficiency of a contribution increases and its marginal private benefit decreases in this order. If peer punishment increases cooperation, does it increase provision of the more efficient global PG or the less efficient set of local PGs? How does the answer depend on the beneficiary group size of the local PGs? We hypothesize that for small beneficiary groups with high per-person benefits, peer punishment might be so effective in enforcing local cooperation that it could be difficult to transition toward global cooperation, and vice versa. Study 3 addresses the documented puzzle that conditional cooperation (CC, meaning the more the others contribute to a PG on average, the more one contributes) exists if the PG game is played with unpaid computerized contributors. We hypothesize that this puzzle can be explained by how CC is usually measured in experiments. We have documented in our previous work that the usual measure is upwardly-biased in human-human interaction and suggested an alternative that is free of the bias. In this study, we propose to apply this alternative measure to human-computer interaction with the aim to see whether it can resolve the puzzle.

• DFG Programme: Research Grants
• International Connection: Czech Republic
• Cooperation Partner: Tomas Miklanek, Ph.D.