Instability of Cooperation Based on False Belief: An Experiment with Artificial Supernatural Punishment
preprint
OA: closed
Abstract
The belief in supernatural punishment, wherein individuals falsely attribute accidental misfortunes to “punishment” by supernatural agents, plays a crucial role in sustaining group cooperation. However, examining the transformation of these beliefs and cooperative behaviors over time has proven challenging. This study proposes the concept of artificial supernatural punishment (ASP) to examine the stability and changes in beliefs and cooperation within an experimental framework. Each participant decides the extent to which they wish to cooperate in a public goods game, after which some participants experience a random reduction of points, unrelated to their cooperative efforts. A total of 179 participants who had registered with a crowdsourcing service were assigned to either a random instruction condition, wherein the reduction target was described as randomly determined, or an ASP instruction condition, wherein the reduction targets were explained as either random or proportional to participants’ cooperation levels. That is, in the latter case, less cooperative participants were more likely to be chosen as targets, leading participants to anticipate a link between low cooperation and reductions, even though the selection was random. As a baseline, 93 additional participants played only the public goods game. Results indicated that the ASP instruction facilitated cooperation in the first game round compared to the random instruction; however, this effect was not sustained through the remaining rounds, and overall cooperation levels did not significantly differ from other conditions. Approaches to mitigate the instability of cooperation dependent on beliefs in supernatural punishment are discussed.
My notes (saved in your browser only)
Citation neighborhood (no data yet)
We don't have any in-corpus citations linked to this paper yet. This is a recent paper (2025) — citers typically take a year or two to land, and the OpenAlex reference graph may still be filling in.
Source provenance
- europepmc
- last seen: 2026-05-20T01:45:00.602351+00:00