News Release

Prosocial influencers can promote societal cooperation

Peer-Reviewed Publication

PNAS Nexus

A modeling study suggests that influential neighbors can be as effective as despotic leaders at promoting social cooperation. Prosocial behaviors can be difficult to sustain in large societies over the long term, as people give into the temptation to defect to strategies that prioritize the wellbeing of themselves and their immediate family. Stefani Crabtree and colleagues constructed a general theoretical framework to explore how cooperation could arise and be maintained in a large society. The authors explore three possible mechanisms for encouraging prosocial actions: neighbors that monitor one another for defection, despotic leaders who monitors and punishes at the whole-society level, and influencers that convince neighbors to cooperate. The authors built mathematical models and simulated individuals playing a common-pool resource game. The simulated society was made up of many types of agents; some agents always cooperated; some always defected; some cooperated and monitored their neighbors for cooperation; and some cooperated only after being caught defecting and punished. In some cases, the authors included influencers, who recruited neighbors to behave prosocially if the influencer found that cooperation resulted in higher individual gains for themselves than defection did. This strategy produced similar results in terms of average agent wealth as the strategies of having a strong leader that prevents detected defectors from ever defecting again. Both influencing and despotic leadership performed much better than monitoring by neighbors. The size of the fine for defecting influenced which strategy was most lucrative, with steep fines promoting cooperation. According to the authors, prosociality can be achieved via a multitude of methods, although punishment of defection is always required. 

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.