Researchers from the Max Planck Institute in Plön show that reputation plays a key role in determining the reward policies people adopt. Using game theory, they explain why individuals learn to use rewards to specifically promote good behavior. Their article appears in Nature Communication.
Often we use positive incentives as rewards to promote cooperative behavior. But why do we primarily reward cooperation? Why is defection rarely rewarded? Or more generally, why do we bother to engage in any form of reward in the first place? Theoretical work by researchers Saptarshi Pal and Dr. Christian Hilbe of the Max Planck research group “Dynamics of Social Behaviour” suggests that reputation effects may explain why individuals learn to reward socially.
With tools from evolutionary game theory, researchers are building a model where individuals in a population (players) can adopt different cooperation and reward strategies over time. In this model, the reputation of the players is a key element. Players know, with some certainty (characterized by the transmissibility of population information), how their interaction partners will react to their behavior (i.e., what behaviors they deem worthy of rewards).
If the transmissibility of information is high enough, players learn to reward cooperation. In contrast, without sufficient information about their peers, players refrain from using rewards. The researchers show that these reputation effects also play out in a similar way when individuals interact in groups of more than two individuals.
enriching antisocial
In addition to highlighting the role of reputation in catalyzing cooperation and social reward, scientists identify a few scenarios where antisocial reward can evolve. The antisocial reward requires either the populations to be matched or the rewards to be mutually beneficial to the recipient and the provider of the reward. “These conditions under which people can learn to reward defection are, however, somewhat constraining since they additionally require information to be scarce,” adds Saptarshi Pal.
The results of this study suggest that rewards are effective in promoting cooperation only when they can induce individuals to act opportunistically. These opportunistic actors cooperate only when they anticipate a reward for their cooperation. Greater transmissibility of information increases both the incentive to reward others for their cooperation and the incentive to cooperate in the first place.
Overall, the model suggests that when people reward cooperation in an environment of high information transmissibility, they ultimately benefit themselves. This interpretation eliminates altruism from social reward – people may not use rewards to improve the well-being of others, but to help themselves.
Picnickers beware: Incentives to Co-op are upon us
Saptarshi Pal et al, Reputation effects drive co-evolution of cooperation and social reward, Nature Communication (2022). DOI: 10.1038/s41467-022-33551-y
Provided by the Max Planck Society
Quote: Why do we learn to reward cooperation? (2022, October 27) retrieved October 28, 2022 from https://phys.org/news/2022-10-reward-cooperation.html
This document is subject to copyright. Except for fair use for purposes of private study or research, no part may be reproduced without written permission. The content is provided for information only.
#learn #reward #cooperation