How to evaluate one's behavior toward 'bad' individuals? Exploring good social norms in promoting cooperation in spatial public goods games. (23rd September 2020)

• Record Type: Journal Article
• Authors: Quan, Ji; Qin, Yu; Zhou, Yawen; Wang, Xianjia; Yang, Jian-Bo
• Is Part Of: Journal of statistical mechanics. (2020:Sep.)
• Publication Date: 2020-09-23
• Subjects: agent-based models -- evolutionary game theory; Statistical mechanics -- Periodicals; Mechanics -- Statistical methods -- Periodicals; 530.1305
• DOI: 10.1088/1742-5468/abb236
• Languages: English
• ISSNs: 1742-5468

Abstract

Abstract: Cooperation in modern society is based on mutual trust and reputation information can well reflect people's social image. Good social norms encourage people to cooperate with high-reputation (or called 'good') individuals and severely punish those who defect 'good' individuals. However, the existing research lacks a unified criterion of how to evaluate the behavior when interacting with low-reputation (or called 'bad') people. Based on public goods games, we focus on exploring second-order social norms that can improve the evolution of cooperation in this study. Parameters to control the reward and punishment intensity through increasing or decreasing one's reputation when interacting with 'bad' individuals are introduced. Simulation results show that this reward and punishment mechanism through reputation can effectively promote the evolution of cooperation in both well-mixed and latticed populations. Under certain high-reputation threshold environments, a well-mixed population structure can even promote cooperation more significantly than a lattice network. However, increasing reward intensity for cooperating with 'bad' individuals cannot further improve cooperation, but in a high-reputation threshold environment, increasing punishment intensity for defecting 'bad' individuals can further improve cooperation. This research extends the use of statistical physics to study the evolution of cooperation from the perspective of reputation-based dynamics.