亲社会行为
公共物品游戏
公共物品
社会困境
人口
社会心理学
困境
囚徒困境
不平等
强互惠
利他主义(生物学)
心理学
微观经济学
博弈论
经济
社会学
非合作博弈
认识论
哲学
人口学
数学分析
数学
作者
Alex McAvoy,Benjamin Allen,Martin A. Nowak
标识
DOI:10.1038/s41562-020-0881-2
摘要
Prosocial behaviours are encountered in the donation game, the prisoner's dilemma, relaxed social dilemmas and public goods games. Many studies assume that the population structure is homogeneous, meaning that all individuals have the same number of interaction partners or that the social good is of one particular type. Here, we explore general evolutionary dynamics for arbitrary spatial structures and social goods. We find that heterogeneous networks, in which some individuals have many more interaction partners than others, can enhance the evolution of prosocial behaviours. However, they often accumulate most of the benefits in the hands of a few highly connected individuals, while many others receive low or negative payoff. Surprisingly, selection can favour producers of social goods even if the total costs exceed the total benefits. In summary, heterogeneous structures have the ability to strongly promote the emergence of prosocial behaviours, but they also create the possibility of generating large inequality. Prosocial behaviours are ubiquitous in nature. These building blocks of cooperative societies can come in many forms, depending on how the underlying social good is produced and distributed. In this study, the authors show that heterogeneous populations can strongly promote the evolution of prosocial behaviours. However, this efficient evolution reveals a thorny side of prosocial behaviours: they generate the possibility of widespread wealth inequality, even to the point of being a detriment to the poorest in the population. The authors provide a general framework that can be used to understand when this harmful prosociality will emerge in a population. These findings suggest that institutional interventions are often essential for maintaining equitable outcomes in heterogeneous societies.
科研通智能强力驱动
Strongly Powered by AbleSci AI