文章基本信息

标题：Pairwise interact-and-imitate dynamics
本地全文：下载
作者：Ennio Bilancini ; Leonardo Boncinelli ; Nicola Campigotto 等
期刊名称：Scientific Reports
电子版ISSN：2045-2322
出版年度：2021
卷号：11
DOI：10.1038/s41598-021-92512-5
语种：English
出版社：Springer Nature
摘要：This paper introduces and studies a class of evolutionary dynamics—pairwise interact-and-imitate dynamics (PIID)—in which agents are matched in pairs, engage in a symmetric game, and imitate the opponent with a probability that depends on the difference in their payoffs. We provide a condition on the underlying game, named supremacy, and show that the population state in which all agents play the supreme strategy is globally asymptotically stable. We extend the framework to allow for payoff uncertainty, and check the robustness of our results to the introduction of some heterogeneity in the revision protocol followed by agents. Finally, we show that PIID can allow the survival of strictly dominated strategies, leads to the emergence of inefficient conventions in social dilemmas, and makes assortment ineffective in promoting cooperation.