KLI for Evolution and Cognition Research, Altenberg, Austria.
J Theor Biol. 2010 May 21;264(2):450-6. doi: 10.1016/j.jtbi.2010.02.015. Epub 2010 Feb 16.
Evolutionary game theory has shown that human cooperation thrives in different types of social interactions with a PD structure. Models treat the cooperative strategies within the different frameworks as discrete entities and sometimes even as contenders. Whereas strong reciprocity was acclaimed as superior to classic reciprocity for its ability to defeat defectors in public goods games, recent experiments and simulations show that costly punishment fails to promote cooperation in the IR and DR games, where classic reciprocity succeeds. My aim is to show that cooperative strategies across frameworks are capable of a unified treatment, for they are governed by a common underlying rule or norm. An analysis of the reputation and action rules that govern some representative cooperative strategies both in models and in economic experiments confirms that the different frameworks share a conditional action rule and several reputation rules. The common conditional rule contains an option between costly punishment and withholding benefits that provides alternative enforcement methods against defectors. Depending on the framework, individuals can switch to the appropriate strategy and method of enforcement. The stability of human cooperation looks more promising if one mechanism controls successful strategies across frameworks.
进化博弈论表明，人类合作在具有 PD 结构的不同类型的社会互动中得以蓬勃发展。模型将不同框架内的合作策略视为离散实体，有时甚至是竞争者。虽然强互惠性因其能够在公共物品博弈中击败背叛者而被称赞为优于经典互惠性，但最近的实验和模拟表明，在 IR 和 DR 博弈中，昂贵的惩罚未能促进合作，而经典互惠性则成功了。我的目的是表明，不同框架内的合作策略能够进行统一处理，因为它们受共同的基本规则或规范支配。对一些具有代表性的合作策略的声誉和行动规则的分析，无论是在模型中还是在经济实验中，都证实了不同框架共享一个条件行动规则和几个声誉规则。共同的条件规则包含了在昂贵的惩罚和保留利益之间的选择，为对抗背叛者提供了替代的执行方法。根据框架的不同，个人可以切换到适当的策略和执行方法。如果一种机制能够控制不同框架下成功的策略，那么人类合作的稳定性看起来就更有希望。
