Evolutionary games and population dynamics

理解合作的进化是进化生物学的核心挑战，最近被《科学》杂志评为25个最重要的悬而未决的科学问题之一。合作在自然界中无处不在，从细菌群体到人类社会。然而，合作是一个难题，而剥削不是。合作者会产生成本，并为他人提供利益，而作弊者则获得利益，但避免成本。因此，达尔文式的选择应该有利于作弊者，尽管一群合作者比一群作弊者过得更好。合作的问题归结为群体和个人之间的利益冲突。已经花费了大量的理论和实验努力来研究能够解决这一困境和支持合作演变的机制。基于进化博弈论的数学框架，这一建议旨在扩展我们对三种不同机制的理解。首先，研究生态和进化动力学之间的相互作用将增强我们对进化过程的理论理解。在空间环境中，生态学和进化之间的反馈似乎通过启动种群密度的稳定空间异质性来支持合作，并暗示了自发栖息地多样化的可能机制。其次，空间扩展的概念很容易推广到任意的人口结构，特别是社会网络。社会互动模式不是一成不变的，而是随着时间的推移而变化的。了解社交网络的动态和新兴结构，为研究eBay、维基百科或公共论坛等虚拟社区的形式和功能创造了良好的机会。第三，惩罚、奖励和声誉在执行或鼓励合作方面的作用是模范的，并与人类和动物种群有关。从人类的角度来看，这为建立制裁机构以及为什么要把大多数人类社会建立在惩罚而不是建立在更具社会吸引力的奖励机制上提供了一种进化的观点。

Understanding the evolution of cooperation represents a core challenge to evolutionary biology and was recently selected among the 25 most important unresolved scientific questions published in Science. Cooperation is ubiquitous in nature, ranging from bacterial colonies to human societies. However, cooperation is a conundrum, whereas exploitation is not. Cooperators incur costs and provide benefits to others, whereas cheaters reap the benefits but avoid the costs. Hence, Darwinian selection should favour cheaters, despite the fact that groups of cooperators are better off than groups of cheaters. The problem of cooperation reduces to a conflict of interest between the group and the individual. Much theoretical and experimental effort has been expended to investigate mechanisms capable of resolving the dilemma and supporting the evolution of cooperation. Based on the mathematical framework of evolutionary game theory, this proposal aims at extending our understanding of three distinct mechanisms. First, investigating the interplay between ecological and evolutionary dynamics will enhance our theoretical understanding of evolutionary processes. In spatial settings, the feedback between ecology and evolution appears to support cooperation by initiating stable spatial heterogeneity in population densities and suggests a possible mechanism for spontaneous habitat diversification. Second, the concept of spatial extension is readily generalized to arbitrary population structures and, in particular, to social networks. Social interaction patterns are not static but change over time. Understanding the dynamics and the emerging structures of social networks creates promising opportunities to study form and function, for example, of virtual communities, such as eBay, Wikipedia or public forums. Third, the role of punishment, reward and reputation in enforcing or encouraging cooperation is modeled and related to human and animal populations. In a human context, this provides an evolutionary perspective on the establishment of sanctioning institutions as well as on the reasons for building most human societies on punishment instead of on socially more appealing rewarding mechanisms.