Theories of Cooperative Behavior III

这个奖项是基于成就的更新。首席研究员之前的研究已经建立了在简单环境中开始和维持合作的条件。之前研究的环境包括两人囚徒困境和n人规范博弈。在过去的五年中，这项工作在国际政治经济学、反垄断法、政府监管、经济理论、人工智能等领域得到了广泛的应用。目前的研究调查了集体属性（如合作或冲突）是如何从复杂群体结构中相互适应的个体的行为中产生的。该方法是建立一系列平行的全种群自适应模型。这些模型将是并行的，因为所有的个体将同时处于活动状态。模型将是适应性的，因为个体将根据自己的目标，调整自己的行为，以适应他们从他人那里体验到的影响。而不是假设调整过程是基于理性的计算，研究者将假设它是基于从经验中学习或相对成功的策略的差异生存。这项研究对社会科学很重要，因为标准的理论和数学方法传统上假设一次只发生一到两件事。在政治和社会上，许多事件同时发生。特别是，许多人做出相互依赖的选择，影响彼此的环境。因此，对平行适应模型的理解有可能增强我们对各种政治过程的洞察力，例如：政治领导，公众参与，国内对外交政策的影响，国际政治经济中的政权形成，以及国际安全事务中的联盟政治。虽然这个项目可能不会对这些领域中的任何一个提供明确的治疗方法，但它承诺促进对所有这些领域共同的某些特征的欣赏。

This award is an Accomplishment Based Renewal. Previous research by the principal investigator has established the conditions under which cooperation can get started and maintain itself in simple environments. The environments previously examined include the two-person Prisoner's Dilemma, and an n-person norms game. In the last five years, this work has seen numerous applications to international political economy, antitrust law, government regulation, economic theory, artificial intelligence, and other fields. The current research investigates how collective properties (such as cooperation or conflict) emerge from the actions of individuals who adapt to each other within a complex population structure. The method is to develop a series of parallel adaptive models of whole populations. The models will be parallel in that all the individuals will be active at one time. The models will be adaptive in that the individuals will adjust their behavior, in light of their own goals, to the effects that they are experiencing from others. Rather than assume that the adjustment process is based upon rational calculation, the investigator will assume it is based upon learning from experience or differential survival of relatively successful strategies. This research is important for the social sciences because standard theoretical and mathematical approaches have traditionally assumed that only one or two things are happening at a time. In politics and society a great many events occur at once. Im particular, many people make interdependent choices that influence each other's environments. Thus an understanding of parallel adaptive models has the potential to enhance our insight into a wide variety of political processes such as: political leadership, public participation, domestic influences on foreign policy, regime formation in international political economy, and alliance politics in international security affairs. While this project may not provide definitive treatments of any one of these areas, it promises to promote the appreciation of certain features that are common to all of them.