Collaborative Research: Individual and Team Behavior in Strategic Interactions

经济学家使用博弈论(战略互动理论)来研究各种各样的政策问题，例如如何防止大公司之间的串通，如何为各种目的设计拍卖(例如电波权利)，以及如何限制银行挤兑和市场恐慌等失灵现象。在许多这样的环境中，决策是由团队做出的，而不是个人--大公司的董事会、国会委员会等--但相对较少的人知道，在战略互动中，团队做出的决策与个人有什么不同。该项目的研究人员将进行精心设计的实验，以确定群体如何同意与其他群体合作，群体在产生更有效的结果方面比个人表现得更好(或更差)的机制，并比较群体与个人如何从过去的经验中学习在相关情况下做出选择。更好地理解团队如何做出决策，以及这些决策与个人做出的决策有何不同，将有助于对博弈论的基础理解。就更广泛的影响而言，当实体的决策由团队而不是个人做出时，这项研究将使经济学家能够更好地预测这些实体的行为。心理学家研究重复的囚徒-S困境游戏时通常会发现，团队的合作程度不如个体，这就是所谓的不连续效应。Kagel和McGee(2016)未能在球队之间没有沟通的情况下，在有限重复的超级比赛中发现中断效应。涉及具有概率停止规则的重复PD游戏的多个游戏的实验将在对手之间有和没有沟通的情况下进行，以理解为什么在某些情况下，群体似乎比个人更不合作，但在另一些情况下则不是。第二组实验将探索团队与个人在协调博弈中达成有效结果的相对效率，在协调博弈中，有效均衡是相当具有挑战性的。最后一组实验将检验团队和个人在感知不同类型的游戏相似程度上是否存在差异。

Economists use game theory (the theory of strategic interactions) to study a wide variety of policy issues such as how to prevent collusion between large firms, how to design auctions for various purposes (e.g. air wave rights), and how to limit dysfunctional phenomena such as bank runs and market panics. In many of these settings, decisions are made by teams rather than individuals - the board of a large corporation, a congressional committee, etc. - but relatively little is known about how decisions made by teams differ from individuals in strategic interactions. Researchers on this project will conduct carefully designed experiments to identify how groups agree to cooperate with other groups, the mechanism by which groups perform better (or worse) than individuals at generating more efficient outcomes, and to compare how groups, as opposed to individuals, learn to make choices in related situations from past experience. Better understanding how teams make decisions, and how these decisions differ from ones made by individuals, will contribute to foundational understanding of game theory. In terms of broader impacts, this research will enable economists to better predict behavior of entities when the decisions of those entities are made by teams rather than individuals. Psychologists studying repeated prisoner?s dilemma games typically find that teams are less cooperative than individuals, referred to as the ?discontinuity effect?. Kagel and McGee (2016) fail to find a discontinuity effect across finitely repeated super-games with no communication between teams. Experiments involving multiple plays of repeated PD games with a probabilistic stopping rule will be run with and without communication between opponents to understand why groups seem less cooperative that individuals in some setting but not in others. A second set of experiment will explore the relative efficacy of communication for teams vs. individuals in reaching the efficient outcome in a coordination game where the efficient equilibrium is quite challenging. A final set of experiments will examine whether teams and individuals differ in the extent to which they perceive different types of games as similar.