Behavioral Foundations of Conflict Resolution

该项目旨在更好地了解冲突行为，特别是预防和解决冲突。通过使用精心设计的经济实验，这些项目专注于识别在实验室外很难识别的冲突行为的特定方面。因此，该项目将为使用实地数据的多项经验研究提供补充证据。该项目依赖于三个经济实验，这些实验以不同的方式研究冲突行为和解决机制。在所有实验中，冲突游戏使用Tullock(1980)2人寻租模型的变体，在该模型中，参与者竞争固定的奖品。双方都可以在这场比赛中投入资源，他们获奖的可能性取决于他们投资的相对规模。投入到比赛中的资源是沉没成本，因此每一项投资都明确地降低了整体效率。大量的实验研究都使用这一博弈来研究冲突行为，其中最有力的两个发现是，投资水平大大超过了标准的纳什均衡预测，以及被试之间的投资水平存在很大的异质性。第一个实验将武器装备和图洛克竞赛中的冲突启动分开。这是研究这一区别的第一次实验，尽管在许多实际冲突中，决定的分离是一个关键特征。试验性的处理方式不同，关于冲突另一方的武器装备和奖品的分配情况各不相同。实验二研究了冲突干预的跨期影响。在图洛克大赛中，每个玩家与同一个对手进行50轮的互动。拟议的实验性治疗方法不同，第三方是否、何时以及如何干预冲突。该项目解决了相关文献中的另一个缺陷。以前的文章只研究了当实验和嵌入的冲突开始时启动的干预机制。因此，这些干预相当于预防性行动。第三个实验解决了该领域经常观察到的问题，即寻租精英对群体之间冲突的激烈程度起着至关重要的作用。在实验中，两名选手一组一组地比赛。在每一组中，只有一名玩家(“士兵”)进行任何冲突支出，而另一名玩家(“精英”)在成功的情况下从这笔支出中受益。实验治疗改变了精英玩家的强制潜力和介入的第三方的资源。

The project aims to provide a better understanding of conflict behavior and in particular conflict prevention and resolution. By using carefully designed economic experiments the projects focuses on the identification of specific facets of conflict behavior that are hard to identify outside a laboratory. The project will therefore provide complementary evidence to the multiple empirical studies that use field data. The project relies on three economic experiments that study conflict behavior and resolution mechanisms in different ways. In all experiments the conflict game uses variations of the Tullock (1980) 2 player rent seeking model in which the players compete for a fixed prize. Both can invest resources into this contest and their probability of winning the prize is determined by the relative size of their investments. The resources invested into the contest are sunk costs and thus every investment unambiguously decreases overall efficiency. A great number of experimental studies has used this game to study conflict behavior and the two most robust findings are that investment levels strongly exceed the standard Nash equilibrium prediction and that there is great heterogeneity in investment levels between subjects. The first experiment separates armament and conflict initiation in the Tullock contest . This is the first experiment that studies this distinction even though the separation of the decisions ius a crucial featuer in many actual conflicts. The experimental treatments vary the information about the armament of the other conflict party and the distribution of the prize. The second experiment studies the intertemporal impact of conflict intervention. Each player interacts with the same opponent for 50 rounds in a Tullock Contest. The proposed experimental treatments vary whether, when, and how a third party intervenes in a conflict. The project addresses another shortcoming in the relevant literature. Previous contributions only study intervention mechanisms that kick in when the experiment and the embedded conflict starts. Hence these interventions are rather preventive actions. The third experiment addresses frequent observations from the field that rent-seeking elites play a crucial role for the intensity of conflicts between groups. In the experiment groups of two players compete against each other. Within each group only one player (the "soldier") makes any conflict expenditure while the other player (the "elite") benefits from this expenditure in case of success. The experimental treatments vary the coercive potential of the elite player and the resources of an intervening third party.