ABSTRACT In tacit coordination games people manage to converge on prominent solutions, which are known as focal points. There is still no accepted explanation of how players manage to converge on the same solution. It could be that the limited explanatory power arises from the fact that existing theories rely on pure strategies to describe behaviour. The aim of the current study is to construct a cognitive model that more accurately describes human behaviour in tacit coordination games. To this end we constructed individual strategic profiles that take into account the subjective preferences of individual players regarding the prominent selection rules. Subsequently, the individual profiles were clustered to gain insights regarding different types of coordinators. By using machine learning and statistical methods we were able to demonstrate, for the first time, the relationship between different types of strategic profiles and coordination ability. The results of this study demonstrate the importance of constructing a descriptive behavioural model that may potentially improve prediction of human decision-making in the context of human-machine interaction.

中文翻译：

---

隐性协调博弈中的个人战略概况

摘要 在默契协调游戏中，人们设法集中在突出的解决方案上，这些解决方案被称为焦点。对于玩家如何设法收敛到相同的解决方案，仍然没有公认的解释。有限的解释力可能源于现有理论依赖于纯策略来描述行为这一事实。当前研究的目的是构建一个认知模型，以更准确地描述默会协调游戏中的人类行为。为此，我们构建了个人战略概况，考虑到个人玩家对突出选择规则的主观偏好。随后，个人资料被聚集在一起，以获得关于不同类型协调员的见解。通过使用机器学习和统计方法，我们能够证明，首次研究了不同类型战略概况与协调能力之间的关系。这项研究的结果证明了构建描述性行为模型的重要性，该模型可能会在人机交互的背景下改善对人类决策的预测。