Experimental Implications of Theoretic Models: Workshop

研究人员正在举办一个研讨会，作为政治学项目理论模型的实验含义活动的一部分。他们专注于对重复的社会困境博弈进行正式分析和经验分析之间的联系。在社会两难博弈的背景下理解合作，对于理解替代体制安排如何在各种政治环境中促进集体行动至关重要。大量的实验证据与现有的严格基于自私偏好的非合作博弈论模型的预测不一致。近年来，学者们转向效用的另一种表征，试图捕捉个体之间的动机异质性。虽然已经提出了多个模型，但似乎还没有一个模型来解释由多个实验研究人员获得的丰富数据，这些实验研究人员使用许多变量的各种规格进行了大量重复的社会困境实验，这些变量包括：如果所有参与者合作，获得的集体利益的大小；个人对彼此身份的了解程度；参与的人数；受试者之间沟通的性质；以及相互制裁的可能性。虽然已经产生了许多理论上的可能性，但还没有人能够用涵盖所有这些条件的汇总数据集来准确地测试他或她的想法，因为这样的数据集还没有被放在一起并公开提供。鉴于正式模型的多样性和已经进行的大量实验，创建一个多校区、多学科的研究工作组的时机已经成熟，以：计划构建一个易于使用的一致数据库，其中包含实验研究人员贡献的许多数据集；审查最近解释重复社会困境中合作水平的正式模型，以检查理论结构以及从替代模型得出的预测的异同；计划对这些数据进行进一步分析，以便全面测试模型；改进更好的理论模型，以解释重复社会困境中的合作；根据上述步骤的结果，为进一步开展理论和实证研究提供一个总体的科学计划。研究人员将于2003年1月召开研究工作组第一次会议，本项目主要集中在这次会议的计划上。

The researchers are holding a workshop as part of the Political Science Program's Experimental Implications of Theoretic Models activity. They are focusing on the linkage between formal and empirical analyses of repeated social dilemma games. Understanding cooperation in the context of social dilemma games is fundamental to understanding how alternative institutional arrangements may foster collective action in a wide diversity of political settings. An abundance of experimental evidence is inconsistent with predictions from existing non-cooperative game-theoretic models based strictly on self-regarding preferences. In recent years, scholars have turned to alternative representations of utility in an attempt to capture motivational heterogeneity across individuals. While multiple models have been posited, no one model yet appears to explain the rich data that has been obtained by multiple experimental researchers who have conducted a very large number o repeated social dilemma experiments using various specifications of many variables including: The size of the collective benefit obtained if all participants cooperated; The level of knowledge that individuals have about each other's identity; The number of participants involved; The nature of communication among subjects; and, The possibility of sanctioning one another.While many theoretical possibilities have been generated, no one has been able to test his or her ideas against an aggregated data set covering all of these conditions precisely, because such a data set has not yet been put together and made publicly available. Given the diversity of formal models and large number of experiments that have already been conducted, the time is ripe to create a multicampus, multidisciplinary Research Work Team to:plan the construction of an easy-to-use consistent database containing many data sets contributed by experimental researchers;review recent formal models explaining levels of cooperation in repeated social dilemmas in order to examine similarities and differences in theoretical structure as well as in predictions derived from alternative models;plan further analyses of these data in order to test the models comprehensively;refine better theoretical models to explain cooperation in repeated social dilemmas;provide a general scientific plan for developing further theoretical and empirical research in light of the results of the above steps.The researchers are holding the first meeting of the Research Work Team in January, 2003, and this project focuses primarily on the plans for this meeting.