Electronic Feedback Mechanisms and Indirect Reciprocity

在线声誉或“反馈”机制，如ebay和亚马逊所使用的机制，旨在通过提供更传统的紧密联系的群体中可用的信息，减轻与陌生人之间的在线交流相关的道德风险问题，在这些群体中，成员经常参与彼此的交易。然而，信任和可信赖性仍然是一个备受关注的问题；例如，美国司法部最近的一项调查（2002年）指出，即使在有在线反馈系统的网站上，网络欺诈的水平也很高。由于其他控制方法（如法律执行）成本高昂，因此研究改进反馈机制对在线市场的长期扩张至关重要。还有一些与间接互惠的本质密切相关的基础研究问题（“我信任你，因为之前你是值得别人信任的”）。’)，以及间接互惠与直接互惠的关系（我信任你，因为以前你对我是值得信任的。”）。这些问题的重要性超越了在线市场，延伸到传统市场中的信任和可信赖性，以及在一般社会互动中建立声誉的作用。拟议的研究由三个相互关联的研究组成。他们都使用实验性的在线市场来研究重要的理论问题。第一项研究调查了在线市场阻碍或增强反馈机制绩效的方面，包括社会关系的作用，以及反馈分数对定价和潜在贸易伙伴数量的影响。第二项研究调查了支持间接互惠所必需的信息量。一些模型表明，在某些情况下，声誉建设可以对关于一个人声誉的不完整信息非常有效，但这一点尚未得到证实。第三个调查领域涉及发展对直接互惠和间接互惠之间差异的更精细的理论解释。一个实验将调查为什么现在的模型没有捕捉到在以前的调查中发现的差异，并且从结果数据中得到的想法将用于开发正式的模型。就市场设计而言，这项研究的目的是就如何改进反馈系统提供政策建议。就我们对间接和直接互惠的基本理解而言，其目的是更好地理解人们如何在决策中使用有关他人声誉的信息，这是一个对整个社会的机构和组织都很重要的问题。

Online reputation or "feedback" mechanisms, such as those used by eBay.com and Amazon.com, aim to mitigate the moral hazard problems associated with online exchange among strangers, by providing the type of information available in more traditional close-knit groups, where members are frequently involved in one another's dealings. Trust and trustworthiness are nevertheless a much-noted problem; a recent U.S. Department of Justice survey (2002), for example, cites high levels of online fraud even at sites with online feedback systems. Since alternative methods of control, such as legal enforcement, are expensive, research into improving feedback mechanisms is important to the long-term expansion of online markets. There are also closely associated basic research questions concerning the nature of indirect reciprocity ('I trust you because before you were trustworthy with others.'), and the relationship of indirect reciprocity to direct reciprocity (I trust you because before you were trustworthy with me.'). The significance of these issues goes beyond online markets, extending to trust and trustworthiness in traditional markets and to the role of reputation building in general social interactions.The proposed research is comprised of three inter-related studies. Each uses an experimental online market to investigate important theoretical questions. The first study investigates the aspects of online markets that hinder or enhance feedback mechanism performance including the role of social ties, and the impact of feedback scores on pricing and number of potential trading partners. The second study investigates the amount of information necessary to support indirect reciprocity. Some models suggest that reputation building can, under certain circumstances, be quite robust to incomplete information about a person's reputation but this is unproven. The third area of investigation concerns development of a more refined theoretical explanation of the difference between direct and indirect reciprocity. An experiment will investigate why present models do not capture differences uncovered in previous investigations, and ideas from the resulting data will used to develop a formal model. The aim of the research, in terms of market design, is to provide policy recommendations on how feedback systems can be improved. The aim, in terms of our basic understanding of indirect and direct reciprocity, is to provide a better understanding of how people use information about the reputations of others in their decision making, an issue of importance to institutions and organizations throughout society.