Continuous Games in the Laboratory

该奖项是根据2009年美国复苏和再投资法案（公法111-5）资助的。该奖项为加州大学圣克鲁斯分校LEEPS实验室进行经济实验的新软件开发提供资金。连续游戏（ConG）软件使研究人员能够研究具有明确财务激励的人类玩家如何在连续时间内进行战略互动，当他们有连续的选择范围时，以及当他们与数十名其他人类玩家匿名互动时。LEEPS的研究人员与美国和世界各地的其他实验室合作，正在使用新软件对各种不同的战略互动进行原创研究。这些游戏包括冲突游戏（如囚徒困境和社会困境）和协作游戏。未经检验的理论预测在连续时间比在标准离散时间设置更有效的结果。某些其他类型的战略互动被预测会导致随机行为(？混合策略？)，但在标准环境下，预测效果并不理想。当玩家可以在连续时间内调整他们的混合时，他们可能会在《ConG》中做得更好。在其他类型的互动中，战略特征会随着时间的推移而逐渐改变，一些理论预测了锁定效应。最后，ConG将使一些重要的应用游戏的调查？例如策略性定价、数量设定和地点选择？在更逼真，连续时间设置。完整的软件包将通过名为EconPort的国家数字图书馆设施提供给研究界。研究人员只需下载软件并重新设置参数，就能进行多种新的战略互动。他们还可以通过编程玩家屏幕的图像元素和阶段游戏的收益函数来运行全新且完全不同的连续游戏；更深层次的编程（例如，消息传递和定时）已经被处理好了。使用标准的支持java的web浏览器，研究人员将能够运行各种异步选择的阶段游戏，响应时间不到十分之一秒。这项工作将直接使国家研究界受益，特别是因为博弈论现在在经济学、政治学、社会学、进化生物学、生态学和计算机科学中发挥着核心作用。在许多情况下，连续时间比离散时间能更好地描述世界，可以作为现场应用和政策的更好指导。因此，关于行为如何在连续环境中展开的证据将使博弈论更接近该领域，并提高其应用价值。具体细节不可能事先知道，但可以想象的间接影响包括更有效地利用拥挤的计算机资源，以及更复杂的反垄断政策。

This award is funded under the American Recovery and Reinvestment Act of 2009(Public Law 111-5).This award funds the development of new software for conducting economic experiments at the LEEPS lab of the University of California at Santa Cruz. The Continuous Games (ConG) software enables researchers to investigate how human players with clear financial incentives interact strategically in continuous time, when they have continuous ranges of choices, and when they interact anonymously with dozens of other human players.In partnership with other laboratories in the US and around the world, LEEPS researchers are using the new software to conduct original research on a variety of different strategic interactions. These include games of conflict (e.g., prisoner's dilemma and social dilemmas) and games of coordination. Untested theory predicts much more efficient outcomes in continuous time than in standard discrete time settings. Certain other sorts of strategic interaction are predicted to lead to random behavior (?mixed strategies?), but in standard settings the predictions have not fared very well. They may do better with ConG, when players can adjust their mixtures in continuous time. In other sorts of interaction, the strategic character gradually changes over time, and some theories predict lock-in effects. Finally, ConG will enable investigation of some important applied games ? such as strategic pricing, quantity setting and location choice ? in more realistic, continuous time settings.The completed software package will be made available to the research community via the National Digital Library facility called EconPort. Researchers will be able to run many sorts of new strategic interactions merely by downloading the software and resetting the parameters. They also will be able to run new and quite different continuous games by programming only the graphic elements of the player screens and the stage game payoff functions; the deeper programming (e.g., of message passing and timing) will already be taken care of. Using standard Java-enabled web browsers, researchers will be able to run a wide variety of stage games with asynchronous choices and response times of less than a tenth of a second. The work will benefit the national research community directly, especially since game theory now plays a central role in economics, political science, sociology, evolutionary biology, ecology and computer science. In many settings, continuous time is a better description of the world than discrete time and may serve as a better guide to field applications and policy. Evidence on how behavior unfolds in continuous settings will therefore bring game theory closer to the field and enhance its value in applications. Specifics are impossible to know in advance, but conceivable indirect impacts range from more efficient use of congestible computer resources to more sophisticated antitrust policy.