从风险决策中的多次博弈到单次博弈:量变还是质变?  

From Multiple-play to Single-play in Risky Decision Making:Quantitative Change or Qualitative Change?

作　　者：孙红月[1,2] 苏寅[1,2] 周坤[3] 李纾[1] 

机构地区：[1]中国科学院心理研究所行为科学重点实验室,北京100101 [2]中国科学院研究生院,北京100049 [3]中国民航大学安全科学与工程学院,天津300300

出　　处：《心理科学进展》2011年第10期1417-1425,共9页Advances in Psychological Science

基　　金：国家重点基础研究发展计划(2011CB711002);国家自然科学基金面上项目(70871110);中国科学院知识创新工程重要方向项目(KSCX2-EW-J-8);北京市重点学科建设项目资助

摘　　要：诺贝尔经济学奖获得者Samuelson于1963年发现人们在单次和多次博弈条件下决策行为不一致。文章综述了两种博弈条件下人们决策行为的差异并质疑了这种差异的传统理论解释机制。描述或预测决策行为的风险决策理论其实只采用了一种评价法则——期望法则,始终没有跳出"最大化"的窠臼。基于实验证据,我们推测,多次博弈时人们遵守了期望法则,而单次博弈时人们所遵循的是非补偿性法则。从多次博弈到单次博弈,不单单是一种博弈次数上的变化(量变),而是代表了从期望法则(补偿性法则)到非补偿性法则两种策略之间的转变(质变)。最后,文章介绍了单次、多次博弈问题在医疗、应急管理以及投资领域的体现,并呼吁更多的研究者关注单次、多次博弈问题。In risky decision making,the subtle interplay between single-play and multiple-play is well illustrated by the literature that has followed an early paper by Paul Samuelson（1963）.The article summarized the difference between responses for single-play and multiple-play and questioned traditional theoretical explanations.Most risky decision-making theories are based on the same assumption：a rational decision maker behaves as if he or she were attempting to maximize some kind of expectation.However,based on a combination of research evidence,maybe we could infer that the multiple-play and single-play gambles were in fact so utterly different that the short run one was perhaps not based on the expectation rule.From multiple-play to single-play,it is likely to represent a strategy shift from expectation rule to non-compensatory rule（qualitative change） rather than pure change on the number of gamble times（quantitative change）.The status and perspectives on the applied researches of multiple-play were introduced and the research prospect was put forward as well.

关 键 词：多次博弈 单次博弈 风险决策 期望法则 

分 类 号：B849[哲学宗教—应用心理学] C93[哲学宗教—心理学]