摘要
本研究运用定量和定性的方法探讨了动态适应性学习条件下两难行为的理性特点及其影响因素。72名杭州本科生参加了本实验,结果发现:1、在有限次重复的两难博弈实验中,人们的两难行为不符合纳什博弈理论,进一步的定性分析指出,这种两难行为体现出“社会理性”的特征;2、在高奖罚的条件下,人们更容易做出接近“纳什均衡点”决策;3、群体规模和反馈方式对人们的对策行为影响不显著,但是,进一步的定性分析部分支持了群体规模的社会理性解释。
This study was designed to examine the rationality of the game behavior in social dilemma in a feedback learning situation with a "social rationality" perspective. 72 students from universities in Hangzhou participated in the experiment. The results indicated that 1) based on the adaptive learning from the feedback of previous trials, players' behavior approached gradually to the sub-optimal solution, but never conformed to the predicion of Nash Equilibrium. Further qualitative analysis revealed the social rationality underlying the improvement of the dilemma behavior. 2) game behavior was affected by the pay-off matrix. Players receiving high pay-off tended to make the choice closer to the solution predicated by the Nash Equilibrium than those with low pay-off. 3) Group size and feedback information had little effect on the game behavior. Protocol analysis partially supported our social rationality explanation of the group size effect on dilemma behavior.
出处
《心理科学》
CSSCI
CSCD
北大核心
2006年第2期290-293,共4页
Journal of Psychological Science
基金
国家自然科学基金资助(项目批准号:70471054)
关键词
纳什均衡
社会理性
社会两难行为
Nash Equilibrium, social rationality, game behavior in social dilemma