PSYCH OpenIR  > 社会与工程心理学研究室
过分自信与绩效的关系:过分自信为何如此普遍?
其他题名Overconfidence and Performance: Why Overconfidence Is So Pervasive In Self-Judgments?
毕研玲
学位类型博士
导师李纾
2011-06
学位授予单位中国科学院研究生院
学位授予地点北京
学位专业心理学
关键词过分自信 绩效 过分自信的普遍性 进化
摘要在人们的决策中,过分自信是最为普遍的现象之一。研究表明,大多数人都会对自己的能力以及未来的前景表现得过于乐观。作为一种决策偏差,人们常常将过分自信和失败联系到一起。甚至有学者认为,在判断与决策领域,没有任何问题比过分自信更具有潜在的破坏力。但从进化观点看,若过分自信真的具有这么强的破坏力,那么,进化的力量早应将其灭除。所以,对于一个追求正确,零偏差的群体来讲,为何过分自信偏差还会如此普遍,依然是个未解之谜。结合目前研究中过分自信对绩效的影响并不明确,本篇研究从二者关系着眼,从过分自信对当下任务绩效和后继任务绩效的影响进行研究,对这一问题进行了部分回答。
研究一使用反馈、提出反面观点、事件启动三种方法尝试对过分自信进行操纵,结果发现提出反面观点法不能有效降低过分自信,但反馈法有效降低了被试对一般性知识问题的过分自信。事件启动是一种操纵过分自信方法的新尝试,研究结果发现被试在回忆自己成功和失败的事件后,过分自信无显著差异,但回忆他人成功事件后的过分自信显著高于回忆他人失败事件后的过分自信。
研究二和研究三从过分自信对当下绩效的影响展开:
研究二发现,在竞争中,过分自信会为个体带来更多利益。个体通过其较高过分自信,可以迫使对手退出竞争,从而占有更多资源。面对高过分自信的对手,被试倾向于相信对手具有较高能力,从而放弃竞争使对手获得较多利益。研究三发现,由于高过分自信被试对自己的数学推理能力具有较高评价,并因此在非常难的数学推理任务中能够坚持较长时间。
研究四从过分自信对后继绩效产生的影响展开:
研究四采用英文单词学习任务,探讨过分自信水平对被试得到自己成绩反馈后,在第二阶段学习中的成绩的影响。结果表明,大多数被试高估了自己的成绩,即出现了过分自信。且过分自信者在第二阶段学习中取得了更好成绩,该效应是通过对成绩不满意的中介作用实现的。
综上,本研究结果不仅增进了现有关于过分自信的认识,让人们认识到过分自信作为一种偏差为人们带来的不仅仅只是危害,还为过分自信为什么会普遍存在于决策中这一问题做出部分回答。
其他摘要One of the most robust finding in the judgment and decision-making literature is overconfidence. People are typically overconfident. They often overestimated actual ability, performance, level of control, or chance of success of themselves. However, as a decision bias, overconfidence has often been blamed for high-profile disasters. It was even recognized that no problem in judgment and decision making is more potentially catastrophic than overconfidence. One objection to this argument is that evolutionary pressures would have rendered overconfidence extinct if it was such a maladaptive and damaging behavior in the real world. So it remains a puzzle why such a false belief was so pervasive in a population of competing accurate beliefs. Combining the unclear conclusion about the effect of overconfidence on performance, it was hoped that the present research which aimed to explore the effect of overconfidence on performance could provide evidence on this point.
As a pilot study, study 1 tested three methods of manipulating overconfidence-“consider the opposite” strategy, feedback and event priming. The findings indicated that: 1) “consider the opposite” strategy could not reduce overconfidence; 2) feedback could significantly reduce the general knowledge overconfidence; 3) there was no significant difference in neither general knowledge overconfidence, nor peer comparison overconfidence after the participants wrote down their own successful and failing events; participants’ general knowledge overconfidence, after they wrote down someone’s successful event, is significant higher than those who wrote down someone’s failing event.
Study 2 and study 3 are designed to explore the relationship between overconfidence and the current performance:
Study 2 indicated that overconfidence actually maximizes individual resources during competing. Furthermore, we found that the effects of overconfidence on the overall benefits were mediated by peer-rated competence.
Study 3 found greater overconfidence led to more persistent efforts to find solutions for very difficult mathematics reasoning items and the experimental manipulation established the causal priority of overconfidence. The effect could be explained by the self-rated competence.
Study 4 is designed to explore whether overconfidence facilitates subsequent performance. We found that most of the participants exaggerated their scores and the exaggeration led to the following affective costs (dissatisfaction). Moreover, overconfidence facilitates the subsequent performance and the effect was mediated by the amount of dissatisfaction.
Across these studies we found consistent supporting evidence for our primary hypothesis that overconfidence helps individuals increase performance. These findings make important contributions to the literature on overconfidence and speak to the pervasiveness of overconfidence. The fact that there are performance benefits of overconfidence to overconfident populations may be one of the reasons that why overconfidence was so pervasive in self-judgment.
学科领域应用心理学
语种中文
文献类型学位论文
条目标识符http://ir.psych.ac.cn/handle/311026/20064
专题社会与工程心理学研究室
作者单位中国科学院心理研究所
推荐引用方式
GB/T 7714
毕研玲. 过分自信与绩效的关系:过分自信为何如此普遍?[D]. 北京. 中国科学院研究生院,2011.
条目包含的文件
文件名称/大小 文献类型 版本类型 开放类型 使用许可
过分自信与绩效的关系:过分自信为何如此普(743KB)学位论文 限制开放CC BY-NC-SA浏览 请求全文
个性服务
推荐该条目
保存到收藏夹
查看访问统计
导出为Endnote文件
谷歌学术
谷歌学术中相似的文章
[毕研玲]的文章
百度学术
百度学术中相似的文章
[毕研玲]的文章
必应学术
必应学术中相似的文章
[毕研玲]的文章
相关权益政策
暂无数据
收藏/分享
文件名: 过分自信与绩效的关系:过分自信为何如此普遍?[2].pdf
格式: Adobe PDF
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。