论文题目：评价性反馈的神经机制以及自尊水平对其影响的研究作者简介：潘晓红，女，1967年 1 月 5 日生，2004年从师于 心理学院（院、系、所）姚力 教授，于2007年7月毕业于北京师范大学 基础心理学 专业并获 教育学博士学位。摘 要关键词：评价性反馈、自尊、脑机制、皮层中线结构评价性反馈是日常生活中常见的现象，特别是在发展教育过程中经常可以见到。与绩效反馈不同，评价性反馈除了提供与任务结果 (performance) 有关的正确或错误信息外（客观信息），还携带着与任务完成无关的社会性信息（主观信息），这些社会信息通常与个体的自我概念有关。目前，认知神经科学对绩效反馈的神经机制研究的比较多，还没有对评价性反馈这种携带了社会信息的反馈的神经机制进行研究，本研究对此做了尝试。自尊是个体对自我的一种情感，属于主体自我的范畴，是人格与社会心理学研究的对象之一。心理学的研究已经发现，自尊水平会影响到个体对信息编码和提取，本研究欲探寻个体的自尊水平对评价性反馈的神经机制是否有影响？通过什么方式产生影响？本研究实验分为两个部分：实验一，利用脑电技术研究了评价性反馈在时间进程上的加工特点，以及自尊水平是如何影响评价性反馈的加工。采用经典的ERP分析方法分析了FRN和P300两个成分，结果发现：FRN波幅受到评价性反馈效价的影响，表现为消极反馈引发的FRN显著大于积极反馈引发的. 经LORETA源定位分析，发现积极和消极两个条件下FRN产生于前扣带皮层。本研究还发现对于中性反馈也产生了FRN，而且中性反馈引发的FRN波幅最大，对其进行源定位分析，发现其发生源位于左侧额下回和右侧腹外侧前额叶。由此推测FRN可能并不单纯地反应对事件结果好或坏的评价过程。与先前绩效反馈的发现不同，评价性反馈产生了两个明显的P300成分（P3a和P3b），LORETA分析进一步发现评价性反馈的早期ERP可能产生于楔前叶，我们推测这一现象可能是由于评价性反馈携带的社会信息引发了被试对自我概念的加工。同时还发现自尊水平影响个体对评价性反馈的加工，主要体现在对中性反馈的赋值上。实验结果发现具有高水平自尊的被试可能给中性反馈赋予积极效价。实验二，为fMRI实验，探寻评价性反馈的神经机制以及自尊水平对评价性反馈加工机制的影响。实验的行为结果表明，被试对自我有正向偏差，而且高水平自尊组的这种偏差显著大于中低自尊组。同时高水平自尊组表现出显著的忽视消极反馈的现象。神经成像结果发现，评价性反馈显著引发皮层中线结构（CMS）的活动，该结构被许多神经成像研究证实与自我相关的信息加工有关，同时以往那些在绩效反馈中被显著激活的脑区在我们的研究中并未表现出明显的激活。因此我们可以推测，评价性反馈触发了被试的自我概念，同时削弱了被试对绩效结果信息的加工。自尊水平可以对评价性反馈的加工造成影响，并且具有不同自尊水平的被试可能会对中性反馈赋值不同。本研究结果表明，（1）评价性反馈与绩效反馈不同，它能引发被试对相关自我概念的加工，同时削弱被试对绩效结果信息的提取。这个发现有助于解释为什么现实生活中评价性反馈往往不能促进学生的学习进步。（2）自尊水平显著地影响被试对评价性反馈的加工，同时具有不同自尊水平的人对中性反馈的赋值明显不同。这个发现有助于澄清人格和社会心理学中一些基本理论问题。

Theme: The Neural Basis of Evaluation Feedback, Self-esteem and Their InteractionAuthor: Panxiaohong　, female, 　born in 1967　, became student of Professor YaoLi in 2004, graduated at Beijing Normal University in July 2007 obtaining a doctoral degree in education.　 AbstractKey words: Keywords: evaluation feedback, self-esteem, brain mechanism, cortex midline structures (CMS)Evaluative feedback is a common phenomenon in our daily lives, especially in the evolution of education. Unlike the performance feedback, evaluative feedback not only provides right or wrong information (objective Information) related to performance, but also carries social information (subjective Information) unrelevant to the task accomplishment which have close relationship with individual self-concept. Currently, accumulating cognitive neuroscience have focused on the mechanism of performance feedback, however, little attention has been paid to the nerval mechanism of evaluative feedback as the carrier of social information. Our research is considered to be the initial exploration.Self-esteem as a kind of emotion from the feeling of oneself is belonged to subject-self, and always attracts the attention of the researchers from personality and social psychology. Psychologic studies have already demonstrated that the level of self-esteem has great impact on the coding and extraction of individual information. This study was emphasized on the influence of individual’s self-esteem level on the nerve mechanism of evaluative feedback. If this is the case, how this influence exerted?Our research mainly includes the following two parts: firstly, electroencephalogram (EEG) was employed to study the characteristics of evaluative feedback process in the course of time, and the impaction of self-esteem level on such process. Feedback related negative (FRN) and P300 were analyzed using the classic ERP analysis method. Results showed that the magnitude of FRN was greatly influenced by the evaluation feedback, and negative feedback always induced more FRN compared to that of positive feedback. Under the analysis of LORETA, FRN was generated at anterior cingulate cortex under both positive and negative conditions. Another interesting findings in our study was the generation of FRN even in the case of neutral feedback, and which induced the largest magnitude of FRN. Further analysis using LORETA presented that this phenomenon was derived from the left inferior frontal gyrus and right ventrolateral prefrontal lobe. Therefore, we supposed that FRN likely does not simply reflect the processing of good or bad evaluations. Unlike performance feedback, evaluative feedback always generated two components of P300 (P3a and P3b). Analysis from LORETA further demonstrated that the early ERP of evaluative feedback was originated from the precuneus, and this enlighten us to suppose that it is the social information carried by evaluative feedback that led to the process of self concept by subjects. Our research also discovered that self-esteem level had great impact on the process of evaluative feedback reflected by the imposed values by neutral feedback. Our experiments found that subjects with high self-esteem probably produced positive valence to the neutral feedback. The aim of second experiment was to explore the nerval mechanism of evaluative feedback and the influence of self-esteem level on the process mechanism of evaluative feedback. Results from behavior data indicated that subject had positive deviation on self-image, and this was much larger in the high self-esteem group than that of low self-esteem group. Furthermore, high self-esteem group show the significant ignoration of the negative feedback. Results from fMRI showed that evaluative feedback can induce activities in cortex midline structures (CMS) which is considered to be the involvement of the self-related information process. However, brain areas stimulated by performance feedback were not stimulated significantly in our study, all of which indicated that evaluative feedback can induce individual’s self-concept with the attenuation on the process of performance. Self-esteem levels have important effect on the process of evaluative feedback, and subjects with various self-esteem levels probably yielded different valence for neutral feedback.Our study make constribution to the following two issues: firstly, unlike performance feedback, evaluative feedback always illuminate subjects’ self-concept, and meanwhile diminish the extraction of performance information. This result served to our understanding about why evaluative feedback often fails to promote students’ advancement in the daily life. Secondly, self-esteem levels have great effects on the process of evaluation feedback, and people often led different values to neutral feedback. This important finding would facilitate us to clarify some fundamental theories in personality and social psychology.