目的：以往研究表明，完美主义可能会损害个体执行功能的子成分，且不同完美主义类型的执行功能子成分可能存在特定的差异或受损模式。因此，本研究使用停止信号任务、任务转换范式、和N-back任务分别测量执行功能的抑制、转换、刷新三个子成分，对不同类型完美主义者的执行功能子成分特征进行系统分析，考察不同类型完美主义者是否存在不同的执行功能子成分受损模式，并结合事件相关电位（Event-Related Potentials，ERPs）技术，探讨其受损的电生理机制。

方法: 研究一在高校大学生中招募被试，共有677名被试完成Frost多维完美主义量表（Frost Multidimensional Perfectionism Scale，FMPS）、流调中心用抑郁量表（Center for Epidemiologic Studies Depression Scale，CES-D）、状态-特质焦虑量表（State-Trait Anxiety Inventory，STAI）及压力知觉量表（Perceived Stress Scale，PSS）。基于Frost多维完美主义量表进行聚类分析，筛选出适应性完美主义者（适应不良组）228人、适应不良完美主义者（适应性组）146人以及非完美主义者（非完美组）303人。在研究一的大样本中招募被试进一步参加脑电实验。研究二采用停止信号任务评估三组被试的抑制功能，最终共有67名被试完成任务且数据质量较好，其中适应不良组20人，适应性23人，非完美主义组24人。研究三采用任务转换范式评估三组被试的转换功能，最终共有76名被试完成任务且数据质量较好，其中适应不良组25人，适应性25人，非完美主义组26人。研究四采用N-back任务评估三组被试的刷新功能，最终共有75名被试完成任务且数据质量较好，其中适应不良组24人，适应性25人，非完美主义组26人。采用卡方检验和方差分析对被试的人口学数据及心理量表数据进行分析，采用方差分析对适应性完美主义者、适应不良完美主义者以及非完美主义者的行为学及脑电数据进行分析。

结果：研究一：在大样本和参加ERPs实验的被试中，适应不良完美主义组、适应性完美主义组及非完美主义组在FMPS、CES-D、STAI以及PSS上得分均存在显著组间差异（p <0.05）。适应不良完美主义组在各量表上得分最高，非完美主义组得分最低。研究二：在停止信号任务中，三组被试在行为学指标上差异不显著（p >0.05）。适应不良完美主义组在FZ电极处诱发的P2波幅显著小于非完美主义组的P2波幅（p =0.017）。研究三：在任务转换范式中，行为学数据的组间差异不显著。三组被试的P2波幅组间效应边缘显著（p =0.085），重复任务下，FZ和CZ处P2三组被试波幅具有显著组间差异的趋势（p =0.008），适应不良完美主义组诱发的P2波幅最小，适应性完美主义组次之，非完美主义组最大。研究四：在N-back任务中，行为学数据的组间差异不显著。脑电成分上，在 0-back条件下，适应不良完美主义组在PZ电极处诱发的N2波幅显著小于非完美主义组的N2波幅（p = 0.016）。

结论：（1）适应不良完美主义者在早期接收刺激阶段认知资源调动不足可能导致其抑制功能存在缺陷，适应性完美主义者抑制功能并未表现出缺陷。（2）适应不良及适应性完美主义者注意资源控制以及对靶刺激的觉察和识别能力的减弱可能导致其转换功能存在缺陷，且与适应性完美主义者相比，适应不良完美主义者缺陷更严重。（3）适应不良完美主义者耗费更多的认知资源去对刺激进行识别和保持注意维持可能导致其刷新功能存在缺陷。

Objective: Previous studies have shown that perfectionism may damage individual executive function sub-components, and there may be specific differences or impaired patterns in the sub-components of executive function among different types of perfectionism. This study used stop signal task, task switching paradigm, and N-back task to measure the three sub-components of executive function, namely inhibition, switching, and refreshing, in different types of perfectionists, and systematically analyzed the executive function characteristics of different types of perfectionists. Event-Related Potentials (ERPs) technology was also adopted to explore the differences in executive function among different types of perfectionists and their corresponding electrophysiological mechanisms.

Methods: In the first study, a total of 677 subjects were recruited from college students to complete the Frost Multi-dimensional Perfectionism Scale (FMPS), the Center for Epidemiologic Studies Depression Scale (CES-D), the State-Trait Anxiety Inventory (STAI) and Perceived Stress Scale (PSS). A cluster analysis based on the Frost Multidimensional Perfectionism Scale was performed to screen out 228 adaptive perfectionists, 146 non-adaptive perfectionists and 303 non-perfectionists. In the large sample of Study 1, subjects were recruited to further participate in EEG experiments. In Study 2, a stop-signal task was used to evaluate the inhibitory function of the three groups of subjects. In the end, a total of 67 subjects completed the task with good data quality, including 20 in the maladaptive group, 23 in the adaptive group, and 24 in the non-perfection group. Study 3 used the task switching paradigm to evaluate the switching function of the three groups of subjects. In the end, a total of 76 subjects completed the task with good data quality, including 25 in the maladaptive group, 25 in the adaptive group, and 26 in the non-perfection group. In Study 4, the N-back task was used to evaluate the refresh function of the three groups of subjects. In the end, a total of 75 subjects completed the task with good data quality, including 24 in the maladaptive group, 25 in the adaptive group, and 26 in the non-perfection group.

Results: Study 1: In the large sample and the subjects who participated in the ERPs experiment, there were significant between group differences in the scores of FMPS, CES-D, STAI and PSS among the maladaptive perfectionism group, the adaptive perfectionism group and the non-perfectionism group (p < 0.05). The maladaptive perfectionism group had the highest score on each scale, while the non-perfectionism group had the lowest score. Study 2: In the stop signal task, there was no significant difference in behavioral data among the three groups (p > 0.05). The P2 amplitude evoked at FZ was significantly smaller in the maladaptive perfectionist group than in the non-perfectionist group (p = 0.017). Study 3: In the task-switching paradigm, group differences in behavioral data were not significant. The P2 amplitudes of the three groups were marginally different (p = 0.085). The amplitudes of the P2 groups at the FZ and CZ were significantly different (p = 0.008) under repeated tasks, with the maladaptive perfectionism group showing the smallest amplitudes and the non-perfectionism group the largest. Study 4: In the N-back task, the group differences in behavioral data were not significant. In terms of ERP components, under the 0-back task, the N2 amplitude evoked at PZ was significantly smaller in the maladaptive perfectionist group than in the non-perfectionist group (p = 0.016).

Conclusions: (1) Maladaptive perfectionists may have defects in their inhibitory function due to insufficient cognitive resource mobilization in the early stage of receiving stimulus, while adaptive perfectionists showed no defects in their inhibitory function. (2) Maladaptive and adaptive perfectionists' reduced attention resource control and ability to detect and recognize target stimuli may lead to defects in their conversion function, and maladaptive perfectionists' defects are more serious than those of perfectionists. (3) Maladaptive perfectionists spend more cognitive resources to identify stimuli and maintain attention, which may lead to defects in their refresh function.