Pathological Narcissism and Metacognition of Emotional Face Recognition

Abstract

The aim of this study is to investigate the differential metacognition of emotion recognition in young adults with pathological narcissism as compared with the control subjects by using relative and global measures, and to investigate the relationship between relative meta-accuracy and performance. Among 495 undergraduate students, 36 were selected into the pathological narcissism group and another 36 into the control group based on the scores from the Pathological Narcissism Inventory (PNI) and Symptom Check List 90-Revision (SCL-90-R). Relative meta-accuracy is calculated using the gamma correlation, means of within participant correlations between confidence ratings and correct vs. incorrect responses on the emotion recognition task. Moreover, global meta-accuracy is calculated using the Pearson correlation between questionnaire (Interpersonal Reactivity Index, Mind-Reading Belief Scale) response and proportion correct on the emotion recognition task. In comparison with the control group, the pathological narcissism group showed a reduced metacognition of emotion recognition in the relative meta-accuracy, while no significant differences were indicated in the global meta-accuracy. The differential relative meta-accuracy between two groups come from the condition. In the retrospective condition, the control group rose the relative meta-accuracy compared with the prospective condition, however, the pathological narcissism group does not indicate a difference of the relative meta-accuracy according to the condition. In comparison with the control group, the pathological narcissism group showed a low mean proportion correct on the Emotional Expression Multimorph Task, while no significant differences were indicated in confidence judgments. Moreover, in the high level condition, the mean confidence was high compare with the control group. Individuals with pathological narcissism revealed the high self-confidence even when the emotional awareness was difficult and lack of prudence when the test was accompanied. Such traits seems to have a damaging effect on metacognition of pathological narcissism. Our results showed that in both groups, individuals who were better at the emotion recognition task also tended to be more accurate in their relative metacognitive assessments.