Empathy for pain is a psychological phenomenon in interpersonal interaction. Previous studies have suggested the preference of empathy for pain, especially the priming effect of friends. But how do we perceive the pain of close friend and acquaintance when there is a different degree of overlap with ourselves? Therefore, the aim of the current study is to explore the cognitive mechanism of self-other overlap affecting empathy for pain by event-related potential study. This experiment was designed a 2(empathy judgement: pain, non-pain) × 3(self-other overlap degree: high, medium, low) within-subject study. Self-other overlap was manipulated by the name of the close friend, acquaintance and stranger. Prior to the experiment, the subjects needed to provide the names of close friends and acquaintances according to the instructions and filled in the Inclusion of Other in the Self (IOS) scale and Questionnaire for Personality Trait Assessment. At the beginning of the experiment, all subjects were told that they would first see the name stimuli (close friend, acquaintance, stranger), and followed by picture. The subjects were instructed to image the scene in the picture happening to the person mentioned earlier. Subsequently, they were instructed to immediately response to empathy judgement. There are 6 blocks in this experiment, each block has 60 trails, and 6 conditions are randomly presented, so there is a total of 360 trails. The ERP analysis time course is from the empathy judgment stimulus presentation (0-1000ms), with the first 200ms as the baseline period. Combined with the relevant research and the purpose of this study, the selected ERP components are: N1(120-160ms), N2 (220-280ms), P3 (360-430ms), LPP(450-650ms). In terms of behavioral data, the study found that the reaction time of the close friend condition was significantly smaller than the acquaintances and the accuracy rate of the close friend condition was significantly higher than the acquaintances and strangers. In the ERPs results, there were no significant main effects and interaction effect. N2 amplitude was decreased in the close friend condition of empathy judgement than acquaintances and strangers. The data showed a significant main effect of the amplitude of P3. Pain judgment elicited significant larger P3 and LPP amplitude than that of non-pain judgment. Meanwhile there was a significant interaction of self-other overlap and empathy judgement. Under the painful judgement condition, close friend condition elicited larger amplitude in P3 compared with acquaintances and strangers. The behavioral results showed that individuals were more sensitive to the pain of the close friend than the acquaintance. Close friend with more larger N2 amplitude and he early perceptual processing stage reflected in N2 was more likely to be a threat-detection stage. Furthermore，the increased P3 indicated that self-other overlap factors enhances top-down cognitive control and evaluation processing of empathy for pain. In conclusion, the early perception could discriminate degree of self-other with N2 is more likely to reflect a threat-detection stage and degree of self-other modulated the late cognitive control processing stage of empathy for pain. This study further improved the cognitive mechanism of the influence of self-other overlap on empathy for pain.