In successful social interactions, distinguishing between our own and another person's emotions is important. For individuals, both self-related information (such as self-name) and emotional stimuli with high sociality or adaptive meaning can automatically capture one's attention, leading to prior and deep processing. Previous studies have confirmed that self-relevance can affect the processing of emotional words. However, there is currently no research using name as a self-relevance clue to investigate whether and how name influences the cognitive processing of emotional words. In the current study, we used names as self-relevance clues to explore the dynamic temporal characteristics of self-related information affecting emotional information and its integration mechanism.

In this study, we used ERP technology and created a 3 (name type: self, friend, unknown) × 3 (emotion type: positive, neutral, negative) within-subjects design. A total of 21 college students (9 males, Mage = 20.4) participated in the experiment. Prior to the experiment, we gathered the participants' own names and their best friends' names, then we found a name that we confirmed was unfamiliar to all participants. All names included three characters. The experimental stimuli were 171 two-character words taken from the Chinese Affective Words System, which included 57 positive, 57 negative and 57 neutral words. The participants were asked to silently read the names (self-name, friend-name and unfamiliar name) and emotional words (positive, neutral and negative) presented in succession, unaware that the presented words were emotional words, while their EEG was recorded.

The ERP results showed the following. (1) The main effect of the emotional words was significant in the early processing stage, and the negative words elicited larger EPN amplitudes (200-300 ms) than the positive and neutral words. (2) In the late processing stage, the emotional words were further processed. The negative words elicited reduced N400 (300-450 ms) compared to the neutral words, and they enhanced LPP compared with the neutral and positive words during 450-650 ms. (3) More importantly, a significant interaction between the names and emotional words was discovered in the LPP time window. The negative words paired with self-names elicited significantly more positive LPP than the neutral words paired with self-names, while the negative words paired with friend-names elicited significantly larger LPP than the positive words and neutral words paired with friend-names. Nevertheless, no significant difference was found among the LPP amplitudes elicited by the three kinds of emotional words paired with unknown names.

Overall, our study demonstrates that there is a processing advantage for negative words in different stages, and self-relevance contained in a name can affect the cognitive processing of emotional words, which mainly occurs in the late stage of emotional lexical processing. Importantly, it seems that individuals first filter information according to whether it is self-related and then process the emotional content (especially negative stimuli) related to themselves or their friends more deeply and elaborately. The negative bias can be explained by the fact that negative stimuli are considered to carry greater informational value than positive stimuli. Therefore, negative stimuli automatically capture more attention and cognitive resources than neutral and positive stimuli, resulting in higher order coding.

Keywords： emotional words ; self-relevance ; EPN ; LPP