Abstract: Using classification–verification paradigm, the present study investigated the time-course of stereotype-activated effect by recording both behavioral and ERP data elicited by gender stereotype’s trait-words which were taken as the core of stereotype representation.
Thirty-four young participants (17 female) were recruited in this experiment. The prime stimuli included two gender-category words, “male” and “female”, and the target stimuli were gender-trait-words generally attributed to male or female. Gender-category and gender-trait words constituted congruent and incongruent situations. And the prime-target pairs were divided into four conditions, “male” priming male-trait-words (M-m), “female” priming male-trait-words (F-m), “male” priming female-trait-words (M-f) and “female” priming female-trait-words (F-f). Each trial started with the presentation of a priming word for 300 ms, followed by a fixation cross presented for 500 ms and then the target word appeared for 300 ms and a blank until participants responded; randomly selected inter-trial intervals (ITI) between 600 and 800 ms separated next trial from the response. Participants had to indicate by pressing one of two buttons whether the target matched the priming word or not and they were instructed to response as quickly as possible but avoid mistakes.
For all participants, the response was faster and more accurate in congruent than incongruent condition, and in-group bias was also found. The early ERP components reflecting perceptual processing were not modulated by experimental conditions. Compared with prime-target congruent condition, gender-trait words that mismatched with the gender-priming word elicited a larger N400 over the fronto-central area. Moreover, consistent with behavioral in-group bias, a larger P600 was elicited under the ingroup-category-word priming condition. These results indicated that the stereotype could be activated by social group-category-word, and the N400 effect reflected the integration of the social properties of words and stereotypical knowledge in memory representation. Both the stereotype-activated effect and in-group bias started at the stage of post-perceptual information processing.
The present study investigated the time-course and cognitive mechanism of stereotype-activated effect for the first time, demonstrating the modulation of N400 relevant to gender stereotype-activated effect. These data indicated the possibility of the N400 indexing the cognitive mechanism and representation model of processing stereotypical information.

Key words: stereotype, stereotype-activated effect, gender-category words, gender-trait-words, in-group bias, ERP