Abstract: The long-held view that self-esteem is pancultural has been challenged in the past decade. One of the main reasons is the repeatedly revealed low self-esteem among people from collectivistic culture. With regard to the causes of low self-esteem in collectivistic culture, two arguments have been proposed. One argues that people in collectivistic culture do not have the need to pursue self-esteem or have a weaker need than westerners due to interdependent culture (Heine, Lehman , Markus, Kitayama, 1999). The other acknowledges the cultural differences in self-esteem, but they do not think that people in Eastern culture lack the need of self-esteem and it is modesty that leads to Easterner’s low self-esteem because Eastern culture values modesty very much (Brown, 2003). However, almost all related evidence was based on explicit self-report measures that are sensitive to social desirability and unable to tap implicit processes. The present research aims to examine whether people in collectivistic culture exhibit self-esteem while self-presentation is minimized and while implicit self-esteem is allowed to manifest by using subliminal evaluative priming procedure that is totally free of self-presentation (Nosek, 2002). The expectation is that when effects of self-representation are removed, Easterners would exhibit self-esteem implicitly.
Method
123 college students from East China Normal University in Shanghai participated in the experiments as volunteers. A 2 (priming type: ‘I’ vs. ‘he/she’)*2 (target valence: positive vs. negative)*2 (target type: attributes vs. non-attributes) within-subjects design was employed. All participants completed the computerized tasks in separated cubicles on computer with a refreshment frequency of 72Hz. They were presented a pre-mask stimulus for 56ms (4 frame) first, then priming stimulus for 42 ms, then post-mask stimulus for 56ms, and finally the target stimulus. The participants were required to respond to the positive and negative target stimuli with key ‘E’ and ‘I’ respectively as soon as possible with the least errors. The dependent variable was reaction time. Data from 3 participants were discarded after preliminary data analysis due to high error rates (higher than 20%).
Results
Repeated measure ANOVA revealed a significant two-way interaction between priming type and target valence, F(1,119)=24.16， p<0.001. Students responded faster to positive targets than to negative ones when primed with ‘I’, t(1,119)=6.92，p=<0.001， M1=534.59， M2=516.64. However, when primed with ‘he/she’, students responded at the similar speed to positive and negative targets, t(1,119)=-1.08， p=.28, M1=531.79， M2=535.32 . These findings suggest that Chinese college students possess self-esteem at implicit level. The three-way interaction was not significant, F(1,119)=2.50， p>0.05, which indicated Chinese college students associated self with positive stimuli implicitly no matter the stimuli are attributes or not, providing convergent evidence that self-esteem is alive among Chinese. In summary, the results demonstrate that Chinese college students exhibit self-esteem when self-presentation is minimized and implicit self-evaluation is allowed to manifest.
Conclusion
By employing subliminal evaluative priming procedure that is free of self-presentation, the present research provides compelling evidence that people’s self-esteem in the collectivism culture manifest pronouncedly at implicit level when self-presentation is minimized, supporting that self-esteem is pancultural

Key words: self-esteem, implicit self-esteem, self-presentation, social desirability, subliminal affective priming