Abstract: The self-complexity model proposes that a highly differentiated self-concept is linked to better coping in response to stress and negative events. Subsequent reports, however, revealed that the relationship of self-complexity and well-being was qualified by strong heterogeneity among studies. Various studies have identified problems concerning the consistency and validity of the often-used measure of self-complexity. Addressing these issues, the current work has been conducted with several purposes. First, the study was to provide tests of the moderating effects of different facets of self-complexity. Second, in order to clear up the role of self-complexity, the present study compared the structural measures of self-complexity by Linville to the component measures by Rafaeli-Mor et al. Third, some modifications of the typical cart-sort task, such as the renewal of the trait-pool and the administration of lexical decision tasks and role-centered sorting strategy, were made to precisely elicit the characteristics of adolescents in Chinese cultural context. Fourth, it attempted to explore the relationship between self-complexity and resilience.
Method
A total of 175 high school students, 95 males and 80 females, participated as part of their psychology course requirement (two of them were eliminated because of incomplete fulfillment). Subjects completed measures of stressful life events, self-complexity, depression and resilience in three sessions across 3 weeks. A multiple regression analysis used depression at Time 3 as outcomes, stressful life events at Time 1 and Time 3, self-complexity as predictors, and depression at Time 1 as control variables. In addition, path analysis was conducted to examine the roles of self-complexity and resilience.
Results
Results showed that: 1) the component measures of self-complexity has both greater face validity and substantial internal consistency, as compared to the structural measures; 2) at times of past life stress, high numerosity was found to attenuate adverse effects of negative events on depression, irrespective of the overlap of the self-aspects; In addition, the numerosity×stress interaction was consistent with the stress protection; 3) at times of current stress, a single 3-way interaction emerged. Specifically, adolescents with few and overlapping roles experienced elevated depression when facing high current stress, but lower depression when facing low current stress; adolescents with high and overlapping roles were less prone to depression at times of high current stress, whereas the ones with high and distinct roles seemed to cope the minor current stress better. Thus the results of this study have supported only several of its hypotheses regarding self-complexity as a buffer for the depression.
Conclusions
Results suggested that, adolescents with high and overlapping roles had clear and consistent self-concepts enabling them to function effectively at times of stress; whereas high numerosity and low overlap might reflect fragmentation and a lack of integration among self aspects, which may be one of the cognitive vulnerable factors to negative life events; whereas adolescents with low and overlapping roles expressed a rather simple self-system, whose self-perception may be easily influenced by external events. It may imply that integration and differentiation of self-aspects were the simultaneous functioning of opposite processes in the development of one’s self-concept. In addition, analysis revealed that resilience might be highly influenced by self-complexity

Key words: self-complexity, past life events, current life events, depression, resilience