Two dissociable processes characterize self-development in adolescence

• In Event: Poster Session 07
  In Poster Session: PS 07 Topic - Social, Emotional, Personality

Abstract

Introduction. Self-concept literature proposes that the concept of self becomes more differentiated across adolescence (Harter, 2012), such that adolescents become increasingly able to make self-evaluations that are specific for different domains. Additionally, previous literature suggests that as peers become more important in adolescence, the opinions of peers may become internalized into the own self-concept (Felson, 1985; Gecas, 1982; Harter, 2012). We aimed to test these two hypotheses on the behavioural and neural level.

Methods. Previous studies investigating self-concept development and the underlying neural changes covered a small age-range or were cross-sectional. In this 3-wave longitudinal fMRI study we examined domain differentiation of self-concept in two conditions: direct- and reflected self-evaluations. Participants (189 adolescents, 10-24yr) rated to what extent a trait in academic, physical, or prosocial domain described them (direct self-concept), or to what extent they think peers would say a trait described them (reflected self-concept).

Results. Preregistered (https://osf.io/8gc6x) analyses showed that on the behavioural level, positivity of self-concept follows a negative quadratic development, with this pattern being most prominent in the academic domain (Figure 1a). Additionally, medial prefrontal cortex (mPFC) activation associated with evaluating self-traits shows a quadratic peak in boys, but increases linearly in girls (Figure 1b). To test for the possible internalization of perceived peers’ opinions into the own self-concept, we assessed developmental changes in item-by-item agreement (i.e. individual correlations between matching items on the direct and the reflected self task). Results showed a linear increase in item-item agreement across adolescence, indicating a growing overlap in trait evaluations from one’s own perspective, and the perceived perspective of a peer (Figure 2a). On the neural level, we tested whether activation in temporal-parietal cortex (TPJ: known to be involved in 3rd person perspective-taking and reflected self-evaluations; Pfeifer et al., 2017; Schurz, Radua, Aichhorn, Richlan, & Perner, 2014), would become more specific for reflected self-evaluations with age. Results showed that whereas right TPJ activation becomes relatively stronger for reflected versus direct self, left TPJ activation shows the opposite pattern (Figure 2b). Last, domain-specific differences between direct and reflected self-concept positivity at timepoint 1 predicted self-concept clarity and fear of negative evaluation 1 and 2 years later.

Discussion. This study reveals two dissociable processes of self-development. The mid-adolescent, domain-specific dip in self-concept positivity, co-occurs with linearly increasing agreement of direct and reflected self-evaluations. This is the first large longitudinal study investigating self-concept development across adolescence, providing great insight into the complex processes regarding behavioural and neural development of self in this life stage.

Authors

• Renske Van Der Cruijsen, Erasmus University Rotterdam
  Presenting Author

• Sabine Peters, Leiden University
  Non-Presenting Author

• Eveline A. Crone, Erasmus University
  Non-Presenting Author