Longitudinal Links between Expressive Flexibility and Friendship Quality in Adolescence: The Moderating Effect of Social Anxiety

• In Event: Poster Session 01
  In Poster Session: PS 01 Topic - Social, Emotional, Personality

Abstract

Expressive Flexibility (EF), or the ability to shift between enhancing and suppressing emotional expressions based on situational demands (Bonanno et al., 2004), seems to be interrelated with adolescents’ social adjustment. Sometimes they must exaggerate excitement when receiving unwanted gifts, for example, or avoid showing irritation toward friends, in order to maintain social harmony. Previous longitudinal and experimental research has consistently evidenced the positive associations from healthy peer relationships to youths’ EF (e.g., Wang et al., 2020). Nevertheless, longitudinal links from earlier EF to later social adjustment appear to be relatively modest and unstable (Wang & Hawk, 2019), highlighting the necessity of investigating intervening factors in this process. Social anxiety might be a potential moderator, for example; Socially anxious adolescents may fail to enhance/suppress their expressions at appropriate moments or to a socially acceptable degree (Kashdan & Breen, 2008), thus leading to counterproductive outcomes. With a three-wave longitudinal design, the present study aimed to test the bidirectional links between EF and friendship quality in early adolescence, as well as the potential moderating effect of social anxiety.

Participants from two junior high schools in eastern China (N=274; 49.8% female; Mage=13.57, SD=0.63) were surveyed at three time points with six-month intervals. At each wave, participants completed the Child and Adolescent Flexible Expressiveness Scale (Wang & Hawk, 2020), The Friendship Quality Questionnaire (Parker & Asher, 1993), and the Social Anxiety subscale of the Self-Consciousness Scale (Scheier & Carver, 1985). A fully recursive path model covered all cross-lagged effects between EF, Social Anxiety, and Friendship Quality. The interaction terms between EF and Social Anxiety were added as additional predictors of Friendship Quality at the subsequent wave (see Figure 1). All cross-lagged and stability paths were constrained across waves. The model was estimated in Mplus 7.0 (Muthén & Muthén, 1998-2017), via MplusAutomation (Hallquist & Wiley, 2018) in R.

Consistent with previous studies, earlier Friendship Quality significantly predicted higher EF later on (βs=.16, 95%CIs [.08, .24], SEs=.04, ps<.001). Although EF did not predict later Friendship Quality (ps≥.410), the EF x Social Anxiety interaction significantly predicted later Friendship Quality (βs=-.08, 95%CIs [-.14/-.15, -.01], SEs=.03, ps=.018). We examined this interaction via the Johnson-Neyman technique (Preacher et al., 2006), which showed the changes of associations between EF and Friendship Quality at each level of Social Anxiety. As seen in Figure 2, the slopes relating EF to Friendship Quality became less strongly positive and gradually became negative as Social Anxiety increased. EF had a significant positive effect on Friendship Quality for participants scoring lower than 0.63 SD below the mean of Social Anxiety; for other participants, this effect was not significant.

These results echoed existing research on the importance of a supportive peer context for promoting adolescents’ expressive flexibility. Importantly, however, we also provided novel evidence suggesting that this flexibility is not always socially adaptive. EF contributed to higher friendship quality, but only among adolescents with lower social anxiety. Therefore, practical interventions aiming to improve youths’ expressive flexibility might also need to consider the role of personality characteristics.

Authors

• Yingqian Wang, Central University of Finance and Economics
  Presenting Author

• Skyler T. Hawk, Chinese University of Hong Kong (CUHK)
  Non-Presenting Author

• Susan Branje, Utrecht University
  Non-Presenting Author

• Caspar J. Van Lissa, Universiteit Utrecht
  Non-Presenting Author