Poster #6 - Neural similarity for self and peer predicts adolescents’ risk taking and peer influence

• In Event: Poster Session 19

Abstract

Background: Adolescence is marked by increased risk-taking behaviors and heightened peer influences. Recent literature suggests enormous individual variations in risk taking and peer influence susceptibility in adolescents (Brechwald & Prinstein 2011; Telzer et al., 2021). Nonetheless, it remains largely unknown which underlying brain mechanism makes some youth more susceptible to their peers and take more risks. Social identity and social learning theories suggest that the extent to which individuals identify with and learn from their close others (i.e., self-peer overlap) in risk-taking contexts may underlie the increased risk taking and peer influence susceptibility in adolescents (Kwon & Telzer, 2022). This self-other overlap can be examined via the similarity between neural representations of self and others during decision making using pattern-based analytical techniques such as representational similarity analysis (RSA). The current study uses the RSA approach to examine the neural representations of adolescent risk taking for the self, their peers, and their parents, with a particular focus on how variability in the neural similarity between decision making for self and peers (i.e., self-peer overlap) may be associated with individual differences in adolescents’ risk-taking behaviors and peer influence susceptibility.
Methods: 169 adolescent participants (Mage = 12.90) completed the Cups task during an fMRI scan, in which adolescents made risky decisions for themselves, their parents, and their best friends (see Figure 1. A). After scanning, adolescents completed self-reported surveys measuring their risk-taking propensity (Alexander et al., 1990) and resistance to peer influence (Steinberg, 2007). Trial-based general linear model (GLM) was conducted to extract the beta weight for each condition, then voxel-wise pattern similarity coefficients were calculated for each participant in each region of interest (ROI; see Figure 1. B). The normalized similarity coefficients were used for subsequent correlation analyses. The nucleus accumbens (NACC) and ventromedial prefrontal cortex (vmPFC) were selected as ROIs given their key roles in representing reward value.
Results: We found substantial individual variation in the way adolescents represented themselves and peers while making decisions. Notably, adolescents with greater neural similarity in the NACC for the self and their best friend during decision making were more likely to engage in risk-taking behaviors and be more susceptible to peer influence (see Figure 2). In contrast, this self-other overlap effect in risk taking and peer influence susceptibility was not found in the neural similarity between self and parent, suggesting the specificity of self-peer overlap in the NACC in adolescent risk taking and peer influence.
Discussion: This study used a representational similarity approach to investigate the neurobiological mechanisms underlying adolescents’ risk taking and peer influence susceptibility. Our study findings add to the developmental neuroimaging literature, suggesting that the self-peer overlap in the neural responses of NACC while making risky decisions may represent an individual’s risk-taking behaviors and peer influence susceptibility during adolescence. This self-peer representational similarity in the reward valuation system may indicate equivalent sensitivity to rewards with peers and internalized peer norms about taking risk in adolescence.

Authors

• Junqiang Dai, University of North Carolina at Chapel Hill
  Presenting Author

• Seh-Joo Kwon, University of North Carolina at Chapel Hill

• Mitchell J Prinstein, University of North Carolina at Chapel Hill

• Kristen A Lindquist, University of North Carolina at Chapel Hill

• Eva H Telzer, University of North Carolina at Chapel Hill