Mental Health Predictors of Adolescents' Social Media Engagement and Digital Addiction: An Interaction of Person-Affect-Cognition-Execution Model Approach.

Abstract

Adolescents' pervasive social media use raises growing concerns about its complex links with mental health, yet predictors distinguishing normative engagement from maladaptive digital addiction remain insufficiently understood. Guided by the Interaction of Person-Affect-Cognition-Execution (I-PACE) model, this study examined whether mental health symptoms (anxiety, depression, ADHD, post-traumatic stress symptoms, eating disorder symptoms) and social factors (bullying, help-seeking attitudes, stigma, social support) differentially predicted social media engagement and digital addiction. In a cross-sectional sample of 524 adolescents (51.0 percent female; aged 15-19 years, M = 17.95, SD = 0.91), hierarchical regression analyses with multiple imputation indicated that the full models explained 12.53 percent of the variance in social media engagement (R2 = 0.125) and 26.08 percent of the variance in digital addiction (R2 = 0.261). For social media engagement, higher general anxiety (β = 0.24, p = 0.040) and greater perceived social support (β = 0.13, p = 0.008) were significant positive predictors, while more favorable help-seeking attitudes were negatively associated with engagement (β = -0.10, p = 0.039). For digital addiction, perceived social support was a significant positive predictor (β = 0.14, p = 0.002), while PTSS (β = 0.14, p = 0.058), bullying (β = 0.11, p = 0.055), and ADHD symptoms (β = 0.15, p = 0.092) showed trend-level associations; depressive symptoms were not significant in the fully adjusted model (β = 0.10, p = 0.317). Overall, findings provide partial support for the I-PACE model and suggest that trauma-related symptoms may warrant further investigation as a subclinical vulnerability associated with problematic digital behaviors.