New Study Addresses the Link Between Screen Time and Adolescent Mental Health

Cheng, Q., Panayiotou, M., Finserås, T. R., Andersen, A. I. O., & Humphrey, N. (2025). How do social media use, gaming frequency, and internalizing symptoms predict each other over time in early-to-middle adolescence? Journal of Public Health. https://doi.org/10.1093/pubmed/fdaf150

Introduction

Concerns that social media and gaming are driving adolescent mental health problems have become widespread, despite mixed empirical evidence. Using a rigorous longitudinal design, Cheng and colleagues (2025) address this debate to examine how social media use, gaming frequency, and internalizing symptoms (e.g., anxiety and depression) predict one another across early-to-middle adolescence.

Grounded in uses-and-gratifications theory, the study questions assumptions of harm by considering both directions of influence and distinguishing stable individual differences from within-person effects.

Methods

The study analyzed three annual waves of data from the #BeeWell cohort in Greater Manchester, England (N = 25,629; mean age ≈ 12 years, 7 months). Social media use (hours per weekday), gaming frequency, and internalizing symptoms were self-reported. Random-intercept cross-lagged panel models separated between-person differences from within-person effects, with analyses conducted separately for girls and boys and adjusted for sociodemographic factors. Sensitivity analyses differentiated active and passive social media use.

Results

Across genders, social media use and gaming did not predict later internalizing symptoms. Notable gender-specific patterns emerged: higher gaming frequency predicted reduced later social media use among girls, while higher internalizing symptoms predicted reduced later gaming among boys. These effects were modest but meaningful, and largely replicated in sensitivity analyses.

Discussion

The findings challenge the dominant narrative that adolescent technology use is a primary causal driver of mental health difficulties. Instead, they suggest stability in both digital behaviors and symptoms, with limited evidence of harmful longitudinal effects. The results highlight the importance of analytic approaches that separate within-person effects from between-person differences, and caution against over-interpreting small associations as causal.

Implications for Mentoring Programs

Mentoring programs should avoid framing social media and gaming as inherently harmful. Instead, mentors can focus on understanding individual motivations for digital engagement, monitoring changes in participation that may signal distress, and supporting balanced, purposeful use rather than enforcing blanket restrictions.

Read the full paper here