Bidirectional Influences Between Social Media Use and Adolescent Depression: Testing the Role of Frontoparietal Connectivity - ABSTRACT Social media is now a major source of environmental input for adolescents. Over 90% of U.S. adolescents are regularly using social media, with 45% reporting that they are online “nearly constantly”. While 80% of adolescents report that social media use (SMU) makes them feel “more connected” with their peers, there are nonetheless growing concerns that pathologically heavy use is impacting attentional focus, executive functioning, and social cognition, and driving increases in depression among youth, particularly among girls. Many of the brain regions constituting frontoparietal networks (FPNs) that support attentional focus, executive function, and social cognition have also been implicated in depression. However, there is limited data linking SMU with alterations in the development of FPNs in the context of psychopathology risk. We have previously demonstrated that greater use of negative emotional words on social media apps is associated with greater within-network functional connectivity of both medial and dorsal FPNs; a diagnosis of depression has also been associated with greater functional connectivity of the medial FPN but lower functional connectivity of the dorsal FPN with midcingulo-insular networks. However, no studies to date have examined whether pathological SMU potentiates depression risk (or vice-versa), and the extent to which FPN development is impacted by SMU. Toward this end, we propose to leverage five waves of longitudinal data from the Adolescent Brain Cognitive Development (ABCD) Study (N>10,000, ages 13–18) to test associations among problematic (v. normative) SMU, risk for adolescent depression, and FPN development. Unlike previous studies in this area which focus on self-reported SMU, we will leverage smartphone-based passive sensing technology to ascertain objective measures of total exposure (i.e., amount of time spent) and nature of engagement (e.g., applying sentiment analyses to determine frequency of negative emotional content) with social media apps. These metrics of social media exposure and engagement will be combined with clinical and self-reported assessments of depression and multimodal (resting-state fMRI, diffusion MRI) measures of dorsal, lateral, and medial FPN connectivity to test three independent Aims. Specifically, we will test whether and how SMU shapes adolescent development of FPN connectivity (Aim 1), whether trajectories of FPN connectivity mediate associations between SMU and depressive symptoms and vice-versa (Aim 2), and adopt a data-driven machine learning strategy to test whether SMU and FPN connectivity outperform other factors in predicting adolescent-onset depression (Aim 3). The results of this innovative and scientifically rigorous study will answer critical questions on what constitutes potentially pathological SMU, who may be disproportionately impacted by SMU in the context of depression risk, and identify modifiable neurodevelopmental mechanisms by which SMU may lead to depression. All research findings will be disseminated responsibly to inform guidance for youth and their families, clinicians, educators, technology-based stakeholders, and policymakers.
