The longitudinal impact of screen media activities on brain function, architecture and mental health in early adolescence

Abstract

The increased use of screen media has raised unknown effects on mental health among adolescents. This study aimed to examine the correlational and causal association between screen media activity (SMA) and mental health problems, and the mediating role of brain functions and structures in this relationship. Data from 4557 adolescents (mean age = 9.955 ± 0.164 years) in the Adolescent Brain Cognitive Development (ABCD) study were analysed across four time points: baseline, 1-year, 2-year, and 3-year follow-ups. Linear mixed models assessed SMA's association with mental health indices and the brain's developmental pattern, respectively. Cross-lagged panel models examined the SMA-mental health problems' longitudinal and causal relationship. Mediation analyses explored brain functions and structures as mediators on the SMA-mental health correlation. Baseline SMA positively correlated with internalizing, externalizing, and stress problems; and negatively correlated with brain volume, area and diverse sets of resting-state functional connectivity (RSFC) after three years. Higher baseline SMA associated with increased internalizing (β = 0.030, SE= 0.012, p_fdr = 0.016), and stress problems (β = 0.026, SE = 0.012, p_fdr = 0.037) three years later. The RSFC between the cingulo-opercular network (CON) and the retrosplenial temporal network (RTN) mediated the effects of SMA on externalizing (β = 0.002, p_fdr = 0.042) and stress problems (β = -0.003, p_fdr = 0.022). TV watching predicted higher externalizing problems (β = 0.054, pfdr < 0.001), while video watching predicted increased internalizing (β = 0.061, pfdr < 0.001), externalizing (β = 0.033, pfdr = 0.035), and stress problems (β = 0.060, pfdr < 0.001). The findings indicate the negative impact of SMA, particularly TV and video watching, on adolescent mental health, mediated by changes in CON and RTN functional connectivity. Future research can explore the specific risks associated with video streaming and consider the role of emerging technologies such as virtual reality in SMA on adolescent mental health.

Keywords: Adolescence; Brain functional connectivity; Longitudinal; Screen media activity.

Fig. 1

A) Our study used the ABCD study's multimodal data, which consisted of over 10,000 US adolescents aged 9–10 years. The data of SMA includes 6 types only baseline and 3-year follow-up data were included in this study. The brain imaging data includes structural and functional MRI data were acquired at the 2-year follow-up. Mental health problems, including behavioral and stress-related symptoms, were assessed using the parent-reported Child Behavior Checklist (CBCL) at baseline and 3-year follow-up. Only data from baseline (BL), 2-year follow-up (2YFU), and 3-year follow-up (3YFU) were used in the final analyses; 1-year follow-up (1YFU) data were excluded. B) For each type of SMA and overall SMA, we examined the associations between usage time and mental health data and brain data. C) Bidirectional longitudinal associations between SMA and mental health problems were examined for each type of SMA and overall SMA, using a two-wave cross-lagged panel model (SMA and CBCL measured at baseline and 3-year follow-up). D) For each type of SMA and overall SMA with effects, we examined if neural correlates related to SMA mediated the effects of SMA on behavioural and stress disorders.

Fig. 2

The association results between SMA and brain developmental pattern. A. The association between SMA usage and brain morphology, including cortical volumes, area and thickness. Only significant brain regions are displayed in colour. B. Left is the associations between SMA and RSFCs among 12 cortical networks [auditory network (AN), visual network (VN), sensorimotor hand network (SHN), sensorimotor mouth network (SMN), cingulo-opercular network (CON), cingulo-parietal network (CPN), dorsal attention network (DAN), default mode network (DMN), fronto-parietal network (FPN), retrosplenial temporal network (RTN), salience network (SN), ventral attention network (VAN)]. Right is the associations between SMA and cortico-subcortical RSFCs [10 subcortical regions including cerebellum cortex (Crcx), thalamus (Tha), Hippocampus (Hip), amygdala (Amg), Putamen (Pt), pallidum (Pl), caudate (Cde), nucleus accumbens (NAc), ventral diencephalon (Vtcd) and brainstem (BS)]. (p_fdr < 0.05).