Long-term impact of digital media on brain development in children

Abstract

Digital media (DM) takes an increasingly large part of children's time, yet the long-term effect on brain development remains unclear. We investigated how individual effects of DM use (i.e., using social media, playing video games, or watching television/videos) on the development of the cortex (i.e., global cortical surface area), striatum, and cerebellum in children over 4 years, accounting for both socioeconomic status and genetic predisposition. We used a prospective, multicentre, longitudinal cohort of children from the Adolescent Brain and Cognitive Development Study, aged 9.9 years when entering the study, and who were followed for 4 years. Annually, children reported their DM usage through the Youth Screen Time Survey and underwent brain magnetic resonance imaging scans every 2 years. Quadratic-mixed effect modelling was used to investigate the relationship between individual DM usage and brain development. We found that individual DM usage did not alter the development of cortex or striatum volumes. However, high social media usage was associated with a statistically significant change in the developmental trajectory of cerebellum volumes, and the accumulated effect of high-vs-low social media users on cerebellum volumes over 4 years was only β = - 0.03, which was considered insignificant. Nevertheless, the developmental trend for heavy social media users was accelerated at later time points. This calls for further studies and longer follow-ups on the impact of social media on brain development.

Keywords: Brain; Children; MRI; Polygenic scores; Social media; Videogames.

Figure 1

(a) Plots representing the quadratic effects of age (years) predicting global cortical surface area, striatum and cerebellum volumes in the overall cohort, adjusted for sex, socioeconomic status, polygenic scores for cognitive performance, and 20 principal components, and the grey shade around the regression lines corresponds to a 95% confidence interval of the intercept; (b) sex-stratified developmental trend adjusted for the same covariates as mentioned above. The dots represent the peak age, estimated by the first derivative. The y-axis represents brain structures.

Figure 2

Developmental trajectories and socioeconomic status, (a) global cortical surface area, and (b,c) cerebellum (presented as adjusted and unadjusted for total brain volumes (TBV)). For visual purposes, we present in age (years). socioeconomic status (SES) is categorized into quartiles using ggpredict [quart2] function in R. Children from low levels of socioeconomic status had a relatively smaller global cortical surface area or cerebellum volumes and accelerated maturation of the brain compared to their developing peers. SES socioeconomic status.

Figure 3

Relationship between digital media usage and cerebellum development over time. The interactions presented in (a,c) social media usage (b,d) playing video games, and time² on the cerebellum development; however, they are presented in age (years) for visual purposes. Digital media usage is categorized based on quartiles using ggpredict [quart2] function in R. Children who spent a longer time on social media usage (a,c) had a decrease in cerebellum volume. Similar findings were seen for those who spent on mean levels. In contrast, children who spent a longer time playing video games (b,d) had an increase in cerebellum volume.