• We found a bidirectional relation between social anxiety and digital media preference. • This association only emerged when attention bias was accounted for. • Larger N2 increased the effect of digital media preference on later social anxiety. Digital media is now a ubiquitous context through which youth interact with their peers, raising questions about its role in social anxiety development. Attentional biases toward or away from threatening information have been implicated in the DMU-anxiety link, although no study to date has provided evidence specific to social anxiety and how these patterns unfold across adolescence. We leveraged a longitudinal sample to examine (1) bidirectional relations between preference for using digital media, relative to face-to-face interactions, and social anxiety symptoms and (2) the moderating role of attention bias. Beginning at age 12-14, adolescents ( N = 155) self-reported social anxiety symptoms and digital media communication preferences across four annual timepoints. Additionally, adolescents completed a dot probe task each year while EEG was recorded. Attention bias was indexed using trial-level reaction time measures and event-related potentials capturing attentional selection and discrimination (N170) and cognitive control (N2) in response to threatening versus neutral faces. Univariate multilevel models were used to examine the lagged effect of digital media preference on social anxiety symptoms, and vice versa. Results showed that digital media preference and social anxiety symptoms were significantly predictive of each other only when individual differences in attention bias were accounted for. N2 to threat, but no other attention bias measures, strengthened the association between greater digital media preference and higher subsequent social anxiety symptoms. Our findings highlight the key role of threat-related attentional processing in clarifying links between digital media use and social anxiety symptoms across adolescence.
Politte-Corn et al. (Sun,) studied this question.