Smartphone language and resting‐state EEG indicators of self‐focused attention prospectively predict major depressive disorder risk in adolescents

Background Central to major depressive disorder (MDD) onset and maintenance is maladaptive self‐focused attention, which can be reliably indexed by greater: (a) usage of first‐person singular pronouns (e.g., I ) in natural language and (b) alpha oscillations in resting‐state EEG. Integrating these largely parallel bodies of research, the present study sought to explicate the associations between, and prospective predictive utility of, linguistic and neural indicators of self‐focused attention in adolescents with remitted MDD over 12 months. Methods At baseline, 126 adolescents (ages 13–18) with ( n = 66) and without ( n = 60) remitted MDD completed resting‐state EEG. Retrospective interviews determined the occurrence of major depressive episodes (MDEs) during the follow‐up period. A total of ~2.3 million messages were passively acquired from adolescents' smartphones, on which the proportion of first‐person singular pronouns was derived. Results During the 12 months, 29 (23.0%) participants developed an MDE (28 remitted MDD, 1 control). Cox regression showed that while greater usage of first‐person singular pronouns prior to MDE increased the risk for MDE (hazard ratio [HR] = 2.02, p < .001), greater resting‐state alpha power at baseline decreased the risk for MDE (HR = 0.78, p = .001). Moreover, greater alpha power predicted subsequent first‐person singular pronoun usage ( β = 0.17, p = .004). Mediation analysis indicated a marginal suppression effect (bootstrapped indirect effect p < .10), such that accounting for first‐person singular pronoun usage amplified the association between alpha power and MDE risk. Conclusions Findings highlight functionally distinct alpha mechanisms and provide support for smartphone‐based first‐person singular pronoun usage as a neurobehavioral risk factor and a potentially promising intervention target for adolescent MDD.