A year in the social life of a teenager: Within-person fluctuations in stress, phone communication, and anxiety and depression

Abstract

Stressful life events (SLEs) are strongly associated with the emergence of adolescent anxiety and depression, but the underlying mechanisms remain poorly understood, especially at the within-person level. We investigated how adolescent social communication (i.e., frequency of calls and texts) following SLEs relates to changes in internalizing symptoms in a multi-timescale intensive year-long study (N=30; n=355 monthly observations; n=~5,000 experience-sampling observations). Within-person increases in SLEs were associated with receiving more calls than usual at both monthly- and momentary-levels, and making more calls at the monthly-level. Increased calls were prospectively associated with worsening internalizing symptoms at the monthly-level only, suggesting that SLEs rapidly influences phone communication patterns, but these communication changes may have a more protracted, cumulative influence on internalizing symptoms. Finally, increased incoming calls prospectively mediated the association between SLEs and anxiety at the monthly-level. We identify adolescent social communication fluctuations as a potential mechanism conferring risk for stress-related internalizing psychopathology.

Linguistic measures of psychological distance track symptom levels and treatment outcomes in a large set of psychotherapy transcripts

“…This is because a relationship between linguistic distance and internalizing symptoms could emerge in mixed-effect models either 1) because, as people increase their linguistic distance, their symptoms reduce (a within-person relationship) or 2) because individuals who, overall, have higher linguistic distance have lower symptoms than individuals who, overall, have lower linguistic distance (a between-person relationship). We consequently followed prior work in decomposing linguistic distance into within-person and between-person components and used these components in mixed-effects regressions ( 67 – 72 ). A variable representing the within-person fluctuation in linguistic distance was created by subtracting each individual’s mean linguistic distance score from the score of each of their observations, producing a variable representing within-person deviation, centered around their individual mean.…”

“…However, the appropriate method for bootstrapping multilevel data is not clear, as random samples can be drawn at the participant level, at the observation level, or at both participant and observation levels. We thus used Bayesian analytic procedures—which do not involve bootstrapping methods—for our mediation analyses, to sidestep this issue ( 70 ).…”

Leveraging deep learning models to understand the daily experience of anxiety in teenagers over the course of a year

“…The study inclusion criteria ensured that participants were female, age 15-17, fluent in English, and had access to an individual smartphone. 14 Twelve monthly in-person assessments were conducted on each participant to assess their experience with anxiety and depression over the past month, measured using the GAD-7 and PHQ-9 scales, respectively. 7,15 Within the same year, participants underwent four separate occasions of three weeks worth of monitoring by daily ecological momentary assessments where they would report their feelings of anxiety and depression (each on a 1 to 7 scale) three times a day through the MetricWire app.…”

Linguistic measures of psychological distance track symptom levels and treatment outcomes in a large set of psychotherapy transcripts

“…This is because a relationship between linguistic distance and internalizing symptoms could emerge in mixed-effect models either (i) because as people increase their linguistic distance, their symptoms reduce (a within-person relationship) or (ii) because individuals who overall have higher linguistic distance have lower symptoms than individuals who overall have lower linguistic distance (a between-person relationship). We consequently followed prior work in decomposing linguistic distance into within-person and between-person components and used these components in mixed-effects regressions (69)(70)(71)(72)(73)(74). A variable representing the withinperson fluctuation in temporal distance was created by subtracting each individual's mean temporal distance score from the score of each of their observations, producing a variable representing within-person deviation, centered around their individual mean.…”

“…However, the appropriate method for bootstrapping multilevel data is not clear, as random samples can be drawn at the participant level, at the observation level, or at both participant and observation levels. We thus used Bayesian analytic procedures-which do not involve bootstrapping methods-for our mediation analyses to sidestep this issue (72).…”