Spontaneous thought characteristics are differentially related to heightened negative affect versus blunted positive affect in adolescents: An experience sampling study

Abstract

Background Mind‐wandering has been linked to negative affect (NA) and depressive symptoms in adolescents. However, mind‐wandering is an extremely broad and heterogeneous cognitive construct. Some features of spontaneous thought may be related to increased NA, whereas others may improve affect, or have no emotional influence. We used ecological momentary assessment (EMA) to investigate the characteristics of spontaneous thoughts in adolescents and their differential relations with moment‐to‐moment affect. Method One‐hundred and sixteen adolescents (ages 13–18; Typical Mood [TM] = 58; Low Mood [LM] = 58) completed 5 days (2–3 times/day) of EMA (total 1037 surveys) assessing current positive affect (PA) and NA and dimensions of spontaneous thought. Multilevel models tested the relation between thought characteristics and affect. Results Relative to the TM group, LM adolescents had a higher frequency of mind‐wandering (38% vs. 56%) and negatively‐valanced thoughts during episodes of mind‐wandering (21% vs. 37%). Negatively‐valenced, self‐referential and past‐oriented thoughts were each associated with higher NA, even when controlling for plausible confounds (e.g., engagement in an unpleasant activity or social interaction, depressive symptom severity). In contrast, task‐focused and positively‐valenced thoughts were uniquely linked to higher PA. Conclusion Characteristics of spontaneous thought—including temporal orientation, self‐referential quality, and task‐relatedness—were differentially related to NA versus PA in adolescents. If replicated, these findings could inform more nuanced assessments of and targeted interventions for specific dimensions of mind‐wandering contributing to high NA versus blunted PA in teens.

Revealing subgroup-specific mechanisms of change via moderated mediation: A meditation intervention example.

“…Importantly, this pattern was not observed in the control condition (red line in Figure 2). One possible interpretation of this pattern of findings is that participants in the HMP group with greater baseline perseverative negative thinking (i.e., lower PAI scores) may have had more room for improvement in mindfulness skills focused on cultivating attention to the present moment (in contrast to negative perseverative thoughts which are often focused on the past, e.g., rumination, or future, e.g., worries; Nolen-Hoeksema et al, 2008; Webb, Tierney, et al, 2022). Thus, exposure to mindfulness training through HMP may have allowed those predicted to be most responsive to acquire beneficial mindfulness skills.…”

Dynamic processes in behavioral activation therapy for anhedonic adolescents: Modeling common and patient-specific relations.

“…Thursdays through Sundays was selected to sample mood, cognitions, and behaviors in adolescents on weekdays (i.e., school days) versus weekends, instead of overburdening participants with EMA surveys 7 days a week (for similar EMA designs in youth, see Alarcón et al, 2020; Carper et al, 2022; E. E. Forbes et al, 2009, 2012; Murray et al, 2022; Webb, Israel, et al, 2021; Webb, Tierney, et al, 2022). Surveys were triggered via a time-stratified random sampling approach such that participants were signaled via a popup notification on their phone once at a random time during two timeslots (4 p.m. to 6:30 p.m. and 6:30 p.m. to 9:00 p.m.), with surveys separated by a minimum of 1 hr.…”

“…E. Forbes et al, 2009E. E. Forbes et al, , 2012Murray et al, 2022;Webb, Israel, et al, 2021;Webb, Tierney, et al, 2022). Surveys were triggered via a time-stratified random sampling approach such that participants were signaled via a popup notification on their phone once at a random time during two timeslots (4 p.m. to 6:30 p.m. and 6:30 p.m. to 9:00 p.m.), with surveys separated by a minimum of 1 hr.…”

Effect of subconscious changes in bodily response on thought shifting in people with accurate interoception

“…This is because we would like to stabilize convergence in estimating models with many parameters. There has been an increasing number of attempts to analyze self-reported data on thought using Bayesian statistical modeling 83 , 84 . We used a generalized linear mixed model to account for the distributional characteristics of each acquired dataset.…”