Echoes of Extreme Weather: Multilevel Latent Profiles of Memory Retrieval in the Acute Aftermath of Climate Change–Related Trauma

Abstract

Objective: Climate change threatens mental health through increased extreme weather events (Clarke et al., 2022). Following trauma, a subset will sustain chronic mental health problems, such as posttraumatic stress disorder (PTSD), depression, and alcohol use disorder. Adaptive peri-disaster memory retrieval may facilitate the construction of a recovery-consistent trauma memory (Zoellner et al., 2025). The characteristics of adaptive processing are debated, with some highlighting contextual processing (specific, situational details of event; e.g., Brewin, 2014) and others pointing to semantic processing (efforts to organize, integrate, or make sense of events; e.g., Park, 2022). In the current study, individuals who engaged in latent patterns of greater contextual and semantic peritraumatic memory retrieval over a two-week daily diary period were predicted to report lower re-experiencing symptoms, as well as total PTSD, depression, and alcohol misuse severity at a one-month follow-up. Method: The current study explored the dynamic role of trauma memory retrieval in recent (past three month) extreme weather event survivors (N = 81). Participants reported memory retrieval frequency (intentional, intrusive) and characteristics (distress, vividness, contextual processing, semantic processing) daily for 14 days. Self-reported re-experiencing symptoms, as well as total PTSD, depression, and alcohol use problems were collected at one-month follow-up. Multilevel latent profile analysis (MLPA) was used to observe latent profiles of daily memory retrieval characteristics, and person-level classes of memory retrieval styles. Latent classes were used to predict re-experiencing symptoms, and total PTSD, depression, and alcohol use severity. Results: A four-profile model of daily memory retrieval characteristics was fit, with “High Activation,” “Low Activation,” “Low-Frequency, Semantic,” and “Intrusive, Low Processing” days emerging. Subsequently, four distinct classes of participants were observed, composed of varying frequencies of the latent daily memory retrieval profiles. The class dominated by “High Activation” days (frequent, distressing retrievals; high semantic and contextual processing) reported the highest re-experiencing (d = 1.25 – 1.77), PTSD (d = 1.05 – 1.53), depression (d = 1.05), and alcohol misuse (d = 0.88 – 1.37). Discussion: Peritraumatic memory retrieval characteristics were consistent predictors of psychopathology, with the combination of high memory frequency, distress, vividness, contextual, and semantic processing being associated with worse mental health outcomes one month later. Higher contextual and semantic processing potentially points to less adaptive, ruminative processes. Heterogeneity of memory retrieval characteristics were observed across the two-week assessment period, highlighting the dynamic nature of acute post-event memory retrieval. The development of light-touch, memory-based early interventions offers considerable promise. Such interventions should adopt a “right-sizing” approach, matching individuals’ cognitive styles with tailored memory retrieval guidance.