4 - (OP24) Effects of Digital Autobiographical Memory Specificity Training on Depressive Symptoms and Its Process-based Indicators in Depressed Older Adults

Background: Memory Specificity Training (MeST) targets overgeneral autobiographical memory, a core cognitive feature of depression, and has been shown to improve memory specificity and depressive symptoms. However, the mechanisms underlying these effects remain unclear, and research on mobile-based MeST in older adults is particularly limited. The present study examines the effects of mobile-based MeST in depressed older adults and explores process-based indicators that may account for individual differences in treatment response.

Methods: This study utilized RCT data from 44 older adults (age ≥ 60, M=73.36) randomly assigned to either a mobile MeST group (n=20) or a mobile cognitive training control group (n=24). Both interventions were delivered via a unified mobile app with group-specific modules. Depressive symptoms (CES-D) and memory specificity were measured at baseline, post-intervention, and 1-month follow-up. Mixed-design repeated measures analysis was conducted to examine intervention efficacy. For the MeST group, training logs were analyzed to examine the relationship between depressive symptom changes and indicators including training dosage and mastery efficiency (average attempts to reach maximum autobiographical memory specificity). To evaluate differences by emotional cue types, linear mixed-effects models (LMM) tested the interaction between time and cue valence (positive vs. negative).

Results: Mixed-design analyses showed that the mobile-based MeST group exhibited a significantly greater reduction in depressive symptoms from pre- to post-intervention compared to the cognitive training control group (B = −6.57, p = .02), and this effect was maintained at the one-month follow-up (B = −8.01, p < .001). Autobiographical memory specificity also increased from pre- to post-intervention, with a greater increase observed in the MeST group (B = 1.70, p < .001). Training log analyses indicated that training dosage was not significantly associated with changes in memory specificity or depressive symptoms (all ps > .05). In contrast, higher training mastery—defined as fewer attempts to reach maximum specificity—was associated with greater reductions in depressive symptoms (Wilcoxon W = 23.5, p = .048, g = 0.86). Linear mixed-effects models further revealed that while no cue valence differences were present at baseline, specificity for positive cues was significantly higher than for negative cues immediately post-intervention (estimate = −0.19, p = .007), with no differences observed at follow-up.

Conclusions: The findings suggest that mobile-based MeST may reduce depressive symptoms and enhance autobiographical memory specificity in older adults. Treatment response appeared to be more closely related to mastery efficiency than to training dosage, and the early preferential specification of positive memories may reflect an initial cognitive process underlying symptom change. Despite the exploratory nature of the analyses and small sample size, these results highlight the importance of mechanism-focused research incorporating process-based indicators in future digital intervention studies.