Effects of working memory training on depressive symptoms: a systematic review and meta-analysis of randomized controlled trials

Working memory (WM) deficits are commonly observed across the depression spectrum and may represent a modifiable cognitive mechanism. This systematic review and meta-analysis evaluated the effects of WM training (WMT) on depressive symptoms and WM performance, and examined potential moderators and the association between WM gains and symptom change. Systematic searches of PsycINFO, Embase, and PubMed were conducted in December 2022 and updated in January 2025. Eligible studies were randomized controlled trials (RCTs) using WMT as the sole cognitive intervention with depressive symptoms as outcomes. Two-level random-effects models were used for depressive symptoms and three-level models for WM outcomes. Risk of bias was assessed by the Cochrane Risk of Bias Tool. Twenty-seven RCTs involving 1692 participants were included. WMT produced a small but significant reduction in depressive symptoms at post-training (Hedges' g = 0.20, 95% CI [0.04, 0.35]), though effect was not maintained at follow-up. Moderate improvements in WM were observed at post-training (g = 0.61, 95% CI [0.24, 0.98]) and follow-up (g = 0.72, 95% CI [0.07, 1.38]). Post-training WM gains were positively associated with symptom reductions at follow-up. Moderator analyses indicated that baseline symptom severity and training material valence may influence outcomes, though subgroup effects were not statistically significant and warrant cautious interpretation. These findings suggest that WMT targets a potentially modifiable mechanism in depression, but its clinical utility remains uncertain. Larger RCTs, particularly in clinical populations and with extended follow-ups, are needed to establish more definitive evidence regarding its preventive and therapeutic value. (Protocol registration: PROSPERO, CRD42023400213.).