A meta-analytic review of the relationship between explicit memory bias and depression: Depression features an explicit memory bias that persists beyond a depressive episode.

Abstract

Emotional bias in explicit memory is theorized to play a prominent role in the etiology, maintenance, and recurrence of depression. Even though this cognitive bias is regarded as one of the most robust phenomena in depression, its magnitudeandboundaryconditions indepressionare currentlyunknown.Thisreviewpresents two three-level meta-analyses to estimate the overall effect size and identify moderators of explicit memory bias in depression. Meta-analysis I (153 studies, 686 contrasts) revealed a small overall effect size for naturalistic explicit memory bias in depression, g = 0.241, 95% CI [0.179, 0.304]. The magnitude of the overall effect was moderated by emotional valence of stimuli, operational de fi nition of memory bias, depth of processing during encoding, explicit memory task, and the (non-)verbal nature of stimuli. Equivalent effect sizes were found for minors and adults as well as for clinical and subclinical depression. Remarkably, a nonsigni fi canteffectsizeemergedfor remitteddepression.Followinguponthelatter fi nding,Meta-analysisII (21 studies, 80 contrasts) examined explicit memory bias in remitted depression under naturalistic conditions and under mood/stress induction. Results yielded a nonsigni fi cant overall effect size, g = 0.131, 95% CI [ − 0.045, 0.307], but a signi fi cant effect size for study conditions with mood or stress induction, g = 0.273, 95% CI [0.004, 0.542]. Both meta-analyses indicated high levels of heterogeneity, even after accounting for variation explained by sample and study characteristics. The fi ndings are consistent with the view that depression is characterized by an explicit memory bias that may persist beyond a depressive episode. These fi ndingshaveimplicationsforcognitivetheoriesofvulnerabilitytodepressionaswellasclinicalinterventions.