Mood-congruent recall of affectively toned stimuli: A meta-analytic review

Abstract

A subset of the published research on mood-congruent memory in normal nondepressed, subclinically depressed, clinically depressed, induced depressed, and induced elated persons is examined with meta-analytic techniques. We estimated the magnitude of mood-congruent recall for these mood states, examined their robustness, and studied within each mood state the extent to which the strength of mood-congruent recall was related to self-referenced encoding and mood intensity. Asymmetric recall favoring positive stimuli appears to be part of the normative pattern of memory performance among individuals that have been labeled normal nondepressed ( d h = .15; p < .001); subclinically depressed individuals show symmetric recall of positively and negatively valenced material ( d h = − .02; p > .20). Clinically depressed, induced depressed, and induced elated subjects display mood congruent recall ( d h, = − .19; p < .05; d h = − .12, p < .05; d h = .08; p < .10). With the exception of induced elated mood, effect estimates derived from different studies are robust in that sampling error accounts for the entire variability among effect estimates obtained from different studies. In studies on induced-elated mood, self-referent processing was associated with stronger mood-congruent recall as compared to other studies. Caveats and implications for future research on mood and memory are discussed.