A meta-analytic review of mood-congruent implicit memory in depressed mood
Introduction
A central premise of cognitive models of depression is that depressed mood is linked to a number of alterations in cognitive processes such as attention, self-perception, and memory (e.g., Beck, 1967, Ingram et al., 2011). With regard to memory-related alterations, researchers have reliably found that explicit recall for depression-congruent information is greater than recall for depression-incongruent information among individuals experiencing a depressed mood state (see Matt, Vazquez, & Campbell, 1992). This process is part of a larger phenomenon known as the mood-congruent memory effect which suggests that emotional information congruent with current mood is more likely to be recalled than information incongruent with current mood (Bower, 1981). Depressive mood-congruent memory bias is thought by many researchers to contribute to the onset and maintenance of depressed mood (e.g., Blaney, 1986, Gotlib et al., 1996, Ingram, 1984), although definitive evidence is not yet available to indicate whether depression-congruent memory biases contribute to or are caused by negative mood.
Using a meta-analytic approach, the current study reviews mood-congruent implicit memory effects associated with depressed mood. Following a brief literature review, meta-analytic results are presented. Researchers have traditionally attended to two components of depression-related mood-congruent memory effects: 1) enhanced recall of negative information among individuals exhibiting depressed mood and 2) enhanced recall of positive information among individuals assigned to nondepressed control groups. Therefore, findings associated with both depressive group advantages in the implicit recall of negatively valenced information and also nondepressed group advantages in the implicit recall of positively valenced information are discussed. Also, moderators associated with study participant characteristics and theoretically relevant study variables are explored.
Section snippets
Explicit mood-congruent memory in depressed mood states
Depressed mood-congruent explicit memory biases have been found in many studies for which participants are explicitly asked to recall previously encoded information (see Gotlib et al., 1996 for a review). The meta-analysis of explicit mood-congruent memory by Matt et al. (1992) revealed that individuals with major depressive disorder tended to exhibit preferential recall of negative information, those with dysphoria exhibited preferential recall of neither positive nor negative information, and
Implicit mood-congruent memory in depressed mood states
In a landmark paper, Graf and Schacter (1985) differentiated between implicit and explicit memory processes and described the kinds of recall tasks that would elicit each type of memory process. The authors suggested that explicit memory entails conscious recollection of previously encountered information, whereas implicit memory entails the facilitation of task performance by previously presented information without requiring conscious recollection. During explicit memory tasks, research
Controversy and relevant theoretical frameworks
The extent to which depressed mood is associated with implicit mood-congruent memory biases, and under what conditions these biases are exhibited have been topics of some debate. Explanations for the mixed findings regarding depressed mood and implicit mood-congruent memory, however, have tended to fall into one of three theoretical camps. The arguments of one camp (e.g., Watkins, 2002) are based on the depth of processing perspective (also known as the levels of processing perspective; e.g.,
Current study
Previous studies of associations between depressed mood and implicit mood-congruent memory have been largely inconclusive. Even papers dedicated to qualitatively reviewing these findings have reached divergent conclusions (e.g., Barry et al., 2004, Gotlib et al., 1996, Watkins, 2002). The current study is the first meta-analysis to examine theoretically-relevant study characteristics as potential moderators of the relationship between implicit mood-congruent memory and depressed mood. In
Relevant study acquisition
The PsycInfo and PubMed databases were searched to obtain studies for inclusion in this meta-analysis. Search terms included (but were not limited to) “depression and implicit and recall,” “depression and implicit and memory,” “dysphoria and implicit and recall,” and “dysphoria and implicit and memory.” After conducting the preliminary searches, additional efforts were made to ensure that as many relevant studies as possible were acquired and considered for inclusion. These efforts included
Implicit recall of negative words
For negative stimuli, the distribution of g statistics from all studies exhibited normality. A Q–Q plot reflecting observed values and expected normal values of g confirmed the normality of this distribution and revealed no notable outliers. The negative stimuli G+ estimate (the weighted mean advantage of depressive group for implicit recall of negative words) for the set of all studies (k = 26 observations) was found to be small in magnitude (G+ = .31; SE = .06; 95% CI = .18–.43; r = .16). As expected,
Discussion
The current findings provide support for the transfer appropriate processing framework in depressive implicit mood-congruent memory biases as well as for cognitive theories of depression that emphasize biases towards self-relevant information (e.g., Beck, 1967). Finally, these findings suggest that participant characteristics are important to consider in analyses of implicit mood-congruent memory biases in the context of depressed mood. In particular, participant age, severity of symptoms among
General discussion
In the current study of implicit mood-congruent memory in depressed mood, mean effect size estimates associated with depressive group advantage for implicit recall of negative information were small in magnitude; mean effect size estimates associated with nondepressive group advantage for positive information were very small in magnitude. The match between the conceptual versus perceptual processing required by encoding and recall tasks was associated with implicit mood-congruent memory for
Conclusions
The current meta-analytic study provided the first comprehensive quantitative review of the impact of depressed mood on implicit mood-congruent memory. The conclusions from this quantitative review are considerably less equivocal than those previously offered regarding depressed mood and implicit mood-congruent memory. As predicted, depressed mood was associated with improved implicit recall of negative information and nondepressed group status was associated with improved recall of positive
References (70)
- et al.
Explicit and implicit memory biases in depression and panic disorder
Behaviour Research & Therapy
(2001) - et al.
Mood congruence effect in explicit and implicit memory tasks: A comparison between depressed patients, schizophrenic patients, and controls
European Psychiatry
(1996) - et al.
Intrusive images in PTSD and in traumatised and non-traumatised depressed patients: A cross-sectional clinical study
Behaviour Research & Therapy
(2007) - et al.
Implicit and explicit memory for emotional information in non-clinical subjects
Behaviour Research & Therapy
(1994) - et al.
Implicit and explicit memory for emotioncongruent information in clinical depression and anxiety
Behaviour Research & Therapy
(1995) - et al.
Implicit memory bias in clinical and non-clinical depression
Behaviour Research & Therapy
(1996) - et al.
A transfer appropriate processing approach to investigating implicit memory for emotional words in the cerebral hemispheres
Nueropsychologia
(2005) - et al.
Levels of processing: A framework for memory research
Journal of Verbal Learning & Verbal Behavior
(1972) - et al.
Affective valence of words, implicit and explicit memory in clinical depression
Journal of Affective Disorders
(1995) - et al.
Implicit cognition and the maintenance and treatment of major depression
Cognitive & Behavioral Practice
(2004)