Direct retrieval bias for general and specific memories for negatively valenced cues in major depression

Abstract

M ajor depressive disorder (MDD) is associated with reduced specificity in autobiographical memory. It has been argued that this tendency occurs through a failure of effortful generative retrieval, regardless of valence of cue word. However, we propose that in MDD general memories are likely to be recalled via direct retrieval, and direct retrieval is more likely for negatively valenced cues. To provide a preliminary test of this, a large sample with MDD (N = 298; M age = 47.2) completed the autobiographical memory test and indicated whether retrievals were generative or direct. Categoric and extended memories for negatively valenced cues were more often directly retrieved than generatively retrieved, and more often than direct retrieval for positively valenced cues. In contrast, categoric and extended memories for positively valenced cues were more often generatively retrieved relative to generative retrieval for negatively valenced cues. Relative to non-clinical samples, direct retrieval for negatively valenced cues was high. Retrieval method and valence may be moderating processes in the type of memories recalled. This preliminary work presents the possibility of an extension of theory on retrieval tendencies in MDD.

Direct accessibility for overgeneral memory predicts a worse course of depression: Re-analysis of the online computerized Memory Specificity Training (c-MeST) for major depression study

“…All analyses were conducted using JASP (JASP Team, 2020). Consistent with previous studies (Hallford & Matsumoto, 2022;Matsumoto et al, 2020), AMT performance 11 DIRECTLY RETRIEVED OGM AND DEPRESSION was coded as proportions of directly retrieved or generatively retrieved specific memories or OGMs for each cue valence. As a certain proportion of omission errors were observed, they were removed from the denominator of the calculation.…”

“…While a cross-sectional relationship between directly retrieved general memories for negative cues and depression has been established (Hallford & Matsumoto, 2022;Matsumoto et al, 2020), longitudinal evidence for this has been lacking. In this study, we aimed to examine the hypothesis that directly retrieved general memories for negative cues prospectively predicts high levels of depressive symptoms one month later.…”

“…Two previous studies have examined whether depression is associated with subjective judgments of the retrieval process. One study reported that depression is associated with increased subjective judgements of direct retrieval (Hallford & Matsumoto, 2022), while another study did not support this association (Matsumoto et al, 2022). There is currently no objective measure to distinguish between generative retrieval and direct retrieval.…”

“…The internal reliability in the study was acceptable (MacDonald's w = .75). Consistent with a previous study (Hallford & Matsumoto, 2022), to assess retrieval method the participants answered the following question before providing the memory for each of the cue words: "Did this memory come immediately to mind?". There were two response options: "Yes, I thought about it straight away", or "No, I had to think about it, and search for it in my memory".…”

“…However, recent studies demonstrate that direct retrieval of general memories in response to negative cues are elevated in individuals with MDD (major depressive disorder), a history of MDE (major depressive episode), and nonclinical dysphoria when participants judge retrieval process themselves (Hallford & Matsumoto, 2022;Matsumoto, Takahashi, & Kawaguchi, 2020). In contrast, the generatively-retrieved general memories in response to negative cues are not as clearly associated with depression.…”