Depression Changes Brain Activity During Memory Encoding and Retrieval, fMRI Meta-Analysis Finds
A meta-analysis of 22 fMRI studies finds depression alters brain activity differently during memory encoding and retrieval, depending on the memory type.
Summary
Depression often comes with memory problems, but it has been unclear how brain activity differs while people form and recall memories. This meta-analysis pooled brain-imaging data from 22 fMRI studies comparing people with depression (current, remitted, subthreshold or at risk) to controls. During memory encoding, depression was linked to lower activity in the caudate and in networks for salience, attention control and movement. When recalling personal memories, activity was lower in the right insula and fusiform gyrus and higher in the left anterior cingulate and left middle frontal gyrus. When recalling non-personal information, the right inferior frontal gyrus was more active. The authors read this as impaired salience integration in some stages and a more vigilant, self-monitoring retrieval mode in others. These are group-level associations from pooled imaging data. They do not show cause, and they do not test any treatment.
Detailed Summary
Memory problems are common in depression and can affect daily function, work and quality of life. For readers focused on brain healthspan, understanding which brain systems are involved matters, because depression is also linked to cognitive decline over a lifetime. Whether brain activity differs during forming versus recalling memories has been unclear.
This study ran three coordinate-based meta-analyses using Seed-based d Mapping. It pooled data from 22 fMRI studies comparing controls with people who had current, remitted or subthreshold depression, or who were at risk for it. Analyses were split by stage (encoding versus retrieval) and by content (autobiographical versus non-autobiographical memory).
During encoding (11 studies, N = 542), depression was associated with reduced activity in the caudate and in regions tied to salience processing, frontoparietal control and motor function. During non-autobiographical retrieval (7 studies, N = 345), depression was associated with higher activity in the right inferior frontal gyrus. During autobiographical retrieval (10 studies, N = 433), depression was associated with reduced right insula and fusiform gyrus activity, alongside increased activity in the left anterior cingulate cortex and left middle frontal gyrus. Between-study heterogeneity was low and no evidence of publication bias was found.
The authors suggest depression may involve impaired salience integration during encoding and autobiographical retrieval, while increased frontal activity during non-autobiographical retrieval may reflect a more vigilant, self-monitoring approach. Brain changes therefore look specific to both the stage of memory and its content. This could help refine models of memory impairment and, eventually, guide targeted interventions.
The caveats are important. This is a mechanistic, correlational imaging synthesis in humans. It shows activity differences, not causes, and no intervention was tested. Memory performance, lifespan and function outcomes were not reported in the abstract. Findings need replication and testing in treatment studies before informing care.
Key Findings
- Mechanism (human fMRI): during encoding, depression showed reduced caudate, salience, frontoparietal and motor-region activity (11 studies, N = 542).
- Non-autobiographical retrieval showed higher right inferior frontal gyrus activity in depression (7 studies, N = 345).
- Autobiographical retrieval showed lower right insula and fusiform activity but higher left anterior cingulate and middle frontal activity (10 studies, N = 433).
- Heterogeneity was low and no publication bias was detected across the 22 included fMRI studies.
- No memory performance, treatment, or lifespan outcomes were reported in the abstract; findings are correlational.
Methodology
Three coordinate-based meta-analyses using Seed-based d Mapping pooled peak coordinates from 22 fMRI studies. Groups included controls versus current, remitted, subthreshold or at-risk depression, split by encoding versus retrieval and autobiographical versus non-autobiographical memory. Heterogeneity and publication bias were assessed and reported as low and absent.
Study Limitations
This summary is based on the abstract only, so details on study quality, medication status, age ranges and subgroup effects were unavailable. Coordinate-based meta-analyses pool group-level activation peaks and cannot establish causation or individual-level prediction. Small study counts per analysis limit confidence, and the diverse depression groups (current, remitted, at risk) may blur distinctions.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.
Enter your email to subscribe:
