Skip to content

Exercise showed a small verbal-memory benefit in adults with depression

The average verbal-memory result favored exercise, but uncertainty across settings and no clear visual-memory effect limit the conclusion.

By 100HP editorialAbstract-based explanation checked

Based on the published abstract. The full paper may contain additional methods, results and limitations.

The 30-second takeaway

Exercise showed a small average improvement in verbal memory across randomized trials in adults with depression. However, the prediction interval crossed zero, suggesting the result may not hold consistently in another setting. The authors reported no clear effect on visual memory, so the findings do not establish a general improvement across different forms of memory.

Keep in mind

The abstract repeats part of its results and leaves an additional estimate's outcome unclear. This abstract-based brief therefore uses the clearly identified verbal-memory result and the stated visual-memory conclusion.

THE NUMBERS, WITH CONTEXT

What researchers found

g = 0.17 (95% CI 0.02-0.32)

Standardized effect on verbal memory

Exercise result across randomized trials; comparator and overall assessment timing not reported. The prediction interval crossed zero.

PeerJ, 2026 · Original source ↓

Average benefit versus consistency

The confidence interval around the average verbal-memory estimate excluded zero. The prediction interval, which addresses how results may vary across settings, crossed zero. Those statements can coexist: an average benefit does not guarantee a benefit everywhere.

Subgroups do not set a schedule

The review explored exercise intensity, format and timing. Reported benefits in selected subgroups do not establish an optimal program, particularly when several characteristics were not significant moderators of the verbal-memory effect.

CHECK THE ORIGINAL

The original publication

Effects of exercise interventions on memory in depression: a three-level meta-analysis.

Zhu X, Zhang Y, Liu C et al.
PeerJ · 2026

PubMed ID
41664658
Record checked

AI-assisted research and writing. This explains one selected publication; it is not a complete review of everything known. Our approach.