# Investigating the impact of migraine on short-term and working memory: a systematic review and meta-analysis

## TL;DR

This study finds that migraine does not affect short-term memory but may impair working memory, which could explain reported cognitive difficulties.

## Contribution

The study provides the first meta-analysis specifically comparing short-term and working memory in migraineurs versus controls.

## Key findings

- Migraineurs showed no significant differences in short-term memory compared to healthy controls.
- Working memory performance was significantly worse in migraineurs relative to healthy controls.
- The cognitive issues in migraine may relate to higher-order memory processes rather than memory storage.

## Abstract

To conduct a systematic review and meta-analysis on past research to ascertain the impact that experiencing migraines has on memory. Specifically, this research explores the short-term and working memory domains and where possible delineates between verbal and visual memory.

Migraine is a prevalent neurological disorder that affects up to 15% of the global population. Migraine is comprised of physical symptoms including head pain, nausea, vomiting, and sensitivity to light and sound. There are also various subtypes that are based on specific symptom manifestations such as the presence of visual aura. Furthermore, a commonly reported cognitive symptom is ‘brain fog’, a colloquial term used to describe everyday cognitive challenges particularly in relation to attention and memory. The empirical literature examining memory performance in migraine has produced inconsistent findings: some studies report clear cognitive impairment, whereas others suggest minimal or no deficits. In addition, research to date has been limited in its examination of specific memory domains and has not consistently considered the influence of migraine characteristics. Thus, this systematic review and meta-analysis aimed to investigate the short-term and working memory performance of migraineurs compared to healthy controls. Subsidiary aims included investigating the memory performance of migraineurs with aura and without aura and investigating the impact of migraine on visual and verbal memory performance.

A systematic search was conducted in Embase, CINAHL, MEDLINE, PsycINFO, and Web of Science Core Collection. Papers were screened in two stages (titles and abstracts and full texts) using inclusion criteria relevant to the aims. Of 3880 articles extracted, 16 met the criteria for inclusion. Random effects models were conducted in JASP and utilised Hedge’s g as the effect size. Heterogeneity was evaluated using the Q and I2 statistics. Publication bias was examined using Egger’s regression test and funnel plots.

No significant differences were found between migraineurs and healthy controls in short-term memory performance. Specifically, no significant differences were observed between migraineurs with aura and healthy controls, migraineurs without aura and healthy controls, or migraineurs with and without aura. Similarly, no significant differences were observed in short-term visual or verbal memory performance. In contrast, a significant difference was observed in working memory performance, with migraineurs demonstrating poorer performance relative to healthy controls.

The findings of this systematic review and meta-analysis indicate that short-term memory performance is preserved in adults with migraine, whereas working memory shows a selective impairment. This pattern suggests that migraine is not associated with a global memory deficit, but rather with vulnerability in higher-order memory processes requiring active manipulation and cognitive control. These results help reconcile objective cognitive findings with commonly reported experiences of brain fog and suggest that migraine-related cognitive complaints may reflect difficulties in cognitive efficiency rather than fundamental memory storage deficits during the interictal period.

There was no registration as this paper is not experimental.

The online version contains supplementary material available at 10.1007/s00415-026-13672-w.