Two researchers at Texas State University screened 622 records and pulled 16 studies published since 2021 that tested four whole-food dietary patterns (Mediterranean, Nordic, Okinawan, and plant-based) against cognitive outcomes in adults aged 60 and over. Their conclusion is that eating patterns, rather than isolated nutrients, are consistently associated with better memory, better executive function, and lower dementia incidence, and that the mechanisms run through systemic inflammation, cerebrovascular integrity, the gut microbiome, and circulating endocannabinoid tone. They also report that the size of the benefit depends heavily on who you are, with sex and APOE genotype emerging as the two strongest modifiers.
For twenty years the nutrition-and-brain field has run a recurring experiment: isolate the compound that seems to be doing the work in a healthy diet, put it in a capsule, and give it to older adults in a randomized trial. Vitamin E. B vitamins. Omega-3. Ginkgo. The capsules keep failing. This review is an attempt to say why, and to formalize the alternative.
The argument is about the food matrix. In a whole diet, polyphenols arrive alongside fiber, unsaturated fat arrives alongside the plant sterols and micronutrients that travel with it, and the combination changes the body in ways a single molecule does not. Woolsey and Sen assembled 16 recent studies covering four regional eating patterns and looked for the physiology that connects them. What they found is that the four diets are cosmetically different and mechanistically similar. Nordic eating is built on rye, root vegetables, berries, and fatty fish. Okinawan eating is built on purple sweet potato, soy, and turmeric. Neither looks anything like the olive oil and legumes of southern Europe. All three, plus a well-constructed vegetarian pattern, appear to lower systemic inflammation, improve vascular function in the brain, and shift the gut microbiome in the same direction.
The more interesting thread is the one the authors treat almost as a footnote. A three-year Mediterranean diet intervention found that the benefit was not uniform. Men gained more memory improvement than women. And carriers of the APOE e4 allele, the strongest common genetic risk factor for Alzheimer’s disease, showed a benefit in year one that had reversed direction by year three, while non-carriers kept improving. If that finding replicates, it undermines the premise that there is a single best brain diet, and points toward matching macronutrient ratios to genotype and insulin status.
The honest reading of the evidence is more modest than the paper’s conclusion. Most of the well-powered results are observational. The randomized trials are short, ranging from four weeks to twenty-four weeks, with two exceptions. Where randomized trials do report effects, those effects are small, and in the one trial with a follow-up measurement, the benefit had faded by 48 weeks. The strongest claim this review can actually support is that whole-food patterns are a better bet than supplements alone.
Actionable Insights
An effect size is just a way of asking how big a change is relative to normal variation between people. A value of 0.2 is small, 0.5 is moderate, 0.8 is large.
Nearly every credible result in this review lands between 0.13 and 0.30. That is small. Concretely, an effect of 0.13 to 0.18 standard deviations is roughly what a typical older adult loses to aging in two to four years, so a lifelong dietary pattern may buy you a few years of cognitive time, not a decade. Translated to a clinic-standard MoCA test, an effect of 0.29 is about one point out of thirty. The threshold usually considered clinically meaningful is two points.
What is worth doing, in descending order of evidence quality:
Replace animal fat with vegetable fat. In a 5,944-person cohort followed six years, higher monounsaturated fat intake tracked with a 37 percent lower dementia rate, higher saturated fat with a 56 percent higher rate. Relative risk, not certainty.
Eat nuts, roughly three to seven servings per week. Two large cohorts found slower cognitive decline at this intake.
Everything else here (blueberry powder, grape extract, ketogenic variants) is either underpowered, nonsignificant, or short-term. Treat as unproven.
Context and Source
- Open Access Paper: Dietary approaches to support cognition in older adults: a systematic review
- Institution: School of Health Administration, College of Health Professions, Texas State University, San Marcos, Texas, United States
- Country: United States
- Journal: Frontiers in Nutrition, published 09 July 2026.
- Impact evaluation: The publisher currently lists an Impact Factor of 5.5 and a CiteScore of 9.6 for Frontiers in Nutrition. Independent Scopus-derived estimates place the impact score nearer 5.0 to 5.2, and the journal sits in Q1 for Nutrition and Dietetics. Using the publisher-stated figure: the impact score of this journal is 5.5, evaluated against a typical high-end range of 0 to 60+ for top general science, therefore this is a Medium impact journal.