Ultra-processed foods

In the ELSA-Brasil cohort (10,775 adults followed for ~8 years), participants whose calories came ≥20% from ultra-processed foods showed a 28% faster rate of global cognitive decline and 25% faster executive-function decline than those who ate the least.

Gomes Gonçalves et al., JAMA Neurology, 2023 [7]

Trans fats

Among 1,018 adults, higher dietary trans-fat intake was associated with worse performance on a word-recall memory test; each gram per day of trans fat was linked to roughly 0.76 fewer words recalled, after adjustment for age, education and mood.

Golomb & Bui, PLoS ONE, 2015 [3]

Saturated fat–heavy diets

In the Chicago Health and Aging Project (815 older adults), higher intake of saturated and trans-unsaturated fats was associated with an increased risk of incident Alzheimer's disease, while unsaturated fats were not.

Morris et al., Archives of Neurology, 2003 [2]

High-carbohydrate, high-sugar diets

Among 937 older adults in the Mayo Clinic Study of Aging, those with the highest carbohydrate intake had a higher risk of developing mild cognitive impairment over ~4 years; those with the highest sugar intake also showed elevated risk.

Roberts et al., Journal of Alzheimer's Disease, 2012 [4]

Artificially sweetened drinks

In the Framingham Heart Study Offspring cohort, daily consumption of artificially sweetened soft drinks was associated with a higher risk of stroke and dementia over 10 years. Sugary drinks were not significantly associated in the same analysis, and the authors called for further study.

Pase et al., Stroke, 2017 [6]

High-salt diets

In mice, a high-salt diet reduced blood flow to the brain and impaired memory through an immune signalling pathway that begins in the gut. The effect reversed when salt intake returned to normal. This is animal evidence and has not been confirmed in humans.

Faraco et al., Nature Neuroscience, 2018 [5]

Fried and high-heat processed foods (AGEs)

Advanced glycation end products (AGEs) form when foods are fried, grilled or roasted at high heat. In mice, a diet rich in AGEs led to brain changes resembling those in Alzheimer's disease; in a small human cohort, higher circulating AGEs were associated with cognitive decline.

Cai et al., PNAS, 2014 [8]

Alcohol, including moderate intake

In the Whitehall II study (550 adults, 30-year follow-up), even moderate alcohol consumption (14–21 UK units per week) was associated with a higher risk of hippocampal atrophy, a brain change linked to memory, compared with abstinence.

Topiwala et al., BMJ, 2017 [9]

Mediterranean and MIND patterns

In the Rush Memory and Aging Project (923 older adults), closer adherence to the MIND diet, which emphasizes vegetables, berries, nuts, olive oil, whole grains and fish while limiting red meat, butter, sweets and fried food, was associated with a lower incidence of Alzheimer's disease [10].

Whole-diet patterns in other populations

In the Hisayama Study in Japan (1,006 adults, 15-year follow-up), a dietary pattern high in soybeans, vegetables, seaweed and dairy and low in rice was associated with reduced dementia risk, supporting the idea that overall pattern matters more than any single item [12].

What is not established

No individual food, supplement or "superfood" has been shown in rigorous trials to prevent or reverse memory loss. Claims that a single ingredient causes or cures dementia are not supported by current evidence. Modifiable risk factors identified by the 2020 Lancet Commission include education, hypertension, hearing loss, smoking, obesity, depression, inactivity, diabetes, alcohol, head injury, air pollution and social isolation.