Ultra-processed foods and brain ageing: What the new review really says

In short: a review highlighted on 23 September 2026 in the Annual Review of Nutrition summarises research on ultra-processed foods, cognitive function and brain ageing. Observational studies show concerning associations, but the authors stress unanswered questions and the lack of causal proof. An ingredient list helps with shopping, yet neither one additive nor the word “natural” determines food quality on its own.

What ultra-processed means

The review uses the NOVA framework. Ultra-processed products are industrial formulations that often combine refined ingredients, added sugar, fat or salt and functional additives. Processing itself is not automatically harmful: pasteurising, freezing, milling and fermenting can make food safer, longer-lasting or usable.

Which evidence the review brings together

The authors examine human studies and basic research. Proposed pathways include metabolism, inflammation, vascular health, gut-brain communication and the nutrient density of the overall diet. Several studies associate a higher UPF share with poorer cognitive outcomes. An association is not proof that the degree of processing itself caused the outcome.

Why cause and correlation are difficult to separate

People who consume more UPFs often differ in income, education, sleep, exercise, smoking, existing illness and the rest of their diet. Food intake is frequently measured by questionnaires and can be imprecise. The category also combines very different products. A sugary drink, fortified wholegrain bread and a plant-based substitute may share a formal category despite different composition and use.

What the review does not claim

• No proven cause of dementia: it does not show that one product or additive causes dementia.
• No risk-free threshold: it does not provide an exact number of grams below which every diet is safe.
• No equation of all processing: cooking and preservation methods are not automatically ultra-processing.
• No treatment advice: people with neurological disease should not change treatment because of a food classification.

Clean label is not the same as a good diet

A short, understandable ingredient list can make comparisons easier. “Additive-free”, “natural” or “clean label” does not guarantee low sugar, salt or saturated fat, and it does not prove a brain benefit. Conversely, an authorised additive does not automatically make a product harmful. Ingredients, nutrition table, serving size and the product’s place in the whole diet are more informative together.

Four practical steps without a ban list

• Simplify the base: choose oats, potatoes, pulses, vegetables, fruit, plain yogurt or unsweetened alternatives more often.
• Compare like with like: check sugar, salt, saturated fat and fibre within the same product category.
• Use convenience intelligently: frozen vegetables, canned beans and wholegrain bread can be practical; composition matters.
• Swap gradually: replace one regularly consumed sweet, drink or ready meal instead of labelling the entire shop as good or bad.

Find simple building blocks in our pulse recipes, vegetable recipes and breakfast ideas.

Conclusion: take it seriously without overinterpreting it

The review organises a growing but methodologically difficult body of evidence. The signals justify better long-term studies and controlled trials. They do not justify claiming that one snack makes the brain age faster. For everyday eating, a varied, mainly plant-rich pattern built around minimally processed staples remains a sound guide without condemning every processed food.

Sources and status

• Pennington Biomedical Research Center: What science says about ultraprocessed foods and the aging brain, research release dated 23 September 2026.
• Carmichael et al.: Ultraprocessed Foods and the Aging Brain: State of the Science, Annual Review of Nutrition, 2026.
• PubMed: Bibliographic record and review abstract.
• German Nutrition Society: Food-based recommendations for healthy eating.

Author: Kochzauber editorial team. Information updated: 24 September 2026. This article does not replace individual medical or dietetic advice. The header image is an editorial symbolic image.