A growing body of large-scale research links high consumption of ultra-processed foods to faster cognitive decline and a meaningfully higher risk of dementia. The relationship isn’t simple cause-and-effect proven in a lab, but the epidemiological evidence is consistent and strong enough to take seriously. A 2025 UK Biobank study of more than 58,000 adults found that those with the highest ultra-processed food intake had a 37% greater risk of developing dementia compared to those who ate the least.
What Counts as Ultra-Processed Food
Not all processing is equal. Researchers use a four-tier system called NOVA to classify foods by how much industrial manipulation they’ve undergone. Freezing vegetables or pasteurizing milk counts as minimal processing. Canning beans in salt water or baking bread from flour, water, and yeast makes something “processed.” Neither category raises concern.
The problem category is ultra-processed: foods manufactured largely from industrial ingredients you wouldn’t find in a home kitchen. Think soft drinks, packaged snacks, instant noodles, chicken nuggets, flavored yogurts, breakfast cereals, and most fast food. These products typically contain emulsifiers, artificial flavors, colorings, hydrogenated oils, and high-fructose corn syrup. According to CDC data from 2021 to 2023, Americans get 55% of their daily calories from ultra-processed foods on average. For children and teenagers, that figure rises to nearly 62%.
What the Largest Studies Show
The evidence comes from multiple large, long-running studies that tracked people’s diets and then followed their brain health for years. A study published in JAMA Neurology found that people eating the most ultra-processed food experienced a 28% faster rate of overall cognitive decline and a 25% faster decline in executive function (the mental skills you use for planning, focus, and decision-making) compared to those eating the least.
The 2025 UK Biobank analysis went further. Beyond the 37% increased dementia risk, researchers found that high ultra-processed food intake was tied to a 76% higher risk of Parkinson’s disease. Brain imaging from a subset of over 5,400 participants revealed that heavy ultra-processed food consumption corresponded to widespread reductions in gray matter, the tissue responsible for processing information, memory, and decision-making. The shrinkage appeared across multiple brain regions, including areas beneath the cortex, with particularly notable changes on the right side of the brain.
The vascular dementia connection may be especially strong. When researchers modeled what would happen if people replaced just 10% of their daily ultra-processed food calories with minimally processed alternatives like fresh fruit, vegetables, legumes, or milk, they estimated a 19% lower risk of dementia overall and a 22% lower risk of vascular dementia specifically.
How Ultra-Processed Food Affects the Brain
Several biological pathways likely connect ultra-processed diets to brain decline, and they probably work together rather than in isolation.
Chronic Inflammation
Ultra-processed foods promote low-grade, persistent inflammation throughout the body. Elevated levels of inflammatory markers in the blood are consistently associated with poorer performance on tests of processing speed and executive function. Over years, this kind of simmering inflammation can damage blood vessels in the brain and accelerate the breakdown of neural connections. The brain is especially vulnerable because it consumes a disproportionate share of the body’s blood supply and energy.
Gut Bacteria and Brain Signaling
Your gut bacteria produce short-chain fatty acids when they ferment dietary fiber. These compounds do important work: they strengthen the gut lining, calm immune responses, and support the production of proteins that help brain cells grow and form new connections. Ultra-processed foods are typically stripped of fiber, which starves the beneficial bacteria that produce these compounds. Higher levels of two bacterial families in particular, Bifidobacteria and Lactobacilli, have been linked to lower anxiety and better memory in both animal and human studies. Diets low in fiber tend to shift gut bacteria away from these helpful species and toward populations associated with inflammation.
Nutrient Displacement
When ultra-processed foods make up the majority of your diet, they crowd out the nutrients your brain depends on. A recent analysis of U.S. dietary data found that even when ultra-processed foods are fortified with added vitamins and minerals, people eating these patterns still perform worse on memory tests compared to those getting their nutrients from whole foods. The association held up after adjusting for other health factors, suggesting that fortification doesn’t fully compensate for what’s missing in a highly processed diet. Whole foods deliver fiber, polyphenols, and micronutrients in combinations that appear to work together in ways a fortified cereal or enriched flour can’t replicate.
Correlation vs. Causation
It’s worth being honest about what this evidence can and can’t prove. These are observational studies, meaning researchers watched what people ate and what happened to their brains afterward. They can’t definitively prove that ultra-processed food directly caused the cognitive decline, because people who eat more processed food may also exercise less, sleep worse, earn less, or have other health conditions that independently affect brain health.
Researchers account for these factors statistically, adjusting for age, income, physical activity, smoking, and existing conditions. The associations persist after those adjustments, which strengthens the case. The brain imaging data adds another layer: it shows structural differences in the brains of high consumers that align with the cognitive test results. And the dose-response pattern (more processed food, faster decline) is what you’d expect if the relationship were real. Still, a randomized trial where you assign people to eat ultra-processed food for decades and wait for dementia isn’t feasible or ethical, so observational data is the best evidence we’re likely to get.
Practical Swaps That Move the Needle
The substitution data offers a realistic starting point. You don’t need to overhaul your entire diet. Replacing just 10% of ultra-processed calories with minimally processed alternatives was associated with a meaningful reduction in dementia risk. For someone eating a typical American diet of about 2,000 calories, that’s swapping roughly 110 calories of ultra-processed food, the equivalent of trading a packaged granola bar for a piece of fruit and a handful of nuts, or replacing flavored yogurt with plain yogurt topped with berries.
The highest-impact swaps target the foods you eat most often. If you drink soda daily, switching to water or unsweetened tea removes a consistent source of ultra-processed calories. If packaged snacks are a staple, keeping cut vegetables, cheese, or whole fruit accessible makes the substitution easier to sustain. The goal isn’t perfection. It’s shifting the ratio, gradually, so that your brain gets more of the fiber, healthy fats, and micronutrients it needs to maintain itself over decades.