Your Attention Is Fraying. Your Brain Isn't Broken — It's Eating Food It Never Evolved For.
June 21, 2026 · Nesse & Williams, Why We Get Sick~5 min read
Here is the unsettling part of a study published this June: the people whose attention had quietly dulled weren't eating badly. Many ate well — Mediterranean-style, the diet doctors bless. What separated the sharp from the foggy wasn't whether they ate vegetables. It was how much of their food had been engineered in a factory their ancestors would not recognize as food at all.
What the 2026 numbers actually say
Researchers at Monash University, the University of São Paulo and Deakin University looked at more than 2,100 middle-aged and older Australian adults, none with dementia, and lined up their diets against standardized tests of visual attention and processing speed. The pattern was steady: for every roughly 10% rise in the share of calories from ultra-processed foods — the sodas, packaged snacks, reconstituted meats, the things with ingredient lists like a chemistry exam — focus measurably slipped, and mental processing ran slower. The findings appeared in Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring. Read them precisely: this is a cross-sectional, observational study. It found an association, not a cause. Diet today, cognition today, measured side by side — it cannot tell us which way the arrow points.
In Darwinian medicine, a body tuned for ancestral real food meets ultra-processed food — an input with no evolutionary precedent — and attention may dull: a mismatch, not a malfunction. 2026 finding: across 2,100+ adults, each ~10% more ultra-processed food tracked measurably poorer attention and slower processing, even on otherwise healthy diets. Framework: Nesse & Williams, Why We Get Sick. Popular-science interpretation, not medical advice — consult a qualified doctor; this is one correlational study, association is not cause, defer to the original peer-reviewed research.
The twist: eating "healthy" didn't rescue focus
You'd expect a good overall diet to soak up the damage. It didn't. The dulling of attention showed up even in people whose diets were otherwise high quality — the Mediterranean eaters included. That's the result that should make you sit up, because it breaks the usual bargain. We tell ourselves the salad cancels the soda. Here, it didn't seem to: the ultra-processed share carried its own signal, riding on top of an otherwise sensible plate. Whatever is going on, it isn't simply "they ate junk instead of food." It's that the processed fraction, specifically, tracked the fog.
An evolutionary lens: the mismatch, not a malfunction
Why would the brain respond like this? Nesse and Williams, in Why We Get Sick, hand us a different question to ask. Before deciding an organ has malfunctioned, ask what it was built for — and what environment it's now running in. Their core idea is mismatch: a body tuned over hundreds of thousands of years to a hunter-gatherer world, dropped into an environment only a few thousand years old, and now a few decades into a kind of food that has no evolutionary precedent at all. Ultra-processed food is a novel input. Pre-digested textures, sugar and fat stripped from their fibrous escort, flavors engineered to outrun every satiety signal we own. The book's recurring verdict fits cleanly here: the problem may not be that the brain broke. It may be that the environment changed faster than the brain ever could.
Why "novel" is the operative word
Nesse and Williams are careful, and so should we be. A craving for sugar and fat was once a life-saving adaptation — in a world where both were scarce, the appetite that hunted them down kept you alive. That same appetite, loose in a world of engineered abundance, becomes the trap. The dulled attention, on this reading, isn't the brain failing at its job; it's a system optimized for one diet meeting a diet it was never tested against. Mismatch is a trade-off and a relocation, not a verdict of "broken." And note what the framework refuses to claim: it doesn't prove the food blunted anyone's focus. It offers a plausible story for why a real association might exist — a hypothesis to test, in Nesse and Williams's own provisional spirit, not a fact to bank.
Hold the headline to the skeptic's questions
This is where honesty earns its keep. A correlation between processed food and dimmer focus shows a relationship, not a mechanism — association is not causation, and a cross-sectional snapshot can't rule out that foggier people simply reach for easier food. Could there be a third variable doing the work — sleep, income, the stress that drives both the snack and the scatter? Absolutely; that's exactly the question a good reader asks. The mismatch story is illuminating, but it is a lens, not proof. The grown-up summary is deliberately modest: a robust association, a compelling evolutionary frame for it, and a causal arrow still unproven. Watch this; don't yet bank it.
What this means for you
So what does a curious, careful reader do with this? Less than the headline urges, and more than nothing. First, the proportionate move: the cheapest, lowest-risk experiment is to shift the ratio — more food your great-grandmother would recognize, a smaller engineered fraction — and notice, without grand promises, whether your afternoons feel clearer. Second, the guardrail: this is one observational study; treat "10% more processed food, measurably less focus" as a flag worth watching, not a dose-response law to obey. Third, the non-negotiable: this is popular-science interpretation, not a diagnosis or a meal plan — if your attention or memory genuinely worries you, that's a conversation for a qualified doctor, not a blog. The point of reading the numbers honestly is to act on them gently, and to keep asking the evolutionary question: not "what's wrong with me," but "what is my body running on that it was never built to run on?"
Dulled focus on ultra-processed food may not be a brain that broke — it may be a Stone-Age brain fed an input evolution never tested it against.
A real association and a compelling frame in 2026 — but correlational, not proof. A flag to watch, not a verdict.
This is a popular-science interpretation, not medical advice; it is not a diagnosis or a meal plan — if attention or memory concerns you, consult a qualified doctor. Evolutionary framework paraphrased from Randolph Nesse & George Williams, Why We Get Sick: The New Science of Darwinian Medicine; for the framework, defer to the original book. The 2026 finding refers to a cross-sectional, observational study of more than 2,100 middle-aged and older Australian adults (Monash University, University of São Paulo, Deakin University), published in Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring, reporting that higher ultra-processed food intake was associated with poorer attention and slower processing even among otherwise healthy diets. This is a preliminary, correlational result — association is not causation, and it is not established that the food causes the effect; defer to the original peer-reviewed study.