I’ve eaten the MIND way for years: berries at breakfast, leafy greens I don’t especially like, olive oil poured over things that never asked for it. So when a headline told me that kind of eating could leave my brain looking two and a half years younger on an MRI, I wanted it to be true before I reached the end of the sentence. Then I remembered that researchers have already run the experiment that could prove it. They put 604 older adults on the diet for three years, and the cognitive scores and brain scans came out looking just like the control group’s.
The headline number comes from a study in the Journal of Neurology, Neurosurgery and Psychiatry that followed 1,647 adults from the Framingham Heart Study Offspring cohort, average age 60 at the start. Researchers matched each person’s eating pattern, taken from food questionnaires filled out at check-ups between 1991 and 2001, against at least two brain MRIs shot between 1999 and 2019, with an average follow-up of 12 years. People who scored higher on MIND adherence lost grey matter more slowly. Every 3-point bump in their diet score came with about 0.279 cubic centimeters less grey matter loss per year, which the team translates into roughly 20 percent less age-related decline, or about 2.5 years of delayed brain aging. The fluid-filled ventricles, which swell as the tissue around them wastes away, expanded more slowly too, worth about another year.
Grey matter is the dense outer rind where your neurons do their computing, so slower erosion there is worth seeing on a scan. The MIND diet is scored out of 15. It’s a mash-up of Mediterranean eating and the blood-pressure-lowering DASH plan, built around foods thought to protect the brain. These volunteers averaged just under 7. I’ll admit that detail charmed me: nobody here was a saint about it, and the score still tracked with tissue you can measure.
Then I read the food-by-food breakdown, and it stopped following the diet’s own rulebook. Poultry tracked with slower grey matter loss and slower ventricular swelling, berries with slower ventricular swelling, sweets with faster ventricular swelling and hippocampal shrinkage, and fried fast food with faster loss in the hippocampus, the seahorse-shaped memory hub tucked deep in the temporal lobe. So far, all on script. But cheese, which MIND tells you to limit, came out looking protective on four separate measures. And whole grains, which MIND actively rewards, lined up with faster loss of grey matter and hippocampus and faster ventricular enlargement. Wait, why would the food the diet is built to reward push the scan in the wrong direction? I doubt the oats are rotting anyone’s hippocampus. When the pieces of an eating pattern scatter in directions the pattern never predicted, my first suspect is the people rather than the food. Whoever eats more of one thing is usually living differently in a dozen other ways, and a food-frequency questionnaire can’t pull all of that apart.
The authors list the weaknesses plainly. The design is observational and can’t establish cause, food questionnaires invite recall bias, and the team couldn’t rule out mild cognitive impairment at the first scan. They also couldn’t track how diets shifted over the years or account for genetic risk, and the volunteers were mostly White. One more line caught my eye: the association was strongest in people who were older, more physically active, and not overweight. That pattern reads to me less like a food acting on neurons and more like a healthy lifestyle riding along inside a diet score, because the benefit clusters right where the exercise and the trim waistlines already were.
The diet has also faced the one kind of test that can separate the food from the person eating it. In 2023, a randomized controlled trial in the New England Journal of Medicine, funded by the National Institute on Aging, assigned 604 older adults with a family history of dementia to either the MIND diet or their usual diet. Both groups cut about 250 calories a day, and the researchers followed them for three years. Cognition improved in both arms: global scores rose 0.205 standardized units on MIND and 0.170 on the control diet. The 0.035-unit gap between them wasn’t statistically significant. On MRI, total grey and white matter volume, hippocampal volume and white-matter lesions all came out similar in the two groups.
Both groups also lost weight, about 5.0 kilograms on MIND and 4.8 on the control diet. My hunch, and it is only a hunch, is that the shared calorie cut and weight loss explain why everyone’s scores crept up a little. That would fit the Framingham pattern too, where the diet’s apparent benefit gathered among the lean and the active.
So you have more than a decade of scans showing that people who already eat this way tend to have brains that age a little more gently, and a three-year experiment in which handing the diet to people produced no measurable change in their brains or their thinking beyond what the control group got. A fair defender would say three years is too short for any diet to move brain tissue. That objection doesn’t rescue the headline, though, because the 2.5 years was never measured in anyone’s memory or reasoning. It’s a grey matter slope converted into a brain-age estimate, drawn from people who were also leaner and more active, and then packaged as a promise about what’s on your plate.
I’m keeping the berries and the greens, because nothing here says they hurt, and a plate built on plants, fish and olive oil is a sane way to eat whatever an MRI says. What I’m dropping is the idea that each handful of blueberries buys me brain-years. I won’t pay for a brain-age scan, or for anything sold under a MIND label, to chase a payoff that the one trial built to find it couldn’t see.