Brain health after 40: what counts as a meaningful benefit?

Text written with the help of AI.

A better memory-test score, a different brain signal and slower decline on a clinical scale are all measurable changes. They answer different questions about brain health. Most of the research below concerns adults over 60, so its findings cannot simply be assumed to apply to people in their 40s or 50s.

What changed in the food studies?

In a 2023 randomised trial, 61 healthy adults aged 65 to 80 received wild-blueberry powder, equivalent to 178 grams of fresh berries, or a matched placebo each day for 12 weeks. The blueberry group performed better on immediate word recall and accuracy when switching between tasks. Delayed recall and measured cerebral blood flow showed no significant advantage. That is emerging evidence for benefits on selected tests, not evidence that blueberries prevent dementia or preserve everyday independence. The trial was funded by the Wild Blueberry Association of North America. King’s College London and University of Reading, 2023.

A separate smoothie study measured something further from cognition. In eight healthy men, a banana-based smoothie produced lower circulating flavanol metabolite levels than a capsule control. The study concerned flavan-3-ols; it did not test memory or directly measure whether bananas alter the cognitive effects of blueberry anthocyanins. Ottaviani and colleagues, *Food & Function*, 2023. It raises a question about absorption, but does not establish a brain-health reason to keep bananas out of a blueberry smoothie.

Did neurofeedback beat a credible comparison?

Neurofeedback gives people visual or auditory feedback from recorded brain activity. Learning to change the targeted signal is one outcome. Improving cognition beyond what a comparable activity achieves is another.

In a 2026 trial, 92 healthy adults aged 65 to 75 were assigned to motor imagery neurofeedback, sham feedback or conventional cognitive stimulation. The neurofeedback groups completed ten sessions over ten weeks, and all groups attended physical activity sessions. Some participants learned to regulate the targeted activity, but cognitive improvements were not significantly greater with genuine neurofeedback than with the comparison conditions. University of Valladolid, 2026. The result has a defined scope: it tests this protocol in this population, rather than settling every possible use of neurofeedback.

Evidence for a commercial home system needs the same attention to comparison. A 2022 Myndlift study reported improvements in questionnaires and cognitive measures, but its retrospective design had no control group. Concurrent treatment and other influences could not be separated from the system’s contribution; the authors also disclosed employment or consultancy relationships with the company. Whitehead and colleagues, 2022. For cognitive enhancement in otherwise healthy older adults, the evidence remains emerging. A change in a trained signal is insufficient on its own to support a broader cognitive claim.

A modest benefit can matter within the right population

The DOMINO-AD trial asked a different question. It enrolled 295 people with moderate or severe Alzheimer’s disease who were already taking donepezil. Over 52 weeks, those who continued treatment had an average advantage over those who discontinued it of 1.9 points on a 30-point cognitive scale and 3.0 points on a separate measure of daily functioning. The cognitive difference exceeded the trial’s predefined threshold for clinical importance; the average functional difference did not reach its corresponding threshold. Howard and colleagues, *New England Journal of Medicine*, 2012.

This is established evidence about continuing symptomatic treatment in people with diagnosed disease. An advantage over stopping treatment may mean less deterioration, rather than improvement from a person’s starting point. It says nothing about taking donepezil to enhance cognition while healthy. A decision to continue treatment also depends on an individual’s response and tolerability and requires clinical review.

The 2025 US POINTER trial studied 2,111 adults aged 60 to 79 at increased risk of cognitive decline. Both groups received encouragement concerning physical activity, diet, cognitive and social engagement, and cardiovascular health monitoring. One group received a more structured programme. Over two years, that group improved more on a composite cognitive score, by 0.029 standard deviations per year relative to the self-guided group: a statistically significant but small difference. Baker and colleagues, *JAMA*, 2025.

The trial supports the structured programme’s effect on that measured score relative to its comparison. It cannot identify which component made the difference, and its published cognitive result does not establish fewer dementia diagnoses or longer independent living. Those outcomes require their own evidence.

Match the decision to the evidence

Personal experimentation can tell you whether you enjoy a food or find an activity worthwhile. It cannot establish that either has reduced your future dementia risk. Likewise, an app score may show progress at an app task without showing a benefit in daily life.

When considering a brain-health claim, ask who was studied, what they were compared with, what changed and for how long. A finding about absorption, a trained brain signal, a cognitive test and day-to-day functioning each has its own value. The expectation you attach to an intervention should stay within the outcome its evidence actually measured.

Sources