Scientists Followed 4,000 People for 3 Years — Their Brains Kept Getting Better
We tend to think of brain health as something that becomes harder and harder to improve as we get older. By our 70s, 80s, or 90s, the assumption is often that the goal has shifted from getting better to simply slowing the decline.
A large new study suggests that may be too pessimistic.
Researchers followed 3,966 adults between the ages of 19 and 94, repeatedly measuring different aspects of their brain health over three years. The study, published in Scientific Reports in May 2026, found something particularly encouraging among its oldest participants: people in their 80s and early 90s were still capable of improving their brain health scores, and researchers couldn’t identify an age at which those gains reliably stopped.
The research was conducted through the University of Texas at Dallas’ Center for BrainHealth, with participants drawn from all 50 U.S. states and more than 60 countries. Rather than measuring brain health with a single memory test, researchers used a multidimensional assessment every six months, allowing them to look at how different aspects of cognitive, emotional, social, and daily functioning changed over time.
The findings don’t mean aging has no effect on the brain. But they do challenge the idea that there is a point in later life when meaningful improvement is simply off the table. And some of the patterns researchers uncovered about who improved, when they improved, and what appeared to matter most were particularly surprising.
What the BrainHealth Index actually measures
The BrainHealth Index measures three areas: clarity, connectedness, and emotional balance. Clarity reflects readiness to reason through complex situations and generate new solutions. Connectedness measures social ties and sense of purpose. Emotional balance reflects how well people handle difficult circumstances.
Unlike traditional clinical tools designed primarily to detect deficits or disease, the BrainHealth Index was built to track upward potential across its three components. That design difference is relevant: most standard cognitive assessments are calibrated to identify what is going wrong, not to capture gradual improvement in people who are not yet impaired.
What the study design looked like
Participants completed the BrainHealth Index assessment every six months over a three-year period, producing seven total measurements per participant. That repeated structure allowed researchers to examine gradual trends rather than relying on a single before-and-after comparison, which is a common weakness in shorter studies.
The BrainHealth Project was registered at ClinicalTrials.gov as a prospective, longitudinal, open-label, single-arm interventional study, with recruitment modeled to continue over at least ten years. The 3,966-person cohort analyzed in this paper represents the first major longitudinal output from that initiative – a subset with enough completed assessments to examine multi-year trends.
The daily training: five to fifteen minutes
Participants completed self-paced cognitive training modules lasting five to fifteen minutes. The modules covered strategies for filtering distractions, synthesizing information rather than simply accumulating it, and maintaining focused attention on one task at a time. Training also addressed sleep habits, stress management, and physical activity.
Optional virtual coaching was available every three months, though the core daily activities required no specialist equipment. The inclusion of sleep and stress management alongside attention exercises positions the intervention as a broader lifestyle program, which may partly explain why improvements appeared across dimensions beyond cognitive speed alone.
What the results showed — and who improved most
Results showed sustained improvements in overall BrainHealth Index scores across all age groups. Participants who started with the lowest baseline scores showed the largest raw gains – partly a statistical effect, since people beginning at the bottom have more room to improve. Participants with higher baseline scores also continued to improve across all three years, which is harder to attribute to a simple floor effect.
Frequency of engagement mattered. Participants who used the training modules most consistently recorded the greatest improvements in their BrainHealth Index scores. That dose-response pattern, while not proof of causation in the absence of a control group, is consistent with the activities themselves driving the changes.
The researchers acknowledged the study’s key limitations: there was no randomized control group, and participants were self-selected – adults already motivated to improve their brain health. Without a control group, it is not possible to confirm that the training caused the improvements rather than the act of regular engagement with the platform itself.
The age finding: no ceiling identified
With participants spanning from age 19 to age 94, the researchers had the statistical range to look for an age at which gains stopped or reversed. They did not find one. Adults in their 80s and early 90s continued to record measurable improvements across the three-year period alongside participants several decades younger.
The brain’s capacity to form new connections and reorganize existing ones – neuroplasticity – was once thought to be largely confined to childhood and adolescence. Current evidence suggests this capacity persists into late adulthood, though its character changes: older brains may recruit different neural regions to accomplish the same tasks rather than repeating strategies that worked at younger ages. The BrainHealth Project findings are consistent with that emerging picture, though the observational design means the mechanisms remain unconfirmed.
The researchers behind the study
Dr. Sandra Bond Chapman founded the Center for BrainHealth in 1999 and serves as the Dee Wyly Distinguished Professor in the School of Behavioral and Brain Sciences at the University of Texas at Dallas. As the study’s senior author, she has argued that the medical community has historically waited for cognitive problems to emerge rather than actively building brain capacity before decline sets in.
“For too long, we’ve operated under the outdated notion that we need to wait until something bad happens to our brains before we do anything for them,” Chapman was quoted on Neuroscience News.
Limitations that matter
The absence of a randomized control group is the study’s most significant methodological constraint. Without a parallel group of similar participants who received no intervention, the data cannot establish that the training caused the score improvements. Regular platform use and repeated self-assessment could themselves produce gains, independent of the specific content.
Self-selection bias is a related concern. Adults who voluntarily enroll in a brain health program and sustain participation for three years are unlikely to represent the general population – they may be more educated, more health-conscious, or more cognitively resilient at baseline. The researchers acknowledge this openly in the paper.
The BrainHealth Project’s ongoing design targets 100,000 adult participants over at least a decade, which will eventually generate the longitudinal data needed to address some of these limitations. The current paper is a prospective observational study with an active intervention component, not a controlled experiment.
Why the oldest participants are the most telling detail
That adults in their 20s and 30s can improve cognitive performance with structured training is not a surprising finding. The more notable result is that adults in their 80s and early 90s showed measurable gains across three years, with no identifiable age ceiling. The study’s wide age range – 19 to 94 – gave researchers an unusual opportunity to test whether improvement eventually plateaus, and the data did not reveal a clear cutoff.
Each component of the training program – cognitive strategy practice, sleep, stress management, and social connectedness – is modifiable without medication or clinical intervention. The daily time investment was five to fifteen minutes, and participants who engaged most frequently showed the greatest gains, suggesting regularity matters more than duration.
Read More: 12 Early Warning Signs of Cognitive Decline You Should Know
What this means for you
The study followed 3,966 adults for three years and found sustained brain health improvement across all age groups, with no age ceiling identified. The design does not allow researchers to conclude that specific exercises caused those gains – the lack of a control group is a real limitation – but the pattern held across participants from 19 to 94, including those with the highest baseline scores.
The behaviors associated with improvement in this study – consistent mental engagement, protected attention, adequate sleep, stress management, and social connectedness – are accessible to most people without specialist equipment. The daily time investment was five to fifteen minutes. Participants who engaged most regularly showed the greatest gains.
The BrainHealth Project continues to enroll participants and will track outcomes for at least a decade. An imaging sub-study is underway in Dallas to examine whether behavioral changes and improved scores correspond to structural changes in the brain – evidence that would move the findings from observational to mechanistically grounded. The current data offer a consistent signal across nearly 4,000 participants and three years: brain health improvement remained possible at every stage of adult life measured.
Disclaimer: This information is not intended to be a substitute for professional medical advice, diagnosis, or treatment and is for information only. Always seek the advice of your physician or another qualified health provider with any questions about your medical condition and/or current medication. Do not disregard professional medical advice or delay seeking advice or treatment because of something you have read here.
AI Disclaimer: This article was created with the assistance of AI tools and reviewed by a human editor.
Read More: Can Alzheimer’s Really Spread Through a Blood Transfusion? Scientists Weigh In
