Learning Music and the Ageing Brain: What the Evidence Actually Says

Parents have always asked us whether music lessons make children brighter. Lately we’ve been getting a different question, and it comes from adults asking on their own behalf: if I take up an instrument now, might it help protect my brain later?

It’s a fair question, and a serious one. Dementia is now the leading cause of death in the UK, and most of us have watched someone we love go through it. So it’s worth answering carefully — neither dismissing the idea nor overselling it.

A good place to start is Dr Tommy Wood’s The Stimulated Mind: Future-Proof Your Brain from Dementia and Stay Sharp at Any Age (2026). Wood is an associate professor of paediatrics and neuroscience at the University of Washington who also works as a performance consultant to elite athletes, including in Formula 1. That combination gives the book its character: he thinks about the brain the way a good coach thinks about the body.

The brain isn’t simply running down

Wood’s central argument is that the slow slide in cognitive function many of us expect after middle age is far less inevitable than we assume. Brains remain capable of adapting and improving at almost any age — provided they’re given the right conditions.

He organises those conditions into what he calls the 3-S model: Stimulation, Sleep and Nutrient Supply. And he frames the goal not as “avoiding disease” but as building headroom — a term he borrows from the economist Art De Vany for the spare capacity you can draw on when life throws something difficult at you. More headroom means more buffer before a loss of brain tissue starts to show up as a loss of function in everyday life.

Music sits squarely in the first S.

“Use it or lose it” — but more precisely

In a 2022 paper in Cells, Wood and the neurologist Josh Turknett set out a model they call demand coupling. The idea is that the body maintains tissue in proportion to what’s being asked of it. Muscle you don’t use is expensive to keep, so you don’t keep it. The same logic, they argue, applies to the brain: the growth, repair and housekeeping processes that keep neural tissue healthy are downstream of cognitive demand. Reduce the demand over years and decades, and maintenance is quietly downgraded to match.

This matters because of how most adult lives are shaped. In our twenties and thirties we’re constantly out of our depth — new jobs, new skills, new responsibilities. Then, gradually, we get good at things. Work becomes familiar. Hobbies settle into routines we can already do. The demand curve flattens right at the point in life when we most need it not to.

The useful implication is a demanding one: what appears to protect the brain isn’t activity as such, but genuine difficulty — doing something you are not yet good at, and keeping at it long enough to become good at it.

Why an instrument is an unusually good stimulus

Learning an instrument is a rare hobby in that it loads almost everything at once. You are decoding notation, converting it into precise finger movements, listening to the result, comparing it against what you intended, correcting in real time, holding a pulse, anticipating what’s coming next, drawing on memory, and — in a group — coordinating all of that with other people.

It also has two properties that cognitive training apps struggle to replicate. First, immediate and honest feedback: you can hear when it’s wrong. Second, an effectively infinite difficulty ladder. There is always a harder piece. You cannot exhaust it, which means you never have to stop making progress.

And, unlike most brain-training regimes, people actually keep doing it — because they enjoy it, and because there’s a concert or a rehearsal or a grade exam in the diary.

What the research shows

Wood cites a substantial body of work on music in his chapter on cognitive stimulation. The headlines:

• Verghese and colleagues, New England Journal of Medicine, 2003. In a study following 469 New Yorkers aged over 75, several leisure pursuits were associated with a lower risk of developing dementia — reading, playing board games, dancing, and playing a musical instrument. Crucially, the association held up after excluding people who may already have been in the earliest stages of decline at the start.
• A 2022 meta-analysis in BMC Neurology pooled three long-term cohort studies and found that playing an instrument was associated with roughly a third lower risk of dementia (hazard ratio 0.64). The evidence base is small and the confidence interval wide, so treat the exact figure lightly — but the direction is consistent.
• Rogers and Metzler-Baddeley, Brain and Cognition, 2024. This is the more demanding test: a review of trials in which healthy over-60s with no musical background were assigned to learn an instrument. Learners showed gains in attention, mental flexibility (switching between tasks) and processing speed. There was no measurable effect on working memory or verbal memory, and the authors are candid that many of the individual studies were small or at risk of bias.
• Marie and colleagues, 2023. In a trial of 132 healthy older adults, six months of piano lessons produced increases in cerebellar grey matter and auditory working memory — measurable growth against a background of the general brain shrinkage that comes with age.
• Rogenmoser and colleagues, 2018. Using brain scans to estimate a “brain age”, musicians’ brains were estimated as younger than those of non-musicians of the same chronological age.

To Wood’s credit, he also cites the sceptics — including Sala and Gobet’s meta-analysis, which found little evidence that musical skill in children transfers to unrelated academic abilities. The picture is genuinely mixed, and worth presenting as such.

The honest caveats

We’re a music school, so take this next bit as seriously as the rest.

Most of the long-term evidence is observational. It shows that people who play instruments develop dementia at lower rates — not that playing an instrument is what caused the difference. People who take up music may differ in education, income, social connection, hearing or health in ways researchers can only partly adjust for. And in some cases the arrow may point the other way: very early, undiagnosed decline can be what causes someone to give up a difficult hobby.

The trial evidence is more direct but shorter-term, and it measures cognitive test scores over months rather than diagnoses over decades. Nobody has run — or realistically could run — a thirty-year randomised trial of piano lessons.

So: no one can promise that learning an instrument will prevent dementia, and we won’t. What the evidence supports is more modest and still worth having — better-maintained cognitive function, and a reasonable case for delayed onset.

Music also happens to cover several other risk factors

The 2024 Lancet Commission on dementia concluded that around 45% of cases worldwide are potentially preventable by addressing fourteen modifiable risk factors across the life course. Several of them are things a weekly music lesson quietly delivers: continued education and mental stimulation, social contact rather than isolation, better mood, and — for anyone playing standing up or in a group — a bit more physical activity.

One of that list deserves a direct mention. Untreated hearing loss is among the largest single modifiable risk factors in midlife. If you’re finding it harder to pick out a melody line, or to follow conversation in a busy room, please get your hearing tested. It is one of the most useful things on this entire list, and it’s free.

What this looks like in practice

If the mechanism really is demand, a few things follow:

• Difficulty is the point. Work just past what’s comfortable. If a piece feels easy, it isn’t doing much for you.
• Little and often beats occasional marathons. Fifteen focused minutes most days is worth more than two hours on a Sunday.
• Play with other people. Ensemble playing adds listening, timing and social demand all at once.
• It is genuinely not too late. The trial evidence above was gathered in adults in their sixties, seventies and eighties who had never played a note.
• Keep the difficulty coming. Twenty years of playing the same six pieces is a habit, not a stimulus.

Where we come in

At AS Music School on Cobnar Road we teach adult beginners as well as children — individual lessons on a wide range of instruments, plus group options including Play in a Band and Musical Theatre, and Stave House early years sessions for the very youngest.

If you’ve been telling yourself for years that you’d love to learn but you’ve left it too late, the science is fairly clear that you haven’t. Get in touch at info@asmusicschool.com and we’ll find you a teacher.

References and further reading

• Wood, T. The Stimulated Mind: Future-Proof Your Brain from Dementia and Stay Sharp at Any Age (2026).
• Turknett, J. & Wood, T.R. “Demand Coupling Drives Neurodegeneration: A Model of Age-Related Cognitive Decline and Dementia.” Cells, 2022;11(18):2789.
• Verghese, J. et al. “Leisure Activities and the Risk of Dementia in the Elderly.” New England Journal of Medicine, 2003;348(25):2508–16.
• Arafa, A. et al. “Playing a musical instrument and the risk of dementia among older adults: a systematic review and meta-analysis of prospective cohort studies.” BMC Neurology, 2022;22:395.
• Rogers, F. & Metzler-Baddeley, C. “The effects of musical instrument training on fluid intelligence and executive functions in healthy older adults.” Brain and Cognition, 2024;175:106137.
• Marie, D. et al. “Music interventions in 132 healthy older adults enhance cerebellar grey matter and auditory working memory, despite general brain atrophy.” NeuroImage: Reports, 2023;3(2):100166.
• Rogenmoser, L. et al. “Keeping brains young with making music.” Brain Structure and Function, 2018;223(1):297–305.
• Sala, G. & Gobet, F. “When the music’s over. Does music skill transfer to children’s and young adolescents’ cognitive and academic skills? A meta-analysis.” Educational Research Review, 2017;20:55–67.
• Livingston, G. et al. “Dementia prevention, intervention, and care: 2024 report of the Lancet standing Commission.” The Lancet, 2024;404(10452):572–628.