The short version
- Both a pre-recorded, screen-led music program and conventional music therapy kept autobiographical memory from declining over 10 weeks, while standard-care patients lost ground.
- Neither music arm improved the trial’s primary measures of general cognition, so “music improves cognition in dementia” is not what this study found.
- Only the therapist-led arm showed modest gains in naming, verbal learning and executive function, which is the case for keeping a human in the room.
A group music program delivered on a screen by a non-therapist did as well as a trained music therapist at preserving autobiographical memory in people with dementia over 10 weeks, in a Finnish randomized trial. Neither version improved general cognition.
The design
The trial, published in Alzheimer’s & Dementia: Translational Research & Clinical Interventions by a University of Helsinki and University of Jyväskylä team including Särkämö, is open access. It randomized 166 people with dementia, 91 living at home and 75 in care homes, into three arms:
- Digital music intervention: group sessions using pre-recorded content led on screen, run by a facilitator who was not a music therapist.
- Conventional music therapy: live group sessions with a trained therapist.
- Standard care.
Sessions ran for 10 weeks. The point of the design was practical: music therapists are scarce and expensive, and the question was whether a scalable, screen-delivered version could deliver some of the same benefit across the whole spectrum of dementia care, from community to institution.
The results
On the primary outcomes, measures of general cognition, neither music arm separated from standard care. That is the headline, and it is a null result.
Beneath it, two secondary findings. Conventional music therapy produced modest gains in naming, verbal learning and executive function. And both music arms, the therapist and the screen, preserved autobiographical memory, the ability to recall one’s own life, while the standard-care group declined over the same 10 weeks.
That second finding is the interesting one. The screen version held the line on a memory domain that matters enormously to people with dementia and their families, at a fraction of the cost of a therapist. The therapist version did that and a little more.
The limits
The primary endpoint was null. Any summary that says music “improved cognition” in this trial is wrong; it preserved one domain and, in the therapist arm, nudged a few others.
Ten weeks is short for a progressive disease. Whether a preserved autobiographical memory at week 10 means anything at month 12 is not something this trial can say.
And “preserved” is relative to a control group that declined. The music groups did not improve; they held.
Why it matters
The result sits inside a wider push to understand why music reaches people whose memory is failing.
Medical Xpress reported last month on a study finding that favorite music still evokes a stronger brain response than neutral music in people with dementia, and that the response differs by dementia type. The Finnish trial’s secondary finding is consistent with that: familiar, personally meaningful music seems to reach a memory system that generic cognitive measures do not capture.
PBS NewsHour reported in August on the Mayo Clinic’s use of improv and music to engage dementia patients, a program built on participation rather than passive listening. And Rice University announced in September a $3.9 million NIH grant for a clinical trial, with brain imaging, on whether creating music, not just hearing it, strengthens cognition in older adults. These are reported programs and plans rather than results, but they mark where the field is heading: from listening toward making, and from therapist-only delivery toward formats that can scale.
The Finnish trial is an early data point for the scaling question. A screen can hold autobiographical memory steady. A human can do a little more. For a care home with no music therapist, the first of those is a usable result.
What to watch
Whether the Helsinki group publishes a longer follow-up of the same participants, and whether the Rice trial, when it reports, finds that active music-making moves the general-cognition measures that passive and group listening did not.
