Welcome back to Healthy Innovations! 👋

I’ve just finished Prof. Daisy Fancourt’s Art Cure: The Science of How the Arts Transform Our Health – a fascinating look at what happens when we treat the arts (singing, dancing, needlework, painting, magic and more) as something we can measure, not just feel.

Drawing on a huge body of research, Fancourt maps consistent links between regular arts engagement and a range of benefits, including better cognition and mental wellbeing, less loneliness, and even longer life. They’re correlations rather than guarantees – but the breadth of the pattern is striking.

I can’t recommend Art Cure enough. Just don’t be surprised if you finish it and immediately want to join a choir or book a life‑drawing class!

With that in mind, let’s dive into this week’s topic: the impact that music can have on memory recovery.

Long before anyone studied it, people already behaved as though the arts were medicine: humming to a crying newborn, singing at a deathbed, painting through grief. Daisy Fancourt, Professor of Psychobiology and Epidemiology at UCL and director of the WHO Collaborating Centre on Arts and Health, has spent more than a decade turning that instinct into data. In 2019, she and co-author Saoirse Finn published a WHO scoping review that synthesized evidence from more than 3,000 studies on the arts and health, the largest evidence synthesis of its kind.

Her new book, Art Cure, draws on that work. Two of its central claims:

Fancourt's cross-national analysis, covering 93,263 adults aged 65 and over across 16 countries and five longitudinal studies, found hobby engagement was associated with fewer depressive symptoms and higher self-reported health. It also found greater happiness and life satisfaction, an association that appeared across culturally and institutionally different countries, though it doesn't establish that hobbies caused the outcomes. The breadth of the pattern is itself the finding. In Fancourt's framing, the arts are a modifiable input to health that most systems still don't measure, fund or prescribe as one.

Mamie Baird Magnusson, one of Scotland's most respected journalists, spent eight years living with dementia before she died in 2012, aged 86. Names went first, then the thread of her own life story. But when her daughter Sally sang the hymns and Scottish ballads Mamie had known since childhood, family accounts describe her joining in, word for word, present in a way she hadn't been in months.

Fancourt's review helps explain why familiar music can remain meaningful in dementia, though the biology is more distributed and less neatly preserved than popular accounts suggest. Musical memory draws on networks involved in auditory processing, emotion, reward, language and autobiographical recall. Some components may stay accessible even when recent memory and word retrieval are impaired, and singing in particular has been linked to gains in autobiographical recall and engagement in several studies, though results vary by the intervention used and what's actually being measured.

Some studies report that music-based interventions in care and hospital settings may:

Reviews in this area consistently flag small, inconsistently designed studies and a lack of high-quality controlled trials, so these are isolated findings worth watching rather than general, settled effects. For someone like Mamie, even a modest reduction in distress during something as ordinary as a bath wasn't an academic outcome. It was the difference between a bad afternoon and a good one.

The part of Fancourt's work that gets less attention is what happens decades earlier. Her review describes cognitive reserve: the brain's resilience to age-related change, built through sustained cultural and creative engagement over a lifetime.

This is one of the arguments Fancourt develops in Art Cure: creative hobbies may help the brain stay resilient, framed as one of the more modifiable factors available, not a cure.

The newest attempt to build on this evidence is virtual reality.

The Wayback Project, made by UK advertising creatives whose own parents had Alzheimer's, uses 360-degree VR film to recreate scenes such as the 1953 coronation street parties, complete with period music, viewable through a basic cardboard headset.

In Canada, Health PEI said in 2020 that it had become the first Canadian government health authority to install Rendever's VR platform across its hospitals and long-term care homes, letting residents revisit a childhood street or a fishing boat from decades earlier. Rendever now says its platform is used by senior living and healthcare partners across North America.

The idea is that personal, autobiographically specific footage may engage similar memory and emotion systems to a familiar hymn, though that's a hypothesis rather than an established equivalence.

What's clearer is the evidence gap:

That gap argues for rigor, not against VR.

Fancourt's own career suggests the standard to apply: treat a headset, like a choir or a painting class, as an intervention that has to earn its evidence rather than borrow credibility from adjacent research. The emerging pattern suggests personal relevance may matter here too, a resident's own wedding venue rather than a stock coronation scene, though that needs direct testing rather than assumption.

Mamie Magnusson's family reached her through a song long before anyone had the data to explain why it worked. Fancourt has spent her career narrowing the gap between that kind of moment and the evidence to back it, headset included.

🧠 Limbic tackles one of the NHS’s oldest bottlenecks: getting people into mental health treatment before they give up waiting. NHS Talking Therapies services have long relied on a lengthy paper referral form as the first hurdle, and a meaningful share of people drop out before ever reaching a clinician. Limbic replaces that form with a conversational AI intake that asks the same clinical questions in plain language, then assesses risk against validated scales including PHQ‑9 and GAD‑7. Anyone flagged as high‑risk is routed straight to a human crisis team rather than left in a queue.

Built specifically for NHS mental health services rather than as a consumer app, it has been validated at genuine scale: about 93% diagnostic accuracy across common disorders including depression, GAD and PTSD, and around 45% fewer changes to treatment allocation once a clinician reviews the case, suggesting the initial triage is landing correctly. It now handles assessments for about 40% of NHS Talking Therapies services, with more than 650,000 patients assessed to date. For anyone tracking how AI actually earns a place inside a public health system rather than just pitching one, Limbic is a useful test case.

🦗 Why eating bugs feels wrong. Scientists in Spain studied hardened tartar from 745 ancient teeth, some over 33,000 years old, and found something surprising: people in northern Europe and Asia carry a gene change, dating back about 9,000 years to when farming began, that makes it harder to digest the crunchy shells of insects.

Neanderthals seem to have had no such trouble. Their tooth tartar contains DNA from flies and mosquitoes at levels similar to modern chimpanzees, and researchers think they were eating meat that already had fly grubs in it, possibly from carcasses stored in ponds or marshy ground where mosquitoes also bred.

People living near the tropics still carry the gene version built for digesting insects easily, which lines up with insects staying a normal food source there while it faded out further north. With over 1,600 edible insect species now being looked at as a sustainable food source, that old genetic split may end up mattering more than anyone expected.

Thank you for reading the Healthy Innovations newsletter!

Keep an eye out for next week’s issue, where I will highlight the healthcare innovations you need to know about.

Have a great week!

Alison ✨

P.S. If you enjoyed reading the Healthy Innovations newsletter, please subscribe so I know the content is valuable to you!