Thread by @realscientists

Happy Thursday, good people of twitter!!! We're going to chit and chat about music, health, and (shocker) BRAINS. Are you excited? I'M EXCITED. BUT I USUALLY AM ABOUT SUCH MATTERS.

Yesterday we took a deep dive into the neuroscience of music and language:

External Tweet loading...
If nothing shows, it may have been deleted
by @realscientists view original on Twitter

External Tweet loading...
If nothing shows, it may have been deleted
by @realscientists view original on Twitter

And today, we're going to talk about how illnesses manifest in brain activity, and what music has to do with it.

We need to start by capturing brain activity when the brain isn't doing stuff. This gives us a baseline with which to study the brain when it's doing stuff. These are called resting state studies, and they're super neat!

Through resting state studies, researchers found evidence of a widely-distributed network that's involved when you're awake but not focused on a task.

We talked about this a bit on Brainvengers day: it's called the default mode network, and it's responsible for keeping you ready to act/react.

The Default Mode Network:

The default mode network is also involved in *daydreaming* which, scientifically, is pretty interesting. You're building new and novel information out of pre-existing information to test out what might work if you encounter this situation in real life.

(even if that situation is very unlikely)

So here's where the gap is. On one side, we have resting state studies and doing-stuff studies with healthy adults, and on the other side, we have the beginnings of resting state studies with clinical populations.

It's still early days with neuroimaging and clinical populations, for numerous reasons. One, these healthy baseline studies are essential, and neuroscience takes a Very Long Time.

Second, the ethics of imaging populations who may not be able to give consent is massively important to consider. Good science is ethical science.

Thirdly, logistics. EEG can be portable and is minimally invasive, but when you get into fMRI and PET studies, you need to get people to a lab, into a scanner, injections may be involved, etc. etc. It's difficult, but not impossible, and VERY worthwhile, But again, it takes time.

So, to date, a lot of work with clinical populations has been based on observational/behavioural/psychological assessments. I'm going to talk about dementia and Alzheimer's here, but if you have expertise in other areas, GET IN HERE.

Shifting gears slightly over to music. If you search "Music and Alzheimer's" on youtube, you'll get heaps of videos of individuals singing, dancing, humming, etc. to music. This is amazing!

When I was still in clinical practice, many of my clients were in moderate to advanced stages of dementia, and had limited language use. These communicative difficulties and institutionalization in general often led (understandably) to confusion, frustration, and aggression.

But with music, many of them were still able to sing, keep beat, play familiar instruments, and even try unfamiliar instruments. Individuals who were in very late-stage Alzheimer's, which is accompanied by a loss of affect (facial expressions) would smile and hum.

I had clients who were non-verbal for whom English was a 2nd/3rd/4th language sing entire songs in English, then and start telling me about where they heard the song.

Me, throughout most of my practice:

These were incredible experiences, but when I went searching for ways to describe what was happening to the individuals' families, my colleagues, and people who funded my work, I came up short. I had lots of stories, but not a lot of explanations.

A very real issue in music therapy is one of results-based funding. Funders want to see results, and music therapy (and many other therapies) suffers from a lack of tools to measure its efficacy.

SO. We're back over to research. Music therapy research with individuals with dementia can focus on a few things that have measures already in place, namely behavioural symptoms (aggression, confusion), and measure how these symptoms change with/without music therapy.

Again, these measures may not be the most sensitive, and this recent reviews have highlighted problems in the methodology. Recent Abstract here: onlinelibrary.wiley.com/doi/full/10.10…
Older review here: pdfs.semanticscholar.org/2973/01b1c9524…

My initial move, being a GIANT NERD, was to do a literature search to see what was going on in the brain. I found lots of information on healthy adults and resting states, a few studies on music listening, a tiny pocket on improv, and nothing on music and Alzheimer's and brains.

So that's how I ended up here.

Here's where the exploration begins. So if we have observational evidence of music being accessible to individuals with dementia on one side, and evidence of music activating a giant swath of brain areas on the other side, what time is it?
Answer:

If we go back to resting state studies, we would expect to see fewer and shorter connections between areas in the default mode network compared to healthy older adults:
pnas.org/content/101/13…

This gives us a starting point for studying music, and NOW, we can go back to the music-language tree from yesterday!

At the root level, our auditory processing network changes as we age, losing high frequency information, and requiring more effort to isolate and integrate relevant information from complex signals

If this interests you, the field of multi-sensory integration is doing A LOT with aging, and it's fascinating. Review here:
ncbi.nlm.nih.gov/pmc/articles/P…

Building up, syntactic processing changes with age, so from here, we could predict smaller responses from musical syntactic areas in older adults:
s3.amazonaws.com/academia.edu.d…

If we have any machine learning or murder mystery nerds in the house, this is a fantastic paper studying writing changes using the works of Agatha Christie, PD James, and Iris Murdoch:
academic.oup.com/dsh/article/26…

We also might predict a stronger response for positively-valenced (happy) music for older individuals:
journals.plos.org/plosone/articl…

Production-wise, we might expect to see greater activation in motor cortices for individuals who had a history of music lessons. Remember: procedural memory (aka muscle memory) is harder to learn, but also harder to forget.
My very rusty piano-playing muscles:

For individual and social improvisation, we could see more activity from the ventral (lower) portions of the pre-frontal cortex:
s3.amazonaws.com/academia.edu.d…

but this is the *extra* exciting bit...WE WON'T KNOW UNTIL WE TRY IT. And we haven't tried it.

Yet.

And that's my plan for the next few years, provided I don't run out of tea.