Science Friday - Piano AI, Giraffes, Alzheimer’s, Mime Psychology. April 9, 2021, Part 2

Episode Date: April 9, 2021

New AI Composes Songs From Silent Performance Videos There have been many awkward attempts in the quest to train algorithms to do what humans can. Music is a prime example. It turns out that the process of turning the individual notes of a composed piece into a fully expressive performance—complete with changes in loudness and mood—is not easy to automate.  But a team at the University of Washington has been closing in on a way to get close, in research they presented at a machine learning conference late last year. Their AI tool called “Audeo,” combining the words “audio” and “video,” watches a silent video of a piano performance. Then, using only the visual information, Audeo produces music with the expressiveness and interpretative idiosyncrasies of the musician it just watched.  Producer Christie Taylor talks to lead author Eli Shlizerman about how one trains an algorithm to make art, and how such tools could help make music both more accessible, and easier to engage with. A Daring Rescue Highlights Giraffes’ Silent Extinction For the past several months, a daring and unprecedented rescue mission has been underway in western Kenya. Local conservationists have been slowly puzzling out how to ferry nine stranded giraffes trapped on a flooded peninsula back to the mainland. The team rescued the most vulnerable first by sedating them for the duration of the journey. But for others they tried a less dramatic approach—coaxing the giraffe with food onto a wooden barge. “We called it the girRAFT,” said David O’Connor, president of the non-profit group Save Giraffes Now. “Some were better sailors than others.” This week, the final four Rothschild giraffes will be moved to safety. It was a valiant, months-long effort, for the sake of nine giraffes. But this small tower—the technical word for a group of giraffes—represents one percent of the total population of its species. There are only about 800 northern giraffes left in Africa. O’Connor calls this charismatic animal’s decline a “silent extinction.” He joins Science Friday to talk about why giraffe populations are plummeting, and why we should be paying attention. Untangling Alzheimer’s Connection To Insulin Resistance Over the past two decades, research into the degenerative dementia of Alzheimer’s disease has been building an interesting case: This crippling brain disease involves some of the same mechanisms and pathologies as Type 2 diabetes—and could in fact represent an insulin resistance of the brain. Even having Type 2 diabetes has been found in some research to increase your risk of Alzheimer’s.  Last month, new research in the journal Alzheimer’s & Dementia looked at the gene expression of cells in the brains of deceased Alzheimer’s patients, and found an additional piece of evidence for this theory. Every type of brain cell the team looked at demonstrated changes consistent with a diminished ability to obtain energy from glucose. The lead author Benjamin Bikman is a physiologist and developmental biologist at Brigham Young University, who also works as a diet coach with a supplement business designed around reducing insulin resistance. He says “the brain is becoming increasingly insulin-resistant. It’s becoming increasingly less able to obtain adequate glucose, and then it becomes more reliant on ketones [for energy].” Ketones, a product of burning fat, are also harder for the body to make when it is insulin resistant, which Bikman says can lead the brain into a chronic energy deficit. Ira talks to Alzheimer’s researcher Shannon Macauley, who was not involved in the new research, about how energy systems shape brain health, why they could be driving Alzheimer’s, and this might lead to new treatments. The Mime And The Mind When you watch a mime pull an invisible rope or run into an invisible wall you as the viewer are tricked into visualizing something that isn’t there. But is it all in the mime? Or does the mind play a role?  Chaz Firestone, assistant professor of psychological and brain sciences at Johns Hopkins University joins Ira to discuss his latest research on how the mind “helps” us see these invisible objects.   Subscribe to this podcast. Plus, to stay updated on all things science, sign up for Science Friday's newsletters.

This is Science Friday. I'm Ira Flato. And yes, it's Friday. Best day of the week. We finally get to put the product of all our hard work onto the airwaves, the intertubes. It's almost time to relax a bit. And when I want to take my mind off an exhausting week, I turn, like many, I'm sure, to the restorative power of music. And sci-fi producer Christy Taylor, Ailer agrees with me, right, Christy? Absolutely, Aira. There's nothing I love more than settling down with a good piece of music after a hard day. And actually, I came with something to share with you, a piece of music that I've been listening to a lot lately. That is nice. Relaxing. A little moody, maybe.

That's a piece called Vulse Sentimentale, which means sentimental waltz by the Russian composer, Piotr, Iqovsky. I have another surprise for you, which is that that is not a human, pianist that you hear playing that music right now. What is it? What is it? A Martian? Who's playing the music? It's actually an AI named Adeo, watching a video of a person, in this case a pianist named Paul Barton, trying to turn what it sees into the music that you're hearing right now. Wait a bit. So AI is doing all these things with medicine and computers, and now it's making our music. Where will this end? I mean, that's amazing, right? Doesn't that sound expressive and emotional and nuanced, kind of like a person having feelings as they play the piano?

Yeah, too much so for my taste, I think. Well, this is all coming from work by Dr. Ellie Schleeserman, who is an assistant professor of applied mathematics and electrical and computer engineering at the University of Washington. He presented this musical research at a conference on machine learning last year. I heard it. I was intrigued, and I wanted to talk to him more about how and why, since that's a good question you asked Ira, he decided to make this. Christy, I am all ears. Get it? Music joke?

Sorry, it's the weekend coming up. The first thing I asked Dr. Schleeserman was just what was going on in that piece of music we just heard. Odell can look at a silent performance video, learning from a top view camera of any piano piece, and then generate the music that corresponds to that performance. It's a computational system that has an input without, any audio, just video frames. And the output of this computational system is a sound,

audio waveform that we can all hear and interpret. So just to give our listeners a sense of what originally this piece might have sounded like, here's the actual human pianist, Paul Barton, playing that piece himself. It's an amazing piece. And then let's go back to the AI version from Adeo. I'm listening to the human versus the AI version, and I do hear some differences, but what are the important similarities in that for you? First of all, we recognize that this is Val Sentimentale, right?

And we also enjoy the music. So that's what we wanted to create. We wanted to see how a computational system will represent movements, something which is pretty mechanical into sounds that we can interpret, we can enjoy, we can imagine, along with the video stream, how these two streams will join together into a beautiful piece. And the differences in terms of the generated output, of course, the computational system is not perfect, and anyone can hear that.

But also, like, each performance will be a different performance. If you will take one musician, they will play it in one way, another musician will play it in another way. So what goes into music that we hear and kind of have an interpretation of, That was the intention of our study to really understand how we can frame it in a computational way. And I just want to understand, too, your AI, Adeo had never actually, quote, heard Val Sentimentale before, right? It's generating it completely fresh based on the input it's seeing from the fingertips, right? That's correct.

Adeo was trained on datasets that are classical music, but also some modern music as well. What it tried to establish is the connection between the movements of fingers and the keyboard that it sees to how to translate them to music. And then it was able to generalize and to translate this information to new music. That's what we often do in AI systems. We train them on particular data sets and then we see how they generalize, especially in such a task called an unsurricular. supervised task where we don't tell the system what to do. It is supposed to generate the music eventually. So, Ellie, a thing I didn't realize before is that a synthesizer isn't just one generic thing.

You can use a different synthesizer software to get a different sound. So here's a sample of Adeo using a slightly different synthesizer software to interpret the same video. That sounds so much cleaner to me, but it's also missing something, right? Yes. The kind of sounds more mechanic, right? This is the interesting component of music. Music starts from the score sheet, and then the music theory,

how to play an instrument, will also include how to move the hands correctly and position the hands and to perform all the notes and so on. So this is the music theory. But when you try to translate it to computational system, these rules are not necessarily the best way to represent the music, or to let the computer to learn how to represent the music.

I must imagine that this work involves a lot of sitting at your computer and coding, but also listening to music, right? Is it pleasant or is it tedious? Yeah, so initial outputs from audio weren't musical pieces that you would immediately recognize as music that would be played by a musician. And that's the challenge as well. So music is very sensitive. This is what we discovered in the beginning of the study.

Even if we are making very small mistakes, even if we are doing some of the phases in the computational system with more kind of noise or inaccuracies, we will eventually generate something that wouldn't sound like realistic music. And the challenge is really to see what are the components, what is this inner representation within an AI system that could enable such a transformation that will sound realistic to us

and will capture all the components of music within it. Have you talked to Paul Barton about how he feels? Like, is he worried that he's going to be replaced by an AI because you can replicate his music without hearing him play it? I haven't talked with Paul about it, but Paul Barton is such a virtuoso. I don't think an AI system can replace Paul in terms of his skills. It's more about enabling ways of interacting with music and also interpreting music. Placing camera on top of a piano, that could be like a pretty simple thing.

But what could it allow us to do? So for example, we could play the piano but then hear a different output that is generated from the system that is powered by audio. For example, we could hear the same music as a violin or a cotto in the system. instrument, for example. We could also get some cues from the system of how we are playing, are we playing in the right way and so on. Beyond that, accessibility is a major issue. We are interacting quite a lot with different tools. One of them is the computer. We interact with the computational systems quite a lot, but with mechanical instruments like the piano or other, we really need to be like next to them, right? So in order to generate music.

So how can we generate music and interact with music in a virtual environment? That's very interesting and compelling in terms of the implications of audio. So if you can teach an AI to play the piano, can you teach it to play a different kind of instrument, like a flute or a violin, where you don't necessarily just plunk a finger on a key and have one set note, right? There's a lot more shifting around with other instruments. Can you still do something with that? For some instruments, we think that we could do a single thing. we could do a similar setup.

We chose the piano for several reasons. One of the reasons is the keyboard is well defined. It has a defined number of keys, putting a camera on top of a piano that doesn't move. It's a pretty easy task to do. And musicians like Paul Barton did that and provided to us this visual stream. But also the piano is an interesting instrument

in terms of being polyphonic. So it can create many, many tones and many sounds at the same time. This is what makes piano a unique instrument in terms of computational investigations. On one hand, the instrument is well-defined, and we can monitor the activity how the instrument is being played.

But on the other hand, the output is so variable and so interesting, we can generate so many music outputs out of it that this was a challenge for us. Do you play the piano yourself, Ellie? I used to study piano when I was a child and liked it, but due to not having access to an instrument, and honestly also the daunting amount of practice as a child, it's a difficult task, right?

So like every day you need to play at least one hour. So this eventually made me stop taking lessons. I still have an urge to play the piano, and once in a while I would sit down and play a simple piece, and I would really enjoy that. And I think these experiences motivated me to think, like, why it is so challenging to interact with music and how we can make it easier and kind of like keep those experiences, even with other systems like computational systems with the computer, iPad, or, any other setup. Well, I love that you found a way to be surrounded by music without having to practice all the time. Well, the practice is still needed. Thank you so much for your time today, Ellie. Thank you so much. Dr. Eli Schleeserman is an assistant professor of applied mathematics and electrical and computer engineering at the University of Washington in Seattle. I'm Christy

Taylor. Thank you, Christy. That was a truly harmonious story. Dare I say, It's intermission. And when we come back, the charismatic giraffe is suffering a silent extinction. We'll talk about it. Everything you've ever wanted to know about giraffes. Stay with us. This is Science Friday. I'm Ira Flato.

Ever heard a giraffe roundup? For the past several months, a daring and unprecedented rescue mission has been underway in western Kenya. Local conservationists have been slowly puzzling. how to ferry nine-stranded giraffes trapped on a flooded peninsula back to the mainland. They rescued the most vulnerable first, forced to sedate them for a duration of the journey, but for others they tried a less dramatic approach, coaxing the giraffe with food onto a wooden log barge. And this week, the last four giraffes will be moved to safety. A valiant months-long effort all for nine giraffes.

but this small pod represents 1% of the total population of its species. Only 800 northern giraffe remain in Africa. My next guest calls this the silent extinction. Giraff populations are plummeting, all while we haven't been paying attention. Dr. David O'Connor, conservation researcher and president of the nonprofit organization Save Giraffes Now. Dr. O'Connor, welcome to Science Friday. Thank you so much, Ira.

And thanks so much for having us and for focusing on giraffe. Give me an idea of what this looks like. The peninsula where the giraffe were stranded? Yeah. So it's, yeah, it was a peninsula going into Lake Beringo in western Kenya. It was about 100 acres in size, this peninsula. And, you know, nearly 10 years ago, eight giraffe were brought to this peninsula to reintroduce them into the region after 70 years of being a country.

stink there. And they were reintroduced to the peninsula because the community and the wildlife authorities thought that, you know, that's the best place we can reintroduce them. They'll be safe there. We can defend them from poachers there. And then as conditions improve, they can easily access the mainland. Well, unfortunately, the lake had other ideas and decided to rise about 40 to 50 feet, completely flooding the peninsula and creating this island that we're left with today. Wow. You know, I did realize how few giraffes there were. Yeah, unfortunately, when you put all the giraffe together in Africa, there's only about 100,000 left. That's our best estimate.

And, you know, that sounds like a lot. But if you compare to, say, elephants, there's about 400 to 500,000 elephants left. So, you know, there really are very, very few giraffe left. And of the four species, three are considered endangered or critically endangered. And so we really do have to act now to try to save these giraffe across the continent. Okay, you say that the giraffe are experiencing a silent extinction. What exactly do you mean by that? It's really, as you said in your introduction, that, you know,

unfortunately, they're slipping away, but they're not getting the attention. It's not really raising the alarm bells the way perhaps it should be in the public consciousness. You know, I think people are familiar with, you know, what's happening with rhino generally or elephant or even, you know, lions or panda. but giraffe are so famous, you know, you see them in kids' books or on television or in cartoons, but nobody really knows what's happening with them in the wild. Are there certain species of giraffe that are more endangered than others? Yeah, the northern giraffe species that you mentioned are the most imperiled, and then the reticulated and the Maasai giraffe are also endangered.

Thankfully, the giraffe across southern Africa, the southern giraffe, they're actually doing okay as a species, and their populations are stable. So, you know, conservation-wise, we're really focused on those three other species. You talk about the articulated giraffe and the other species. I don't think most of us realize that giraffe can look different from one another. Yeah, yeah, they can. And, you know, generally speaking, they look giraffe-ish, you know, with the long neck and the long leg, legs. But, yeah, you know, when you put them side by side, they're really quite.

quite different. And genetically speaking, the difference between them as a species are the same as a brown bear is from a polar bear, say. Wow. Does that mean that they diverge from some common ancestry relatively recently? It's actually further back than we think because they, you know, they are quite distant, but some of them do still maintain the ability to hybridize and produce a hybrid calf. So they're, you know, they're not, you know, all that far back. But they did come from a common ancestor. And I think it was mostly geographical separation that sent them on their different species lines. You said that we know very little about giraffe.

Maybe it's easier to ask what we don't know about them. Yeah. You know, that's a really good question. And it's one of the reasons I got interested in giraffe because we know so very little about them. You know, we don't know with great certainty there are numbers. We have estimates, like I've said, but we don't know a great certainty. We don't really know where they are. We don't know how many are left in Ethiopia or where they are in South Sudan or in Somalia, for instance.

We don't know their home range sizes. We don't know their seasonal migration routes. We don't know their social structures in terms of how they heard together or even how they communicate. Do they communicate? They're largely silent. And I would argue we still don't know the reason why they have a long neck. Yeah, that is really interesting. I've always wondered about that long neck. You know, there's that whole Lamarck thing about the draft had long neck, right? I also find it really interesting that they have only seven vertebrae in their neck like we do, but they're much bigger vertebrae, right? Yeah, yeah, isn't that bananas? I love that fact. They just have really big vertebrae. And, you know, they also have a really unique blood flow system, which I find fascinating. They're sort of like,

To me, you know, like a carefully engineered race car or a Formula One car, like they're very precision engineered beings. You know, because sometimes, you know, when you stand up too quickly, sometimes we feel a bit woozy. Well, if you can imagine, you know, if you're drinking as a giraffe at the level of your feet and something startles you and you whip your head up through an arc of, you know, 19 feet, why don't they pass out every time they stand up straight? Right, right, yeah. So they have this unique blood flow system with sets of valves to stop, you know,

the blood backflowing quickly or, you know, going all the way to their head and having a hemorrhage. So, you know, they have like compression socks basically built into their legs. I mean, they're absolutely fascinating. They're really, and they're the watchtowers of the savannah. You know, other species depend on them and will herd around them at the bottom of their feet, you know, because they'll be the first ones to see danger. Wow. Can they use their necks aggressively or to fight with or fend off people, people, other animals that want to,

attack then? Yeah, yeah. Well, Ira, you laugh, you know, by fending off people, but I have been kicked by a giraffe and broke my ribs. Oh, is that right? Yeah, they do fend off people, too. Mostly, they use their legs for defense against other species. Their kick can kill a lion, for instance. But for their necks, and that's one of the thoughts about how they got the long neck, is they use it in fights, either to gain a sort of hierarchy of dominance or also for access to females. And it's usually the males that fight. And what they do is, they swing their heads like a mallet and bash each other on their sides with their heads. And, you know, they use their necks to sort of swing through that motion.

And the sound of the impact, it's like this big, boom. And you can feel it in your chest. So they really are hammering each other with their heads. It's quite a spectacle. You know, we've been having fun and talking about how the giraffe used parts of their bodies. but the serious question about their extinction, I'm wondering whether we don't notice much, why it's such a silent extinction,

because we don't think of poachers, right, selling ivory and killing them for that reason. Yeah, yeah, you're exactly right. And it's, you know, and it's also a function of not many people are working on them or just watching them. So you also don't have many people, you don't have many eyes on them

to kind of raise that alarm. But for giraffe, when it comes to poaching, they're not poached for those trophy items like rhino horn or ivory that we've heard about. But they're poached more for bushmeat, and it's mostly in-country. And it's unfortunately, when it gets to the scale of being commercialized,

where you have people going out on motorbikes and gangs with automatic weapons and the meat's being taken to commercial centers and being sold that way, that's when it's not sustainable. and the population just gets hammered. Tell me more about your non-profit Save Giraffes Now. How many projects, how many savings, rescues have you had to do?

Yeah, so right now, Save Draths Now is working across nine countries in Africa with partners. And we focus on three main priority areas that we think impact and help Giraff, you know, now. It's in our name. We read our action-oriented. So we're very much focused on rescues and rewilding, as you say. We also focus on anti-poaching, which is hugely important. And then we focus on community-led projects where we're working with local communities to monitor giraffe. And then also we're setting up giraffe rescue centers for giraffe calves that have become orphaned,

either their parents or mother has been poached or killed, or they've become very far separated from the herd and they can't find their way back. So, you know, with this current reintroduction project off the island, it's right now we're moving nine giraffe off the island and we'll also reintroduce another, say, 12 back into this area to start this breeding population in the Rucco Conservancy where they disappeared 70 years ago. But with the orphan program, you know, we're able to save and release back to the wild, you know, tens of giraffe. And we're growing that in more country as resources allowed. So that program where you're rescuing the drafts now because the lake was flooding, that has been successful. Yeah, thankfully it has. There were nine draft trapped on this island. And as of today, we have rescued five successfully. And there are four more left on the island that are actually being rescued this week. I was meant to be there, but with COVID restrictions in Kenya at the moment, I wasn't able to go.

But the team on the ground there are excellent. So, you know, we're all watching our phone. for updates from Kenya to see how that's going. So what's the plan? Do you rescue them and you wait for the waters to recede and then put more giraffe back there? Well, you know, that was our big problem is how do you get the world's tallest animal off an island when it doesn't swim? You know, the lake levels at the time when we started this were rising about six inches a day. And there was one giraffe particularly that was trapped alone on this sliver of island that was really going to be flooded. So we had to move quite quickly.

And so together with the community and the wildlife authorities and our partners, we created this raft. And we're actually sailing the giraffe off the island, if you can believe. Yeah, I'm trying to think this giraffe, I'm thinking, is top heavy on a raft, right? Yeah. Well, what worst combination could you have? Exactly, yeah. Yeah, but the community, they built this raft, which we've nicknamed the giraffe, And it's a real engineering marvel, actually.

They put together 60 steel, basically oil drums together and then put the planks and stuff on top of it and create this very, very stable square base. And then we built up the sides, you know, about 10 feet so the draft don't go into the water. And it's been amazing. We've been taking the giraffe one by one off. some I'll admit are better sailors than others

but it's been so far very very successful no giraffe have got injured no people have got injured and they've all joined up on the mainland together forming a new herd so you sort of had a Maguire the whole thing to create the raft

right yeah I mean yeah you're right if you can imagine you know Rucco is very remote and you know you have to sail everything in by boat most of the times you know because the roads aren't impossible in the rain season, et cetera. And, you know, the team there are just incredible, so resourceful and talented to be able to create this engineering marvel that's now saving these giraffes.

This is Science Friday from WNYC Studios. Talking with Dr. David O'Connor, president of saved giraffes now and a researcher associated with the Smithsonian Institution. This is, like, I've never heard of a giraffe rescue or the need to rescue giraffe. You were talking a bit before. People eat the giraffes. Is that the problem? Yeah, in certain places. So, you know, there's four main reasons for the decline of giraffe across Africa.

At the top of the list is loss of habitat and habitat degradation and fragmentation. That affects so many species. Then there is, you know, prolonged droughts and the less dependable rains that are now we're seeing because of climate chaos and climate change. change. And then in certain countries in Africa, its poaching is very, very heavy and taking a very heavy toll on these giraff. And it is for meat. And it's, you know, maybe if it was, you know, at a village or a family level, you know, where they really are desperate for protein, that might be, you know, sustainable. But when it's commercialized in these areas and where you, people coming in on motorbikes or with automatic weapons and just plowing down the giraffe and bringing the meat into

town, into commercial centers, that's when you really, really see the populations going way down. And is a giraffe called a pod instead of a herd? Is that what? It's actually, believe it or not, called a tower. Sounds right. A tower. Whoever came up with that deserves some kind of award, I think. And what is a typical tower composed of one male, many females? How does that work? Well, as we were saying at the beginning, we don't know. We still don't know. We still don't know because if you look at a giraffe tower and say you start with you know five giraffe in this tower it's not like a herd of sheep say where there's 10 sheep and they go around they all stay together and that's it that's the herd in a tower you'll start with five draft suddenly two might wander in

out of nowhere and they join up for a while then one might move off somewhere else and go wander off into the distance so they're not stable they're they're a fission fusion type dynamic we call it and but mostly I would say they're female dominated these towers. And, you know, they, they, they come together and have semi-stable relationships and then also for the, for their calves. And then basically, the mature males go from female tower to female tower to female tower, basically trying to get lucky, I think. And when you bring in the, the giraffe, the tower, to the breeding area that what used to be, the peninsula now is the island which will become a peninsula again.

How do you know how many of what kind to bring in there? Right. So we knew historically that this area of Kenya had a type of northern giraffe called the Ross Child's giraffe, also known as the Nubian giraffe. So we knew that they were evolved for that area. And it was very much 70 years ago, the center of their range in Africa. And unfortunately, they were wiped out by poaching. and being caught in human conflict over 70 years ago.

And so we knew that it was Roshchez that had to go back in. And with the Rucco Community Conservancy there, that's 44,000 acres. They were very keen to bring back giraffe to their land. And so we thought this is the perfect opportunity to get giraffe back in onto the mainland, off the island, and then also bring in some more to start a breeding population and just saving these draft from the island, our first step. This has been fascinating, Dr. O'Connor, for all of us. I learned so much about giraffes and about your rescue efforts.

Thank you very much for taking time to be with us today. Thank you so much. It's been a pleasure. Dr. David O'Connor, conservation researcher, president of the non-profit saved giraffes now. We're going to take a short break, and when we come back, we're going to talk about Alzheimer's disease. and a new study that links it to lifestyle and nutrition. Stay with us. We'll be right back after this break.

This is Science Friday. I'm Ira Flato. Alzheimer's disease is still such a mystery. There are still no known ways to stop this degenerative disease. Available drugs, well, they work only to slow symptoms. But there is a potentially fruitful area. Researchers have been investigating the connection between type 2 diabetes, insulin resistance and Alzheimer's for decades. It's known, for example, that having diabetes

elevates your risk of Alzheimer's. The question is why there is an intriguing clue. A team at Brigham Young University and Washington University in St. Louis found that brain cells in regions most affected by Alzheimer's seem to show impaired ability to metabolize sugars. Here's lead author, Dr. Benjamin We identified every relevant gene involved in glucose metabolism and all of the different cells of the hippocampus, and every gene in every cell was compromised or significantly down. This finding supports characterizing Alzheimer's as a metabolic disorder, suggesting that lifestyle or dietary changes may help prevent or treat Alzheimer's. As the brain is becoming increasingly insulin resistant, it's becoming increasingly less able to get adequate glucose.

So you end up in this sort of tragic scenario where the body is flush with glucose, but the brain can't use it. Dr. Benjamin Bickman. Here to update and talk more about the research into Alzheimer's and blood sugar is Dr. Shannon McCauley, assistant professor of geriatric medicine. She studies the connection between Alzheimer's disease and type 2. diabetes at Wake Forest School of Medicine in Winston-Salem, North Carolina. And I should note, Dr. McCauley is not affiliated with the new research I just mentioned. Welcome back, Shannon. Thank you. It's a pleasure to be here. First of all, what did this team look at in this new research?

Can you fill us in on that? Absolutely. So this team took post-mortem brain samples and looked at how the genes change that really regularly brain metabolism in a few key regions in the brain. And I think what they found is that glucose, which is a fuel for our brain, was unbelievably impaired in the regions that are prone to Alzheimer's pathology, such as amyloid plaques and neurofibrillary tangles. And what they also found is it wasn't just the neurons getting sick. In fact, they found that oligodendrocytes, which are our mom, myelinating cells of the brain were the ones that seemed to be unbelievably impaired using their

approach. So one, it really reinforces the idea that as we develop Alzheimer's disease, that metabolic dysfunction could be playing a causal role in this disease, but also that cell types in these brain regions vulnerable to Alzheimer's pathology are impacted. Why couldn't those brain cells get the glucose they need? So it's hard to know from this. study and a rich literature over the last 20 years, whether it's the transport of glucose into the brain that's impaired, or whether there's an intrinsic deficit in these different cell types to use glucose properly. What we do know is that the regions that are prone to Alzheimer's disease have a really high reliance on glucose-4 fuel. They're very active regions, their well-connected

regions, and those network failures are typical of Alzheimer's disease. And so, So what this study really showed was that they do not process glucose correctly, and how much of that is a cause or an effect still needs to be parsed out with future studies. And as someone who's looking for this connection between Alzheimer's and insulin resistance, what excites you about a study like this and why? For a long time, individuals just thought that the misuse of fuel or the hypometabolism we see in Alzheimer's, which is a key feature that most people use with neuroimaging was just because the neurons were dying or the neurons were dysfunctional.

And what really excites me about this study is it's really showing that perhaps we don't need to focus as much on the neurons, but we need to look at the glia. The glia are the cells that support the neurons, but they're way more than that. So the oligodendrocytes are myelinating cells, the astrocytes and the microglia have a very complex role and inflammation and just normal brain function. And I think what's really rich about this study is it shows that these other cell types are contributing actively to the energy deficit we see in Alzheimer's disease. You know, I've heard years ago people would talk about Alzheimer's as possibly being, quote, type three diabetes because of this insulin resistance and the same sort of jargon

gets thrown around with Alzheimer's this way. Could there be a connection like that? Absolutely. And I think there's there's a couple things that happen with type 2 diabetes. So you have this peripheral insulin resistance that happens in your muscle, your fat, your liver, where your body just doesn't respond to high levels of blood sugar and it becomes toxic to your body. The brain is a very interesting organ and how it uses energy, how it responds to insulin. And insulin serves kind of a double role. It's not only important for managing glucose, but it also seems to to be somewhat of a trophic factor that supports synaptic health. And we know neurons need to fire, have proper brain function.

And insulin really supports that, both from a metabolic standpoint, but then from almost a synaptic plasticity standpoint. And I think that there's a lot of information coming out that is getting at the core of what insulin resistance in the brain means, what does this mean in the context of Alzheimer's, and maybe it could be something we could target, either pharmacologically or through lifestyle interventions. You know, the authors are quoted in a press release saying that, quote, because the average person is eating insulin spiking food so frequently, I look at these findings as a problem we've created and that we're making worse, unquote. Is it too simple to say that eating differently could

solve Alzheimer's? No, I don't think it's too simple per se. And I'll qualify by saying that I don't know that changing our diet is going to ward off all the aspects of Alzheimer's. Alzheimer's is really the confluence of many different mechanisms over a lifetime. But what I will say is that bad eating habits, poor sleep, inactivity, all comes down to the fact that we're not taking in energy properly or using energy properly. And this is really targeting these key networks in our brain to wear them out. So I do think that lifestyle modifications can all kind of keep these networks healthier, longer, and if nothing else, increase our quality of life so we can age in a healthy way.

You know, what I have not heard you say, and what I have not heard from reading the research we're talking about, is the word plaques. And, I mean, that's all we hear about Alzheimer's is clearing the plaques in the brain. Is that not necessary? So you can't have Alzheimer's disease with our current definition without considering amyloid plaques and neurofibulary tangles. At autopsy, that's how we diagnose Alzheimer's disease. There is a lot of work, including work from my own lab that has shown that if you have high glucose levels or hyperglycemia, this makes more amyloid beta and pushes the formation of amyloid plaques. So one way that poor diet can be linked to poor brain health is through generating too much abeta and forming too much amyloid plaques. But the thing is, is that clearing amyloid plaques is insufficient for solving the complexity of Alzheimer's disease. And again, one of the things that's coming out with a lot of work nowadays is that what's the role of glia?

What are the astrocytes, the microglia, and the ligandrocytes doing to contribute to this? And those cell types are very reliant on glucose. They have a lot of different roles. And so it could be that those cells are the ones that really are driving this metabolic phenotype. And it might be if we can get those cells to respond better, you might not have as many amyloid plaques or neurofibrillary tangles. You're of the opinion then that if we follow this metabolism road, we might get some really good results about understanding Alzheimer's.

Absolutely. And I think you can see that with, again, lifestyle interventions, as well as some clinical trials that are repurposing diabetic medications like metformin. Metformin is used to treat type 2 diabetes, but it's also being looked as potentially having a helpful role in Alzheimer's disease. And I think that there's a lot of great medicines out there that people are taking that could help or hinder

the development of Alzheimer's disease. And this is an avenue that we need to investigate further. I think we definitely have realized that removing amyloid plaques is not going to be our cure all for Alzheimer's. And so I really appreciate the NIH's investment in diverse approaches. There's a lot where we're using genetics studies to drive some new avenues into neuroinflammation and what the microglia and astrocytes are doing. So I think that we're going to see a lot of really innovative approaches to treating Alzheimer's disease in the next five to 10 years. Dr. McCauley,

thank you for this really informative discussion. My pleasure. Thank you. Thanks for having me. Dr. Shannon McCauley, an assistant professor of geriatric medicine. She studies the connection between Alzheimer's disease and type 2 diabetes at Wake Forest School of Medicine, Winston-Salem, North Carolina. Don't you love to watch a mime? When you watch a mime, you know, someone seemingly pulling an invisible rope or running into an invisible wall, you're tricked into visualizing something that isn't there. But why does it work so magically? Is it all in the mime or all in the mind? That was the question my next guest set out to study in the science of miming. Dr. Chas Firestone is an assistant professor of psychological and brain sciences and director of the perception and mind lab at Johns Hopkins University in Baltimore. Welcome to Science Friday. Thanks for having me. It's good to be with you, Ira.

So what did you discover? How is the mind helping you see what's not really there? Well, what we were interested to know in our study is not just whether you infer the presence of objects that mimes create, but how? What kind of process in your mind is responsible for giving rise to those impressions you get when a mime seems to be trapped in a box or pulling a rope? So, for example, right now, you know, you're understanding the words I'm saying. And if I were to ask you, hey, Ira, could you, could you stop understanding English, you know, stop understanding the words I'm saying? That's not something that you can do. Just like you can't mentally command your body to stop digesting food or to stop sweating when it's hot or something like that. And so what we wanted to know is when your mind decides for you that there must be some invisible robe or box or wall, is it more like understanding English?

Is it more like something that happens automatically? or is it more like something that you can control? This is Science Friday from WNYC Studios. And so you started doing some experiments. Please describe them for us. So what we did is we set up some conditions where we showed people videos of someone. It happened to be me, actually,

interacting with some invisible objects. And we asked whether the influence that those videos has on people is automatic, whether people can't help but make the inferences that the actor who's interacting with the invisible surfaces seems to be wanting them to make. And then after people watch these videos, what we had happened was a line would appear on the screen in front of them. And they would just have to tell us what the orientation of the line was. So it could either be a vertical line or a horizontal line. And the idea is that sometimes the subject would see the actor step on an invisible box, which has a kind of like horizontal surface.

and sometimes they would see the actor bump into an invisible wall, which has a vertical surface. And sometimes, after seeing the actor bump into the invisible wall, a vertical line would appear, just like the wall, but sometimes a horizontal line would appear. And your job is just to tell us what the orientation of the line is. Forget about the guy who ran into the wall in the box. Just tell us, was the line that showed up horizontal, or was the line that showed up vertical? So it turns out that you're faster to say that a horizontal line is horizontal, if just before it you saw a person interact with a horizontal surface, like a box.

And you're slower to say it's horizontal if you first saw an actor interact with a vertical surface and vice versa. And so, in other words, you can't even inhibit your own mind's inclination to create for you the impressions of these surfaces. What if you happen to have a mime, I'm not talking about you personally, if your mind was not really good, does the brain still help you to see the object? I think it's the case that a mime does have to be pretty compelling in order to make you experience these objects. And one way that maybe we can see this for ourselves is just maybe you've ever tried to do this, you know, at a party or something. Hey, look at me. I'm pretending to be a mime. And how impressive are we when we do this? Not very impressive. Not so good. And so just like we might fail to, you know, impress our party guests if we ever would try this as like, you know, just for fun. There's a kind of sweet spot or tipping point where the interaction with this invisible surface

really does have to be sufficiently realistic in order for your mind to do this automatically. You know, this reminds me of a famous experiment about colors and words. Do you know what I'm talking about? I do know what you're talking about. This is a beautiful and classic experiment. The effect you're referring to is called the Stroop effect. The way that the Stroop effect works is you're going to be shown a word, and the word is going to appear written in an ink of a certain color.

And your job is just to name the color of the ink and forget what the word says. So for example, maybe you'll see the word red written in red ink, but maybe you'll see the word red written in blue ink. And the strip effect is the finding that if you ask someone to just report the color of the ink and not pay attention to the meaning of the word, they have a hard time doing that. So it's hard to say the word blue when the word red is written in blue ink. it's harder to do that than when the word red is written in redding. And our study is kind of capitalizing on that similar sort of congruence and incongruence,

that match and mismatch, where it's easy to report the orientation of the line when it matches the invisible surface that you've just seen. And it's harder to do it when it doesn't. And so just like in the Stroop effect, we learn that reading is automatic in a certain way. You can't help but read the words that are in front of you. We learn something similar about miming in our study. We learned that you can't help but infer the surfer the search. surfaces that a person is portraying for you with their behavior.

And this is because we have so many life experiences, I would imagine. We know how things should be. To be honest, it's not something that we actually know the answer to. So here's one possibility. It's possible that over the course of your life, kind of like you just said, you have experience seeing how objects interact with one another, and you start to learn regularities about how that's true. And then you come to form expectations about those regularities, say, as a mature adult. But it's also possible that this kind of knowledge is built into you, that even as a young baby, as an infant who can't even speak yet, you might be sensitive to this kind of thing. And in fact, a really interesting research direction that we could take is to ask whether

an infant shows something similar like this. Does an infant get surprised if someone seems to bump into something that the infant can't see? And if they do, then that might tell us that actually this knowledge is actually a lot more primitive than you might imagine, not something you had to apply. acquire over the course of your life, but something that might actually be kind of built into who you are. Yeah, that would be a very interesting experiment. I'd like you to do that for us. Would you please? I'd love to. In fact, you know, I'm recently the father of a new baby, so maybe I'll try it on

him. And you can tell us when you come back next time. How about that? Fantastic. Dr. Chas Firestone, assistant professor of psychological and brain sciences and director of the perception and mind lab at Johns Hopkins University. Thank you for taking time to be with us today. Thanks. It was great to be with you. That's about all the time we have for this hour. If you missed any part of the program or you'd like to hear it again, subscribe to our podcasts. And on the Science Friday Vox Pop app this week, Earth Day is right around the corner. How do you plan to celebrate? Tell us about it. That's on the SciFri Vox Pop app wherever you get your apps. Have a great weekend. I'm Ira Flato.