University Place

Are You Really Sleeping? Sleep and the Brain

– Welcome, everyone, to Wednesday Nite @ the Lab.

I’m Tom Zinnen. I work here at the UW-Madison Biotechnology Center. I also work for UW-Extension Cooperative Extension, and on behalf of those folks and our other co-organizers, Wisconsin Public Television, the Wisconsin Alumni Association, and the UW-Madison Science Alliance, thanks again for coming to Wednesday Nite @ the Lab.

We do this every Wednesday night, 50 times a year.

Tonight, it’s my pleasure to introduce to you Stephanie Jones. She’s with the UW’s Institute for Sleep and Consciousness.

She’s gonna talk to us about “Are Your Really Sleeping? “Perspectives from the Brain.”

Sleep is something we all try to do 1/3 of our life.

It’s a pretty amazing time in neuroscience to think about what we can learn about the brain and the interaction of the brain with the rest of the body when sleeping is going on. So, I’m looking forward to this.

Stephanie was born in Newburyport, Massachusetts and went to high school there. She got her undergraduate degree at the University of Massachusetts Amherst, got her master’s degree here at UW-Madison, and her PhD here in neuroscience. And as I mentioned, she’s with the Institute for Sleep and Consciousness at the School of Medicine and Public Health.

Please join me in welcoming Stephanie Jones to Wednesday Nite @ the Lab.

[audience applauds]

Well, thank you all for coming, and thank you, Tom, for inviting me.

We use this really cool technology in the lab. It’s a 256-electrode head net, and I was gonna have it right here on a disembodied head, and I forgot it.

[audience laughs]

Sorry about that.

So what I’m gonna talk about is the work that we do at the Wisconsin Institute for Sleep and Consciousness.

I would love to say that this work is all mine and I’m a genius and I did it all on my own, but, alas, that would be a lie. I’m gonna present a body of work that was produced here over the course of maybe 10, 15 years. It involves a huge number of people, but there are a few that deserve honorable mention.

The center director is Giulio Tononi. He is arguably one of the world’s leading sleep research experts. He also is a consciousness researcher. Also arguably the world’s leading theorist of consciousness. Nobel Prize caliber, but don’t tell him I ever said that.

His partner, research partner, is Chiara Cirelli.

So the hypothesis I’m gonna present to you was developed by the two of these folks together. So they are the two main PIs in the group.

And then Ruth Benca also did a lot of work on the clinical component of this body of research. She is now gone at the University of California Irvine, but she has made substantive contributions to this work.

And also Brady Riedner, who is my closest colleague. Brady and I do a lot of work on the clinical end together. He runs through all of this work because he’s sort of the technical guy. He is responsible for developing a lot of the analytical tools we use to do this research.

The Wisconsin Institute for Sleep and Consciousness was, up until April, the Wisconsin Center for Sleep Medicine and Sleep Research. So we’ve had a leadership change and a title change.

Scientists aren’t generally trained to speak to lay audiences. So, this is kind of an experiment for me. But I do feel like it’s an important thing to do because, after all, you’re all paying my salary, right? I’m out there at Research Park partying on federal tax dollars.

[audience laughs]

So, I feel like, you know, I have an obligation. We have an obligation to present the work we do in some way that’s understandable. I don’t know if what I’m gonna show you is understandable. That’s the problem. So, I would love it if you would all come up after and say to me, this was too much, this was too little, this was difficult to understand, etc. I have a really thick skin, so you don’t have to worry about hurting my feelings. I’d like to continue to do this kind of thing, so I’d like to hear from you.

We all know what sleep is, right? Or we think we do. We all do it.

But sleep is a state of reduced responsiveness to the environment that’s readily reversible. So we measure sleep in the mammal with electrophysiology, you know, technology.

But, basically, sleep is fundamentally, first and foremost, a behavior. And we can distinguish this in all animals. It’s rapidly reversible, or readily reversible. That distinguishes it from a state like coma or hibernation.

As a little fun fact, animals who hibernate will come out of hibernation to sleep, in fact. And they come out in a state of sleep deprivation.

So, I got a quote. Everybody starts a talk with a quote, right?

“Sleep is a criminal waste of time “inherited from our cave days.”

Bam! Does anybody know who said that?

[man mumbles]

Thomas Edison. – Thomas Edison. You get a bag of candy.

[audience laughs]

He said this around the time he was inventing the light bulb. And this is the time when we started invading the night, right? What was this? The late 1800s.

We start invading the night and giving short shrift to sleep. We’re controlling our environment.

Some predictions say we now sleep anywhere from one to three hours less. So you can imagine in a place like Wisconsin, when the days are really short, we slept longer. But now we control our environment. We don’t sleep so long. Edison would say that that was a wonderful thing. Do we think he’s right? Probably not, right? I think I’m gonna try to convince you that that’s not correct.

That we shouldn’t be giving short shrift to sleep.

And why not? Well, I’m gonna argue that sleep is incredibly important. If it weren’t, we wouldn’t spend about 35% of our time asleep. We sleep more than we do pretty much anything else. Despite, you know, pharmaceutical companies trying to dispense with it.

This is some data from the Bureau of Labor Statistics. So if you live 77.8 years, you will spend 27 years sleeping. If you work until you’re 65, you’ll spend 10 years working. So sleep is an incredibly important thing.

When I was looking for some of these statistics, I also learned that [laughs] women spend 17 years on a diet.

[audience laughs]

I was like, what? And then I thought, yeah, I’m on a diet right now, of course.

So, why else am I gonna tell you sleep is important? Sleep is absolutely universal. So all animals do it, from the lowly fruit fly, birds do it, and it’s so important that animals have evolved really special adaptations to it, right?

This here, this is so gross. This is a parrot fish, and he secretes a ball of mucus when he sleeps, right? I’ll keep my duvet.

[audience laughs]

Dolphins and many species of birds will sleep unihemispherically. That means half the brain will go to sleep.

It’s important in aquatic mammals, as you can imagine, because they have to keep moving, right? So seals do this, dolphins do this. There are all kinds of interesting adaptations. And I think that argues for sleep being fundamentally univ..important, right? If nature could have dispensed with it, I think it sure would have.

Sleep is also dangerous. I mean, from the perspective of an animal, right, they put themselves at significant predation risk. I mean, to some extent we do too. And sleep is dangerous for us insofar as it comes at a huge opportunity cost. We can’t do all the things that we want to do to make ourselves successful as a species. Reproduce. Most of us don’t do reproduction during sleep. We don’t work. We don’t do anything important.

And another thing that argues for the importance of sleep is what happens when we don’t get enough?

[audience laughs]

We’re all familiar with this. In light of the conversation about sleep’s role in, you know, cognitive function, memory consolidation, I like to highlight that sleep is also fundamentally involved in emotion regulation, right? You know that if you haven’t slept and your office mate is tap, tap, tapping on his keyboard, you just want to light him on fire, right? For everything he does. I’m thinking about my colleague Brady when I talk about that.

This is also particularly robust in children. I’ll come back to that. That’s a theme for me, children, and the importance of sleep in that population.

When children are sleep deprived, we all know the toddler, you know, screaming, they have profound difficulties with emotion regulation. But it does continue across the lifespan.

And, also, of course, not sleeping has profound effects on cognition and performance. So, when we’re sleepy, our brains function less well. We don’t remember things. We don’t perform. We’re impulsive. We make bad decisions. 26% of drivers report falling asleep at the wheel once to twice a week. Something like 38% of all auto crashes are associated with drowsy driving. And some major disasters, Chernobyl and the space shuttle Challenger disaster, were allegedly associated with sleep loss.

So, what have I just argued for you there? That sleep is really important. It’s really fundamental, and we can’t dispense with it. So, why do we do it, though?

It’s almost an embarrassment in biology. So we have come so far, but we can’t answer this very simple question: Why do we sleep? If it’s this important, we should know why. Nobody agrees on the answer to this question. So what I’m gonna present to you is a hypothesis. I might even call it a theory, which was developed over the years by those people I outlined for you, Giulio Tononi, Chiara Cirelli.

And the argument is that sleep, in its simplest form, sleep is the price we pay for learning, okay? So I’m gonna outline this in really basic terms before I then get into showing you all this data. And if it’s too much data, you can blame Tom because he told me that you all wanted to see data, data, data.

[audience laughs]

So I am gonna bring it on. I’m gonna define a couple of things. So, I’m gonna use the word plasticity a lot.

The brain is moldable across the lifespan. It’s more moldable when you’re young, as you can imagine. You know, you come out of the womb; you become bombarded by experience. It’s that experience that builds and optimizes a brain. So, plasticity is the ability of that brain to modify its structure and function following experience, okay?

This is a fundamental biological feature, but, remarkably, we haven’t known this to be true for very long. We used to think that, you came out and things just stayed as they were. Little people got bigger and that was it.

So, plasticity is also the foundation for learning, okay?

So, this hypothesis is called the synaptic homeostasis hypothesis. We call it SHY. Basically, what it says is this. You walk around during the day and you are learning. So, I’m gonna show you experiments where we formalize learning, right?

But just the act of living and walking about is learning. It’s acquiring experience and modifying your behavior, okay? So you walk around and you learn, and you learn and you learn and you learn. And at a certain point your brain kind of just taps out, right? Learning is an expensive thing. And then when you go to sleep, what sleep does is it takes down all the noise of the day and it preserves the important stuff.

It seems like a really simple idea. And I think one of the beauties of it is that it is such a simple idea. But believe it or not, it’s controversial. So not every sleep researcher will agree with all the things that I’m about to tell you, right? They’re wrong.

[audience laughs]

So, some scientists believe that you actually, sort of, learn while you sleep. That when you’re sleeping connections get stronger. Your brain is kind of reflecting on experience and optimizing itself. Yeah, yeah.. There’s not a lot of evidence for it, but people do persist in their beliefs.

Okay, so here’s a basic principle. Wake is always associated with learning, right? That makes perfect sense to everybody? Yeah, because if you didn’t learn while you were awake, imagine if you learned while you were asleep.

So, you’re offline, and your brain is creating a conscious world, right? It does, while you dream, even though you’re not aware of it. You can’t sample your world. You would just be taking information generated in this private offline world. It doesn’t make any sense, right? So, you learn while you’re awake. You perceive the world, and you change your behavior accordingly.

From the neuron’s perspective – these are units of information transfer in the brain – from the perspective of a neuron, learning is, essentially, getting, increasing your relationship with your neighbor. It’s increasing synaptic strength, okay? So, across the day you’re learning, and learning is increasing strength between synapses, right? That’s how you build a big brain that has learned a lot, right?

As you’re going across your day and you’re learning and your synapses are getting stronger and stronger, right? At a certain point you tap out. It becomes too much. You get saturated, right? Because synapses, in fact, are really expensive things.

Even under optimal conditions, the brain is about 2% of our body mass and it consumes about 25% of our resources, right? In terms of glucose, it has incredible energetic demands. And synapses, in particular, so all those resources, most of those resources go to signaling between neurons, not just housekeeping functions, right?

So if you’re learning and strengthening synapses all day long, you can imagine a point at which resources are depleted, but also space. I mean, we don’t have unlimited space in this skull, right? We don’t have the capacity to continue to grow and grow and grow. So, at a certain point something has to happen, or you will just explode. So, this is where sleep comes in.

Sleep comes in and it downscales synapses. So, the things that are important have been tagged as such because from a neuron’s perspective, learning is firing or not firing, right? So, if something’s important to you during the day, your neuron is somehow guided by this is important, right? So, they fire, they fire, they fire. They get stronger. And then, so sleep comes in and it just says, globally, I’m gonna take down all this noise, and what’s important will be left behind.

I’m gonna show you in detail some experimental modeling of this and also talk specifically about the kinds of sleep that does this, the kind of sleep.

So, in this way, sleep improves memory, right? Sleep aids in consolidation of memory. And , um.all the things that we do, insight, just extraction, everybody knows the sense of you wake up refreshed. Sometimes you’ve worked on a problem and it seems solved, right? So what sleep has done is it has, it’s increased the signal-to-noise ratio. It’s taken down all the nonsense and left behind all the good stuff. Is everybody buying my story so far?

[audience mumbles]

Good.

Okay, so what have I said? Sleep serves an absolutely essential function. I showed you a lot of evidence to support that, right? Waking is associated with learning, which is represented in our brain by increased synaptic strength. Sleep removes all the noise of the day by downscaling, okay? Just globally downscaling synapses. And then here’s the other thing. Sleep also is a reset button. That’s how I like to think of it. You know how Staples has that easy button? I think of sleep as the reset button because there, at some point, you know, as I described, you become saturated. Your brain no longer has the capacity. There’s not enough room for new synapses or new learning. There aren’t resources to support it. So, sleep comes in and does a reset. And you wake up the next day refreshed and energized, ready to learn anew.

I’m gonna describe the kind of sleep that does this, that hits the reset button for us, okay? So, wake plasticity and the sleep reset.

I’m gonna talk about slow waves. So, you probably all have heard of the different kinds of sleep, right? Does everybody know what kinds of sleep we have? No? – [man] REM. – REM, rapid eye movement sleep. And non-REM, non-rapid eye movement sleep. Non-rapid eye movement sleep takes up about, in a healthy adult, 75% of sleep time.

Okay, so this gentleman, Simone, wearing this head net was the fantastic piece of technology I failed to bring. Thats, um…people, believe it or not, sleep with that net on, quite comfortably. I’ve done it myself.

Um, so here is a night of sleep I’m showing you. This is from a standard EEG set. This is recorded from the scalp, right? These little squiggly lines, in fact, convey a lot of important information. And over in the right-hand corner is, um. a hypnogram. That just shows you how those stages go across the night.

So, a typical night of sleep, you, um, go from stage one, two, three, four. Actually, now we only do stages one, two, and three. It’s just convention. And then you will have a bout of REM sleep. And you can see that as, across the night, REM sleep gets longer and longer. That black bar, um, represents the duration of REM.

So, you descend into sleep like this. Stage one, stage two. You know, stage one is that stage where you’ll sometimes have what’s called hypnic jerks. That sense of falling, right? And at this point, um, your consciousness gets a little bit wobbly. You might be in a, in a, um kind of fugue state, is what I like to call it. And then you descend into stage two. The hallmark of stage two, sleep spindles and K-complexes. Those allegedly have some special functions for learning and memory. And then we get into what’s called slow-wave activity. Delta wave. Delta waves. This is arguably the most important kind of sleep. And this is the kind of sleep I’m talking about when I say sleep hits the reset button.

And people are fond of– Oh, and this below is REM sleep, which looks a whole lot like waking. Actually, in this image not. But REM sleep basically looks like an awake brain. It’s an activated high-frequency signal.

People often ask me if REM sleep is the deepest sleep. In fact, no. The deepest sleep that you go into is this sleep here. It changes across development, like everything does, right? So, when you’re young, when you’re a child, children have, um, enormous, long, slow waves. It takes up a larger proportion of their sleep. Ah, and then as you age, like with everything, these, sadly, go away. In men, it’s like falling off a cliff. And in women, it’s a gradual decline. But, so, these big important waves change across the developmental spectrum.

Oh, this we’ve just determined does indeed work. So, I’m showing you, basically, three layers of the cortex, okay? What you’re gonna see here is the firing of neurons when they’re awake and then when they go to sleep.

So, you see the fast spiking firing. Those flashing, those are neurons. This is the local field potential, okay? Now we’re gonna put the model to sleepsoonly-ish That is sleep.

So those are what’s called off periods. So, this is when you see those great big waves on the surface of the cortex using EEG, that’s what it looks like if you were to record intracellularly, right? Neurons will all go off together and then fire together, off together, fire together. So, it’s like a symphony all coherently singing together.

Okay, so now I’m gonna have to orient you to the topoplots. So, we use a technology called high density EEG. This is 256 electrodes recorded from the scalp in that net that we just saw.

Slow-wave activity is this special sleep I’m talking about. I’m gonna refer to it as slow-wave activity much of the time, okay?

And we know a lot of things about this sleep. When you look at a topoplot, the red parts, where it’s red suggests that power in this low frequency slow-wave activity domain is strongest there, okay? So, when I point to this right here, you know that there’s a lot of slow-wave activity going on in that region frontally. In fact, that’s what the topography of slow-wave activity looks like, if you just take a freeze frame.

Um, so what I’m showing you here is this is slow-wave activity when following a night of sleep deprivation, okay? So, this is following a night of sleep deprivation. There’s a whole lot of power in that first hour of sleep.

This is slow-wave activity on just a regular night’s sleep.

This is slow-wave activity on a night when you’ve had a nap during the day.

And this, I totally forgot what that is. It doesn’t matter.

[audience laughs]

Because my point is this, um – slow-wave activity is homeostatically regulated, right? So, the more you stay awake, the more power you have here, and this will become important as we go along, okay?

So, um, if you are sleep deprived, you don’t really, you dont really increase the duration of sleep when, when you have a sleep epoch. What you, um, increase is the power or the intensity of sleep, and that is reflected in this, in this very special slow-wave activity.

Okay, now we’re gonna get to learning. Now, remember, what I said was that we, um, learn all day long, right? Right now we’re all learning. Um, but in a lab environment, you have to formalize it somehow to test it, right?

And so what I’m gonna talk to you about is local learning and sleep need, okay? And how we know that they’re linked. So how we know that sleep, particularly slow-wave activity, is involved in the process of learning, okay?

So, I’m gonna talk about, um, some experiments that we have done to show evidence that slow-wave activity is important for regulating synaptic plasticity. I’m gonna show you three different models.

Remember I said that learning is associated with increases in synaptic strength, right? And that sleep is associated with downscaling the unimportant stuff.

Okay, so I’m gonna show you three different experiments here.

The first one, in this task what you do is we train people to rotate– It’s an implicit learning task. We train them to rotate a cursor, um, and vary it by a certain number of degrees. So, essentially, they are figuring out that there’s some rotation, but they’re not aware of what that is.

The reason we use this particular task is because it, we know that it’s underpinned by a certain region of the contralateral motor cortex, right? So, it’s a well-defined task. We can look at where the learning is happening, essentially.

So, and what we find is that when people do this task, slow-wave activity is increased in the contralateral motor cortex, right? So, we know that the task activates this contralateral motor cortex, contralateral to the side that you’re using, right? Because your brain is, um – and what we see is that slow-wave activity increases locally in that region, right?

And, in fact, we know it’s not just a motor effect because the percent change in slow-wave activity, those who sleep more intensely learn better. So, the performance is associated with that regional sleep intensity.

Humans have one big central sleep period. That’s how we are built. So, you can split up the sleep period, but it’s not ideal. It’s not your best sleep.

But provided you are getting slow-wave sleep, presumably you can learn as well. So, we do see when people take naps during the day that nap is associated with performance improvement in all kinds of tasks, not just motor tasks, but, you know, any task you learn in a lab. You learn word matching, right? So, naps will improve performance.

We can learn during the day. Of course, we can. If I gave you a list right now and said come back in five minutes and, you know, do you remember this, this, this, this, you would. But if I allowed you to sleep on it, you would, youd learn it better, basically. We can learn during the day, but sleep optimizes it.

The harder you’re awake, the harder you sleep. Right?

So, um, and if you do a lot of learning during the day – right? -your brain will sleep harder, more intensely. Um, and there’s nothing you can do about it. Biology drives that, so if you stay awake, which I, I, I think is equivalent to learning, you will sleep harder.

Now, what I’ve shown you in this task is that if you tax a particular brain region hard, it will sleep hard, right? This special sleep. This downscaling sleep. Non-REM slow-wave activity, which is the dark red stuff. This is sleep intensity.

Okay, so basically what I’m showing you here is that we are working this particular motor cortex really hard. And then what is happening is the brain in that region is sleeping really hard. So, of course, keep in mind the whole brain is asleep here, right? But what we have done is we have taxed it. So, we have shown that if you tax it really hard, you can make it sleep in a special way.

So, you know, if you’re learning to play the piano, for example, your whole brain will be asleep, but particular regions of it will sleep harder than other regions because you’ve taxed it.

Does that, does that help everybody?

And it’s associated with performance improvement. So, if you were scientists, you would say to me, “Yeah, but how do you know it’s just not a motor issue? “That you’re just moving that arm? And” But, in fact, what we know is that, the betthe people who do better, who improve on that rotation task, right? They sleep harder.

So there’s a direct relationship between how well you do and how hard you sleep. It seems to be that sleep is linking those two things fundamentally.

Um, we can also just use something called TMS, transcranial magnetic stimulation. It’s a big coil. It goes click, click, click. It almost blows your eardrums out. And we can tax the motor cortex with this, right? Basically what we do is we’re making neurons fire like crazy, experimentally, okay? I’m inducing the brain to work really hard. Not with a task, but basically I’m inducing synaptic strength. Learning, right? With this big magnetic coil.

And then what I see is, once again, when I tax that brain really hard, it sleeps really hard in that really small region. So keep in mind the whole brain is asleep, but I’m taxing it hard and it’s sleeping hard.

We also see this in animals. I mean, I could show you, I could show you data from fruit flies to support this argument. So, this is a rat. He does a he does a reaching task. He reaches for a sugar pellet. So, it’s the same kind of thing. The contralateral motor cortex is involved in this task. As he reaches, he learns to reach, the contralateral motor cortex sleeps harder and harder. The better he does, the harder that part of his brain sleeps.

So, what I’m telling you here is that sleep is, um, can be local, and it’s use dependent. So, if you work it hard, it sleeps hard, and that can happen on a regional basis.

So, what I’m saying is stronger synapses equals better performance and more slow-wave activity, right?

Remember, waking is associated with stronger synapses because that’s the foundation of learning. So, I’ve just shown you data in a task in a person, a task in a rat, and then just inducing it with a magnetic coil. If I make synapses stronger, I make the brain sleep harder in that region.

I can also do the opposite. If I immobilize your arm, the contralateral the motor cortex that underpins that arm won’t sleep as hard. We put someone in a cast for a week, believe it or not, or a sling and they don’t use their arm, what you see is that the brain there isn’t tired. The synapses aren’t getting stronger, so it doesn’t sleep so well. It doesn’t need to sleep as well is the argument.

So weaker synapses equals less slow-wave activity. So, the argument I’ve built here for you is that sleep is directly involved in regulating plasticity at night, right? If you learn hard, you sleep hard. And sleep has some fundamental role in that process.

Now, I said at the outset, I think, that there’s a price to pay for this. So, remember, we walk around all day. We’re learning, and at a certain point we tap out. There’s not enough space. There aren’t enough resources. So, what there is a price to pay for this plasticity. And what is that price?

So, the synaptic homeostasis hypothesis basically says synapses are expensive, there’s not space to continue to learn and learn and learn. At a certain point you saturate, you can’t do it anymore, so you have to sleep. And, in fact, you do HAVE to sleep. There isnt, you know, um, sleep is a biological imperative. So, people who say, “Oh, I don’t need to sleep,” or, you know, long-haul truck drivers saying, “No, no, no I’m fine.” They’re not fine, in fact. They are sleeping.

So, what I’ve shown you, I hope, is that there’s evidence that a brain sleeps locally. So, you can be a behaving person and parts of your brain will go offline. Like I showed you withwhen we were looking at those local field potentials, right? What it, what it looks like is just a bunch of firing and then stop, bunch of firing, stop. I’m gonna show you that again.

So, this is a wake brain. All these neurons in your cortex are firing happily. Then the neuromodulatory environment changes, and voilthat’s what a neuron looks like, or a set of neurons look like when it goes to sleep. So, it’s an off period, right?

So, what peopleWhat happens to people when they think they’re awake is, in fact, often they’re having off periods. That’s the price you pay for plasticity. That’s the price you pay for extending wakefulness beyond its natural duration. Or, you know, so I’m gonna show you some evidence for that, okay?

So, this is data from rats, okay? These are just freely moving animals. Ah, the longer you keep them awake, they will continue to move around and look like wakeful animals, right?

This is the local field potential while they’re awake. These are just all cells firing, firing, firing, okay?

This is what happens during sleep deprivation.

So, this is an awake moving animal. He’s having a great big off period. This is non-REM sleep, that special sleep. This is what the EEG looks like when you have an off period.

So those great, big slow waves that I’m talking about, that slow-wave activity, if we were to record from the neuron, again, it looks like this.

The neuron goes off.

So, what you see on the cortex is a great big slow wave, and what you see in the cell is off.

So, in a freely moving rat who is behaving, we’re keeping him awake in nice, friendly ways. They like to be engaged. If you play with them forever, they’ll stay awake, but while they’re moving around, they’ll have off periods, okay?

Remember, go back to this. So, we all know this feeling that when you haven’t slept, you’re tired, you’re crabby, you make poor decisions. What is the foundation for that? Can anybody guess I’m gonna say the foundation is for that?

– [Man] Off periods.

Off periods, excellent. Wow, you’re learners.

[audience laughs]

So, the question is, is he awake or asleep?

So, you remember that I showed you this rat earlier performing the reaching task, right? And I said that his contralateral motor cortex would sleep really hard because he had learned well, how to reach, how to grab, okay?

So, this is what happens. If you keep him awake and keep him behaving on this task, you continue to make him reach. He will get to the point where he’s so tired, he’s still so motivated to get this though, right? It’s a sugar pellet, by the way, they love it. He will continue to reach, but every time he makes a mistake, guess what that’s associated with?

[audience mumbles]

Off periods. So, he’s awake, he’s moving around. He’s even reaching. But when he drops…

I mean it’s a fairly complex behavior for a rat, right?

When he drops or has a miss, he has to squeeze his hand through this little, you can’t really tell. But so those mistakes are associated with an off period.

So, this is in a freely behaving animal. He’s awake, he seems alert, but, in fact, not so much.

I will tell you that we’ve done a similar thing in the peoples.

Two tasks.

We keep people up for 40 hours doing two different things. Driving simulation, right? And we know that, um, driving simulation activates a particular network. I mean, it activates a lot of networks, but, um, this kind of posterior, um, motor cortex, okay? This parietal network. Driving simulation generally activates this. So, people are up for 40 hours, and they’re pretending to drive, okay?

The other people are up for 40 hours and they’re listening to audiobooks. This,um, in general, activates another cortical network.

So, we can differentiate between where the brain is being activated during these tasks, okay?

So, during the audiobook, we’re, um, activating this frontal auditory network. During the driving simulation, it’s this posterior network.

So, they stay up for 40 hours, and they do this. They’re not allowed to do the other thing, right? Theyre not allowedso, if you’re listening to your book, you’re not allowed to be messing around with your hands.

Guess what happens?

If you keep them up long enough doing this task, the sameall the things that I’ve said apply. So, um, they will sleep harder in that region once we let them go, right? Go to sleep.

So, the entire brain will go to sleep, but we will be able to see evidence of regions sleeping more intensely with that special slow-wave activity.

But, also, we will see during the day, okay, in these awake, freely moving people behaving, we will see lapses, off periods. Now, we can’t record them intracellularly, but I think I’ve shown you enough.

What we will see is great big slow waves appearing on the cortex. So, people are awake, but they will have lapses, um, and those lapses are associated with they’re regionally specific, right? We’ve taxed a specific region. And we will see slow waves appearing only there.

So, the rest of the EEG is activated. So, remember, I’m covering the head with 256 electrodes. So, it’s a full helmet, let’s say. Um, and what you see is the appearance of slowing, local slowing, on the, the region that has been overly taxed.

So, this is, essentially, the foundation for the errors people make when they’re doing things like flying an airplane when they’re sleep deprived or, um, driving a big car. People are notoriously poor at, um, assessing their own level of alertness, let’s say, right?

So, to review, sleep is like a reset button. So, when you go to sleep, the brain downscales all the unnecessary business of the day, and it preserves the stuff that’s important.

In that waySo, waking is associated with synaptic strengthening. This is learning, right? And this effect can be local, as I’ve shown you.

Um, without the b without sleep, the brain becomes saturated, and this learning can no longer happen. And also, when you get saturated, sleep becomes absolutely imperative.

So, you may be awake, you may be driving, but, in fact, there are parts of your brain that are saying, “Oops, can’t do it anymore.” That’s the price of plasticity.

So, sleep downscales synapses leaving what’s important, and it’s in this way that sleep, ah, improves memory.

It consolidates memory by getting rid of the noise and leaving the signal, basically. The signal that you’ve worked so hard to drive in during the day. And then it frees up resources in space for the next day. That’s how it’s the reset button.

We also, in the lab, study lots of sleep disorders. One of them, which you might be familiar with, is obstructive sleep apnea. A lot of people have it. A lot of men have it. After 45, I think you can argue that all men have a little, regardless of weight. Weight is a big predictor of whether or not you will have obstructive sleep apnea.

It’s basically a crowding of the upper airway, um, or a floppiness, um, and it leads to secession of breathing during sleep.

It’s associated with all kinds of cognitive effects. All kinds of bad effects. I mean, so it’s rough on the cardiac system, but it’s particularly hard on the brain.

Um, and behaviorally, a person with sleep apnea people adapt to it very quickly. They think that they’re okay. This is why, I think, compliance with the treatment regime is so low. People say, “Oh, no, no, I’m fine.” But, of course, I just argue that, no, they’re not fine. They think they’re fine, but they’re having off periods like crazy, right?

Obstructive sleep apnea.

This is what a night of sleep looks like.

So, remember I was showing you that the big red stuff, where we had a lot of that important sleep?

In obstructive sleep apnea, these big slow waves aren’t here.

And then, more importantly, we have studied every possible population you can imagine. And, in fact, my special interest is in development.

Children.

Children um, often either snore or have sleep apnea. As you can imagine with the growing um, obesity epidemic, sleep apnea in children is getting much more pervasive, right?

Ah, children also often snore because they’re little and they haven’t, they havent yet grown into their airways. So, they can have a really crowded airway, not because they’re overweight but they have big tonsils, big adenoids. It’s a fairly common thing.

And, see, here’s the puzzle over, over snoring in kids. We, we always argue that snoring and OSA in kids is relatively benign, and the reason we say that is because when you look at a regular standard EEG in children with this disorder If you look at an adult, the sleep looks like a mess. You can see them arousing all night long, okay? It’s a hideous mess.

when you look at a child, though, he doesn’t look a mess. And people have argued, well, sleep is so fundamental to development, it’s so important to building and optimizing that brain that it they don’t have disrupted sleep.

But the puzzler is, then why do they have such significant behavioral effects? Ah, even a snoring child who does not have

I should have said that obstructive sleep apnea is associated with oxygen desaturations, right? But the brain is darn good at protecting itself from subtle changes in blood oxygenation.

Um, but in snoring, there’s no, theres no pause in breathing. There’s no stopping. It’s just airwave resistance.

But it turns out that kids who snore have behavioral effects, learning effects that are almost as significant as kids with OSA. How could that possibly be?

Their sleep looks completely normal. But here’s the thing, just like we see in adults

This is the adult data that I just showed you. Just pay attention to this top line here.

So, remember that good stuff? The slow-wave activity in the adults isn’t getting back there, right? It isn’t getting back there in kids who snore either.

Even though in this person the whole global EEG is a hideous mess and we can say that person has 500 arousals in a night.

With kids with OSA, mmm, sleep doesn’t look that bad. But it turns out that they’re not getting sleep back there. And even in kids who snore, they’re not getting sleep back there. And why should we care about this?

We know that sleep mirrors, sleep intensity, the hardness, the deepness of sleep, mirrors synaptic strength, right?

I, I saidthat’s what I was showing you when I was showing you those learning tasks. Remember, learning is associated with synaptic strengthening.

The other thing that we know, though, is that when you build a brain in a child, you always build it from the back. Brain development goes from back to front, in to out, right?

So, a five, six-year-old child, which is the age of the children I just showed you, is still in the middle of building.

And this is a high density EEG in typically developing children.

And what we see is that over the course of the lifespan, sleep, this power, this special sleep, is strongest where they are building synapses, okay?

So, little people, their slow-wave activity?

For us, it’s in the front.

For them, most of the power is in the back.

So, slow-wave activity mirrors brain development, synaptic development, as we would expect it would be based on all of the things that I just said.

Um, so they’re building a brain in the back. You know, the parietal cortex. They’re working hard to build that region.

But in snoring and OSA, sleep isn’t getting back there. So, it’s not doing its job. So those kids are not building an optimal brain.

We don’t know what the long-term consequences are, but you could imagine.

In older adults, this is true too.

Why do we care in older adults that sleep apnea affects posterior sleep intensity?

We care for a lot of reasons, but one notable reason is, um, the relationship between OSA and Alzheimer’s disease and, um, amyloid deposit.

So, it turns out that amyloid, preferentially, deposits right back here, right?

We know that OSA always makes Alzheimer’s, and even dementia, worse.

So, one potential mechanism for that relationship, this is speculative here, all I’m saying to you is that sleep doesn’t get back here. Amyloid deposits back here.

We know that not only does sleep work on the synapses back here, but there’s something remarkable called the glymphatic system that works during sleep. You probably heard about this. This is all in the news.

It’s, ah, its basically the garbage truck of the brain. It clears out metabolic waste.

We’ve only just discovered this.

It clears out amyloid, in particular.

Um, so in people with OSA and Alzheimer’s, sleep isn’t getting back here. Amyloid is. So, sleep may not be doing its job in part because the glymphatic system isn’t optimized. So, what are some potential mechanisms for this? Why is sleep not getting there?

One thing I didn’t tell you about those big slow waves is that they are, um, a truly dynamic phenomenon. So, the brain doesn’t go to sleep all at once, just like it doesntits sleep intensity isn’t uniform. The brain doesn’t just fall asleep.

Um, slow waves have particular preferential points of origin. They kind of start in the front and in the middle, and they make their way back.

So, if you are arousing across the night, as you would be in OSA, right? It’s conceivable that sleep just doesn’t get back there. Slow waves can’t travel fast enough to compete with all the arousals that you’re having.

Another, um, sleep disorder that plagues a lot of people is insomnia, primary insomnia. There’s never been a neural signature of, of this disorder.

Um. What, what we’ve seen recently is that during sleep in people with insomnia, there is local waking in the sensory cortices.

So, this red stuff, now I’m changing the whole scene, this red stuff is alpha. Alpha is what you primarily see in a waking brain.

So, people with insomnia report, you know, not sleeping well, but if you look at them, they seem to sleep fine.

But it’s probably the case – it looks to us that the case is that, you know, the sensory and auditory cortices are still monitoring the environment. So alpha is a signature of wakefulness, and they are kind of quasi-awake.

I, you know, we study all kinds of sleep disorders, but I will stop there. So, that’s it, that’s my story. Thank you for listening.

[audience applauds]