GUEST SERIES | Dr. Matt Walker: The Biology of Sleep & Your Unique Sleep Needs

Huberman opens the first episode of a six-episode series on sleep with Dr. Matthew Walker, whom he identifies as professor of neuroscience and psychology and director of the Center for Sleep Science at the University of California, Berkeley, and author of the best-selling book Why We Sleep. He says the series will cover essentially all aspects of sleep and provide numerous practical tools, naming the biology of sleep and the different sleep stages, why sleep is so important for mental and physical health, emotionality, learning and neuroplasticity, the timing of lighting, temperature, exercise and eating, alcohol, cannabis, supplements and drugs, naps, dreaming, and lucid dreaming. Episode one, he says, focuses specifically on why sleep is so important, what happens when we do not get enough sleep or enough quality sleep, the various sleep stages, and a very specific formula that everyone should know for themselves called QQRT, which stands for quality, quantity, regularity, and timing of sleep. The supplied excerpt breaks off as he begins to enumerate the four factors.

After brief pleasantries, Huberman says that in this series we are going to go a lot deeper and remarks that Walker looks very well rested; Walker replies that he actually slept pretty well last night despite it being a foreign location, in the same time zone, which he says helps just astronomically. Huberman then deliberately declines to ask what a great night's sleep is for Walker, because he is pretty sure the answer will involve some individual differences that people need to pay attention to in terms of what is optimal sleep, and instead starts with the basics: What is sleep?

Walker defines sleep, at least in humans and in fact in all mammalian species, as broadly separated into two main types. Non-rapid eye movement sleep, or non-REM, is further subdivided into four separate stages unimaginatively called stages one through four increasing in their depth of sleep: stages one and two are light non-REM and stages three and four the really deep sleep. Rapid eye movement sleep, or REM, is named not after the popular Michael Stipe band of the 1990s but because of these bizarre horizontal shuttling eye movements that occur during this stage, hence the rapid eye movements. He calls REM, depending on your definition, the principal stage in which we dream, but qualifies it immediately: if your definition is quite loose, meaning any reported mental activity when you wake up, then it turns out that we dream in almost every stage of sleep. He says he will describe REM sleep from here on in as perhaps dream sleep and explicitly flags that he will make that faux pas. The two types, he says, play out in a beautiful battle for brain domination throughout the night, with that cerebral war won and lost on average for the average adult every 90 minutes and then replayed every 90 minutes; the excerpt cuts off as he begins to describe what this cycling creates.

Asked whether the cycle is consistent within an individual, Walker says it is relatively stable within an individual, and that the size of the difference from one individual to the next is much bigger than the size of the difference within an individual from one night to the next. Huberman offers the analogy of a healthy menstrual cycle in a woman, which he says can range from as short as 24 days to 31 days and still be considered a healthy cycle that's regular, and which will change across the lifespan but stay pretty consistent within a given woman for a good number of years while varying quite a bit between women; Walker agrees, saying Immensely. The point is framed as a claim about relative magnitudes, not about fixed values.

Walker raises sex-specific differences: on average, men, if you look at them, will have a sleep cycle that's about 15 to 20 minutes longer than women, which on a 90-minute average is actually quite a lot. He says he brings this point up because people send him claims, probably on social media, asking whether it is true that you really have to structure your wake-up time at these very distinct 90-minute on the clock marks, when the clock strikes the 90-minute midnight, and set your alarm accordingly.

Huberman anticipates that Walker will tell him it's wrong, presumably, and spells out the rationale of those devices: one would be better off waking up at the end of a 90-minute cycle as opposed to in the middle of a 90-minute cycle, even if it means getting less total sleep, because waking at the end of a 90-minute cycle allows one to be more alert upon waking. Walker sketches the strong version of the claim, that there's something uniquely special about the completion of a 90-minute cycle that will have you ejected out of sleep feeling like an Energizer bunny kind of, describing it as some of the claims that they make.

But that's nonsense

False. So sleep for as much as you possibly can sleep. Don't terminate that sleep artificially on the basis of anyone telling you that there is this kind of Da Vinci Code magic 90 minutes. That's unfortunately not true.

Walker's rebuttal is also a self-correction: he says he has been guilty of saying it's a 90-minute cycle and repeating that, so if I didn't know any better, I would believe that, and he adds that he is not trying to chastise anyone, only asking people to be aware of that and not to worry or stress about this unique 90-minute cycle. He notes that there are some products out there that say they're going to time you on your 90-minute cycle and wake you up, and says I would probably stay a little bit clear of some of those. The disagreement is with a popular clock-timed simplification rather than with cycling itself, which he had earlier described only as holding on average for the average adult.

Huberman asks what to do after six hours of sleep when he wakes and feels he could go back to sleep, provided that my work schedule allows for that, or whether after you've gotten a certain amount of sleep it's a good idea to get up and go. Walker's initial answer is to hold tight and stay in bed with an asterisk that I'll come back to, and see if you can get back to sleep, with different ways of helping you do that to be discussed. The asterisk is that if you're then in bed for the next 45, 50 minutes wide awake, the danger, and it doesn't happen to everyone, is that you start to associate this thing called your bed with this thing called wakefulness and not sleep.

Walker anchors this in cognitive behavioral therapy for insomnia, where one of the things that we do is we try to prevent you from spending long periods of time awake. He offers it as probably about a 25-minute rule of thumb and immediately corrects himself: It's not a rule, it's a rule of thumb. The condition is hedged as well: if after about 25 minutes you just can't seem to catch it, and this is happening frequently, he would just be mindful of you then starting to build a bonded association in your brain that your bed is also the place of being awake.

His analogy is that you would never sit at the dinner table waiting to get hungry, so why would you lie in bed waiting to get sleepy, and the answer is that you shouldn't, so we need to break that association. He says there's nothing stopping you from saying There's still sleep in me. I know that there is, getting out of bed, going to a different room, reading a book or listening to a podcast, and going back to bed only when I feel sleepy because my schedule allows for it; he calls that the best way I would tell you if you still think there's sleep there on the table to try and get it back.

Huberman calls knowing that there's sort of a conditioned place effect of being awake in bed immensely valuable, says he gets pretty good sleep most of the time but has had phases of life, including recently, where sleep has been challenging, and describes the words in his mind as Here's the battleground as he heads toward bed, a night of going to sleep, waking up, going to sleep, waking up; he defers the business of continuity of sleep to later. Walker connects this to people who say they are so surprised because they are watching television and falling asleep on the couch and then get into bed and are wide awake, saying that's because, in part, you've built this connection in your brain. He describes the battleground as almost this adversarial thing which in some ways infers that at that point you feel as though your sleep controls you, and it is a miserable feeling; in CBT-I, gradually over time, at that point now you control your sleep, your sleep doesn't control you, and that is such a freedom when you get it.

Huberman says he has prided himself his whole life on being able to sleep anytime, anywhere, which he learned from his bulldog Costello, or maybe that's what brought us together; Walker says he is so sad he's not here around anymore with us and calls him the best sleep ambassador, if there's a poster child for good sleep, and Huberman replies he's here in spirit, sleeping. Huberman then restates the takeaway for people to hear: If you can't fall asleep or if you wake up in the middle of the night and you can't fall back asleep pretty quickly, after about 20 minutes or so, probably best to get out of bed. Notably, his restatement uses about 20 minutes or so and probably, whereas Walker's earlier figure was probably about a 25-minute rule of thumb that he explicitly says is not a rule; the two numbers are left unreconciled in the exchange.

Huberman asks Walker to flesh out the physiology and neurochemistry and touch on the dream features of each of the four stages, especially the deeper stages three and four and REM sleep, not just at the level of rapid eye movements but in terms of the types of dreams or the characteristics of kind of bodily state. Walker describes going into the lab or looking at sleep traces from his sleep center with bewildered awe, and explains that in a laboratory you look like a spaghetti monster, with all of these electrodes on your head, things above your eyes and things on your body, essentially measuring three main signals: electrical brain activity, muscle activity, and eye movement activity. Those three, he says, are necessary for me to know, are you awake, are you in sleep, and if you're in sleep, which stage of sleep you're in.

The three signals, as Walker lays them out: 1. Electrical brain activity, which at stage two he reads from what we call the EEG or the electroencephalogram. 2. Muscle activity. 3. Eye movement activity. Together these resolve whether you are awake, whether you are in sleep, and which stage of sleep you are in.

Stage two non-REM is marked by sleep spindles, which he calls beautiful, short, synchronous bursts of electrical activity that last for about a second to two seconds, maybe a little longer, bursting at what we call a frequency of somewhere between 12 to 15 hertz, meaning these brain waves are going up and down 12 to 15 times per second. By contrast, when we're awake, brainwave activity can be going up and down maybe 20, 30, 40 times per second, very fast and frenetic and actually very chaotic electrical brain activity, but as we're going into these lighter stages the brain starts to slow down. Going down into deeper non-REM sleep, the speed of oscillation, of going up and down, is maybe just one or two times per second, which he calls incredibly slow.

Asked whether this is whole brain activity or localized activity, Walker begins with So we'll come onto this and offers a football stadium analogy: a single microphone dangling over the middle of the stadium picks up the summed voices of the 60,000, 70,000 people underneath, just as an electrode on your head measures the summed activity of hundreds of thousands of neurons underneath, though they have now started to use maybe 100, 200 electrodes on your head to pick these up in local territories of your brain. He stresses that the powerful slow brainwaves of deep non-REM stages three and four are not just slow activity and that the brain is not dormant: the size of the waves, what we call the amplitude, is almost quadruple, maybe 10X the size of the brainwaves when you are awake. His beach analogy contrasts the very quick, small, random waves when you're awake with the kind of epic 20, 30-foot waves that come in very slowly in Hawaii, and he says sleep spindles keep coming riding on top of those big slow waves, referencing the Sleep Sonification Project for what you would hear now.

Huberman takes that description to answer the question he was going to ask, whether the pattern of brain activity occurs in similar or identical form during any waking states, and understands the answer to be no: if I understand correctly, this is a very, very specialized brain state unique to sleep, unique to a specific portion of sleep, and that begs the question, what is it doing?

And I will say, well, it turns out that they're all important. You need all of them.