Making Sense - Sam Harris - #267 — The Kingdom of Sleep

00:00:00.000 Welcome to the Making Sense Podcast. This is Sam Harris. Just a note to say that if

00:00:12.180 you're hearing this, you are not currently on our subscriber feed and will only be hearing

00:00:16.320 the first part of this conversation. In order to access full episodes of the Making Sense

00:00:20.820 Podcast, you'll need to subscribe at samharris.org. There you'll find our private RSS feed to

00:00:26.420 add to your favorite podcatcher, along with other subscriber-only content. We don't run ads on the

00:00:31.580 podcast, and therefore it's made possible entirely through the support of our subscribers. So if you

00:00:36.360 enjoy what we're doing here, please consider becoming one. Today I'm speaking with Matthew

00:00:48.500 Walker. Matt is a professor of neuroscience and psychology at UC Berkeley and the director of

00:00:54.580 its Sleep and Neuroimaging Lab, and he's also a former professor of psychiatry at Harvard University.

00:01:00.680 He has published over 100 scientific studies and has appeared on 60 Minutes, Nova, BBC News,

00:01:08.080 and many other outlets. His first book, Why We Sleep, has been an international bestseller,

00:01:13.800 and he also hosts his own podcast, the Matt Walker Podcast. I've been wanting to speak to Matt for

00:01:20.980 quite some time because, as you'll hear, I've been increasingly worried about the quality of my own

00:01:25.520 sleep. I'm late to the party here, but now I'm convinced of the importance of sleeping well most

00:01:33.200 nights. And Matt and I get into all the details here about the nature and importance of sleep.

00:01:40.080 We discuss sleep and consciousness, the stages of sleep, sleep regularity, light and temperature,

00:01:47.820 the evolutionary origins of sleep, the generally doomed attempt to reduce one's need for sleep,

00:01:55.920 the connection between deficiencies in sleep and all-cause mortality, Alzheimer's disease, diabetes,

00:02:04.540 obesity, and heart disease, the role that sleep plays in learning and memory and mental health,

00:02:10.400 heart rate variability, REM sleep behavior disorder, and various parasomnias. We discuss lucid dreaming,

00:02:19.680 dreams as a kind of therapy, the connection between meditation and sleep, the various forms of insomnia,

00:02:27.580 and there are practical tips for what to do about them strewn throughout our conversation.

00:02:31.980 We discuss sleep hygiene, caffeine and alcohol, sleep efficiency, bedtime restriction,

00:02:40.920 napping, and finally sleep tracking. And as we'll hear on that final topic of sleep tracking,

00:02:48.700 Matt and I discover that each of us is associated with the company Aura that makes a sleep tracking ring.

00:02:54.920 I am a minor investor in the company, and Matt is its scientific advisor. Neither of us knew about

00:03:01.220 the connection before we started talking, and you'll hear I have a bit of a love-hate relationship

00:03:06.040 with my own Aura ring. It is a remarkable device, but I may have what Matt calls orthosomnia,

00:03:12.940 which is an overabundance of concern about my sleep data. In any case, make of that what you will,

00:03:19.160 and I hope you find this conversation useful as it runs nearly four hours. And now I bring you Matthew

00:03:26.340 Walker. I am here with Matthew Walker. Matt, thanks for joining me. It's a delight and a privilege to be

00:03:39.560 speaking with you, Sam. Thanks for having me. So you've written a book, Why We Sleep, that seems to

00:03:45.860 have gotten into the hands, if not the brains of more or less everyone. And now you have your own

00:03:53.040 podcast, the Matt Walker podcast. And you have been on many, many podcasts that I've noticed talking

00:04:01.040 about the science of sleep and seemingly almost single-handedly making people newly aware of the

00:04:13.060 importance of sleep in their lives, both from the side of, you know, physical health and mental health,

00:04:20.140 emotional regulation, really just across the board when you're talking about human well-being.

00:04:25.960 The difference between good and bad sleep seems paramount. And I must say, I have really neglected

00:04:33.460 sleep as a variable for most of my life. In fact, I think I was early in life toyed with the, you know,

00:04:43.140 fairly crazy ideal of limiting sleep so as to boost productivity. And we'll get into all of that. But

00:04:50.880 before we dive into the specific chapters of our conversation here, perhaps you can introduce

00:04:56.700 yourself, your background intellectually and academically, and just tell us how you came to

00:05:03.240 focus on sleep.

00:05:06.440 I wish I could take the compliment of bringing sleep back onto the public awareness map. I stand on the

00:05:13.580 shoulders of many of my colleagues and they are astronomically wonderful. So I try to do my part.

00:05:19.640 In terms of my background, I am a professor of neuroscience and psychology at the University

00:05:26.100 of California, Berkeley in America. And I've really tried to dedicate myself to understanding the

00:05:35.420 question of why we sleep for the past 20 years. I think like most people, I am an accidental sleep

00:05:42.800 researcher. I often think, you know, when kids are young and the teacher says, tell me what

00:05:49.620 you would like to be when you grow up. No one's shooting their hand up in the classroom and

00:05:54.320 saying, I desperately want to be a sleep researcher.

00:05:58.020 Yeah. And I can attest that when I started my neuroscience PhD, someone from a sleep lab,

00:06:04.620 I forget who, tried to recruit me to their lab. And I thought, why would I want to study sleep?

00:06:11.800 I had no interest at that point. And, you know, now I feel some chagrin over that dismissal because

00:06:19.940 it is increasingly fascinating. And as I said, consequential.

00:06:24.280 And in some ways, I, you know, I don't blame you. Maybe at the time, certainly even 20 years ago,

00:06:28.860 one could argue it's almost academic suicide to suggest that you want to become a sleep researcher

00:06:34.600 and not necessarily truthful, but some would argue that it was almost a charlatan science

00:06:41.480 to begin with. And of course it is. It's the most bizarre, strange, illogical, irrational,

00:06:48.620 from an evolutionary perspective, idiotic thing that an organism can do. And you're going to

00:06:55.120 leverage an entire academic career on that platform. Good luck and good night would be the,

00:07:00.720 I think, the tagline. But I was studying for my PhD, people with different forms of dementia.

00:07:09.280 And I was using brainwave patterns to try and differentially diagnose them very early on in

00:07:14.060 the course of dementia. And I was failing miserably, couldn't get any good results.

00:07:19.480 And one weekend, I had this little igloo of journals that I would retreat to, which tells you

00:07:25.000 everything about my social life. And I started to learn that some of those dementias

00:07:30.540 would eat away at sleep centers and other forms of the dementias would not, because there are many

00:07:35.820 different forms of dementia. So I realized I was measuring my patients at the wrong time,

00:07:39.960 which was when they were awake, and I should be measuring them when they were asleep.

00:07:44.100 I started doing that. I got some fantastic results. And at that point, I started to ask the question,

00:07:51.520 I wonder if these sleep disruptions and impairments are not a consequence of the dementia.

00:07:58.780 They're not a symptom of the dementia. Maybe they are a cause of the dementia. But I realized 20 years

00:08:05.320 ago, no one could answer a very fundamental question, which was, why do we sleep? And I think

00:08:10.860 the crass answer at that time was that we sleep to cure sleepiness, which is the fatuous equivalent of

00:08:18.660 saying, I eat to cure hunger. It tells you nothing about the unique benefits.

00:08:23.240 But then I started to explore this thing called sleep, and I fell absolutely in love with it. And to

00:08:33.220 this day, 20 years on, I still think it is the most beguiling thing in science. It is a love affair that's

00:08:39.840 not left me for all of those decades. And I remain an amorous partner to its wonderful gifts, both

00:08:48.520 nightly as a practice and also from an intellectual and academic and research perspective. Does that

00:08:55.380 give some background?

00:08:56.000 Yeah, yeah. Well, if I can follow your romantic analogy here, sleep is a fairly coy mistress for

00:09:04.300 many of us. And this, you know, speaking personally, this has always been not even on the back burner for

00:09:13.780 me as a problem to solve in my life. I just, I've accustomed myself to sleeping badly and just accepting

00:09:23.000 on some level that I sleep badly. And so, encountering your work is fairly arresting to someone in my

00:09:31.440 condition, because the stakes, as we will elucidate here, are incredibly high, given the connection between

00:09:39.500 sleep and health. So, I wanted to, at the outset, address the component of worry here, worry about

00:09:48.720 sleep, because many people listening to us will also recognize in themselves that their sleep is

00:09:55.340 far from ideal. And to add a layer of worry to that is obviously counterproductive when the goal is to

00:10:05.560 make it easier to sleep soundly and on some better schedule in general. So, can you address this effect

00:10:14.420 that our conversation is likely to have, especially when we're talking about possible links between

00:10:20.900 poor sleep and dementia and, you know, all the rest? It's just, it's very easy to begin to treat

00:10:28.440 this as some kind of medical emergency in the offing. What do you have to say as a, by way of

00:10:35.220 guidance or caution on that point?

00:10:37.660 In some ways, it's a rock and a hard place that I've found myself in. And this is something that

00:10:43.920 I've learned since publishing the book. And I think it's something that I've corrected in my

00:10:48.600 communication to the public. As I was writing the book at the time, at least within the public sphere,

00:10:54.300 as you mentioned, sleep was the neglected step system in the health conversation of today. And it

00:11:00.380 was that way. And I was so familiar, as all of my colleagues were, with the disease and the sickness

00:11:07.500 and the suffering that was happening because of this sleep deficiency that was so pernicious throughout

00:11:14.120 most first world nations, that I wanted to try to, no pun intended for either this podcast or the

00:11:24.060 topic, but sort of wake people up to the fact of the importance of sleep. And I think that in my

00:11:30.300 communications and maybe even in segments of the book, I was perhaps heavy handed. And I had neglected

00:11:38.580 to recognize the concern for the sleep anxious and those who are having sleep difficulty. And I've

00:11:46.560 since become so much more sensitive to that. And I can't deny the science. I can't not tell you

00:11:56.920 about the links between insufficient sleep and Alzheimer's disease, obesity, diabetes, cardiovascular

00:12:04.580 disease, depression, anxiety, even suicide, some forms of cancer. But I also don't want people to

00:12:12.520 become overly anxious. But how do you do that? How do you find that sweet spot? And so for me, it's

00:12:20.540 been a real lesson and a lesson also because I am no poster child for sleep. I have had my battles and

00:12:27.560 I did not mention them in the book. And I think I should have. I'm being personally open. I'm a very

00:12:33.200 private person. I've had at least three bouts of insomnia during my lifetime and they were vicious.

00:12:39.740 And just because you know a little about sleep doesn't mean as though you are immune to its

00:12:45.860 vagaries. It is a mistress that can be very fickle. So I think for this podcast, it's important to keep

00:12:54.620 in mind two things. First, everyone has a bad night of sleep. And if you're there at night struggling to fall

00:13:03.480 asleep, don't worry. Even with all of the facts and the science that we will discuss, it's not the worst

00:13:11.020 thing in the world. The second thing is that if you are persistently and continuously chronically

00:13:17.020 struggling to sleep, you don't have to because there are efficacious treatments, many of them

00:13:23.860 non-pharmacological, which is great, that can help course correct. In fact, even in older adults where

00:13:31.540 you think there is no hope at all for a solid night of sleep, those therapies, many of them seem to be

00:13:38.160 beneficial to restoring some degree of good sleep. So you don't have to suffer in the nighttime silence

00:13:44.820 that there is benefit there. I think that that's perhaps the best way to approach it with sensitivity,

00:13:51.960 compassion, understanding, but truthfulness to the science. You know, I wouldn't want to make people

00:13:59.240 nervous about, you know, eating so precisely that it doesn't change their blood sugar, set them on a path

00:14:06.900 towards, you know, pre-diabetes or type 2 diabetes. And where you become so obsessive and anxious that

00:14:14.520 food and the joy and pleasures of eating start to fail. I also don't want to do that with sleep,

00:14:21.700 but I equally don't want to tell you that it's fine just to eat a pint of ice cream every night and that

00:14:27.140 your blood sugar won't suffer. I'll tell you about that science too. Yeah, yeah. Well, so that's great

00:14:33.280 by way of introduction. And we will get into all of the aspects here, including all of the practical

00:14:39.200 recommendations you have for improving sleep and bypassing any perverse cul-de-sac of worry about

00:14:47.260 sleep that can get in the way of that project. So let's just begin. Let's jump into our first chapter

00:14:53.480 here on what sleep is, even before answering the question that is the title of your book about why we

00:15:01.500 sleep. What is sleep? From a functional perspective, I think the headline statement you could argue is

00:15:09.300 that sleep physiologically, at least, is perhaps the single most effective thing that we can do every

00:15:14.500 day to reset the health of our brain and our body. And that's not to dismiss food or nutrition or

00:15:22.280 exercise. But if you were to take you, Sam Harris, and I were to deprive you of food for 24 hours,

00:15:30.300 deprive you of water for 24 hours, deprive you of physical activity for 24 hours, or deprive you of

00:15:35.960 sleep for 24 hours, and I were to look across your brain and your body and see which one demonstrates

00:15:41.700 the more demonstrable impairment, by a very large margin, it's sleep. But I don't want to sort of do

00:15:48.520 that Coke, Pepsi, Dr. Pepper, I'm still missing one, I can't think of it, challenge. So you could ask

00:15:55.620 from a functional perspective what sleep is. You can also ask what is sleep as a process that unfolds

00:16:03.340 across the night in terms of its architecture. And then you can also ask and debate what is sleep as a

00:16:10.520 conscious state versus a non-conscious state. And so I'm happy to maybe speak about how sleep

00:16:18.120 unfolds, since that may be the logical entry point, or just go straight into how we can noodle and

00:16:24.420 wrestle with the idea of it being a conscious versus non-conscious state, which can get us into

00:16:28.420 tautological waters. But you tell me which of those two perhaps would be best to start with or fruitful

00:16:34.120 for you. Yeah, well, the question of whether it's conscious is, and I know I've spoken about this

00:16:40.940 elsewhere, is very difficult to resolve just because it's difficult to discriminate an interruption

00:16:47.840 in consciousness from a mere failure of memory. So for instance, dreams are routinely conscious,

00:16:55.020 but it's also possible to have dreams and not recall them at all. And then one could wonder

00:17:00.800 whether those dreams, you know, whether those stages of REM sleep were actually associated with

00:17:06.020 conscious dreaming. And one could wonder that the state of deep sleep is also a state of conscious

00:17:13.540 enjoyment of something quite formless and profound, but there's just no memory of it. And so we read it

00:17:22.800 as just a loss of experience for that period. And so I don't know how we would, I mean, I'm happy to hear

00:17:27.640 anything you think on that topic, but I'm unaware of anything that would resolve that for us.

00:17:33.320 I think it's a very elegant point, which is we rely for that question in part subjectively from the

00:17:42.460 sleeper themselves, a report of whether or not they were experiencing anything going through their mind

00:17:48.720 just before we woke them up and said, you know, what were you having as an experience? And that suffers from

00:17:56.180 the failures of memory, which we know happen. Just because you don't remember your dreams doesn't

00:18:02.380 mean that you weren't dreaming. I think one way that you can get closer, but we will still fail

00:18:07.700 is to split that question apart on the basis of perception, which is to say, depending on your,

00:18:19.200 I mean, behaviorally, the way that we define sleep in other species where we can't, for example,

00:18:23.660 stick electrodes on them is as a condition in which the organism stops responding to the outside world,

00:18:30.820 which is about perception. Does this mean that we are not conscious during sleep because we typically

00:18:36.240 stop responding to the outside world in all stages of sleep? And that depends on your definition of

00:18:42.200 consciousness, but we stop interacting with, and for the most part, perceiving the outside world,

00:18:49.720 which some would argue is a loss of consciousness, or at least a shift towards non-consciousness.

00:18:56.480 But I'd counter argue that we don't entirely stop perceiving the outside world. So for example,

00:19:03.420 I can have electrodes on your head and I can play sounds while you're asleep that don't wake you up.

00:19:09.240 And I can still see that the brain at some level is processing those sounds in a way that is not

00:19:15.360 dissimilar to the way it does when we're awake, consciously perceiving those sounds. We can do

00:19:21.120 fMRI studies and we can play those sounds as you're sleeping in the MRI scanner. It's hard to believe

00:19:27.180 that people can, but they do sleep in the scanner. And you can see that there are different ways of

00:19:32.480 perception. There's a great study that looked at new mothers. And what they found was that when they

00:19:38.300 played the cry of that infant versus another sound, even though they remained asleep, it was a very

00:19:45.680 different network, a salience network activated in response to the child of that mother versus

00:19:52.560 another sound of equal volume, et cetera. So there's definitely some degree of processing and

00:19:57.700 discriminatory processing, but I still don't think it's the same non-conscious state as anesthesia,

00:20:06.740 meaning that there is still some degree of perception of the outside world during deep

00:20:12.440 sleep. In other words, what we call extraception, the ability to focus or sense the outside world.

00:20:18.040 Well, there's got to be just based on the fact that you can wake somebody up from deep sleep,

00:20:23.840 you know, so that that's got to get in somehow.

00:20:25.900 That's exactly it. Yeah. Yeah. I think you, you exactly predicted where that conversation was going,

00:20:31.120 which is that no matter what stage you're in, sleep at least is a condition in which it is

00:20:36.700 environmentally reversible. For example, if a sound is loud enough or if someone were to pinch your

00:20:42.220 skin hard enough, which would be a desperately cruel thing, wouldn't it, to do when someone's

00:20:48.120 asleep, you would wake up from sleep, which is to say that in sleep, we are unresponsive,

00:20:53.780 but that state of unresponsivity is reversible. Now that's not true of anesthesia or death,

00:21:00.440 for as best we can tell. So I think it's very hard to argue then that we don't have a very

00:21:07.980 substantive yet qualitatively different form of consciousness when we dream, especially during

00:21:12.300 when we go into REM sleep dreaming. So I think we can get a little bit closer to a dissection of

00:21:17.940 what do we think of as the state of conscious processing during sleep. But I still feel as though

00:21:25.700 I don't see data that can really solidly give us one argument in either favor, conscious,

00:21:33.360 non-conscious state. Yeah. I would just add here that conversely, there are states of meditation or

00:21:40.820 drug intoxication where someone is also totally unresponsive to the outside world, but all too

00:21:48.580 conscious of something, right? I mean, in terms of their subjective report, once they come back from

00:21:54.560 those experiences. So there's kind of a double dissociation here. So I think responsiveness to

00:22:00.500 stimuli isn't the cut we need. We need, we obviously need the neural correlate of consciousness where we

00:22:07.340 can just scan your brain and say, you know, by some methodology and say, okay, this is the footprint

00:22:12.720 of consciousness in the human brain. And it winks out in this condition, let's say general anesthesia,

00:22:19.780 and it's attenuated to this degree in this stage of sleep. But unfortunately, we don't have that yet.

00:22:26.460 And I think there are conceptual and operational limits to our getting it. Again, the role of self-report

00:22:34.800 is always potentially confounding and seditious here because you can just, you know, we just need a

00:22:42.360 sufficient cohort of people who are reporting things that occurred in the chapter that we're deeming

00:22:50.220 to be unconscious. And, you know, either we're going to think they're delusional or they're lying or

00:22:57.320 they're in some other way wrong, or we're going to, or that's going to erode our confidence that

00:23:03.440 really the lights are out during that epoch.

00:23:06.180 I think self-report speaking about fickle mistresses is so prone to all of those errors.

00:23:13.120 Okay. So with that caveat in mind, let's launch into, it would be good to just give us the structure

00:23:19.580 of sleep here in human beings. You can say anything else you want about other animals, but what is sleep

00:23:27.200 for people?

00:23:29.040 Sleep, at least in human beings and in fact, in all mammalian species,

00:23:35.140 as long as they are land dwelling, there's a caveat there too, is broadly separated into two

00:23:40.620 main types. On the one hand, we have non-rapid eye movement sleep or non-REM sleep for short.

00:23:47.480 And on the other hand, we have rapid eye movement sleep or REM sleep. I often want to make people

00:23:54.400 clear on the fact that that's named not after the popular 1990s Michael Stipe pop band, but because

00:24:01.000 of these bizarre horizontal shuttling movements that occurred during this stage of sleep, that's where

00:24:07.280 it gets its definitional name from. And coming back to non-REM sleep, which I always feel sorry for,

00:24:14.940 by the way, isn't it sad to be defined by something that you're not? You are not REM sleep.

00:24:23.400 Well, I guess in this case, you're deep and light.

00:24:26.740 That's correct. So non-REM sleep is then further subdivided into four separate stages, increasing in

00:24:35.700 their depth of sleep. So stages one and two are what we would consider, or your sleep tracker will

00:24:40.660 probably try to tell you, are the light stages of non-REM sleep. Whereas stages three and four,

00:24:46.600 that's the really deep non-REM sleep. And REM sleep then is the stage in which we principally dream,

00:24:56.660 depending on your definition. Dreaming isn't exclusive to REM sleep, but for what most people

00:25:02.540 would say in the lay public, this is dreaming, what they're really referring to are the bizarre,

00:25:10.600 narrative, hallucinogenic, emotional, memory-laden experiences that come from this thing called REM

00:25:19.240 sleep. So those two types of sleep, non-REM and REM, will play out effectively in a battle for brain

00:25:27.820 domination throughout the night. And that cerebral war between non-REM and REM, in humans at least,

00:25:35.540 and it's different for different species, will last about 90 minutes. And that creates,

00:25:40.980 for the average adult, a prototypical 90-minute cycle where you go into non-REM sleep and then

00:25:46.760 you go into REM sleep. But what changes, however, is the ratio of non-REM to REM within those 90-minute

00:25:55.820 cycles as you move across the night. So in other words, in the first half of the night, the majority of

00:26:02.720 those 90-minute cycles are going to be comprised of lots of non-REM sleep, particularly deep non-REM

00:26:09.440 sleep. But as you push through to the second half of the night, that sort of seesaw balance shifts

00:26:16.220 over and those 90-minute cycles are comprised of much more rapid eye movement sleep and very little

00:26:23.320 deep sleep. And that has some consequences that we can also talk about. But I would probably mention

00:26:33.140 also every one of those stages of sleep, or almost all of those stages of sleep, we have now learned

00:26:40.460 are important. There is no one more important stage of sleep than the other. Now you can argue,

00:26:47.820 well, what are you talking about importance? Are you talking about mortality risk and death? And we can

00:26:51.820 use that as a filter to debate that as well. But overall, different stages of sleep provide

00:26:58.200 different functions for the brain and the body at different times of night. So we need all of those

00:27:04.640 stages. And is it true that we generally wake up, however briefly and indiscernible, after each of

00:27:13.640 these 90-minute phases? You get through your REM period and then there's a brief awakening?

00:27:18.800 That's absolutely, yeah. You definitely need to be a sleep researcher, take a sabbatical and...

00:27:25.460 Build me a time machine and I'll go back and have the conversation differently.

00:27:29.800 So we do know that usually at the end of every one of those 90-minute sleep cycles, at the end of

00:27:36.980 each of those REM phases, there is a brief termination of sleep where we wake up. And in part,

00:27:43.920 we think that that's perhaps because of the need to maneuver the body and change the body's

00:27:51.360 position. And so we have these brief awakenings. They're usually so brief that most of us don't

00:27:56.620 recall them. They're not imprinted in memory. But everyone will typically have a brief awakening and

00:28:02.760 then a movement episode after where they shift position.

00:28:06.060 Right. And we'll talk about sleep tracking and the tools that are available to do that

00:28:12.560 personally beyond going into a sleep lab and getting totally hooked up. But viewing these stages

00:28:20.980 in their totality, you've said that each is indispensable, but it does seem, at least in the

00:28:29.760 way one communicates the imperative to get all of these stages. Most of us are not deficient in

00:28:37.280 the stages of light sleep. And it's really the stages of REM and deep sleep that are marketed as

00:28:44.680 truly restorative, right? And those are the areas of real deficiency. I mean, so for instance,

00:28:50.560 if someone was sleeping six hours, but they got very long epochs of deep sleep and REM sleep,

00:29:02.140 would that strike you as a much healthier profile than someone sleeping six hours,

00:29:07.240 but it's mostly devoted to the stages one and two of light sleep?

00:29:12.180 Yes. I think that that's fair to say. We do need stage two as well. We've discovered that stage two

00:29:18.380 non-REM sleep is associated with certain forms of memory and memory processing. And there is a

00:29:25.400 particular electrical feature of stage two non-REM sleep, which continues on into deep non-REM sleep

00:29:32.520 stages three and four called sleep spindles, which are these beautiful little champagne cork

00:29:38.920 synchronous bursts of electrical activity that happen during stage two non-REM sleep. And then

00:29:45.220 stages three and four, they last for about a second and a second and a half. And they seem to be

00:29:50.880 critical for a number of different processes of both the brain and they seem to transact or be at

00:29:56.540 least associated with several benefits for the body. But overall, I would say that it's very difficult

00:30:02.920 to have a night where you're not transitioning because when you go down into deep non-REM sleep,

00:30:09.320 you have to progress through stage two. And when you're coming out of deep non-REM sleep,

00:30:15.220 you have to progress through stage two non-REM sleep again, the lighter form of non-REM sleep

00:30:20.200 before you get up into REM sleep. And so it would probably be rather difficult. You can manipulate

00:30:27.960 conditions in which this can happen, which I won't bore you with, but where you could have the scenario

00:30:33.940 that you described, but for the most part, you're still going to get that stage two non-REM sleep.

00:30:38.980 Yet what you said is correct.

00:30:41.720 Well, this is where I'll seed you with practical questions throughout. But the first that comes to

00:30:48.820 mind here is what are the implications of waking with an alarm clock versus waking with the change

00:30:58.520 in lighting conditions born of sunlight coming through the window? I guess there's the implication

00:31:04.440 of, uh, of using a sleep mask or a, uh, or blackout curtains where you're not getting those

00:31:11.160 environmental light cues. I can imagine, you know, if you're unlucky, your alarm clock rings

00:31:17.300 when you're in, in stage four sleep, say, and you're brought out of that in a less than ideal way.

00:31:24.020 What are, what, what are those, what are those effects and what do you actually recommend if

00:31:28.760 a person's schedule allows for it? What, what do you recommend as a, as a mode of, of waking up in

00:31:34.420 the morning? Unless you are waking up within the first couple of hours of sleep, it's unlikely that

00:31:42.620 your alarm would wake you up in the deep stages of non-REM sleep. That's not true. However, if you

00:31:48.360 take an afternoon nap and that nap lasts a little bit too long and by too long, what I mean is you're

00:31:55.440 going past that sort of 20 to 25 minutes and you're starting to go down into the deep sleep

00:31:59.940 and then your alarm wakes you up. Then you almost have this kind of sleep hangover for the next

00:32:08.880 hour or so. Those naps are terrible. Yeah. Like, you know, it's a change of time zone when you're,

00:32:14.400 when you have terrible jet lag and you decide, okay, there's no way I'm going to make it to

00:32:18.800 the evening. So I got to, I'm going to give myself an hour to sleep here. And that, that waking up

00:32:25.380 from that hour is just about the worst wake up one ever gets. It's pretty grim, isn't it? And it's

00:32:30.720 what we call sleep inertia where you get a state carryover where your brain never typically wakes

00:32:38.280 up from is jolted out of that deep sleep naturalistically from an evolutionary perspective

00:32:42.860 across millions of years. That's not been the case. And so we're not well prepared for

00:32:48.180 recovering from that assault. And therefore we suffer this terrible sleep inertia.

00:32:54.720 So it's not so likely to happen, but when it does happen, it's grim. It can also happen at night

00:33:01.240 when, for example, you get a phone call and all of a sudden it wakes you up at, you know,

00:33:07.220 2.30 or 1.30 in the morning. And once again, you're jolted out from that deep sleep. And yes,

00:33:14.020 you can answer the phone and you can be somewhat responsive, but it is just grim. You're in this

00:33:20.760 total treacle haze of cognitive dysfunction, and it's all you can do to allow words to tumble in

00:33:28.500 some meaningful way, one foot in front of the other out of your mouth.

00:33:31.420 So that is perhaps a less likely circumstance. What would I suggest? It's difficult because one

00:33:41.260 of the critical things that people need to do to get their sleep back on track is the simple act of

00:33:47.100 regularity, which is going to bed and waking up at the same time, no matter whether it's the weekday or

00:33:53.060 the weekend. And for that, we often require an alarm clock. And I also advocate for people not

00:34:02.820 just to have an alarm clock in the morning, but why don't we have a to-bed alarm as well as a to-wake

00:34:08.200 alarm? And it's one way to help keep us on schedule and track. I would say, however, that if you study

00:34:16.220 hunter-gatherer tribes whose way of life hasn't really changed for, you know, hundreds if not thousands of

00:34:21.840 years, they don't seem to wake up in an artificial manner. And if you ask them, you know, do you find

00:34:31.240 ways to force yourself to wake up? They find it a perplexing question. Why would you, why would you

00:34:38.460 terminate something that's not yet complete? It's a little bit like saying, why would you go out to

00:34:42.800 your favorite restaurant, order your favorite dish, have two bites of that dish, and then get up and walk

00:34:48.740 out? You would stay until you're full when you are complete with that meal. And why would we wake up

00:34:56.580 when we are not yet full of the sleep that we need? And Mother Nature will take care of that when it's

00:35:02.440 time to wake up and we've had the sleep that we need. We do. So one way some people will ask me,

00:35:08.320 how do I know if I'm getting enough sleep? It's not the ideal way, but one suggestion is to say,

00:35:14.580 if your alarm clock didn't go off in the morning, would you sleep past that alarm? And if the answer

00:35:18.840 is yes, then you're still carrying some degree of a sleep need, which means that by waking up

00:35:24.100 artificially, you're inducing a sleep debt as a consequence. What about the role of light cues in

00:35:30.600 bringing someone out of sleep? It's, we used to think that light perhaps was the trigger of,

00:35:38.280 or one of the facilitating functions for rising people out from sleep in the morning. And again,

00:35:47.760 by looking at those hunter-gatherer tribes, what we found is that that's not really the case. They

00:35:52.000 often typically will wake up a little bit before the dawn. What seems to be the trigger for the arrival

00:36:00.580 of wakefulness and the termination of sleep is more so temperature, both the internal temperature

00:36:06.420 and the ambient temperature rising, because often they will sleep with the environment,

00:36:13.220 with the ambient temperature, unlike many of us in modernity, where we have a controlled temperature.

00:36:21.060 So that's not to suggest that light can't be a facilitator to help you wake up in the morning.

00:36:27.120 And in fact, I will, I have one of these little smart lights next to my bedside,

00:36:32.480 and I program it to try and say, you know, two minutes before the time that you're supposed to

00:36:38.520 wake up, start to bring light into the room. I would say though, that I do have an alarm myself.

00:36:47.500 My alarm is, and we can get into sort of chronotypes and what your preference is, but

00:36:52.520 my alarm is set for around 7.04 in the morning or at 7.04 in the morning. Not because there's anything

00:36:59.260 special or unique. Please don't go rushing out and changing your wake-up time to that.

00:37:04.160 We're not going to have a chapter on numerology here and the significance of even numbers.

00:37:08.080 The only reason I do that is why not just be idiosyncratic? Why would you set it at,

00:37:13.440 you know, 7.05 or 7 or 7.10? Just why not 7.04? That tells you probably everything about me and

00:37:21.820 why I'm desperately unpopular. But so, but I usually wake up naturally, I would say about 80%

00:37:28.200 of the time I wake up naturally before my alarm clock. So I think there are, one of the worries

00:37:33.980 that people have when I tell them to do the experiment, if you have the luxury and the

00:37:37.500 schedule flexibility to do it, stop your alarm and just sleep in the way that you are, your body

00:37:45.660 wants to sleep. The greatest worry is that, my goodness, I normally wake up at 7 and I'll probably

00:37:52.100 wake up at 9 o'clock in the morning is the first concern. Now, that may be true to begin with for

00:37:58.200 the first few days because you're probably trying to sleep back a debt that you've amassed chronically

00:38:02.900 over weeks, if not months or years. And the second problem is that when people sleep long,

00:38:10.720 they wake up and once again, they have that strange sleep hangover effect where if they get

00:38:15.840 nine hours of sleep, they feel worse than when they get seven hours of sleep. That is typically

00:38:21.200 because you are in the phase of paying back the debt. And if you let that experiment play out for

00:38:27.380 another week, you wash away that sort of pressure to sleep. Now, we can speak about sleep debt and

00:38:33.320 whether you can ever truly pay back the bank or not. But that goes away with time. It's sort of like

00:38:40.380 detoxing from a drug. At first, it's brutal and you have all of these side effects and you have a

00:38:46.680 withdrawal syndrome. And in some ways, that's the withdrawal syndrome where you start sleeping longer.

00:38:51.880 That settles down. It's like a Richter shot and then it finds a sweet spot. And gradually,

00:38:57.320 you will actually acquiesce to your typical sleep need and your sleep profile. Most people don't have

00:39:02.080 the luxury to do that. So light can be helpful. Temperature is one. I also have one of those smart

00:39:08.500 home thermostats. And temperature is critical for sleep. We need to ironically warm up to cool down

00:39:19.220 to fall asleep. And then we need to stay cool to stay asleep. And finally, we need to warm up to wake

00:39:27.760 up. And so you can create a bespoke tailored temperature profile for your night of sleep that

00:39:35.800 can help to some degree. Now, of course, you're under the sheets and the ambient has some role to

00:39:41.640 play, but it's also altered by what's going on locally underneath the sheets too. So you can't

00:39:48.120 control it exquisitely. And that's where smart mattresses are coming in to try and take that out

00:39:52.460 the equation. So those are some of the ways that you can play around with sleep. I do like the idea

00:39:58.480 if you are particularly, if you are a night owl, and you struggle to wake up at the time that society

00:40:07.800 forces you to, which is not in synchrony with your morningness or eveningness preference,

00:40:14.320 you can use light in the morning. But then you can reverse that trick in the evening where you try to

00:40:20.960 ensconce yourself with as much dim light and darkness to help you try to get to bed a little bit

00:40:28.020 earlier. So it's not as though light should be dismissed and, you know, blocking devices,

00:40:34.120 blackout curtains, eye masks, earplugs, sound is another pollution that will disrupt your sleep.

00:40:40.640 I will typically use all of those. I have blackout curtains, I have an eye mask, and then I have

00:40:45.960 earplugs. I think I'm starting to sound like the Woody Allen neurotic of the sleep world, but that's

00:40:52.280 just me.

00:40:52.620 Yes. Well, all we need is one picture of this setup and to completely discredit you as a expert on sleep.

00:40:58.020 Oh, I've been so discredited by lots of different things, but that would, I think, seal the deal.

00:41:02.800 Okay, so let's transition to the question of why we sleep. I think there's probably no real boundary

00:41:10.760 between what sleep is and why we do it conceptually here, at least in places, because part of the story

00:41:18.360 here is the evolutionary question of just why sleep is a thing and how it came to be that animals like

00:41:27.440 ourselves dedicate so much of their lives to this state that seems fairly pointless and even dangerous.

00:41:35.600 I mean, this is the, you can imagine, in civilization, the danger is less salient, but just imagine

00:41:43.320 how precarious it would be to, you know, go out in the woods where there are bears and perhaps several

00:41:51.420 other species that could consider you a meal and to just take eight hours of darkness to be unconscious

00:42:01.120 for. I guess there's a potential evolutionary answer there in that the one thing you're not doing

00:42:07.780 when you're sleeping is stumbling around in the dark where you're not very good at seeing and

00:42:14.060 several other things can see you better than you can see them, but I'm not sure that's an adequate

00:42:18.900 rationale. So let's begin talking about the origins of sleep as we know them or can

00:42:28.860 hypothesize about them. What do you think about why sleep even exists?

00:42:35.780 So far in every species that we've studied to date, sleep or something that looks very much like

00:42:42.100 it seems to exist. And what that has suggested is that sleep evolved with life itself on this planet

00:42:51.060 and has fought its way through heroically every step along the evolutionary pathway.

00:42:56.980 Let's linger on that point. That's very interesting because you can imagine the adaptive benefits that

00:43:03.960 would generally accrue to any species that could just get over its need for sleep. I mean,

00:43:10.860 there would have been, you would think, a selective pressure in the direction of completely erasing

00:43:16.020 sleep. So it suggests that it's rather hard to do.

00:43:19.740 I think it's a beautiful way of thinking about it because from an evolutionary perspective,

00:43:25.280 just as you noted, it is the most idiotic of all things. Firstly, when you're asleep, you're not

00:43:30.540 eating, you're not foraging for food, you're not finding a mate, you're not reproducing, you're not

00:43:35.600 caring for your young. And worst of all, as you noted, you're vulnerable to predation. So on any one

00:43:41.140 of those grounds, but especially all of them as a collective, sleep should have been strongly

00:43:46.920 selected against during the course of evolution. And it's once been said that if sleep doesn't

00:43:53.840 serve an absolutely vital function, then it's the biggest mistake the evolutionary process ever made.

00:43:59.280 And what we've now since learned is that Mother Nature didn't make a spectacular blunder in creating

00:44:04.260 this thing called sleep. But even very old evolutionary, you know, species like earthworms,

00:44:11.520 for example, seem to have periods of, it's called lethargic, or essentially a sleep-like state.

00:44:18.940 You know, this takes sleep back millions of years. Even some bacteria that seem to live at least

00:44:25.220 several days, they will have an active phase and a passive phase, perhaps the precursor to sleep.

00:44:32.280 So you're right, you could well imagine why, if some species had understood a way to circumnavigate

00:44:41.260 its way around the essential need for sleep, it would have dominated for lots of different reasons,

00:44:46.880 at least within its species category. The fact that we haven't seen that yet argues that sleep must

00:44:53.600 be fundamental at the most basic of biological levels. And it's one of the reasons why when people

00:44:59.880 will say to me, well, look, can't you, you know, if you're a doctor training, I think we learned to

00:45:06.640 overcome our need for sleep. We learned to tolerate and deal with insufficient sleep, and you can do

00:45:13.220 that. If you could, trust me, I think Mother, you know, there's some degree of hubris there, which is

00:45:18.820 Mother Nature, if she could have even halved the amount of time that you are vulnerable to all of

00:45:25.860 those vicissitudes of sleep. She certainly would have. And the fact that it's been preserved tells

00:45:34.200 you that doesn't seem to be possible. And within the lifespan, we think that we can come along and

00:45:38.940 within a 10-year training of a career, we could overcome it. It's unlikely to be the case.

00:45:44.480 Actually, we might punctuate this part of the conversation with the cases of various people

00:45:51.460 who, at least by their own testimony, have gone a fair way toward overcoming their personal need

00:45:59.760 for sleep. I think it was Winston Churchill who, during the war years, was sleeping the last 10

00:46:05.960 minutes of every hour or something like that. I don't know if that's apocryphal, but what do we

00:46:11.340 know about anyone's successfully titrating their sleep down to something like a minimum? I'm sure

00:46:21.340 there are genotypes here that we may know something about where people just require less sleep than

00:46:26.360 is normal. But actually, I once had a doctor who claimed to sleep no more than three and a half hours

00:46:34.920 a night. And, you know, whether he was, again, this is before the age of sleep tracking, so he could

00:46:42.320 have been delusional. But what do we know about people who sleep much less than you would recommend?

00:46:50.220 Firstly, from an epidemiological or population-based perspective, which is simply associational,

00:46:55.240 using that sweet spot that we recommend, which is somewhere between seven to nine hours a night for

00:46:59.960 the average adult, once you start to get less than that, the shorter your sleep, the shorter your

00:47:05.380 life. That short sleep predicts or cause mortality. Are there people in history who have claimed to be

00:47:12.620 short sleepers? There are. And Churchill was one. Edison was another, although Edison was a habitual

00:47:19.200 napper during the day, and he used naps and sleep as a creative tool. Then you have Thatcher,

00:47:25.560 Margaret Thatcher, you have Ronald Reagan. Just named two people who ended their lives with

00:47:32.920 Alzheimer's. So that's not a great commercial for their strategies.

00:47:35.960 Exactly where I was going. Yeah. Yeah. You know, they seemed, if on the face of it,

00:47:41.700 to make it through until the, you know, 50s or even 60s. My goodness, there is evidential proof that you

00:47:49.040 can sleep what they claimed to be sleeping, which is four hours a night, and get away with it. And

00:47:54.860 ultimately, what we learned is that one way or another, sleep deficiency seems to get its hux into

00:48:02.940 you, that the elastic band of sleep deprivation can stretch only so far before it snaps. And

00:48:08.760 tragically, for both of those individuals, Thatcher and Reagan, they succumb to the disease of Alzheimer's.

00:48:17.080 And we now know that there are- I now realize we have several files open, but

00:48:20.300 each of these seems important. So on that point, how do we disentangle association and causation here?

00:48:29.380 Because couldn't it also be true that one of the early symptoms of Alzheimer's or, you know,

00:48:34.860 being at a special risk for it is to have one's apparent ability to sleep diminish over the course

00:48:41.880 of one's life, even maybe starting as early as one's 30s or 40s?

00:48:46.780 Yeah, so we can go, Alzheimer's disease is actually a great example. It's probably been,

00:48:52.000 I think, one of the most exciting areas of sleep research in terms of discoveries in the past 10

00:48:57.680 or even five years. We started with just those epidemiological associations, which are simply

00:49:03.260 that, the correlation, the not causation. And what that told us is that people who were

00:49:07.900 if you'd like to continue listening to this conversation, you'll need to subscribe at

00:49:16.020 samharris.org. Once you do, you'll get access to all full-length episodes of the Making Sense

00:49:20.680 podcast, along with other subscriber-only content, including bonus episodes and AMAs and the

00:49:26.960 conversations I've been having on the Waking Up app. The Making Sense podcast is ad-free and relies

00:49:32.260 entirely on listener support, and you can subscribe now at samharris.org.

00:49:37.900 Thank you.

Making Sense - Sam Harris - November 10, 2021

#267 — The Kingdom of Sleep

Episode Stats

Length

Words per Minute

Word Count

Sentence Count

Misogynist Sentences

Summary

Transcript