Book review: Why we sleep

post by ricraz · 2018-09-19T22:36:19.608Z · score: 52 (25 votes) · LW · GW · 15 comments

This is a link post for

I read this book (by a sleep scientist called Matthew Walker) because I knew that it would tell me to sleep more, and I hoped it would cite enough scary statistics that I'd be likely to actually follow through. Well, it worked - I'm keeping a copy on my bedside table for the foreseeable future, just as a reminder. In addition to the exhortations to get more sleep, it contains a variety of other interesting and important facts about sleep.

What is sleep?

What's it good for?

The evolution of sleep

How to sleep better

As you can probably tell from the above, Walker is very much a cheerleader for sleep. This does bias him in some noticeable ways - e.g. his overt scorn towards coffee. He also blurs causation and correlation at some points throughout the book, so I'd be surprised if all of the deleterious effects mentioned above are as significant as he claims. But the overall picture is stark enough that I'm now very worried about the ongoing sleep loss epidemic.


Comments sorted by top scores.

comment by habryka (habryka4) · 2018-09-19T23:18:48.949Z · score: 28 (11 votes) · LW · GW

I think I would love to see an epistemic spot-check in the style of AcesoUnderGlass for this book. If anyone is up for it, I am happy to put out a $50 bounty for anyone writing up such a post and post it to LessWrong.(Independent of quality, as long as it's not completely egregious. If more than one person ends up writing one, then I am happy to pay out the price up to three times)

comment by yagudin · 2018-09-24T11:00:06.549Z · score: 11 (2 votes) · LW · GW

The most in-depth, but a bit outdated (c. 2012) article on sleep is written by Piotr Wozniak, whom you might know as a pioneer of spaced repetition software. The article is ~300 pages long. It includes summary & myths sections which are a bit longer than this post.

comment by jmh · 2018-09-24T16:43:55.064Z · score: 3 (2 votes) · LW · GW

Just a side question here about the "a bit outdated (c. 2012)" note. Is that because you think the science/level of knowledge or some other technology related to such studies is changing that quickly?

Both the reference to additional sources and the review were great. Thanks to both you and ricraz.

comment by yagudin · 2018-09-28T15:08:21.429Z · score: 1 (1 votes) · LW · GW

You are welcome! A general concern about the pace of scientific progress.

comment by yagudin · 2018-11-02T22:39:23.984Z · score: 1 (1 votes) · LW · GW

Wikipedia page for 'Cognitive behavioral therapy for insomnia' is a great source of useful sleep related habits.

comment by Elo · 2018-09-19T23:49:54.566Z · score: 5 (2 votes) · LW · GW

Good book. Would recommend.

My major take away is that sleep problems are usually psychological not Physiological. Usually people don't know how to stop a busy mind and sleep. This leads to endless rumination and being held awake by stress. We don't teach the skill and most people just figure it out on their own.

Also extra sleep tips here:

comment by Sherrinford · 2018-09-20T09:56:24.939Z · score: 4 (2 votes) · LW · GW

Thanks for the interesting review summary. Does he say a bit more about the effects of napping / biphasic sleep vs monophasic sleep?

comment by ricraz · 2018-09-20T18:26:31.397Z · score: 1 (1 votes) · LW · GW

I'll check when I have a copy at hand. From memory, he says that the "biphasic" sleep where you wake up in the middle of the night for an hour or so has no scientific backing. I think his main evidence for the health benefits of of afternoon naps is looking at the rates of heart disesase in Mediterranean countries as their sleep patterns shift. This doesn't seem to be particularly rigorous, but I recall the effect size being pretty large.

comment by habryka (habryka4) · 2018-09-19T23:27:35.540Z · score: 4 (2 votes) · LW · GW

Great post! I generally find book reviews and book summaries quite useful, and am happy to see more of them. On the object level, I am vaguely remembering a study that I can't find right now, that added something interesting to the sleep question, which was something like this:

"We had three test-groups, one of which slept normally at about 8 hours a night, one of which slept for 7 hours a night, and one of which slept for 6 hours a night, for a week. The 7 hour group started out with a similar performance to the 8 hour group, but went down to the performance of the 6 hour group after about 4 days. However, the self-assessment of how sleep-deprived the individuals were was quite accurate for the 6 hour group, but didn't identify any worsening aspects of sleep deprivation for the 7 hour group, even after 4 days of testing. This suggests that subjects are quite bad a assessing mild sleep-deprivation, even after prolonged exposure."

I wonder whether the book made any reference to that study, since I've been looking for a while, and if anyone can find it, than I do remember it being a significant update on how much I trust myself to assess how much sleep I need.

comment by Oren Milman (oren-milman) · 2019-04-08T05:26:54.247Z · score: 1 (1 votes) · LW · GW

In chapter 7 (Too Extreme for the Guinness Book of World Records) Walker mentions a research by David Dinges which seems kind of similar to what you described. I didn't find a reference in the book, but I found this highly cited paper, which seems to me like the one he was referring:

The Cumulative Cost of Additional Wakefulness: Dose-Response Effects on Neurobehavioral Functions and Sleep Physiology From Chronic Sleep Restriction and Total Sleep Deprivation

Walker mentions a research by Gregory Belenky with almost identical results that was published around the same time. I found a highly cited paper which seems to me like the one he was referring:

Patterns of performance degradation and restoration during sleep restriction and subsequent recovery: a sleep dose‐response study


Walker has a sub-title "you do not know how sleep-deprived you are when you are sleep-deprived" in chapter 7, so you can guess what the researches above found.

comment by habryka (habryka4) · 2019-04-08T06:16:04.265Z · score: 2 (1 votes) · LW · GW

Excellent, these really do like the studies that I remember reading. Thank you a lot!

I would be glad to send you $10 via PayPal if you want, since I've been looking for these for quite a while.

comment by Oren Milman (oren-milman) · 2019-04-09T11:02:23.131Z · score: 3 (2 votes) · LW · GW

No thanks. Knowing I helped you out is exactly the right reward for me :)

comment by Oren Milman (oren-milman) · 2019-04-08T07:52:02.732Z · score: 3 (2 votes) · LW · GW

Thanks for the review!

I am only halfway through the book, but I am already firmly resolved to highly prioritize 8 hours of sleep each night.

I would encourage everyone to read the book while experimenting with getting enough sleep consistently (e.g., for a week), and see for yourself whether the short-term benefits are worth it.

Maybe my assessment is not accurate, but it seems to me that "8 hours of sleep each night is extremely beneficial (relative to other stuff you could do)" is not a mainstream belief among Less Wrong readers. (Though I could find examples, e.g., and [LW · GW] .)

I would love to hear that I am completely wrong here, but if I am not, then why is that? Isn't the evidence for the benefits of sleep as overwhelming as the book describes it?

Even after correcting for the book's bias toward favoring sleep (note that I am not qualified to assess the existence/size of such bias), the evidence still seems overwhelming to me.

A few things that I would add to your great summary:


The book gives another argument for the huge benefits of sleep, which I find quite intuitive:

Addressing the question of why we sleep from an evolutionary perspective only compounds the mystery. No matter what vantage point you take, sleep would appear to be the most foolish of biological phenomena. When you are asleep, you cannot gather food. You cannot socialize. You cannot find a mate and reproduce. You cannot nurture or protect your offspring. Worse still, sleep leaves you vulnerable to predation. Sleep is surely one of the most puzzling of all human behaviors.
On any one of these grounds—never mind all of them in combination—there ought to have been a strong evolutionary pressure to prevent the emergence of sleep or anything remotely like it. As one sleep scientist has said, “If sleep does not serve an absolutely vital function, then it is the biggest mistake the evolutionary process has ever made.”
Yet sleep has persisted. Heroically so. Indeed, every species studied to date sleeps.


I think this quote from the book is also of particular interest to Less Wrong readers:

You Do Not Know How Sleep-Deprived You Are When You Are Sleep-Deprived
The third key finding, common to both of these studies, is the one I personally think is the most harmful of all. When participants were asked about their subjective sense of how impaired they were, they consistently underestimated their degree of performance disability.

(As mentioned in another comment, I think the studies that Walker refers to here are these two:

The Cumulative Cost of Additional Wakefulness: Dose-Response Effects on Neurobehavioral Functions and Sleep Physiology From Chronic Sleep Restriction and Total Sleep Deprivation ( )

Patterns of performance degradation and restoration during sleep restriction and subsequent recovery: a sleep dose‐response study ( )


I would summarize the following quote by "You might believe that your current state is close to your optimal state, but you just forgot (or never knew) how much better life could be if you got enough sleep consistently":

Similarly problematic is baseline resetting. With chronic sleep restriction over months or years, an individual will actually acclimate to their impaired performance, lower alertness, and reduced energy levels. That low-level exhaustion becomes their accepted norm, or baseline. Individuals fail to recognize how their perennial state of sleep deficiency has come to compromise their mental aptitude and physical vitality, including the slow accumulation of ill health. A link between the former and latter is rarely made in their mind. Based on epidemiological studies of average sleep time, millions of individuals unwittingly spend years of their life in a sub-optimal state of psychological and physiological functioning, never maximizing their potential of mind or body due to their blind persistence in sleeping too little.
comment by MakoYass · 2018-09-22T23:57:24.211Z · score: 3 (1 votes) · LW · GW
Humans seem to be naturally biphasic: modern hunter-gatherer tribes sleep for 7-8 hours at night, and then nap for 30-60 minutes in the afternoon.

I'm having fun imagining a workplace where this sort of pattern is encouraged. A large part of the difficulty would be dealing with self-destructive work cultures where nobody wants to violate a norm in a way that risks making them look lazy. I'd want to start by having the highest performers try napping after lunch, so that people come to associate it positively with productivity, frame napping as the opposite of lazy, like exercise, it's not work but it's something that hard workers do. But there's a chance the higher-performers are exactly the people who wouldn't benefit from napping- There is such a thing as a short-sleeping gene in humans and a lot of CEOs seem to have it- so starting with them might soil the whole thing.

I'm remembering hearing stories of a lot of workplaces getting nap pods and telling their employees that it is "okay" to nap. I don't think this should be taken seriously. If you don't have enough beds for everyone in the office to sleep, it wont become a norm. It certainly wont become a habit.

I'd want to experiment with assigning a sample of people (or a set of volunteers) to napping every day, that's a design we could take seriously.

comment by SquaredCircle · 2018-09-20T15:51:11.077Z · score: 2 (2 votes) · LW · GW

Thank you for the write up! Matthew Walker was also on The JRE Podcast and has a Talk at Google. Well worth checking out.