Alexey Guzey on sleep: Put this guy on Rogan and give him a Ted talk.

Regular readers of this blog will be familiar with Alexey Guzey as the author of the Nabokovian article from 2019, “Matthew Walker’s “Why We Sleep” Is Riddled with Scientific and Factual Errors.” I don’t know that anything that Guzey writes will top that Kinbotean feat, but I very much enjoyed this new post, Theses on Sleep, where he writes:

In this essay, I [Guzey] question some of the consensus beliefs about sleep, such as the need for at least 7 hours of sleep for adults, harmfulness of acute sleep deprivation, and harmfulness of long-term sleep deprivation and our inability to adapt to it.

It appears that the evidence for all of these beliefs is much weaker than sleep scientists and public health experts want us to believe. In particular, I conclude that it’s plausible that at least acute sleep deprivation is not only not harmful but beneficial in some contexts and that it’s that we are able to adapt to long-term sleep deprivation.

Here’s his key theoretical argument:

Comfortable modern sleep is an unnatural superstimulus. Sleepiness, just like hunger, is normal.

1. Experiencing hunger is normal and does not necessarily imply that you are not eating enough. Never being hungry means you are probably eating too much.

2. Experiencing sleepiness is normal and does not necessarily imply that you are undersleeping. Never being sleepy means you are probably sleeping too much.

Most of us (myself included) eat a lot of junk food and candy if we don’t restrict ourselves. Does this mean that lots of junk food and candy is the “natural” or the “optimal” amount for health?

Obviously, no. Modern junk food and candy are unnatural superstimuli, much tastier and much more abundant than any natural food, so they end up overwhelming our brains with pleasure, especially given that we are bored at work, college, or in high school so much of the day.

What if the only food available to you was junk food and candy?

1. If you don’t eat any, you starve.

2. If you eat just enough to be lean, you’ll keep salivating at the sight of pizzas and ice cream and feel distracted and hungry all the time. Importantly, in this situation, the feeling of hunger does not mean that you should eat more – it’s your brain being overpowered by a superstimulus while being bored.

3. If you eat it as much as you want, you’ll probably eat too much and become fat. And if you eat way too much candy or pizza at once, you’ll be feeling terrible afterwards, however tasty the food was.

Most of us (myself included) sleep 7-9 hours if we don’t have any alarms in the morning and if we get out of bed when we feel like it. Does this mean that 7-9 hours of sleep is the “natural” or the “optimal” amount?

My thesis is: obviously, no. Modern sleep, in its infinite comfort, is an unnatural superstimulus that overwhelms our brains with pleasure and comfort (note: I’m not saying that it’s bad, simply that being in bed today is much more pleasurable than being in “bed” in the past.)

Think about sleep 10,000 years ago. You sleep in a cave, in a hut, or under the sky, with predators and enemy tribes roaming around. You are on a wooden floor, on an animal’s skin, or on the ground. The temperature will probably drop 5-10°C overnight, meaning that if you were comfortable when you were falling asleep, you are going to be freezing when you wake up. Finally, there’s moon shining right at you and all kinds of sounds coming from the forest around you.

In contrast, today: you sleep on your super-comfortable machine-crafted foam of the exact right firmness for you. You are completely safe in your home, protected by thick walls and doors. . . .

It’s an appealing argument, and, as Guzey says, this is only the half of it, since there’s a direct benefit to sleeping less, which is that you get to spend more time awake doing fun things like reading or watching TV or whatever you want, as this is all extra time that you’d otherwise be spending unconscious.

But . . .

I also have some difficulties with Guzey’s argument. Here are three problems I have:

1. Yes, it’s natural in some sense to be hungry and sleepy and, more generally, uncomfortable much of the time. Indeed, I was just thinking about this the other day when standing outside in the cold air but in relative comfort beneath all my layers: when I was a little kid, down jackets didn’t exist. I remember when they came out, when I was about 11, and they were amazing! Before down jackets, we’d be uncomfortably cold all winter, standing in recess or waiting for the bus or whatever. Look: I’m not claiming we were suffering; I had a comfortable middle-class childhood; it’s just that this particular discomfort was part of our lives. It reminds me of reading about old-time people in England who didn’t have instant hot water and would have to take cold baths. Brrrr! I feel so lucky we can take warm showers whenever we want. Anyway, here’s my point: it’s natural to have a baseline level of discomfort—but it’s also natural to not want discomfort. Guzey might call McDonalds, comfy beds, warm showers, and down jackets as unnatural “superstimuli”—but discomfort can be a distraction, no? So I’m not sure I buy his argument. To continue the analogy of hunger: sure, lots of great work has been done by hungry people, but when I’m too hungry I can find it hard to focus. Here’s another analogy: infectious disease. It’s kinda natural to get sick a lot, but when we’re sick it’s hard to do other things.

That said, I have found that it can be easier for me to focus on my work when I stay up late. I don’t think I work very efficiently in the middle of the night, but sometimes it seems that this is really the only way I’ll get a project done. So maybe the discomfort is helping me. Not helping me focus, exactly, but giving me a sense of urgency, a kind of natural deadline that allows me to push through. Contrary to what Guzey writes, I wouldn’t say that staying up late helped “my creative juices start flowing”—I don’t think I’m any more creative in the middle of the night, actually I think I’m less creative at that time—but I sometimes am able to push things through. For me, the bottleneck is not creativity, as I have zillions of ideas every day; the bottleneck is working them through one at a time.

2. Regarding Guzey’s analogy to food: Sure, but lots of people are exposed to modern junk food and don’t get fat! The point is that this superstimulus doesn’t have to be bad. I wear down jackets in the winter, and that works too.

3. I personally find sleep very enjoyable. I guess this is the “superstimulus” thing, but if I put on my economist hat for a moment: if there’s something I like and it doesn’t have negative externalities, then this is a good thing! At least, that should be our first approximation.

Guzey also has some arguments regarding sleep and depression; I have neither intuition nor research experience in that area so I’ll offer no comments either way.

Less sleep, more leisure

To me, Guzey’s strongest argument in favor of sleeping less is opportunity cost. Yes, I find sleep to be very pleasant—but I should be comparing it to the alternative! For example, sleep is more pleasant than working or doing the dishes—but I already have to work and do the dishes. If I sleep an hour a day less, the extra hour I get is all gravy. I can do whatever I want with that hour. The tradeoff is the momentary unpleasantness of the being woken by the alarm clock, versus the extra hour I can spend reading or watching TV or blogging or whatever.

I do think this opportunity cost argument has been undersold. Indeed, the you-can-get-by-on-less-hours-of-sleep gurus often seem to be making a productivity argument: get less sleep and you can get more work done and become a success in life! But I think that’s a weak argument because it’s tying the less-sleep thing to the more-work thing. It’s a terrible sell: sleep less and work more, who wants that? Less sleep = more leisure, that’s a much more attractive argument.

Appendix: He has no trust in sleep scientists

My favorite part of Guzey’s post is this bit, which I’ll repeat in its entirety:

Why do I [Guzey] bother with all of this theorizing? Why do I think I can discover something about sleep that thousands of them couldn’t discover over many decades?

The reason is that I have approximately 0 trust in the integrity of the field of sleep science.

As you might be aware, 2 years ago I wrote a detailed criticism of the book Why We Sleep written by a Professor of Neuroscience at psychology at UC Berkeley, the world’s leading sleep researcher and the most famous expert on sleep, and the founder and director of the Center for Human Sleep Science at UC Berkeley, Matthew Walker.

Here are just a few of biggest issues (there were many more) with the book.

Walker wrote: “Routinely sleeping less than six or seven hours a night demolishes your immune system, more than doubling your risk of cancer”, despite there being no evidence that cancer in general and sleep are related. There are obviously no RCTs on this, and, in fact, there’s not even a correlation between general cancer risk and sleep duration.

Walker falsified a graph from an academic study in the book.

Walker outright fakes data to support his “sleep epidemic” argument. The data on sleep duration Walker presents on the graph below simply does not exist:

Here’s some actual data on sleep duration over time:

By the time my review was published, the book had sold hundreds of thousands if not millions of copies and was praised by the New York Times, The Guardian, and many other highly-respected papers. It was named one of NPR’s favorite books of 2017 while Walker went on a full-blown podcast tour.

Did any sleep scientists voice the concerns they with the book or with Walker? No. They were too busy listening to his keynote at the Cognitive Neuroscience Society 2019 meeting.

Did any sleep scientists voice their concerns after I published my essay detailing its errors and fabrications? No (unless you count people replying to me on Twitter as “voicing a concern”).

Did Walker lose his status in the community, his NIH grants, or any of his appointments? No, no, and no.

I don’t believe that a community of scientists that refuses to police fraud and of which Walker is a foremost representative could be a community of scientists that would produce a trustworthy and dependable body of scientific work.

I just love how direct Guzey is: “Did any sleep scientists voice the concerns they with the book or with Walker? No. They were too busy listening to his keynote at the Cognitive Neuroscience Society 2019 meeting.”

And I agree with his conclusion: “I don’t believe that a community of scientists that refuses to police fraud and of which Walker is a foremost representative could be a community of scientists that would produce a trustworthy and dependable body of scientific work.”

This is not a personal criticism of the community: we all know that whistleblowers are special people, and it would seem silly to consider it a moral failure that Walker’s colleagues in the psychology department at his university don’t, for example, threaten to walk off their jobs en masse unless their employer does something about his scientific misconduct. After all, you don’t see me offering to quit Columbia in protest against our continuing employment of Dr. Oz. There are indeed strong moral arguments for keeping your head down and doing your work rather than spending your time fighting the system. And then there’s the Javert paradox. So, sure, the individuals who do sleep science could be doing great work. But, yeah, if the sleep-science community, as a community, is cool with Walker, not just as an oddball crank or as a fun cheerleader but as a representative of the field . . . yeah, that does seem to be a problem. It doesn’t mean the work is wrong, but I don’t think we “have no choice but to accept that the major conclusions of these studies are true,” as the saying goes.

P.S. Interesting discussion in comments. I appreciate the commenters who agree with Guzey and also the commenters who are completely unimpressed by his arguments. One lesson I take from this is that sleep experiences vary a lot between people, and over the lifespan of individual people, which implies that (a) there’s a limit to how much we can generalize from our own experience, but (b) we can learn something by introspection if we reflect upon the variation we’ve experienced in our lives.

As noted in the above post, I think that Guzey’s opportunity-cost argument is interesting, even if it does not have universal validity; indeed, Guzey himself mentions the example of operating heavy machinery.

Finally, one of my frustrations with Walker is not just that he makes stuff up, but that he’s in a non-discussion framework: it’s all set up as, this is The Truth and nobody’s allowed to question it. That’s just annoying. So Walker has a problem with accepting uncertainty as well as with embracing variation.

For reasons discussed many times on this blog (for example, search on “16”), variation is typically hard to study. So I don’t have any easy answers here. I like how Guzey’s opening up the conversation, and I also appreciate the criticisms of Guzey’s conjectures. It’s too bad the world of Ted / NPR / Rogan / University of California isn’t open in this way.

50 thoughts on “Alexey Guzey on sleep: Put this guy on Rogan and give him a Ted talk.

  1. Completely coincidentally, the NYT today delivered the following thread about sleep

    https://www.nytimes.com/2022/02/12/style/segmented-sleep.html?campaign_id=9&emc=edit_nn_20220212&instance_id=52992&nl=the-morning&regi_id=77532059&segment_id=82522&te=1&user_id=d7e3e90dc8fbbcc2d51df749fc62495f

    “Unbeknown to Ms. Rafea, she had naturally reverted back to a sleep cycle that was believed to be standard in multiple cultures in the late Middle Ages through the early 19th century.”

    “During that time, many people went to sleep around sundown and woke three to four hours later. They socialized, read books, had small meals and tried to conceive children for the next hour or two before going back for a second sleep for another three to four hours. It was only when artificial light was introduced that people began forcing themselves to sleep through the night, said A. Roger Ekirch, a professor of history at Virginia Tech and the author of ‘The Great Sleep Transformation.’”

    “We don’t really know the long-term impacts of segmented sleep because we don’t really have much data on it,” said Matthew Ebben, an associate professor of psychology in clinical neurology at Weill Cornell Medicine and NewYork-Presbyterian.

    Reference is made to
    https://academic.oup.com/sleep/article/39/3/715/2454050

    which says, “I [A. Roger Ekirch] was particularly surprised by the discovery reported by Yetish et al.1 that the members of all three of these equatorial societies did not “regularly awaken for extended periods in the middle of the night.” In short, these individuals did not experience a “bimodal sleep pattern.” The authors conclude, “by extension,” that this pattern was “probably not present before humans migrated into Western Europe. Rather, this pattern may have been a consequence of longer winter nights in higher latitudes.” Not only is this broad inference highly questionable, but significant historical and ethnographic evidence also exists to suggest the prevalence of segmented sleep in preindustrial equatorial cultures.”

    • Paul:

      Not a coincidence at all! It’s well known that the editorial board of the NYT reads this blog in order to decide what are the important stories to write on. Statistical Modeling, Causal Inference, and Social Science is the Matt Drudge of the 2020s.

      Soon we can expect an in-depth NYT series of interviews of Trump voters in diners in Niles, Michigan.

  2. I hate “theoretical argument by analogy”. It’s so easy to support any arbitrary position. By the same argument:

    Experiencing death is normal and does not mean that you are unhealthy. Never dying probably means you are living too much.

    Meanwhile, “opportunity cost” is largest just before a college exam. Studies show that you learn better if you get adequate amounts of sleep. But if you have not yet studied, and your choice is study or sleep, you’ll learn more by studying than by sleeping. I’ve never seen a study that tried to identify the optimal mix, given a limited amount of time to do both.

    • The value of Guzey offering some theoretical arguments by analogy is that it neutralizes the intuitions you have from the *previous* analogies. (And diet, exercise, sleep, and psychology are full of these ‘folk’ theories.) It’s sort of like what you should do if someone quotes a proverb at you, “he who hesitates is lost” – shoot back, “look before you leap”. And then hopefully they will realize that the evidential value of a witty saying is nil. Aporia is better than arrogance.

  3. When I was young I worked summers as a commercial fisherman, and got a lot of experience with being short of sleep. I could stay sort of functional with four solid hours plus naps, but I could function better with more sleep. The same seemed true for the other people on the boats. I think the lack of sleep is part of the reason that fishing is the second most dangerous job (after logging).

    I think the analogy with hunger is not good, at least in my experience. I get hungrier if there is an opportunity to get something good to eat, and less hungry if I know I’m not going to get anything to eat. Sleep does not work like that.

    I agree with Guzey’s point about scientific fields that tolerate nonsense, since I’ve worked in one, but just because Walker’s science is bad doesn’t mean that you can get by on too little sleep.

      • Roy,
        I’ve worked a lot on environmental (or instream) flow assessment, a backwater branch of applied ecology that has long tolerated nonsense, especially a bogus method called the physical habitat simulation system (PHABSIM) or variations of it. The basic problem is that people keep trying to come up with simple solutions to a really hard problem.

  4. It is funny he brings this up:

    1. Experiencing hunger is normal and does not necessarily imply that you are not eating enough. Never being hungry means you are probably eating too much.

    I mentioned on here before that most people today have no idea what hunger actually feels like. What they think of as hunger is actually carbohydrate cravings. Anyone can prove this to themselves with about 2 weeks of a low carb (under ~50 g per day) and high fat diet.

    I’m not saying that is necessarily a healthy diet, or anything like that. Just that you will find out that those “hunger pangs” are not actually hunger. They are a withdrawal symptom.

  5. Guzey seems to be offering a plausibility argument: reasons to elevate a hypothesis into the set of ideas worth considering. This is a far cry from saying you have any sort of compelling evidence that the hypothesis is true; it’s more of an argument for why the hypothesis might be worth investigating.

  6. Alexey’s got it backwards. The scene he’s demonstrating – caves, predators etc. – is not the natural state of things. This is, this where we are now. It just took some time to get to this, to break free from the un-naturalness. Ironically enough, also the state before that, when we were amobeas wandering around was also natural. But that exact period in time that Alexey is describing wasn’t natural.

    A common misunderstanding, though, so I don’t blame him.

  7. I’m having trouble with the Kinbote reference. Guzey seems like the enthusiastic and perhaps obsessed annotator. But Kinbote is perhaps the most unreliable narrator in literature.

  8. Without spending too much time in the weeds here there’s a few things not being considered that are worth mentioning. The first is that even ‘primative’ cultures did not necessarily sleep without comfort, but adapted for example by appointmenting guards to act as lookouts. The second is humans used to be more polyphasic in their sleep and would sleep during the day to some degree. Third lots of people talk about sleep outside the context of circadian rhythms, but they’re not insignificant *especially* in relation to sleep and it makes everything harder to frame.

    Lastly, on the topic of Walker’s fame and the lack of policing. I do question to some degree his fame *within* the sleep field. I am admittedly a no one in this field and not on human side of it, but I had honestly never heard of him. This is not to say my N=1 position is right. But sleep is a small field and he is not the only one giving big talks of the like. Academics don’t think too much of them. Im not sure many sleep scientist would have even read his book to begin within given that is for lay people. In my experience people don’t do that. Surely it’s not good and I’m not defending Walker, but most academics find it not worth the effort to even call out their abusive colleagues ime. Thus, I think this might be more reflective of greater cultural phenomenon in academia.

  9. 《You sleep in a cave, in a hut, or under the sky, with predators and enemy tribes roaming around. You are on a wooden floor, on an animal’s skin, or on the ground. The temperature will probably drop 5-10°C overnight, meaning that if you were comfortable when you were falling asleep, you are going to be freezing when you wake up. Finally, there’s moon shining right at you and all kinds of sounds coming from the forest around you.》

    Have you camped out for a week or two, and not wanted to go back? Or am I just an outlier to be trimmed?

  10. I’ve always thought a simple exploration of the proposed dangers of chronic sleep deprivation would be to regress lifespan against family size. If sleep loss is so deadly, people with multiple kids should have shorter life spans, higher rates of cancer, etc.

    One aspect of pre/non-industrialized sleeping habits not mentioned is that co-sleeping was/is more common, and hence sleep disturbance. I lived in west Africa for a time and women would typically sleep in a single bed with multiple small children. Straw mattress, mosquitoes, multiple kids. Not a recipe for uninterrupted sleep!

  11. “You sleep in a cave, in a hut, or under the sky, with predators and enemy tribes roaming around…”

    ???? Europeans by that time were quite sophisticated, employing an invention called “fire” to keep warm in the winter. Summer temps don’t drop 5-10°C night even in the arctic. Ooo…just a little short on actual reality.

    This goes right along with his n=1 experiment on sleep deprivation and cognition, from which he drew ridiculous conclusions about sleep that his data had not the slightest prayer of supporting. Hardly a plausible replacement even for Walker, who can at least grasp what data actually *could* support his claims, even if has to add some data here or remove some there to make it work.

    • MJ,
      1. You say: ” Europeans by that time were quite sophisticated, employing an invention called “fire” to keep warm in the winter. ” As almost anyone who has ever camped outdoors can attest, it is not generally feasible to use fire to maintain a constant or close to constant temperature at night. Unless tended constantly, a fire will start to die after a while and will emit much less heat. If your sleeping bag isn’t warm enough you will find that the cold wakes you up after a few hours; you then feed the fire and try to go back to sleep. Sleeping in an uninsulated cabin. You can do a lot with heavy blankets but I think it’s the case that 10,000 years ago people in many parts of the world would get cold at night on many nights of the year.
      Also, depending on the outdoor temperature it can take a lot of fuel to keep even a small hut reasonably warm. Here’s a transcript that touches on the amount of fuel needed in Montana, for example https://www.thirteen.org/wnet/frontierhouse/project/series2.html There is a strong incentive to use as little firewood as possible and thus often be rather cold.

      2. You say “Summer temps don’t drop 5-10°C night even in the arctic.” That is an amusingly false statement…amusing because it’s so badly false and obviously false, and also so easily shown to be false. Here https://en.wikipedia.org/wiki/Diurnal_air_temperature_variation is a relevant Wikipedia page. Temperatures drop at night and often don’t stop until they reach the dew point — at that point, condensation of water vapor into water releases sufficient energy to stabilize the temperature. Note the scale on the temperature scale for July: in large parts of the country the temperature drops by more than 10C on during even an _average_ night.

      I have nothing to say about Guzey, who could be very right or very wrong or anywhere in between as far as I know, but I couldn’t let your absurdities pass without comment.

  12. Yeah, a lot of this argument sounds like nonsense to me, sorry. If the goal is thought provoking trolling of the consensus, sure, fine, I guess. If the goal is to generate hypotheses that are plausible, no. I’m in my late 30s. Due to a variety of lifestyle changes over the years I can speak to several very different sleep patterns, and the ones that involve routine undersleep definitely have been a huge drag on productivity. Now it’s true that some people are less sensitive to this and seem to have lower sleep requirements but I know very few people in that category.

  13. Interesting discussion all around. I appreciate the commenters who agree with Guzey and also the commenters who are completely unimpressed by his arguments. One lesson I take from this is that sleep experiences vary a lot between people, and over the lifespan of individual people, which implies that (a) there’s a limit to how much we can generalize from our own experience, but (b) we can learn something by introspection if we reflect upon the variation we’ve experienced in our lives.

    As noted in the above post, I think that Guzey’s opportunity-cost argument is interesting, even if it does not have universal validity; indeed, Guzey himself mentions the example of operating heavy machinery.

    Finally, one of my frustrations with Walker is not just that he makes stuff up, but that he’s in a non-discussion framework: it’s all set up as, this is The Truth and nobody’s allowed to question it. That’s just annoying. So Walker has a problem with accepting uncertainty as well as with embracing variation.

    For reasons discussed many times on this blog (for example, search on “16”), variation is typically hard to study. So I don’t have any easy answers here. I like how Guzey’s opening up the conversation, and I also appreciate the criticisms of Guzey’s conjectures. It’s too bad the world of Ted / NPR / Rogan / University of California isn’t open in this way.

    • “One lesson I take from this is that sleep experiences vary a lot between people”

      HAH! There was a sleep section in a recent Science (last November, maybe), and the complete lack of consideration for individual differences drove me screaming mad. (Another things that irritated was that they’ve basically figured out that cycles of neural activity generate cycles of increased/decreased blood flow, and the associated cycles of increased/decreased blood volume (is probably what) drives CSF flow during sleep. But they couldn’t come up with the word “pump”.)

      A close friend in high school got by on 4 hours or so of sleep and got way more done than I did (he wrote a full-blown Lisp system for our IBM 1130 in 1971/72, I wrote a life program and started a chess program). He became one of the top Comp. Sci. types of our generation. I didn’t.

      • I’ve known several people who were proud that they got by on much much less sleep than most, and they were all quite accomplished. One was a billionaire who came close to giving a huge chunk of money to UCLA for brain research, but that fell apart when they realized what he really wanted was for them to figure out why his brain worked so much “better” than everyone else’s and why everyone else didn’t go without sleep the way he did. The answer was pretty simple – he was Bipolar Type II and spent most of his time hypomanic. He considered his occasional deep crashes into anergy and depression an anomaly, which he kept hidden via a very loyal entourage who covered for him in those times, and considered his manic/hypomanic periods “normal.”

        Anyway, the point about variation between individuals in need for sleep is very clear, and like you I want to scream any advice or study that purports to describe the ideal amount of sleep for most people.

        • To your last point:
          Ditto for the ideal amount of sleep, but if we take baseline measures for each individual, it’s quite easy to isolate the effect of sleep deprivation. However, I believe this whole topic is about hair splitting.
          It is obvious that evolution had enough time to make anything perfect and more efficient, did not get rid of sleep/hibernation.
          Also, in highly cognitive demanding tasks, such as flying where top attention and situation awareness is necessary, tons of very good research has demonstrated what happens not only if one is sleep deprived, but tired or otherwise distracted. Studies utilizing activities that one could do ‘in their sleep’ are all over the place.

  14. This reminds of the famous data-set sleepstudy that Douglas Bates used to use to illustrate how to use linear mixed models. Something pertinent even in that tiny, toy data-set is that the by-subject intercept sd estimate is *larger* than the residual sd estimate (look at the random effects).

    > summary(m)
    Linear mixed model fit by REML [‘lmerMod’]
    Formula: log(Reaction) ~ Days + (1 | Subject)
    Data: sleepstudy

    Scaled residuals:
    Min 1Q Median 3Q Max
    -3.3244 -0.5512 0.0602 0.5520 3.7035

    Random effects:
    Groups Name Variance Std.Dev.
    Subject (Intercept) 0.016414 0.12812
    Residual 0.009435 0.09714
    Number of obs: 180, groups: Subject, 18

    Fixed effects:
    Estimate Std. Error t value
    (Intercept) 5.530065 0.033060 167.27
    Days 0.033668 0.002521 13.36

    Correlation of Fixed Effects:
    (Intr)
    Days -0.343

    This is true even after you add random slopes by subject:

    > summary(m)
    Linear mixed model fit by REML [‘lmerMod’]
    Formula: log(Reaction) ~ Days + (1 + Days | Subject)
    Data: sleepstudy

    Scaled residuals:
    Min 1Q Median 3Q Max
    -3.9149 -0.5044 0.0406 0.5250 4.4564

    Random effects:
    Groups Name Variance Std.Dev. Corr
    Subject (Intercept) 0.0108545 0.10419
    Days 0.0003269 0.01808 -0.05
    Residual 0.0065873 0.08116
    Number of obs: 180, groups: Subject, 18

    Fixed effects:
    Estimate Std. Error t value
    (Intercept) 5.530065 0.027008 204.754
    Days 0.033668 0.004754 7.082

    Correlation of Fixed Effects:
    (Intr)
    Days -0.193

    Maybe Guzey should study statistics a little. As Blastland and Spiegelhalter wrote in the Norm Chronicles, “The mean is an abstraction. The reality is variation.” Especially when the mean comes from n=1. :)

  15. I dunno, a lot of animals sleep a lot.

    I did a bit of Googling starting with “how much do chimpanzees sleep in the wild,” and it gave me numbers pretty close to the contemporary human 8 hours/day. So that makes me a bit skeptical of the idea that somehow this number is an artifact of modern civilization.

  16. I’m unimpressed with Guzey’s approach to “debunking” conventional sleep science wisdom. When he says there’s no evidence that lack of sleep has potential serious consequences, I wonder what world he’s living in. On the subject of anecdotal self reports as evidence of anything, I’ve been a part of several of naturalistic experiments in sleep deprivation. His idea that significantly less sleep is fine, and maybe even optimal, is bullshit. Going from fourth-year med student (fairly normal sleep with semi-frequent late-night study/cram sessions) to internship (months of continuous sleep deprivation, sometimes severe) is a move from a stressful but exhilarating world to a condition of chronic irritability, confusion, and resentment. It induces a state of apathy about anything except the task at hand and the illusory hope for an hour of uninterrupted sleep. It is only through force of will that one functions at all, and the idea of doing something creative or pleasurable isn’t ever a consideration. Every intern seems to have the same experience. Significant sleep deprivation is unhealthy and miserable.

    I moved from medicine to a career where periodic work “crunches” are part of the game, where work weeks go for 40-50 hours to 60-90 hours per week. This is in a technical and artistic field, and I’ve seen again and again at rapid falloff in human performance and efficiency (and especially creativity!) once the workload seriously starts eating into sleep time.

    Finally, I’ve recently become a father. My smart watch informs me that my deep sleep has plummeted for months on end. I didn’t need the watch to tell me though – stumbling around like a zombie and feeling like I’d suddenly aged 5 years was clear enough.

    There’s no doubt that a hard deadline or external pressure can lead to more productivity and sometimes bursts of creativity. But to assume it isn’t the deadline/pressure, but working through the night to meet that goal, is a silly mistake.

    I do think all this is worthy of frank discussion. I hope that the widespread use of sleep tracking smart watches and things like the Oura ring will lead to some thoughtful research on the connection between different sleep patterns and various disruptions and productivity/health. My impression when I was doing neuroscience research was that sleep research didn’t necessarily attract that sharpest researches, in part because the methodology was difficult or squishy. Hopefully that is changing now that widespread quality data on actual sleep in large numbers of people is potentially easily studied.

    • Yeah, I agree. Personally I don’t buy into theories about physical health benefits of precise, specific amounts of sleep. But the mental health benefits are clear to me. When I am sleep deprived, I am an asshole! “Leisure time” while sleep deprived is also often awful. If I’m depriving myself of sleep to doomscroll on twitter, that’s not some kind of opportunity cost I am avoiding, that’s for me actually *worse* than if the sleep time simply disappeared.

      Guzey might argue that being sleep deprived and hungry is the natural state of humanity, and he might even be right about it, but I prefer a life that is not solitary, poor, nasty, brutish, and short.

  17. “that we are able to adapt to long-term sleep deprivation”

    I dunno… has he ever had kids? It takes a while for kids to actually “sleep through the night”, and sleep deprived parents do not operate on all cylinders.

  18. Given that Guzey has zero training or background in sleep medicine why should anyone take any of his “theories” seriously? Additionally, given that Andrew is listed as an advisor for Guzey’s company, how might Andrew’s perspective on this be biased by his role advising Guzey? The military has done an astounding amount of research on the deleterious effects of sleep deprivation on myriad outcomes, but it seems that Guzey is able to wave a wand and ignore countless RCTs (since that is one of his critiques) about how sleep affects all sorts out outcomes.

    • Billy:

      I haven’t done much advising of Guzey. He asked me if I’d be on the advisory board of his organization and I said sure. I thought his work was thoughtful so I wanted to be helpful. I guess if I was paid then maybe I should declare some sort of conflict of interest when linking to him.

      I don’t know anything about the sleep literature—but I don’t really need to know anything about it to see that the Why We Sleep guy was misrepresenting data.

      I don’t doubt that lots of high-quality work has been done on sleep. If you have a particular reference or set of references that would contradict Guzey’s claims, could you please share them in comments? That would be helpful.

      • Alhola, P., & Polo-Kantola, P. (2007). Sleep deprivation: Impact on cognitive performance. Neuropsychiatric Disease and Treatment, 3(5), 553–567.
        https://pubmed.ncbi.nlm.nih.gov/19300585/

        Moore, B. A., Tison, L. M., Palacios, J. G., Peterson, A. L., & Mysliwiec, V. (2021). Incidence of insomnia and obstructive sleep apnea in active duty United States military service members. Sleep, zsab024. Advance online publication.
        https://pubmed.ncbi.nlm.nih.gov/33532830/

        Janna Mantua, Alexxa F Bessey, Carolyn A Mickelson, Jake J Choynowski, Jeremy J Noble, Tina M Burke, Ashlee B McKeon, Walter J Sowden, Sleep and high-risk behavior in military service members: a mega-analysis of four diverse U.S. Army units, Sleep, Volume 44, Issue 4, April 2021, zsaa221, https://doi.org/10.1093/sleep/zsaa221

        Institute of Medicine (US) Committee on Military Nutrition Research; Marriott BM, editor. Food Components to Enhance Performance: An Evaluation of Potential Performance-Enhancing Food Components for Operational Rations. Washington (DC): National Academies Press (US); 1994. 7, The Effects of Sleep Deprivation on Performance During Continuous Combat Operations. Available from: https://www.ncbi.nlm.nih.gov/books/NBK209071/

        Troxel, W. M., Shih, R. A., Pedersen, E. R., Geyer, L., Fisher, M. P., Griffin, B. A., Haas, A. C., Kurz, J., & Steinberg, P. S. (2015). Sleep in the Military: Promoting Healthy Sleep Among U.S. Servicemembers. Rand health quarterly, 5(2), 19. https://www.rand.org/content/dam/rand/pubs/research_reports/RR700/RR739/RAND_RR739.pdf

        Good, C. H., Brager, A. J., Capaldi, V. F., & Mysliwiec, V. (2020). Sleep in the United States Military. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology, 45(1), 176–191. https://doi.org/10.1038/s41386-019-0431-7

        Luxton DD, Greenburg D, Ryan J, Niven A, Wheeler G, Mysliwiec V. Prevalence and impact of short sleep duration in redeployed OIF soldiers. Sleep. 2011 Sep 1;34(9):1189-95. doi: 10.5665/SLEEP.1236. PMID: 21886356; PMCID: PMC3157660.

        Bonnet MH, Arand DL. Clinical effects of sleep fragmentation versus sleep deprivation. Sleep Med Rev. 2003 Aug;7(4):297-310. doi: 10.1053/smrv.2001.0245. PMID: 14505597.

        Shattuck, N. L., & Brown, S. A. (2013). Wounded in action: what the sleep community can learn from sleep disorders of US military service members. Sleep, 36(2), 159–160. https://doi.org/10.5665/sleep.2356

        Mysliwiec, V., McGraw, L., Pierce, R., Smith, P., Trapp, B., & Roth, B. J. (2013). Sleep disorders and associated medical comorbidities in active duty military personnel. Sleep, 36(2), 167–174. https://doi.org/10.5665/sleep.2364

        As a disclaimer, I’ve not reviewed these studies, however, I’ve consumed a substantial enough amount of the literature related to sleep deprivation and narcolepsy to know that Guzey is pushing a ridiculous “hypothesis” that has no grounding in any medical science. They’ve done studies related to sleep deprivation and the effects of stimulants going back to world war II. I’m sure if he was a competent researcher he would have been able to find plenty of literature, and quite a few well-designed experimental studies, to have prevented him from formulating his purely econ skewed view of sleep. I’m not saying that there is anything wrong with pointing out flaws in someone else’s research that is clearly problematic, but advancing a ridiculous and wholly unsupported claim doesn’t do much to advance knowledge in the space.

  19. Nit: The down jacket was invented in 1936, and entered fashion around ’39. I expect you noticed them on one of the occasions where better down production caused a price drop.

    • John:

      Yeah, I guess so. I can’t remember the exact year, maybe it was 1976. Before 1976 nobody had down jackets; I’d never seen one or heard of such a thing. After 1976 everyone had one. It was a qualitative difference. All of a sudden we weren’t cold all winter long! A couple years later the price of down jackets really dropped and you could get one for $25.

  20. It seems strange to talk about sleep in a vacuum and leisure in a vacuum, rather than seeing the first as a means to the second. If you’re too tired, you can’t enjoy leisure. It’s not one or the other.

    If you’re underslept (whatever that means for you — judge by feel), you will also perform less well than you could at virtually everything.

    I know nothing of Guzey’s work habits, but engineers in general seem to believe themselves to be robots, immune to the normal dips in functioning that come with barely sleeping and working for sixteen-hour stretches. What if they just stopped after eight, went to bed, and the next day did work that was twice as good in half the time? I find it odd that their work involves creative insight and demands quality, yet many of them act as though the ideal schedule for it is that of a drudge or a minder of an assembly-line. What about quality, eh?

    It seems like they have never tested the counter-thesis, of quality work done in short but intensely-focused sessions. Instead tech culture views every coding project as an ultra-marathon, and treats over-long break-free sessions as some kind of juice for creativity.

    Your brain isn’t so different from your muscles. If coding requires creativity and problem-solving, rather than just raw hours, then it’s probably not so different from lifting weights. Nobody would say that the best time to try to set a new PR is after having lifted weights for five hours straight (if you could even do that). Why would the same not hold for cognitive pursuits? It smells to me like mind-body dualism for people who treat themselves as quasi-robots because of what they do for a living.

    Same thing for ignoring or downplaying the importance of exercise and nutrition, when anyone who’s experimented with these things can tell you that they make you perform better mentally as well. The truth is most engineers are too smart for their own good, and they can afford to burn off that extra IQ with bad habits since their work doesn’t demand the full breadth of their intelligence. One can only imagine how well Musk might perform under the best of conditions if he can do what he does on two hours sleep (or did during his companies’ ascents). As you put off sleep, your work gets slower and shittier but your tired brain is too, you know, drunk to notice it.

    No slight meant to engineers, by the way. It’s just a problem for any group of people whose self-model is a brain in a vat, lol. Your brain is not immune from what you do to your body. This lesson is not rammed home until you reach your 50s and 60s but after that point those who have not taken care begin to decline. This seems like a big thing to overlook for smart people who care about their cognition.

Leave a Reply

Your email address will not be published. Required fields are marked *