I’m not sold on the conclusions right now but I think this raises a number of excellent points (particularly the analogy with fasting) and I’m really looking forward to an extended discussion of it.
The fasting analogy is interesting, as is the analogy with exercise—some kinds of activities are beneficial in the long-run even when they are damaging/unpleasant in the short run. But surely these are exceptions to the general rule, right?
Besides exercise, it’s not good to repeatedly injure yourself and then have the wounds heal. (Exercise is essentially the small, specific subtype of “injury” which is actually good for the body in the long term.)
Getting sick with a cold or flu is good at building immunity to that kind of virus when it comes around a second time, but aside from immunity concerns, it would be better for your health to never become sick at all. (As with viruses, the same goes for diseases caused by parasites or bacteria.) Especially as a young child, getting badly sick can impact your development and later IQ / income / etc substantially. Getting mildly sick is probably mildly bad for those same metrics.
One of the reasons junk food is bad is because it has lots of quickly-absorbed sugars, which rush into your bloodstream and force your insulin/glycogen system to ramp up quickly and do a lot of work. Over the long term, putting all this stress on your body’s ability to absorb sugars is though to reduce your body’s insulin sensitivity, leading to metabolic disorders like prediabetes. So, chronic consumption of junk food is bad—but should I prefer a totally healthy “low-glycemic index” diet? Or a mostly low-glycemic index diet but I occasionally consume a blast of sugary sweets to “exercise” my insulin system? I don’t think science has given us a real answer here, but most doctors would probably recoil in horror at the idea of “exercising” one’s metabolic system by occasionally binging junk food, and I’d be inclined to agree with them.
Some kinds of psychological stress and trauma are probably beneficial in the long run (for instance, working hard on a project to meet a deadline and feeling invigorated + learning better productivity skills as a result), while other kinds are probably just bad.
Basically, it seems like there are plenty of examples on both sides, and I can’t figure out any general rule that would let me predict ahead of time which seemingly-bad behaviors/stressors are secretly good or not. The examples of exercise and fasting are helpful reminders to keep an open mind, but I don’t think they can tell us much more than that when we’re trying to figure out how to think about sleep.
(Similarly, some things about the modern world are “superstimulus”—like junk food. But others are just progress—like the fact that I can afford a healthy diet with lots of meat and vegetables if I so choose, while my agrarian ancestors got much more of their calories from samey, not-very-nutritious grains. I don’t know if comfortable beds are a superstimulus encouraging us to oversleep harmfully or just modern progress enabling us to higher-quality sleep. But I do appreciate that the “superstimulus” hypothesis is reasonable and encourages us to keep an open mind.)
I am much more sold on “variety is good for humans, and mild-moderate deprivation and excess is variety” than “humans should permanently run on much less sleep than they think they need” or “sleepiness is a lie”.
One tricky thing here is humans aren’t actually guaranteed to have a pareto optimim, or to have a path that gets all good things. It seems really plausible childhood illnesses damages development and IQ, and lack of childhood illness causes allergies and immune vulnerability later (I think think the hygiene/old friends hypothesis is largely correct, even if it doesn’t support eating dirt in particular), and there isn’t an ideal level that gets you your max IQ and no allergies. Variety is something of a hack to get some of both and also create a discovery process for what you need more at a particular moment.
Most doctors would probably recoil in horror at the idea of “exercising” one’s metabolic system by occasionally binging junk food, and I’d be inclined to agree with them.
In the US, pregnant women get a glucose tolerance test to check for gestational diabetes, which involves drinking 100g of glucose on an empty stomach and seeing how quickly it gets processed. But that’s once or twice per pregnancy.
The most valuable contribution of the OP for me was in breaking down “sleep research” into more fine-grained hypotheses.
Does sleeping X hours per night cause an increase in [specific physical or mental health risk]?
A need for a different word for “the physiological states and dynamics associated with various sleep stages” as opposed to the psychological experience of being asleep. Let’s call these “physical sleep” vs “sleep experiences.” Also, we may want to introduce a category of “stimulating experiences” vs “non-stimulating experiences,” the latter of which includes sleep, meditation, perhaps a stroll around the park.
The possibility of interventions that control or reproduce any downstream impacts of physical sleep or sleep experiences on mortality/injury, cognition, and awake experiences (ie meditation as substitute for sleep experiences in memory consolidation).
The possibility that it is in fact non stimulating experiences that are neglected in our culture, and that our understanding of these dynamics is so poor that we uniquely assign their benefits to sleep.
As an analogy, maybe sleep is like lemons, and what we need is vitamin C. Our culture currently assumes there’s a unique power in lemons that can be found nowhere else in nature, but also that anything that we call “lemon juice” (lime juice, lemon juice piped through copper tubing) will give at least some of the benefits. This despite reports from polar explorers that bear liver replicates the effects just as well, but they’re weird crazy polar explorers and the admiral of the navy says limes are fine.
I think Guzey should also look into Buddhist monks, another group who habitually get less than 8 hours of sleep and are highly involved with non stimulating and transcendent experiences. It’s actually interesting to me that so little of the chatter around meditation and enlightenment addresses the role of low-sleep regimens in this context. I’d be curious to learn more.
I’m not sold on the conclusions right now but I think this raises a number of excellent points (particularly the analogy with fasting) and I’m really looking forward to an extended discussion of it.
The fasting analogy is interesting, as is the analogy with exercise—some kinds of activities are beneficial in the long-run even when they are damaging/unpleasant in the short run. But surely these are exceptions to the general rule, right?
Besides exercise, it’s not good to repeatedly injure yourself and then have the wounds heal. (Exercise is essentially the small, specific subtype of “injury” which is actually good for the body in the long term.)
Getting sick with a cold or flu is good at building immunity to that kind of virus when it comes around a second time, but aside from immunity concerns, it would be better for your health to never become sick at all. (As with viruses, the same goes for diseases caused by parasites or bacteria.) Especially as a young child, getting badly sick can impact your development and later IQ / income / etc substantially. Getting mildly sick is probably mildly bad for those same metrics.
On the other hand, I enjoyed your post a few months ago examining whether letting kids play outside and get dirty is helpful for calibrating their immune systems and reducing allergies later in life. It seems like the “hygiene hypothesis” is less firmly established than I thought, but if true it would be an example something else like exercise and fasting where injury/stress leads to long-term benefit.
One of the reasons junk food is bad is because it has lots of quickly-absorbed sugars, which rush into your bloodstream and force your insulin/glycogen system to ramp up quickly and do a lot of work. Over the long term, putting all this stress on your body’s ability to absorb sugars is though to reduce your body’s insulin sensitivity, leading to metabolic disorders like prediabetes. So, chronic consumption of junk food is bad—but should I prefer a totally healthy “low-glycemic index” diet? Or a mostly low-glycemic index diet but I occasionally consume a blast of sugary sweets to “exercise” my insulin system? I don’t think science has given us a real answer here, but most doctors would probably recoil in horror at the idea of “exercising” one’s metabolic system by occasionally binging junk food, and I’d be inclined to agree with them.
Some kinds of psychological stress and trauma are probably beneficial in the long run (for instance, working hard on a project to meet a deadline and feeling invigorated + learning better productivity skills as a result), while other kinds are probably just bad.
Basically, it seems like there are plenty of examples on both sides, and I can’t figure out any general rule that would let me predict ahead of time which seemingly-bad behaviors/stressors are secretly good or not. The examples of exercise and fasting are helpful reminders to keep an open mind, but I don’t think they can tell us much more than that when we’re trying to figure out how to think about sleep.
(Similarly, some things about the modern world are “superstimulus”—like junk food. But others are just progress—like the fact that I can afford a healthy diet with lots of meat and vegetables if I so choose, while my agrarian ancestors got much more of their calories from samey, not-very-nutritious grains. I don’t know if comfortable beds are a superstimulus encouraging us to oversleep harmfully or just modern progress enabling us to higher-quality sleep. But I do appreciate that the “superstimulus” hypothesis is reasonable and encourages us to keep an open mind.)
I am much more sold on “variety is good for humans, and mild-moderate deprivation and excess is variety” than “humans should permanently run on much less sleep than they think they need” or “sleepiness is a lie”.
One tricky thing here is humans aren’t actually guaranteed to have a pareto optimim, or to have a path that gets all good things. It seems really plausible childhood illnesses damages development and IQ, and lack of childhood illness causes allergies and immune vulnerability later (I think think the hygiene/old friends hypothesis is largely correct, even if it doesn’t support eating dirt in particular), and there isn’t an ideal level that gets you your max IQ and no allergies. Variety is something of a hack to get some of both and also create a discovery process for what you need more at a particular moment.
In the US, pregnant women get a glucose tolerance test to check for gestational diabetes, which involves drinking 100g of glucose on an empty stomach and seeing how quickly it gets processed. But that’s once or twice per pregnancy.
The most valuable contribution of the OP for me was in breaking down “sleep research” into more fine-grained hypotheses.
Does sleeping X hours per night cause an increase in [specific physical or mental health risk]?
A need for a different word for “the physiological states and dynamics associated with various sleep stages” as opposed to the psychological experience of being asleep. Let’s call these “physical sleep” vs “sleep experiences.” Also, we may want to introduce a category of “stimulating experiences” vs “non-stimulating experiences,” the latter of which includes sleep, meditation, perhaps a stroll around the park.
The possibility of interventions that control or reproduce any downstream impacts of physical sleep or sleep experiences on mortality/injury, cognition, and awake experiences (ie meditation as substitute for sleep experiences in memory consolidation).
The possibility that it is in fact non stimulating experiences that are neglected in our culture, and that our understanding of these dynamics is so poor that we uniquely assign their benefits to sleep.
As an analogy, maybe sleep is like lemons, and what we need is vitamin C. Our culture currently assumes there’s a unique power in lemons that can be found nowhere else in nature, but also that anything that we call “lemon juice” (lime juice, lemon juice piped through copper tubing) will give at least some of the benefits. This despite reports from polar explorers that bear liver replicates the effects just as well, but they’re weird crazy polar explorers and the admiral of the navy says limes are fine.
I think Guzey should also look into Buddhist monks, another group who habitually get less than 8 hours of sleep and are highly involved with non stimulating and transcendent experiences. It’s actually interesting to me that so little of the chatter around meditation and enlightenment addresses the role of low-sleep regimens in this context. I’d be curious to learn more.
Oh man this is so much better than my curation notice.
I may well have skipped this if it wasn’t for your curation notice. I take your judgment on what’s worth reading pretty seriously!
neat, sounds like a success story all around.
What’s a curation notice?
It’s a comment indicating that a moderator has curated something:
https://www.lesswrong.com/posts/HvcZmKS43SLCbJvRb/theses-on-sleep?commentId=hMvSncYmxMukMJm4s