Came here from Adam's article! I actually have spent the last couple of months fixing my insomnia/sleep issues, so I have a lot of thoughts.
Have you heard about the sleep doctor Alan Walker (he wrote a book Why We Sleep and also went on a bunch of podcasts)? A lot of people identify as short sleepers, but almost all of them are not actually short sleepers and will suffer health consequences https://www.youtube.com/shorts/gDYYqXcYuJQ. I don’t think the recommendation of 7-9 hours denies that a distribution exists, but rather that the distribution for most people is in that 7-9 range, instead of a larger range. Many people identify as short sleepers because their lifestyle determines how much they sleep, and then after years, they think their average well-rested state is close to their max well-rested state, but it's been years since they had weeks of 8-9 hours of good quality sleep and forgot/never knew their true well-rested state, which can be a lot higher.
From my time fixing my sleep, I do strongly believe that lifestyle affects how much sleep the average American gets rather than Americans seemingly getting the amount of sleep they actually need. Sleep duration directly affects sleep quality, and sleep duration/quality are affected by so many things like caffeine intake/timing, dinner timing (digestion lowers sleep quality), late exercise spiking adrenaline/cortisol, vitamin D levels, melatonin levels, browsing constantly not only exposing us to blue-light keeping us up, but contributing to insomnia since the only time our brain has time to process the day is when we try to go to sleep but can’t, the stress of modern life/work/inflation. I feel like the average American will definitely have one of these things delaying/disrupting the quality of their sleep.
We can have some sort of empirical measure of quality sleep by measuring REM/deep sleep with a sleep tracker. You’ll likely find on your 5-hour night that you got significantly less REM and less deep sleep. If you’re doing things like dealing with a breakup (rem helps mental health/processing events), or going to the gym (deep sleep affects hormones), the consistency of getting the right amount of sleep becomes really apparent, you get gains much easier with 15 hours of deep sleep over a 10 day period vs probably half the deep sleep on a more random schedule.
I actually really identify with you, about letting my lifestyle determine my sleep, naturally sleeping at different times because I stayed up playing games or hanging with friends, waking up randomly, and living this fun lifestyle from middle-school into my adult working life. But now I know that the time I wake up is more influenced by my circadian rhythm, rather than when my body determined that I got “enough” sleep. And over the years my memory has been getting worse which I thought was normal, I never put on muscle that easily, always had trouble with willpower with my diet. But getting a consistent sleep schedule has literally reversed all of that.
I see that you’re content with your sleep patterns being determined by your lifestyle, but definitely curious to see if you'd experience a lot of benefits if you experimented with a boring sleep schedule for a couple weeks lol
I haven't heard of Alan walker! Thanks for the tip :)
Re: many people identifying as short sleepers, if you notice the NHANES data had (relative to what you would expect for a perfect standard deviation) slightly more short sleepers than long sleepers (which I think points to your theory being true).
Re: empirical measures, I don't actually understand most of the measures you mentioned, but I think I resonate on a personal level. Deep sleep feels good, sometimes I wake up feeling healed, and I've definitely resolved some personal issues in dreams.
Re: sleep patterns and lifestyle, I cut a part of the blog post where I talked about my sleep history but I might write it up as a second post. TL;DR: My sleep has varied A LOT over my life and I've tried most things (from not sleeping at all to boring sleep to bi-phasic sleep).
I think it’s actually Matthew Walker. Andrew Gelman and co at Columbia stats have questioned his methodology, but I think his recommendations (haven’t read his book) are in the mainstream of sleep advice.
Back in 1967 my family moved from Canada to Florida. I thought to myself, now I have no need to sleep under a blanket. So I didn't. I then asked myself about the sheet. I was able to train myself to sleep with no covering, but it took over a month and my sleep was very disturbed for that month or so that it took me to train myself. I slept with no covering for about 12 years until I got married.
I sleep with at least a sheet, even though I've lived in Florida since 1976, mainly because I don't want things crawling on me (e.g., palmetto bugs) or biting me (e.g., mosquitos). I have animals (dog & cat) and my house is not insect-proof.
Yeah - I would take insecticide and spray it around the sides of my matress (not on the top where I slept). That stopped the crawlies. And there were times palmetto beetles flew over and landed on me. Yuk!
"We go back to basics and acknowledge how little we know. There are many situations, such as what amount of sleep is best for you specifically, where you can see the results clearly for yourself without any need for big datasets and science machines."
I think this is a great example of a broader problem in science. Science (and scientists) are very well equipped to answer the question of: "Based on survey results, how does hours-slept correlate with survey-taker perceptions of their own health and happiness?". On the other hand, science is not well-equipped to answer the question of: "How many hours of sleep should I get tonight?". In fact, the institution of science in modern society is entirely incapable of answering that question, as you lay out in your article. I think this understanding comes naturally to some scientists (though not all), but is completely lost on the general population who is unfamiliar with academia or research. This is not to say the first question is not worthy of study, only that we should be very careful when we apply the answer of the first question to the second question.
"the institution of science in modern society is entirely incapable of answering [the] question [how many hours of sleep I should get tonight]"
I don't think that's true. Sure, there will not be on singular number fitting everyone all the time. But that just means we're talking about a more complex function. Surely there are let's say 10 or 20 (time-dependent?) variables which would predict the amount of sleep a given person needs on a given night with extremely high accuracy. It's not completely random, after all. Using self-experimentation is not really different from a scientist observing and suggesting ways for a single person to figure out how much they should sleep - in the former case, you're simply both scientist and test subject. And humans don't vary so much that there would be 8 billion different "types" of sensible sleep habits - many will bunch together.
So, if we can use science to find the variables and their relationships to sleep, then I think we can easily answer the second question well in the vast majority of questions. Moreover, this probably holds even more strongly for aspects
which vary less between humans than need for sleep (hehe) does.
How do you craft this function? Using survey results? Using biomarkers?
The question is, "how many hours of sleep should I get tonight?"
Not "How many hours of sleep yields optimal survey results or biomarker serum concentrations".
Who is to say that I want to be alert and awake tomorrow? Maybe I have a flight I want to sleep through.
Who is to say I value alertness at all-- maybe my culture values extra working hours at the expense of alertness during the day.
Maybe my son has a piano recital tonight and I'd rather trade some hours of sleep to support him at a pivotal moment.
You've missed the point entirely. There are value-judgements and impossibly unique circumstances inherent to the second question. Science can not provide the answer because you first must change the question before answering it.
You also are refuting my quote by claiming that the practice of "science", ie data collection, analysis, correlation etc, could figure out this hypothetical function. You might feel that even with the hypotheticals I posed above, there should be some achievable function which captures these. This might be true, but I never claimed differently. I can assure you, the "institution of science in modern society", as I said in the quote, will never yield this function. No one seems to be researching the optimal amount of sleep you should get if you want to be tired the next day because you have a flight you want to sleep through, but not so tired where you aren't functioning at all because you will have to go to a relative's piano recital when you land.
I myself am a scientist and engineer. My most lucrative skill is applying answers from the first question to the second question. Of course those survey results are a good baseline, but normal people seem to be increasingly misunderstanding that science is virtually incapable of providing answers to the second type of questions, only approximations. Often times, these approximations are insanely simplified and only have value in the aggregate.
> The question is, "how many hours of sleep should I get tonight?"
> No one seems to be researching the optimal amount of sleep you should get if you want to be tired the next day because you have a flight you want to sleep through, but not so tired where you aren't functioning at all because you will have to go to a relative's piano recital when you land.
You must admit that you have pivoted significantly from the initial question (in a way I perceive to be somewhat intellectually dishonest even, to be frank). I don't think that science can answer a "should" question without any conditionals attached. But the is-ought distinction was neither discussed in the post, nor your first comment, nor my reply. So I must wonder what you meant then, when you wrote "science [...] is entirely incapable of answering that question, *as you lay out in your article*." (emphasis mine)
What the article does state is that using non-individualized advice on sleep duration is probably not very helpful. It also explicitly encourages self-experimentation, but to what end? Surely, the goal is something like "to understand how you personal sleep system *generally* works and how to function optimally given this".
Your points about the flight and the piano recital appear to be total red herrings to me. Science couldn't answer the should-question, even if exactly 8 hours total sleep would be "optimal" always for everyone. And why *should* anyone research the optimal amount of sleep you should get for a hyper-specific individual situation? I don't see this as a criticism of the modern institution of science at all, but as a sensible use of everyone's time. Probably even yours! After all, *you* could do that research by testing/tracking sleep techniques and duration each time you have a flight + recital combo. I reckon it doesn't come up often enough for you to bother, or to even have a chance to gather some data.
Lastly, notice that even in that situation, science could provide some guidelines. It could sensibly answer the question of "how much sleep should I get today, GIVEN that I want to wake up well-rested tomorrow". And then you can adjust *that number* based on your personal circumstances and situations, if needed.
> How do you craft this function? Using survey results? Using biomarkers?
Whichever. I think a sensible goal metric would be something like "perceived restedness after waking up" or whatever. So, just use that as a survey result to this end and you're done. Otherwise, if you test this and the question turns out to perform poorly, you can always pick a different metric (like biomarkers, or cognitive performance on a test at the end of the day, or a different survey question like "when did you first start feeling tired today?")
First, I was a family doctor for 40 years, and always tried to tell people to adopt an approach similar to yours. Accept that sleep varies from one day to the next, sleep when you are tired, and decide whether or not to get up when you wake up. Any attempt to “standardize” or regulate sleep will be doomed to failure.
Second, I retired a couple of years ago and went from a tightly regulated/scheduled life to one which is completely unregulated and unscheduled. My sleep has improved enormously. I go to bed when I want to, much later than I used to. I set an alarm simply so that, when I wake up in the middle of the night, I can tell myself the alarm hasn’t gone off yet, so no need to even think about checking the time or getting up. I still wake up before the alarm, but choose to stay in bed longer, even when it goes off. I’m probably averaging 6-7 hours of sleep a night, far less than what I used to think I needed. Most days I feel great. Some days I take a nap, if I feel like it.
I guess the lesson in this is that the laissez-faire approach works great if you have no specific obligations to be anywhere at any specific time. Trying to fit sleep into a societally-imposed schedule is counterproductive, at best. Sleep “disorders” probably reflect the fact that human physiology, sleep included, can’t be expected to conform to the averages.
I'm currently crunching some initial numbers that will become an update post of sorts. Are you ok if I cite this comment in its entirety in that post? I think it gets at the spirit of what I'm trying to convey really nicely and opens the discussion of sleep to obligations like jobs instead of obligations like "sleeping exactly 8 hours a night."
There have been actual scientific studies done on melatonin use, and the most interesting melatonin studies are ones that measure the amount of melatonin in any given capsule that you might buy at the store. A 2023 study from the University of Mississippi found that 22 out of 25 melatonin gummy products contained different amounts of melatonin than the label claimed, with one gummy containing more than three times the amount and another containing none at all. Other studies have found that melatonin supplements can contain other substances, like serotonin, that may be concerning. Because the FDA can't regulate supplements like melatonin, some say that the pills may not contain what the label says, and that melatonin is a psychoactive hormone that shouldn't be taken without a doctor's guidance.
The only studies that show a positive correlation between melatonin supplement and sleep had so few participants that they aren't even relevant.
I wrote this same article about sleep, except I wrote mine about infant sleep. I love the idea that you know you best. Averages are informative, but they're not definitive.
I also linked from Adam’s article - I wanted to add an anecdote re “programming” oneself to wake at a certain time. I regrettably forget where, but many years ago, I heard/read the following: if you want to wake up at a certain time, picture yourself on a cliff above the ocean. As clearly as possible, imagine yourself carving your desired wake-up time on a rock, then throw the rock over the cliff and into the water. For a while, I did this pretty often, and as I recall, it worked astonishingly well, consistently allowing me to wake at the time I had “programmed”.
I really need to revisit this, partially because it was a much gentler way to wake than an alarm, and because it raises some questions: assuming it does work as well as I remember, does that mean that the somewhat arbitrary division of the day into 24 hours is etched into my brain? Or is some part of my brain counting down? On a related note, is it more difficult to wake at odd times, like 7:47, than times like 7:30 or 8:00?
It’s kind of tangential to the main point of your article (sorry!) but does bear on it somewhat. Also brings up an issue involving a basic criterion for the “right” amount of sleep, the feeling of restedness: it does seem to be the case that the manner of waking, and when in the sleep cycle waking happens, also have an effect on that feeling.
I’d be interested to hear whether anyone else has ever heard of, or has experience with, a technique like this.
I have always believed that the idea of the normal distribution, though scientifically well studied, is epistemically underappreciated. I am a chronic sleeper, perhaps an hypersomniac. A secondary study I'd like to suggest is to compare those who fall on the extremes of either sides of the sleep distribution with certain traits such as energy level, moodiness, and emotional liability.
I’ve been a bad sleeper since childhood (and I’m well into middle age now). I don’t know how much I need, but I know that I rarely get enough—I remember the rare occasions when I sleep well years later. I do everything you’re supposed to (I get a lot of exercise, don’t use screens in the evening, spend time outside, very little alcohol, etc) but I still sleep poorly, often *very* poorly (like last night) and will try any (non-pharmaceutical) thing to make it better. So I signed up for your study. Thanks!
I've always needed 7 hours of sleep a night. I remember in early teenage years I'd get 7 hours of sleep a night. In late teenage years I got into the habit of staying up really late each night but sleeping in very late at the weekend, so I'd only get 5 hours each weeknight but 12 hours each weekend night. And then after many months of this working fine I realised 5*5+2*12=49=7*7. I was still getting 7 hours a night on average.
Now 25 years later I'm still getting 7 hours a night on average. And I was fine. Until a couple of years ago when I got diagnosed with Long COVID / chronic fatigue; but even now I'm taking extra breaks and doing more resting, I'm still sleeping around 7 hours a night.
Came here from Adam's article! I actually have spent the last couple of months fixing my insomnia/sleep issues, so I have a lot of thoughts.
Have you heard about the sleep doctor Alan Walker (he wrote a book Why We Sleep and also went on a bunch of podcasts)? A lot of people identify as short sleepers, but almost all of them are not actually short sleepers and will suffer health consequences https://www.youtube.com/shorts/gDYYqXcYuJQ. I don’t think the recommendation of 7-9 hours denies that a distribution exists, but rather that the distribution for most people is in that 7-9 range, instead of a larger range. Many people identify as short sleepers because their lifestyle determines how much they sleep, and then after years, they think their average well-rested state is close to their max well-rested state, but it's been years since they had weeks of 8-9 hours of good quality sleep and forgot/never knew their true well-rested state, which can be a lot higher.
From my time fixing my sleep, I do strongly believe that lifestyle affects how much sleep the average American gets rather than Americans seemingly getting the amount of sleep they actually need. Sleep duration directly affects sleep quality, and sleep duration/quality are affected by so many things like caffeine intake/timing, dinner timing (digestion lowers sleep quality), late exercise spiking adrenaline/cortisol, vitamin D levels, melatonin levels, browsing constantly not only exposing us to blue-light keeping us up, but contributing to insomnia since the only time our brain has time to process the day is when we try to go to sleep but can’t, the stress of modern life/work/inflation. I feel like the average American will definitely have one of these things delaying/disrupting the quality of their sleep.
We can have some sort of empirical measure of quality sleep by measuring REM/deep sleep with a sleep tracker. You’ll likely find on your 5-hour night that you got significantly less REM and less deep sleep. If you’re doing things like dealing with a breakup (rem helps mental health/processing events), or going to the gym (deep sleep affects hormones), the consistency of getting the right amount of sleep becomes really apparent, you get gains much easier with 15 hours of deep sleep over a 10 day period vs probably half the deep sleep on a more random schedule.
I actually really identify with you, about letting my lifestyle determine my sleep, naturally sleeping at different times because I stayed up playing games or hanging with friends, waking up randomly, and living this fun lifestyle from middle-school into my adult working life. But now I know that the time I wake up is more influenced by my circadian rhythm, rather than when my body determined that I got “enough” sleep. And over the years my memory has been getting worse which I thought was normal, I never put on muscle that easily, always had trouble with willpower with my diet. But getting a consistent sleep schedule has literally reversed all of that.
I see that you’re content with your sleep patterns being determined by your lifestyle, but definitely curious to see if you'd experience a lot of benefits if you experimented with a boring sleep schedule for a couple weeks lol
I haven't heard of Alan walker! Thanks for the tip :)
Re: many people identifying as short sleepers, if you notice the NHANES data had (relative to what you would expect for a perfect standard deviation) slightly more short sleepers than long sleepers (which I think points to your theory being true).
Re: empirical measures, I don't actually understand most of the measures you mentioned, but I think I resonate on a personal level. Deep sleep feels good, sometimes I wake up feeling healed, and I've definitely resolved some personal issues in dreams.
Re: sleep patterns and lifestyle, I cut a part of the blog post where I talked about my sleep history but I might write it up as a second post. TL;DR: My sleep has varied A LOT over my life and I've tried most things (from not sleeping at all to boring sleep to bi-phasic sleep).
I think it’s actually Matthew Walker. Andrew Gelman and co at Columbia stats have questioned his methodology, but I think his recommendations (haven’t read his book) are in the mainstream of sleep advice.
I can’t sleep if I don’t have a blanket on me. I assume this is because I’ve always slept with a blanket.
I m need at least a sheet. I remember as a kid thinking exposed skin would be vulnerable to the little darkness demons.
oh yeah you can't let those demons get you!
Back in 1967 my family moved from Canada to Florida. I thought to myself, now I have no need to sleep under a blanket. So I didn't. I then asked myself about the sheet. I was able to train myself to sleep with no covering, but it took over a month and my sleep was very disturbed for that month or so that it took me to train myself. I slept with no covering for about 12 years until I got married.
I sleep with at least a sheet, even though I've lived in Florida since 1976, mainly because I don't want things crawling on me (e.g., palmetto bugs) or biting me (e.g., mosquitos). I have animals (dog & cat) and my house is not insect-proof.
Yeah - I would take insecticide and spray it around the sides of my matress (not on the top where I slept). That stopped the crawlies. And there were times palmetto beetles flew over and landed on me. Yuk!
Very nice article, thank you.
"We go back to basics and acknowledge how little we know. There are many situations, such as what amount of sleep is best for you specifically, where you can see the results clearly for yourself without any need for big datasets and science machines."
I think this is a great example of a broader problem in science. Science (and scientists) are very well equipped to answer the question of: "Based on survey results, how does hours-slept correlate with survey-taker perceptions of their own health and happiness?". On the other hand, science is not well-equipped to answer the question of: "How many hours of sleep should I get tonight?". In fact, the institution of science in modern society is entirely incapable of answering that question, as you lay out in your article. I think this understanding comes naturally to some scientists (though not all), but is completely lost on the general population who is unfamiliar with academia or research. This is not to say the first question is not worthy of study, only that we should be very careful when we apply the answer of the first question to the second question.
"the institution of science in modern society is entirely incapable of answering [the] question [how many hours of sleep I should get tonight]"
I don't think that's true. Sure, there will not be on singular number fitting everyone all the time. But that just means we're talking about a more complex function. Surely there are let's say 10 or 20 (time-dependent?) variables which would predict the amount of sleep a given person needs on a given night with extremely high accuracy. It's not completely random, after all. Using self-experimentation is not really different from a scientist observing and suggesting ways for a single person to figure out how much they should sleep - in the former case, you're simply both scientist and test subject. And humans don't vary so much that there would be 8 billion different "types" of sensible sleep habits - many will bunch together.
So, if we can use science to find the variables and their relationships to sleep, then I think we can easily answer the second question well in the vast majority of questions. Moreover, this probably holds even more strongly for aspects
which vary less between humans than need for sleep (hehe) does.
How do you craft this function? Using survey results? Using biomarkers?
The question is, "how many hours of sleep should I get tonight?"
Not "How many hours of sleep yields optimal survey results or biomarker serum concentrations".
Who is to say that I want to be alert and awake tomorrow? Maybe I have a flight I want to sleep through.
Who is to say I value alertness at all-- maybe my culture values extra working hours at the expense of alertness during the day.
Maybe my son has a piano recital tonight and I'd rather trade some hours of sleep to support him at a pivotal moment.
You've missed the point entirely. There are value-judgements and impossibly unique circumstances inherent to the second question. Science can not provide the answer because you first must change the question before answering it.
You also are refuting my quote by claiming that the practice of "science", ie data collection, analysis, correlation etc, could figure out this hypothetical function. You might feel that even with the hypotheticals I posed above, there should be some achievable function which captures these. This might be true, but I never claimed differently. I can assure you, the "institution of science in modern society", as I said in the quote, will never yield this function. No one seems to be researching the optimal amount of sleep you should get if you want to be tired the next day because you have a flight you want to sleep through, but not so tired where you aren't functioning at all because you will have to go to a relative's piano recital when you land.
I myself am a scientist and engineer. My most lucrative skill is applying answers from the first question to the second question. Of course those survey results are a good baseline, but normal people seem to be increasingly misunderstanding that science is virtually incapable of providing answers to the second type of questions, only approximations. Often times, these approximations are insanely simplified and only have value in the aggregate.
> The question is, "how many hours of sleep should I get tonight?"
> No one seems to be researching the optimal amount of sleep you should get if you want to be tired the next day because you have a flight you want to sleep through, but not so tired where you aren't functioning at all because you will have to go to a relative's piano recital when you land.
You must admit that you have pivoted significantly from the initial question (in a way I perceive to be somewhat intellectually dishonest even, to be frank). I don't think that science can answer a "should" question without any conditionals attached. But the is-ought distinction was neither discussed in the post, nor your first comment, nor my reply. So I must wonder what you meant then, when you wrote "science [...] is entirely incapable of answering that question, *as you lay out in your article*." (emphasis mine)
What the article does state is that using non-individualized advice on sleep duration is probably not very helpful. It also explicitly encourages self-experimentation, but to what end? Surely, the goal is something like "to understand how you personal sleep system *generally* works and how to function optimally given this".
Your points about the flight and the piano recital appear to be total red herrings to me. Science couldn't answer the should-question, even if exactly 8 hours total sleep would be "optimal" always for everyone. And why *should* anyone research the optimal amount of sleep you should get for a hyper-specific individual situation? I don't see this as a criticism of the modern institution of science at all, but as a sensible use of everyone's time. Probably even yours! After all, *you* could do that research by testing/tracking sleep techniques and duration each time you have a flight + recital combo. I reckon it doesn't come up often enough for you to bother, or to even have a chance to gather some data.
Lastly, notice that even in that situation, science could provide some guidelines. It could sensibly answer the question of "how much sleep should I get today, GIVEN that I want to wake up well-rested tomorrow". And then you can adjust *that number* based on your personal circumstances and situations, if needed.
> How do you craft this function? Using survey results? Using biomarkers?
Whichever. I think a sensible goal metric would be something like "perceived restedness after waking up" or whatever. So, just use that as a survey result to this end and you're done. Otherwise, if you test this and the question turns out to perform poorly, you can always pick a different metric (like biomarkers, or cognitive performance on a test at the end of the day, or a different survey question like "when did you first start feeling tired today?")
"What the article does state is that using non-individualized advice on sleep duration is probably not very helpful."
"I don't think that science can answer a "should" question without any conditionals attached."
Then we are in agreement.
Two thoughts….
First, I was a family doctor for 40 years, and always tried to tell people to adopt an approach similar to yours. Accept that sleep varies from one day to the next, sleep when you are tired, and decide whether or not to get up when you wake up. Any attempt to “standardize” or regulate sleep will be doomed to failure.
Second, I retired a couple of years ago and went from a tightly regulated/scheduled life to one which is completely unregulated and unscheduled. My sleep has improved enormously. I go to bed when I want to, much later than I used to. I set an alarm simply so that, when I wake up in the middle of the night, I can tell myself the alarm hasn’t gone off yet, so no need to even think about checking the time or getting up. I still wake up before the alarm, but choose to stay in bed longer, even when it goes off. I’m probably averaging 6-7 hours of sleep a night, far less than what I used to think I needed. Most days I feel great. Some days I take a nap, if I feel like it.
I guess the lesson in this is that the laissez-faire approach works great if you have no specific obligations to be anywhere at any specific time. Trying to fit sleep into a societally-imposed schedule is counterproductive, at best. Sleep “disorders” probably reflect the fact that human physiology, sleep included, can’t be expected to conform to the averages.
Hi Rick,
I'm currently crunching some initial numbers that will become an update post of sorts. Are you ok if I cite this comment in its entirety in that post? I think it gets at the spirit of what I'm trying to convey really nicely and opens the discussion of sleep to obligations like jobs instead of obligations like "sleeping exactly 8 hours a night."
Thanks for the comment,
Violet
Sure, feel free to quote me.
There have been actual scientific studies done on melatonin use, and the most interesting melatonin studies are ones that measure the amount of melatonin in any given capsule that you might buy at the store. A 2023 study from the University of Mississippi found that 22 out of 25 melatonin gummy products contained different amounts of melatonin than the label claimed, with one gummy containing more than three times the amount and another containing none at all. Other studies have found that melatonin supplements can contain other substances, like serotonin, that may be concerning. Because the FDA can't regulate supplements like melatonin, some say that the pills may not contain what the label says, and that melatonin is a psychoactive hormone that shouldn't be taken without a doctor's guidance.
The only studies that show a positive correlation between melatonin supplement and sleep had so few participants that they aren't even relevant.
I wrote this same article about sleep, except I wrote mine about infant sleep. I love the idea that you know you best. Averages are informative, but they're not definitive.
https://open.substack.com/pub/beckdelahoy/p/some-babies-just-dont-need-much-sleep?utm_source=share&utm_medium=android&r=21hm6o
I also linked from Adam’s article - I wanted to add an anecdote re “programming” oneself to wake at a certain time. I regrettably forget where, but many years ago, I heard/read the following: if you want to wake up at a certain time, picture yourself on a cliff above the ocean. As clearly as possible, imagine yourself carving your desired wake-up time on a rock, then throw the rock over the cliff and into the water. For a while, I did this pretty often, and as I recall, it worked astonishingly well, consistently allowing me to wake at the time I had “programmed”.
I really need to revisit this, partially because it was a much gentler way to wake than an alarm, and because it raises some questions: assuming it does work as well as I remember, does that mean that the somewhat arbitrary division of the day into 24 hours is etched into my brain? Or is some part of my brain counting down? On a related note, is it more difficult to wake at odd times, like 7:47, than times like 7:30 or 8:00?
It’s kind of tangential to the main point of your article (sorry!) but does bear on it somewhat. Also brings up an issue involving a basic criterion for the “right” amount of sleep, the feeling of restedness: it does seem to be the case that the manner of waking, and when in the sleep cycle waking happens, also have an effect on that feeling.
I’d be interested to hear whether anyone else has ever heard of, or has experience with, a technique like this.
I have always believed that the idea of the normal distribution, though scientifically well studied, is epistemically underappreciated. I am a chronic sleeper, perhaps an hypersomniac. A secondary study I'd like to suggest is to compare those who fall on the extremes of either sides of the sleep distribution with certain traits such as energy level, moodiness, and emotional liability.
I’ve been a bad sleeper since childhood (and I’m well into middle age now). I don’t know how much I need, but I know that I rarely get enough—I remember the rare occasions when I sleep well years later. I do everything you’re supposed to (I get a lot of exercise, don’t use screens in the evening, spend time outside, very little alcohol, etc) but I still sleep poorly, often *very* poorly (like last night) and will try any (non-pharmaceutical) thing to make it better. So I signed up for your study. Thanks!
I've always needed 7 hours of sleep a night. I remember in early teenage years I'd get 7 hours of sleep a night. In late teenage years I got into the habit of staying up really late each night but sleeping in very late at the weekend, so I'd only get 5 hours each weeknight but 12 hours each weekend night. And then after many months of this working fine I realised 5*5+2*12=49=7*7. I was still getting 7 hours a night on average.
Now 25 years later I'm still getting 7 hours a night on average. And I was fine. Until a couple of years ago when I got diagnosed with Long COVID / chronic fatigue; but even now I'm taking extra breaks and doing more resting, I'm still sleeping around 7 hours a night.