Hacker News new | past | comments | ask | show | jobs | submit login
The myth of the eight-hour sleep (bbc.co.uk)
607 points by gps408 on Feb 22, 2012 | hide | past | favorite | 161 comments

I am really into 'do what works for you'. You may find that that sleeping in two 4 hour blocks changes your life. You feel alive!

Alternatively you may find yourself more tired. Personally I like sleeping 8-9 hours a night. I find myself fairly alert a few minutes after I wake up and I can start my day. It certainly doesn't feel 'unnatural' to me.

I am also a big fan of sleeping when other people sleep so I can enjoy time with friends and family. Unusual sleep patterns typically mean missing out on some of this time.

We often miss our sleep because we think we are missing on things.

You don't feel like you over slept after 9 hours of sleep?

Either way I have to agree with doing what works for you. I always made an attempt to sleep a full 8 hours and failed. I get by on 4-5 hours of sleep a night and feel fine!

I find my acceptable waking points occur in 1.5 hour sleep cycles, and how many I need depends on how mentally taxing the stuff I'm working on is. For ordinary work, stuff I know how to do fairly well already, I need 8 hours. If I'm learning something new or putting in long hours that require concentration, I'll need 9.5 hours. If I'm thrown into a completely new situation - say, I need to decide the architecture of my startup and learn 3-4 new frameworks all at once, or I've just moved cross-country and started a new job - I'll need 11 hours. I can also function on 6.5 hours but feel pretty crummy, and there's a noticeable performance penalty for anything that requires concentration (math, computer programming, etc). On 4 hours of sleep, I'm a zombie - I do okay with everyday tasks and social interaction (with people I know), but can't really learn anything new and retain it. Curiously, I feel better on 4 hours of sleep than on 6.5, 9.5, or 11 - I'm more alive and less groggy, I just notice that my performance on external measures declines.

I recall in the 2010 there were a bunch of athletes that all said the secret to their success - besides training hard - was getting 9 hours of sleep at night. IIRC both Sasha Cohen and Shaun White mentioned it.

I think I'd also read brain research that REM sleep (which tends to be back-loaded towards the end of the night) is when your brain converts experiences from that day from short-term memory into long-term skillsets. That'd be consistent with my experience that more mentally demanding tasks require more sleep, and the experience of those Olympic athletes, and that I don't really suffer a short-term penalty from lack of sleep, it's more a long-term problem.

But how do you do this? You wake up and go to sleep again and again? I wake up naturally after 7 or 8 hours, would be kinda hard for me to sleep 11.5 hours.

I've never had problems sleeping up to 12 hours. Might be a personal thing, or related to whether you have accumulated sleep debt.

I find that I'll often come up to a point of half-consciousness at about 8/9.5/11 hours, but as long as I don't sit up or get out of bed or start checking my e-mail on my phone, it's pretty easy to fall back asleep.

Whether I want to depends on how much mental activity I've been doing; I'm far more likely to think "That was a nice dream, I don't wanna get out of bed yet" if I've got a thorny challenging problem in my head than if I'm like "Gotta get into work and start coding."

I think it also depends how much light you have in the room. I think you would find it easier to sleep longer if the room was still dark after 8 hours.

I feel similarly to the OP, in that I need usually more than the 8 hours recommended. It sometimes seems people are critical when I tell them I feel fully refreshed only after 9+ hours of sleep, usually implying that I'm just lazy. Like most things in life, a person's optimal amount of sleep varies. Some people can make do on 5-6, whereas some need 9 or more. I'm personally toward the 9+ side. Definitely a "what works for you" thing.

As for segmented sleep, I'm not sure I could ever do that. If I even nap for like 30-60 minutes after 7pm, the chances of me falling asleep by midnight are pretty much nil. I'd be screwed if I tried to sleep from 8-12 and then again at 2-3am.

My sleep patterns are a lot like the parent commenter. I usually sleep about 7 to 8 hours, but occasionally sleep 9 or so. Do I feel like I've overslept? Sure in that I've "lost" an hour or two that I could have used. But my body doesn't feel like it's overslept. When it does, it tells me. With a half day of headaches.

'oversleep' is often dehydration.

What do you mean? Am I just dehydrated when I feel overslept, or oversleeping causes dehydration? The latter makes little sense to me, but I'm asking for clarification because this issue happens to me quite often.

> oversleeping causes dehydration? The latter makes little sense to me

Well, when you're awake you will get thirsty and drink if you overheat. If you're asleep you won't.

both. the longer you are asleep the more dehydrated you are, exacerbating grogginess.

What works for you isn't necessarily what works for someone else. 4-5 hours a night turns me into a zombie after about two days. My optimal duration seems to be about 8.5 hours, up to 9 hours if I heavily exerted myself the previous day.

I once got caught up on sleep over a period of several weeks. I then started falling into the segmented sleep pattern like in the article.

I didn't get out of bed when I woke up, but I found I was extremely lucid. As soon as I could use the time for problem solving, I stopped waking up :(

When you work extremely hard, beat the odds, you are too tired and haven't slept, and then manage to reach a milestone.

You suddenly begin to feel a sweet sense of pain. The feeling is just so beautiful that it feels amazing.

I've had that feeling quite a few number of times. And I will do anything to feel like that again.

Lucid dreaming?

There is only one study in this article, and it involves how a group of people adapted to a 14-hour sleep pattern. Other than that, there are no studies of importance here, nothing that confirms concretely that this kind of segmented sleep is effective for humans. It is based on historical hearsay but cannot make a prescriptive judgment. The evidence purely anecdotal. Please be careful in reading things like this that you do not immediately form a blind belief or justification.

I'm going to say something not so geeky but I think babies also sleep in segments and for shorter periods. We mainly adapt their "natural" habits to the society we're living. Is it for the better or not, I don't know.

They do, however that is as much to do with the need for frequent feeding given their tiny stomachs!

> Please be careful in reading

You should take your own advice. Bi-Modal sleep has a lot more evidence than just that, and the article includes some from historians, which I guess you missed.

This sleep pattern is well accepted, and has been known for decades. Not sure why they are writing an article about it now, except maybe the author heard about it and decided it would make an interesting article.

So quote them. Xurinos is rightly pointing out the lack of referencing.

By the way, I worked in sleep medicine, first as a medical tech, and then in technical support for the medical equipment. I never came across the idea that split sleep sessions was right or healthy, and never heard the sleep physicians or clinical specialists mention such a thing. It is far from "well accepted and known for decades", at least up until I left the industry in 2009 (though I spent less time with the specialists from about 2006).

Xurinos is presenting the correct scientific position; you are presenting the antithesis of science: "Yeah, you're wrong, everyone knows it and has always known it, I'm sure there's some research somewhere".

It's interesting that the references of yours that are scientific studies are all in the last 6 years (which is when I say I was more out of the loop). This is hardly "known and accepted for decades".

It was originally published in 1993, and you make it seem like if you didn't know about it it didn't exist.

Although in your defense, it's not widely known. (Most people never even think to ask, although once they do the evidence is compelling.)

... 'it' refers to which reference?

ref 1 seems to be a series of excerpts of articles ref 2 is a literature review of historical acounts

neither ref 1 nor 2 is a scientific study, and according to the dates come from around 2000

ref 3 wiki article referencing articles from 2005, ref 4 is science journalism and refers to a 1993 study

remaining refs are all studies from 2006+

So, yes, I didn't know it existed. I've never heard sleep professionals talk about it in my career to 2006ish. Good sleep was pretty universally considered a contiguous segment of uninterrupted sleep with particular cycles by sleep practitioners. My problem is not with the content - I'm open to learning about new things about stuff I 'know' - but with your description of it being something well-known and universally accepted for a long time.

I think it's fascinating, and thank you for providing references. But it's really a very new avenue of formally studied biology.

> and universally accepted

I didn't mean among practitioners, I meant among research, i.e. that there isn't significant criticism, which to me makes it accepted. It's normal for there to be huge lag between research and practice, especially since research is frequently wrong.

I remember reading about this many many years ago, and it was written as if it had been known for even longer before that. But I guess the recent references seem to start with 1993 which is long, but not as long as I thought it was.

I'll have to eat humble pie here. The kind of comment you made initially has always been suited to the criticisms I've raised, in my previous experience. It's unusual for there to be substance behind that style of claim - so I apologise for my abrasiveness.

Thank you for the links. It's a fascinating topic. Sleep medicine and the theory behind it is really, really interesting. Actually conducting sleep studies - attaching things, watching people sleep all night, then finishing off your shift smelling people's morning breath... is one of the most boring jobs in the world :)

Also, don't Spanish people and sailors sleep in segmented blocks? Not exactly newsworthy, then.

Sailors' sleep patterns are mostly dictated by their watch schedule. Different ships have different schedules (eg, three eight-hour shifts, or four six-hour shifts, etc) and basically you try to get sleep whenever possible.

I sailed on a ship that had six four-hour watches with three watch groups. That meant that one group would stand two 4-hour watches per 24-hour period (eg, 4am-8am then 4pm-8pm). With less than eight hours between watches, you had no choice but to sleep in two different blocks each day.

The effects of such a schedule and standing watch in a windowless engine room deep in the hull of the ship are quite interesting. For example, people (myself included) would completely forget whether it is PM or AM. If you occasionally go outside--rather than straight from the engine room to your bunk--you never knew whether to expect it to be light or dark out. I often went out expecting to reminisce under the star-laden sky, only to find out it's broad daylight out.

No, we don't. In Spain, some people take a quick 30-minute nap after lunch, but I wouldn't call that sleeping in segmented blocks.

Link to relevant article for people who are interested: https://en.wikipedia.org/wiki/Siesta.

Siesta (midday nap) is very common in Bolivia, Mexico and other Latin American countries, but I'd hesitate to call it segmented sleep.

Not to detract from your main point, but I don't think Siesta is very common in Mexico and most other Latin American countries. I was born in Cuba, lived there for five years and lived in Mexico for about three or four. I don't know of a single person who regularly slept after noon.

In fact, I had an opportunity to live in Barcelona for three months and while a lot of businesses do close down from noon to three, I got the impression that people were not using that time to sleep, and that the idea of siesta in Catalunya is on the decline.

Ok, I did work for Bolivia for a time, and our Bolivian office closed for siesta, as did our main customer (the La Paz city council).

Same for a more rural area of Mexico (Toluca surroundings), they also did siestas there.

Googling a bit, acording to Time magazine, siesta was struck down in Mexico in 1944, so it's more a myth than reality now I guess (probably subsists in some more rural areas).


My impression was that siesta still persisted in some countries, but that it tends to disappear as they conform to western work hours, and, as the Time article states, it does involve four commutes instead of two, so it's not workable in urban areas (btw, people who siesta still work the same amount of hours, I'm not saying they work less).

Observation: In the Philippines, most parents force their kids to take a siesta. These are taken around 1 or 2 in the afternoon. However, some parents fail to wake their kids up after 3 hours of sleep. In this case, these kids are up till the wee hours of the night, making their sleeping habits more complicated than it should be.

Regarding the article, I guess one should just do what works for them (their body). Some people find it necessary to get 8 hours of sleep, and some are okay with six.

I don't see how your post could be "also" to the parent post: It's saying to be skeptical of the claims due to lack of evidence, while your post contrarily implies that this is already established knowledge.

I think he's saying that the conclusion from the historical study (people used to have bi-modal sleep patterns) isn't exactly revolutionary either. So it's in line with the parent post in that there's not really anything noteworthy in the article.

Lots of things are noteworthy for people who don't already know them.

Like people now who just believe blindly now about the way they feel they should live their life by the norms?

It goes both ways.

The New York Times took on this topic a few years back in a very good article that argued that the whole idea of the 8-hour sleep was invented by the mattress industry (and other purveyors of sleep products), and that humans don't need anywhere near 8 hours of continuous sleep.

Ironically, all of the industry's marketing makes people anxious about getting enough sleep--and makes it harder for them to get to sleep (thus propagating the need for more expensive mattresses and pillows.)


Western consumerism is going to change human lives for bad.

Then again, it's already changed human lives for the better as well. You take the good, you take the bad, you take them both and there you have... the facts of life.

Oh my God. This could change my life.

One lifelong problem I've got is sleeping. I have great trouble falling asleep, and my sleeping patterns are very uncommon: some days I sleep 12 hours, some 5, but I'm usually very tired because of this. However, I used to think I had a "gift" of being really creative and having my best ideas just before sleep, just waking up, and during insomnia episodes. However, I've discovered that many suffer from this phenomena (anyone here?). This article could explain a lot!

I only found out recently (well into my adult life) that I am hyper-sensitive to caffeine. One cup of coffee in the morning will keep me up till 3 am at night.

This tricked me into thinking that I'm an extreme night owl, and it kept me from ever getting really good sleep.

Now, I've quit caffeinated beverages entirely, and I get tired around 11:30 pm every night, and I sleep fantastically well. I wake up naturally around 9:30 every morning and feel great.

All that is to say ... consider caffeine and other "drugs". You might be surprised how much they're affecting you.

Similar here, but I've found that as long as I don't drink caffeine after 3pm, I sleep well. Any later than that and I wake at 3am or so. Made such a difference when I discovered this... I lost a serious amount of grumpiness.

I must be seriously broken. I can even enjoy a cup of coffee at the bedside and still sleep. Not that I do that often.

Falling asleep for me is easy. I pretty much always wake up about four hours later. Then sometimes I remain wide awake for another few hours. After which I'm ready for sleep again. The second bout is pretty glorious. This does mean however, that I need to pretty much put 11 hours aside each night, if I want to play to my bodily demands.

My other half rather annoyingly insists everyone should wake when they rise. Which just isn't good for my mental health - or my programming for that matter!

i'm basically immune to caffeine, afaict.

i frequently come home late, drink a diet coke or whatever, and head straight to bed. i've gone back and forth between drinking a six pack of diet mountain dew a day and nothing and never noticed a difference. i can slam two bottles of Bawls (160mg) or eat 3 Fooshes (http://www.thinkgeek.com/caffeine/candy/6e27, >300mg total) and still take (and want) a nap.

if i gave myself anything close the amount of sleep i appear to be capable of taking, i'd never be up to get anything done....

I doubt you're immune, just addicted. You say you can go a day without caffeine. I challenge you to go 3 days with absolutely no caffeine. Absolutely nothing, so no decaf, no chocolate. I bet it makes you feel horrible. If so, you're addicted, like the vast majority.

Regardless of the nature of addiction in his specific case, I can assure you there are people who don't get the rush from coffee (even without prior exposure, that can possibly increase your level of tolerance).

Me for example, I feel very sleepy after drinking coffee. This was against my expectation and I blogged about it the second time it happened to me: http://0x40.blogspot.com/2007/08/wheres-that-forward-biased-...

I was born in India and my family does not drink coffee, hence I was never exposed to it during my childhood, except now and then, at parties.

It was only at work that I tried it out regularly and because of the way it affected me, I had to give it up.

I even joke that a few cups of coffee would probably help me go to sleep (but I have never tried it out in practice yet)

It has been 4 years since I blogged about it and I would probably have drunk less than 5 cups since then, with none in the last year, so addiction is out of the question.

Sugar works very well to wake me up, but I have to look out for and handle the subsequent crash later.

On an (un)related note, I also have tremendous tolerance to alcohol. I found this out a decade ago when I could legally drink and would experiment with various spirits and liqueur. Since budget was limited, each of us in a group has a limited number of shots, and I used to feel awkward when my friends used to talk about "that dizzy feeling" and I did not experience any. It was only later that I found out what was going on.

This also means my options to reach the "state of bliss" is limited, and the cheapest way for me to reach that state is meditation.

I am not sure if both symptoms are related, but someone who follows this more closely can perhaps comment.

I'm also the kind of person who can drink coffee at night, or can have a couple of cups of coffee after dinner, and not notice any effect on my sleeping. I'm definitely not addicted, because I rarely drink coffee on weekends - so the last three day period I went without coffee was this past weekend.

I usually go to bed between 9:30 and 11:00, and I drink my coffee black if that makes a difference. Maybe it's all the sugar that's keeping people up (or it could be that different people have significantly variant biochemistries; who knew?)

edit: I knew I remembered seeing this...



Abstract: "We explored whether caffeine, and expectation of having consumed caffeine, affects attention, reward responsivity and mood using double-blinded methodology. 88 participants were randomly allocated to ‘drink-type’ (caffeinated/decaffeinated coffee) and ‘expectancy’ (told caffeinated/told decaffeinated coffee) manipulations. Both caffeine and expectation of having consumed caffeine improved attention and psychomotor speed. Expectation enhanced self-reported vigour and reward responsivity. Self-reported depression increased at post-drink for all participants, but less in those receiving or expecting caffeine. These results suggest caffeine expectation can affect mood and performance but do not support a synergistic effect."

Not necessarily. When I was in high school, I drank between 4 to 6 cups of coffee a day. It never affected me. The reason for this is twofold: It's just part of the culture in my family in India, and I really liked the taste.

Fast forward a couple years and I am in college. I hate, absolutely detest the taste of any american coffee. I don't know what's missing, but I only enjoy coffee in India. I went from 4-6 cups a day to about 1 cup a week in a couple weeks. Nowadays, I don't drink much coffee at all, probably 2 or 3 cups a month at the most. But when I am in India, I drink as much as I can. :)

While the parent commenter may indeed be addicted, it's also possible that he is simply unaffected by caffeine. I certainly find that true for me.

Sorry, but enjoying the taste of coffee doesn't disable the chemical addiction and dependency inherent to its chemistry.

Ah, that was an unfortunate ordering of sentences. I was just giving an explanation for my consumption rate, not a reason for why I wasn't affected. I should have proofread that. I'm sorry.

i'll be happy to do so again. (well, not so happy about the chocolate part....) (i hate coffee on taste principles, so decaf isn't even in the picture, all my caffeine these days comes from diet soda and chocolate.) i'll get back to you on the results.

It may be the coffee that's helping you to sleep. Some dutch (based on the region) have the habit of drinking a good cup of coffee in the evening right before going to bed. I tried this a couple of times. Around 5-10 minutes after the coffee I felt incredibly tired and had a really fantastic sleep. It's as if the coffee helped me fall into deeper sleep. However, when I'd wait too long after having the cup of coffee, it'd take me hours until I could finally fall asleep.

Nice to see I'm not alone :) it's impressive to see how different are our sleeping habits. I have to hack myself one of these days.

I have to stop the caffeine around 5pm. If I have a coffee at 6pm or later, I can't fall asleep before midnight.

For me it's around noon, but then again I go to bed at 9pm and get up at 4:30.

For me it's TV. Especially bad for me is watching before going to bed. TV keeps me awake, stimulates me. After watching I've problems falling asleep, and the sleep itself isn't recreative. I don't feel really awake after such a night.

Thanks Charlie! How did you find out that you were hyper-sensitive? In my case, I don't drink colas, but I love that hot coffee in the morning...

Ideally, you should keep a sleep journal for 30 days before you try to change anything about your sleep habits. As to the actual experiment don't drink any tea/coffee/colas for 30 days and see how it effects your sleep patterns.

PS: And I do mean zero for the full 30 days if you have any caffeine you have to start over.

This could be a solution for me.

Drinking tea is part of the culture of my family, so I started drinking it at a very young age (as soon as I could hold a cup?) with 3/4 milk and 1/4 tea. As I got older, the amount of tea increased until there was no milk left. These days, whoever I visit, it seems I can always count on being offered a cup of tea.

When I started having all sorts of issues sleeping at a young age, it just didn't occur to me or anyone that it could be the copious amounts of tea. Doctors ran tests on my blood numerous times and didn't find anything. They asked me all sorts of questions (even if I was pregnant at the tender age of 15, which shocked young innocent me), but caffeine never came up. Strangely, sending me to bed earlier resulted in tossing and turning even longer into the wee hours of the morning. Sleeping pills just resulted in feeling strangely dazed, but still not sleeping.

These days I have stopped drinking tea/caffeine after a certain hour in hopes of getting more sleep, but I still drink a cup of coffee in the morning and I still have troubles getting to sleep (takes a couple of hours no matter hour exhausted I feel). Maybe it's time to cut out the caffeine completely.

Caffeine has really nasty withdrawal symptoms for most people. I highly recommend slowly cutting out caffeine rather than going cold turkey.

For me, it was gradual. I had had random bouts of insomnia for awhile, and realized those correlated with days I had caffeine after 4 PM or so. So I cut back to noon, which improved it, then cut back to the morning, which was even better, and then just cut out caffeine entirely. I've never slept better.

You're definitely not alone. Extraordinary insight and creativity, rich, vivid, inspired and out-of-the-box thinking, are typical for "twilight" states, when you're floating between waking state and sleep; not fully awaken to the outside world, but not fully unconscious either. The only problem is replicating and maintaining these states.

It is sometimes possible to remain fully aware in what is essentially a full-on dream state. This is the best of both worlds, combining the rationality and critical thinking of the waking state, with the unbound creativity of the dream. It's like opening up the firehose of new ideas, while remaining in control.

I am fairly certain that at least some of the unusual claims made by people who practice meditation a lot (such as having access to a higher "wisdom", etc.) are based on gaining a measure of control over these states. For this reason, I think it's worth exploring some of the traditional meditation systems, seen as empirical "brain hacking" techniques. I'd love to see more research done in this field; some of Sam Harris' work is relevant.

Thanks for the insight. As a matter of fact, I do meditate :) From my experience (nothing scientific), to replicate these states, what you need is to shut down your rational mind. You see, everyone has in there mind a small voice going "hey, I like this lamp" or "mmm I forgot to do something". This voice is your rational, conscious part of your "self". However, it's not your entire "self". Meditation aims at quieting this "voice" (it's really hard, and I'm not there yet). Once its quiet, you've got "space" for actual inspiration to come in, and once this happens it's amazing, and can happen in many levels. That's why meditation is a mystical experience. Mystical means that it is not possible to communicate via ideas, just by personal experience. If you get good at it you can actually control you emotional states and LISTEN to them, bringing you a lot of wellbeing in many ways. (ok, I'm waaay off topic, sorry).

That sounds pretty interesting. Any suggestion on books or ofher resources for peope new to meditation? How did you get started with meditation?

I like Alan Watts for meditation.[1] I have some mala beads as he suggests, and find they help me a lot. Its easier to time sessions, just count the beads with your fingers and when you reach the end of the mala, you can stop. They also help one to breath/count without consciously focusing on breathing/counting, or even worse, getting stuck in your head trying to focus on not focusing.

Regarding sleep, read something very interesting the other day about Delayed Sleep Phase Disorder (DSPD).[2] Something that might help is a program called f.lux that reduces the blue light in your monitor in the evening, supposedly making it easier it to sleep. I haven't tried it yet but intend to soon.

1. http://www.mediafire.com/?u53kedhwn5w

2. http://evincarofautumn.blogspot.com/2011/09/how-to-be-produc...

I'm far from being an expert, but I learned this type of meditation http://en.wikipedia.org/wiki/Transcendental_Meditation from a guy my mother hired to teach all my family. There are many different meditations and practices, and I've actually never read a book about just meditation but I recommend books by Osho and Deepak Chopra (just google'em) to have insight on Indian practices and philosophy. I would recommend to reach out any yoga institution around town and ask. It's better to learn from someone than from a book, and, as cliché as it sounds, if you are looking for it, the appropriate practice will come to you.

> as cliché as it sounds, if you are looking for it, the appropriate practice will come to you

Yeah. Keep looking for the right career / hobby / spiritual path / etc., and it will likely come to you. Provided you are sincere and are actually looking.

I've a different bias, myself, but to each his own. The various buddhist techniques also seem well suited for the western skeptics, since they typically don't require adherence to any theology at all. You could practice Zen and be a total agnostic or atheist. "Brain hacking" in its purest form. YMMV.

I find Zen difficult, myself. Feels too... um... dry, for lack of a better word. I require something that engages the emotional apparatus, and the physical base, in a more active fashion. I guess this is like preferring Python over Ruby over Perl over etc.etc.etc. Everyone's different.

Haha I never heard anyone compare Zen to Python. Point granted, anyways.

A few times in my life I've had these between state experiences. I'm not sure what triggered them or why they happened. And I can't recall fully the details of what I saw. However, I can remember the effect it had on my mind.

Once it was morning, but I was not awake. I saw a tablet with writing on it, infinite detail like a fractal going deeper and deeper. But I knew I wasn't awake, my eyes were closed. But the clarity was amazing. My eyes were closed and yet I saw like as if my eyes were open. I told myself this is what having photographic memory must be like.

Another time it was morning again and I thought I was awake. But I saw things floating in the air, like holograms of objects. And somehow I knew, I was still sleeping, but not sleeping. What I saw was a perfect memory recall of my room, a place I had seen thousands of times. But it was all in my mind and my mind could insert impossible objects at will.

Yet another time I heard the most beautiful music. I'm not a musician, and I can't remember anything about the music. I just remember it was amazing, like something bach would create.

These weren't like normal dreams which are forgotten soon after waking. They left a lasting impression, and even though I can't recall the images or the music, somehow I can remember the awe and wonder of the experience.

It is sometimes possible to remain fully aware in what is essentially a full-on dream state

Lucid dreams. I get these quite often, sometimes with full control, most of the time you wake up as soon as you realize you have control.

I sometimes get there by being sleep-deprived. It's pretty hard to get there, though. Sometimes you just fall asleep. And even if you feel creative, you might not have the energy to actually get anything done.

Also, dreams tend to fade rapidly. If you plan to use them to get ideas, it's best to write the ideas down immediately, before the ideas evaporate.

> I sometimes get there by being sleep-deprived.

Ah, yes. It's a little bit different. Clarity, calm, a kind of purely-mental energy, an ease to achieve good focus. Sometimes a positive, sunny feeling (unless you're completely exhausted or stressed, in which case it's nasty); other times quite an intense but smooth joy. But at the purely physical level, batteries are running out.

The twilight state has less clarity (it's not obscure, though), and calm doesn't enter the picture. But it's full of this roiling primeval emotional drive to create (in a very wide, general sense). It's where dreams come from, and most inspiring ideas. Some folks find it easier to experience it if they linger in bed in the morning, trying to avoid yanking themselves out of sleep directly into full waking state. It's a fragile thing, handle it gently.

This happens to me, too, but I'd gladly trade all the good ideas in the world for a decent amount of sleep!

Starting from about age 12, but definitely getting worse as I get older, sleep is something I just can't seem to get right, no matter how I approach it.

I've tried getting eight hours of sleep, and it doesn't work. I do pretty good on six hours, but only for a while. Over the course of a few weeks, six becomes four, which becomes three, which is then two, and suddenly it's 3 days later and I'm still awake.

When it gets to the point that I find myself getting less than four hours of sleep a night, the only way I can force my body to sleep more, despite being very physically tired, is with medication. Of course, most sleep aids can't be taken for long periods of time, so that only helps for a short while.

Sometimes I don't get tired until 4am, and other times it's all I can do to stay awake at 6pm, regardless of whether I got up at 6:30 in the morning, or noon.

I've tried enforcing a rigid schedule, meditation, exercise, various sleep aids, antidepressants (Pazil, Zoloft), just "going with it", and as of late, melatonin supplements at the recommendation of my doctor. They helped for a few weeks, but then that, too, stopped being effective.

The closest I get to a normal sleep schedule is that about half the time, I can go to sleep around 4am and get up at 10am - but that's not really a very practical solution and there are 3-4 days a week where it's just not at all possible from a logistics standpoint.

"I had a "gift" of being really creative and having my best ideas just before sleep, just waking up, and during insomnia episodes."

This is called Hypnogogia http://en.wikipedia.org/wiki/Hypnogogia

The only way I can definitely replicate it is to stay up all night. The next day I will feel a lot more creative, but at the same time a lot less coordinated, frequently tired and not able to concentrate on doing something complicated. Also it's super easy to forget things, so you need to write down all your ideas.

I dunno if you often for instance stay up very very late with friends talking about things, but I think there is a similar effect at some point.

I remember reading about a famous inventor (can't remember who, sorry) who purposefully induced this state by holding something in his hand which would make a noise when let go. While falling asleep, the object (say, a metal ball) would drop (into say, a metal bowl), making enough noise to wake him up just as he was falling asleep. If an idea came he would write it down, if not he would repeat the process.

Salvador Dali:


The comments on that mention Thomas Edison also doing something similar.

I cannot say anything about the sleeping patterns but are you sure that those creative bursts aren't happening just because it is quiet, no distractions, etc.?

Don't think so, because these come during the day in the most unexpected moments (with or without distraction). My theory is that all creativity comes from the subconscious mind, and during nighttime it's the subconscious that takes over. Just before sleeping, a gate opens where rational consciousness and unconsciousness "coexist" and that's when new ideas can be retained. It's like that guy that tried near-death experiences so he could write good ideas down.

It's pretty blissful when the sound of the traffic evaporates. Always the best time for coding for me.

My brother in law is a "sleep scientist" at UPenn.

His recommendation is that sleep cycles typically happen in 4 hour intervals so it's best to sleep 4 or 8 hours a night.

Getting up int he middle of a sleep cycle is often as bad as getting less than 4 hours of sleep.

And going to bed drunk is the worst for your sleep cycle.

I can believe this. One time I had 2 hours to sleep where I normally get 7 or 8 a night. I felt way worse the next day then if I hadn't slept at all. It was like my fight or flight response went haywire and I felt jittery and anxious all day.

> 4 hour intervals

What about those 90 minute cycles that are also commonly cited in popular sleep-advice? Those wouldn't work into a 4hour schedule.

Honestly i don't know what specifically you are referring to.

Are you referring to those sleep cycles where you have 4 90 minute naps a day instead of sleeping once at night?

I'm not the subject matter expert, so I don't want to speak authoritatively:)

He's talking about REM cycles, which last 90 minutes. Generally people recommend that you sleep for an amount of time that is a multiple of 90 minutes so that you do not disrupt one of these cycles.

I believe the grandparent was referring to the same phenomenon described in the relevant Wikipedia article at http://en.wikipedia.org/wiki/Sleep#Sleep_stages

I was talking about whatever contemporarily popular science apps and websites like http://sleepyti.me/ are buying into.

Ah, in that case I don't know, but I will ask.

Their latest research indicates that a full sleep cycle lasts approximately 4 hours, which does fly in the face of the older sleep data.

it's entirely possible that their hypothesis will turn out to be incorrect.

That's really interesting, do they have that published anywhere?

I'll ask him for links to his papers, but the bottom of the corporate site has a bunch of links:


Google ngram explorer has the term 'first sleep' peaking in use in English Language circa 1870, before declining ~80% to present day. http://bit.ly/zt8AfD

Tough to say if this was a 19th century phenomenon or was persistent through history based on the structure of their corpus.

It does not seem to be relevant, as it seems to track along with the general use of the word 'sleep'


Isn't this the proverbial midnight? You went to sleep for about four hours, then woke to have a midnight snack.

Makes me wonder whether we're now (mostly) sleeping through midnight because eating more carbs acclimates us to extremes of blood sugar and insulin levels, letting the midnight food craving pass unnoticed.

It would have been nice if there were some actionable items. What are the best segments? Is there something wrong if I can sleep 8 hours without trouble?

Does anyone have links to more research, perhaps with something a little more actionable of a conclusion?

The article suggests that it's natural to take two four hour sleep sessions, intermediated by a one to two hour break.

I think that this bears out to what many of us are experiencing, and the idea to consider is simply to stop worrying about it and embrace it. The 'second sleep' is likely getting cut short, because we are not budgeting for the break.

I consistently tend to find myself daydreaming during this part of the night, anxious (as the article notes) about not sleeping. So personally, I'm just going to continue doing this without worry, and budget time for it. Duh, right? Why didn't I think of that before? It makes perfect sense now after reading the article.

I may even go wild and actually get up and do things for an hour or two every night, you know, for science.

Many thanks to the poster.

I am going to try for a week or two. Having nearly 12 hours dedicated to work, I only have 4 exhausted hours in the evenings for eating dinner, spending time with my wife, kids and myself.

If my wife and I split our sleep cycle into two 4 hour periods we can better spend my four hours of free time.

The waking period between my commute and first sleep could be spend eating dinner, playing with the kids, and exercise. After my first sleep I can spend time with my wife, and study with a rested mind. Theoretically it seems like a good idea. We'll see how I cope after a week or two of trying.

For a long time I've wanted to cut myself off and figure out a natural sleep cycle. Normally (for better or for worse), I sleep 5-6 hours a night during the week, then get 8-9 on the weekends, but that's definitely shaped to the work week.

Also, maybe interesting - two 4 hour sleep cycle with a couple hours between is what happens naturally to me after a night of too much drinking. I wonder if there's some reason for that.

I believe I've read that all the alcohol processing eventually makes your blood sugar drop very low which is when you wake up (if you're like me, around 6-7am after going to bed at 3). Then of course, you're not very rested from little sleep (and not very high quality sleep) so you probably go back down for a few more hours.

Maybe that could explain why the segmented sleep pattern was considered normal way back when. I don't imagine there was a whole lot for people to do in the evenings besides drink =)

I thought that was common knowledge - that with drinking - you have in general poorer sleep. It's rather insidious.

Following Carnival last night, I suffered this horribly. Thanks for the explanation.

Sleep is weird. If I wake up in the middle of the night and go back to sleep... I feel like ass the next day. If I get less than eight hours I'm pretty tired the next day.

But science says that shouldn't happen!?

Not enough evidence to suggest that there's any kind of regularity to how humans "should" sleep yet.

It is even more weird than that. From my and some friend's experience: if you sleep more then 9 hours straight, instead of usual 6-8 (this happens on weekends usually), I feel really terrible in the morning - not being able to get out of bed, but in the same time not being able to sleep. Finally when I get out of bed - I feel really trashed for next several hours.

So it is not just less than 8 hours makes things worse, same applies to more then 8 hours. So it is either I have to sleep 6-8 hours only, or I have adapted to this schedule so well, that anything else makes me feel terrible.

On the other hand, I have handful examples of me traveling with few flight changes, timezones changes, more than 30 hours in flights/airports, and after getting to my destination I spent full 16 hours day working or meeting with friends just fine. May be a little dizzy. And after consequent 8 hours sleep night - feel great next day.

This suggest, that it (8 hours sleep requirement) is more than just a something "hardwired" in to human's body...

Yes, sleep is weird.

Often when you wake up after sleeping longer than usual (say more than 9 hours) you feel tired because of dehydration. Your body loses a fair bit of water while you sleep, so that might be reason you feel bad after a lie-in.

I have anecdote to confirm this. I stay well hydrated, and drink more water than most people I know, will generally sip on a glass by my bed, and have a glass of water by my bed at night. I have also been able to sleep over 12 hours (generally waking up at least once to use the restroom and water) Yet I feel fantastic when I do such!

This is why, I find that getting up at the same time on the weekend as I do during the work week actually helps keep my sleep patterns somewhat regular. So, although the idea of an occasional Saturday morning sleep-in sounds good, in practic, I'm better off getting up at my usual 4:45am every day and supplementing with a short nap after lunch if necessary.

> If I wake up in the middle of the night and go back to sleep... I feel like ass the next day.

If you wake up because you have to, or if you just happen to wake up in the middle of the night naturally?

Just get rid of your alarm clock, and wake up when your body tells you to.

I tried this for one semester in college, and never went back to an alarm clock. I decided to face the risk of missing classes, exams, meetings, etc. The result? I got up 1-2 hours earlier most days than I did with an alarm clock. I have been living without an alarm clock for 20 years now; in that time I have overslept maybe 5 times, and on those occasions it was pretty clear that my body was fighting off sickness and really needed the sleep. The only time I use an alarm is if I need to get up at 3:45 am to catch a 6am flight.

Alarm clocks are a terrible habit. They wake us up with no regard to where we are in our sleep cycles. When I used an alarm, I used to wake up and go back to sleep until the alarm went off. Then I would hit the snooze button forever. With no alarm, I just get up the first time I wake up in the morning. I almost always wake up at the end of a good sleep cycle.

Getting rid of my alarm clock decreased my overall amount of sleep but greatly increased the quality of my sleep.

I tried this, and I ran into a lot of difficulties. First, I assumed that I could go to bed whenever I wanted and I would wake up on time no matter what. This was not the case. Other times I would wake up on time but go back to sleep. And sometimes I just never woke up on time at all.

This approach works for many (my grandfather used this his whole working life, but not for me, sadly.

Not setting an alarm made me more conscious of my decisions around sleep. If I had to get up at 5:15 for rowing practice, and I had an alarm set, I stayed up as long as I wanted and told myself the alarm would wake me up. When I got rid of the alarm clock, I could no longer fool myself, so I started keeping a more reasonable schedule.

The other thing that helped at first was repeating in my mind, in a relaxed way, what time I wanted to wake up. I found I'd wake up within a few minutes of the time I wanted to wake up, consistently. Our brains keep track of time, and to some degree it seems we can tap into that.

That sounds pretty cool, I'll give it a try!

hm, sorry to ask, but: are you employed?

most folks I know, including me, have serious problems with sleep. it seems like we tap into each corner of human body and understand less or more how it works, but yet there is no real science behind how sleep works and how much of it we need. Heck, afaik scientists still argue why we need sleep at all and whats the main reasoning behind it. But not to brag too much: sometimes it happens to me that I sleep less hours and am more refreshed than sleeping longer, or even legit 7 or 8. I guess I just hit the right spot at the right time and REM phases worked out. If someone comes up with a device that can help you sleep not too long not to short so you always wake up fresh, I think he/she will nail a billion dollar idea!

I am. Actionable item: set your alarm 15 minutes later than you would normally wake up, go to sleep 15 minutes earlier than you normally do. If you do not wake up before the alarm, then leave it as is, and go to sleep 25 minutes earlier the following day. I am a firm believer your body is awake before your alarm rings.

I've found if I go to bed early enough and focus on rising at the same time every day (this includes weekends), after a few weeks, I will begin waking up at that time. It's nice to have an alarm, but waking before the alarm is exceedingly restful.

I'm employed. I don't use alarms. I just go to bed early enough so that I'll naturally want to wake up in time for work.

so you should say that.


> Just get rid of your alarm clock, and wake up when your body tells you to.

means you are telling me and everyone else to get a rid of your [mine] alarm clock.

Guess not everyones body acts the same, right?

Sure, that's the best way, but most people cannot do it; they risk being late for work, or in my case, getting the kids to school. (Actually, to be honest, my only risk is in not having as much time as I'd like before the kids wake up.)

A good second choice seems to be using something like 'Sleep Cycle' for iPhone, or one of these fancy new wristband accelerometer gizmos. I've had good success with Sleep Cycle, myself.

> I've had good success with Sleep Cycle, myself.

I'm really liking the Sleep Cycle app, too. Works great, simple, best 99 cents I've spent. Prior to it, I had the $69 WakeMate wristband, but never warmed up to it – it was uncomfortable, and a pain to set and keep charged.

I have an alarm clock, but it's only for emergencies.

For a long time, I used to wake up about 2 minutes before the alarm. Every single morning.

Lately, though, I've been waking up 20-60 minutes early. I just shrug and decide whether I feel like I need the extra time, or if I should do something with it. I usually get up and do something.

I have never slept for 8hrs, not even close. I sometimes wonder how people can sleep for that long and what their jobs and life's must be like. After 5-7hrs I usually wake up feeling totally refreshed and ready to go (even better after my first coffee ;))

You are lucky :) Some math - 1 hour a day less sleep than "normal" people give you:

1 hr * 365 days / 24 hours in day ~ 15 days.

And these are 24hours day. What it really should be:

1 hr * 365 days / (24 hours in day - 6 hours you sleep on average) ~ 21 days.

Which is, well, in office terms, work month :) congrats, you got 1/12 life for free :)

I don't think sleep can be counted out of life.

Dreams are the most beautiful things to ever happen to humans. I have always lived two lives one in dreams and one while am awake.

And some experiences that I've had dreaming just can't be had while I'm awake.

This is great, I've always worried about being or feeling awake for long periods at night. What doesn't seem clear though is whether I should be setting aside 10 hours a night for sleep (two 4 hour cycles + 2 waking hours in between)

There's a Zen saying relevant to the article (and to much of the discussion here). Rather than wringing your hands over "optimal sleep", I suggest trying it out. If your schedule allows, it can work wonders:

When hungry, eat. When tired, sleep.

Interesting article. I usually sleep from 7pm - 10pm, wake up and do stuff (like read HN), then sleep again from 3am - 8am, with just enough time to get ready for work in the morning.

Not too long ago I suffered from minor insomnia and lucid dreaming. It was awful, to the point where I started to dread bedtime. For me, my diet at the time (fatty foods + sugar + caffeine + alcohol) had a lot to do with it, so once I committed to a healthy diet, my body settled back into a consistent sleep cycle. I guess it's really what works for you.

I experience two sleeps sometimes when I'm not alone or when I drink a lot. If I'm alone, I usually read for an hour, and if I'm not alone, reading is the second choice. I always go back to sleep easily and get my full allotment. Until I read this article, it never occurred to me that I might be able to have that experience every night. Honestly, it sounds wonderful; sign me up to be in the vanguard. Now if only I could figure out how to make it happen when I'm alone and not drunk!

This would be an interesting opportunity for Zeo or Fitbit to chime in on this if they were to mine their data, anonymize it and publish something about their users sleep patterns.

There are a few mobile sleep timer apps that could also contribute data.

According to my Bodymedia Fit, I almost never ever get a true 8 hours or more of sleep. Sure, I lay down at 10 PM or so and get up at 6 AM or later, but if this little device is as accurate as they claim it really opened my eyes that I was never getting 8 hours, when I thought I actually was. It says normally I get between 6 and 7 hours.

Edit: I'm curious what the recommended amount of sleep, or if it really is an individual thing per 24 hour or so period.

There is some problem with the title of the article. It could also mean that you dont need 8 hours of sleep. Other point is the need for sleep varies with age. Babies sleep for very long time like 14 hours; old people need only 5 to 6 hours of sleep. The 8 hour sleep is for youth between 25 and 40 i think. Otherwise, very original point, I have never thought sleeping in 4-hour chunks was natural.

Not that I know anything about this, but isn't it possible that having this segmented sleep could possibly be bad for you? Just because our ancestors slept as such doesn't mean it's necessarily good for you, does it? (That's not to say it couldn't be good, just that I don't understand why the article seems to imply that it necessarily must be good.)

Has anyone ever read http://www.supermemo.com/articles/sleep.htm or put it through any verification? It looks good but I'm no biologist, I've used it as my basis for years to avoid polyphasic schedules and embrace naps and biphasic sleep when my life allows for it.

I have this weird sleep pattern, where I sleep from around 5pm-7pm or 7.30pm without an alarm clock, I am sleeping very deeply, always dreaming and really hard to wake me up (I don't notice the phone ringing right next to me) and then I'm sleeping from 1am to 6am, where I usually don't remember dreaming and where I wake up very easily

Interesting. Some of the lucid dreaming techniques try to capitalize on the split sleep schedule by taking melatonin before the first segment(induces a REM-less, deep sleep for the most part) and other supplements (vitamin B6, etc) before the second segment, resulting in a longer, more vivid dreaming session.

"In the early 1990s, psychiatrist Thomas Wehr conducted an experiment in which a group of people were plunged into darkness for 14 hours every day for a month."

'Plunging' people into 14 hours of darkness per day? Does this sound like a scientifically sound way to determine the natural sleeping patterns of humans?

Subtract artificial light, and how long would you be "plunged" into darkness every night this time of year at your latitude? Enforcing prehistoric lighting conditions at night and observing the resulting sleep pattern seems like a decent of finding someone's "natural" sleep pattern to me...

My sleep is a lot more deep when I regularly sleep around 7 hours. Anything less or over makes me restless. I think I am going to try this. If I get up in middle of night, I will work for an hour or two and when I feel sleepy, I will go to bed.

I've kept a sleep pattern of 4-5 hours by accident, and I don't seem to function any less efficiently than when I sleep 8 or more hours. I do end up taking a nap on occasion, but that's still 6-7 hours, not 8.

Some recent studies showed that an healthy night seems to be no more than 7 hours.

Segmented sleep is interesting but I don't think a majority of people could balance that with family and work.

It rather seems that family and work need to be adjusted to balance them with our natural sleeping rhythm.

Assuming that this is in fact out natural sleeping rhythm.

A great page about sleep cycles is this one http://www.supermemo.com/articles/polyphasic.htm

Hah, the Uberman's sleep schedule (six 20 minuted sessions totaling two hours of sleep per day). I remember trying that for a few days after seeing the kuro5hin article way back when. Supposed to take two weeks to adjust, but it was so painful after a few days that I lost faith and feared I getting caught in a hoax.

Really? I think this is BS... I don't care about statistics or a 1595 painting or whatever.

What I know is if I sleep less than seven hours I feel like crap during the day.

I seem to operate best on a 6 hour block and a 3 hour block. The problem is that doing this my body wants to have a total cycle time around 26 hours.

The guy who did this study spent 20 years. He found 500 'references to segmented sleep'. One of which being a painting depicting people not sleeping and sleeping simultaneously.

That's less than one reference every 2 weeks. Either that guy had the most relaxing job in the world, or it is genuinely difficult to find such references. Assuming the latter to give the researcher the benefit of the doubt here; how deep do you have to look before you start to think that maybe this isn't actually that common after all?

Also, how the bbc title reflects this study is beyond me.

So if one wanted to try getting into this cycle, do you set an alarm after 4 hours or if you wake up at all naturally in the night, stay up?

I'm always tired and sleepy, doesn't matter how many hours I sleep. Taking pills to sleep make me fill dizzy all day long :(

While we are on the subject of sleep, does anyone here experience sleep paralysis?

I had 10 hours last night. 8 til 6

now corporations want you to work more than 9 hours per day.

Especially in Asian countries it is considered normal that your workplace demands more than 8 hours. Becoming like that in the USA as well for some.

In Europe, or at least Nothern Europe, it feels like working days are actually becoming shorter. I seem to recall 37.5hrs/w being standard for most office workers, at least in Sweden. Will have a look and see if I can find a source later!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact