Hacker News new | past | comments | ask | show | jobs | submit login
Sleep Deficit: The Performance Killer (hbr.org)
152 points by hackly on Dec 18, 2011 | hide | past | favorite | 41 comments



I know this is anecdotal but I feel it is worth sharing. I've had insomnia for several years. I've tried treating it with drugs, sticking to routines, exercise, meditation, diet, controlled lighting, and quite a few other methods. Eventually, I just stopped looking at it as a problem. I sleep about 5 to 6 hours a night now and I feel great. I just accepted I don't have the same sleep patterns as everyone else and moved on with my life. I've read about reaction times and how sleep affects cognition but I really don't believe it, for me. I've tested performance by going shooting and recording scores after various amounts of sleep and found that after more than 7 hours of sleep a night I get worse scores.

Science is averages, you might be average or you might not. Test your own performance and find out what you need. If you only need a few hours sleep and your performance is not affected, do that. Many people with insomnia worry about it or try to fix what might not be broken. You might be an outlier and not the average.


> Not every sleeper is the same, of course: Dinges has found that some people who need eight hours will immediately feel the wallop of one four-hour night, while other eight-hour sleepers can handle several four-hour nights before their performance deteriorates. (But deteriorate it will.) There is a small portion of the population — he estimates it at around 5 percent or even less — who, for what researchers think may be genetic reasons, can maintain their performance with five or fewer hours of sleep. (There is also a small percentage who require 9 or 10 hours.)

Don't try this unless you've actually done some measuring and testing; overconfidence and biases are more common than tests:

> Still, while it’s tempting to believe we can train ourselves to be among the five-hour group — we can’t, Dinges says — or that we are naturally those five-hour sleepers, consider a key finding from Van Dongen and Dinges’s study: after just a few days, the four- and six-hour group reported that, yes, they were slightly sleepy. But they insisted they had adjusted to their new state. Even 14 days into the study, they said sleepiness was not affecting them. In fact, their performance had tanked. In other words, the sleep-deprived among us are lousy judges of our own sleep needs. We are not nearly as sharp as we think we are.

https://www.nytimes.com/2011/04/17/magazine/mag-17Sleep-t.ht...


I've suffered from depression, and random crap like insomania. For years, I'd rely on the doctor and follow the medications. These just made my life and sleep much worse.

Then one day, I ate a cookie that had been cooked with marijuana. Suffice to say, I slept a lovely 8 hour sleep that night - and the greatest thing? I've only ever had to use marijuana a _few_ times in the last 3 or so years. So, it seems to work long-term...

I've figured it's probably better for my health to stop switching weird medications every 3 months, and have a brownie every now and again.

I'm a much healther and happier person by far.


Interesting has there been any research done into this application. I have friends who have a little bit and just fall asleep (am not sure the effect on myself as I can't smoke and no one has ever offered me a hash cookie/brownie).


> Interesting has there been any research done into this application.

You mean using marijuana to treat insomnia (or sleep disorders in general)? Quite a lot, actually - marijuana is one of the most studied drugs in existence, and its effects on sleep disorders have been documented thoroughly. It depends on the state, but some of the states with medical marijuana laws accept sleep disorders as a valid use.


Yeah, I think this is fairly common but mostly not emphasized in the current debate because it's not as politically sympathetic a position: terminally ill patients using marijuana to reduce nausea make better example cases.


I have a 26-hour circadian sleep-rhythm, and so each night I go to bed 2 hours later than the night before. It made school years absolutely insufferable, but now I have a job where I can set my own hours and I just go with it. I'm not able to do certain things like shop during my "awake during the night" phase, but it usually works out.

I'm not sure if I have much to contribute other than "me too". I can confirm though that in my experience, drugs, routines, etc just don't work for me.


I was also going to note that I have worked shifts that are hard on sleep. When working an overnight shift, for example, I found it impossible to sleep more htan 4 hours per day (day because you get to sleep maybe 8am).

The funny thing is the impairment isn't linear throughout the day. There are times when it is really bad (1-3am) and times when it is pretty good (dawn, during the daylight hours etc). In general the only times I found myself impaired was at work. I have talked to people who worked in other industry overnight and their experience was the same.

As far as I can tell, without tremendous amounts of caffeine, the brain seems to go partway to sleep, able to recover from approx. half of the outstanding sleep debt per day. Part of the problem is the way in which we use caffeine to compensate.


How can you measure how long your circadian sleep cycle is?


The time you wake up one day, to the time you wake up the next, averaged out over a long period of time. For most people it's right around 24 hours, as it should be.

Circadian technically means every 24 hours, so a better word may have been just "body clock": http://en.wikipedia.org/wiki/Circadian_rhythm#Criteria


There definitely are people who do just fine on 5 to 6 hours a night. I've known a couple.

Alas, I'm not one of them. I really need 8.5 to 9.


I'm exactly the same. I'm in my 40's now but have always been this way; the big difference that age has brought me is that I'm less flexible about it. In college I could get by with a few 3-4 hour nights in a row, as long as I made it up (and all of it) over the weekend, but now I can't carry a deficit nearly as well, so I don't.


I hear you. Even though I am still in my late 20s, I still need 8-9 hours of sleep a night. Anything less than that and I feel like I was hit with a trunk the next day.

I really wish I could get by on 5-7 hours like so many of my friends do, but after a while I just had to accept I couldn't physically do it.


In your late 20s? I raise it to my late 40s.

I often sleep six or seven hours on work days, but I have to recover in the weekend.


I think the best quote about this, during reports from Prof Jim Horne about people people different in this respect: "In a nutshell, if you sleep for eight hours a night go to work and find yourself lolling and drooling on the keyboard, you aren't getting enough. If you're sleeping five hours and running the country, you probably are getting enough." The latter was in reference to Margaret Thatcher who seemed to function fine with about 5 hours sleep at night.


Margaret Thatcher used to take vitamin B shots to compensate. An a chronic insomniac myself, I really wish this macho, superhuman bullshit about sleep would just die already.

I try to get 8 hours and still struggle. Apparently there is a small fraction of people who manage with tiny amounts of sleep because they are geneticically different[1], but there is no use in using that as a standard that 95% of people should live by.

[1] http://www.scientificamerican.com/article.cfm?id=genetic-mut...


Wait, is 5-6 hours a night considered insomnia? Or am I misreading you, and you used to sleep less and have improved?


I used to sleep between 0 and 14 hours a day (completely irregularly). Now, I sleep a fairly regular 5 to 6 hours per day.


Doesn't just apply to adults: highschool students do not get anywhere near enough sleep [1]. Some schools have tried moving to a later opening, which has in some cases led to better academic performance from the students, but most schools still start god-awfully early.

[1] My own experience, may not apply to everyone, everywhere.


Not just your experience: http://www.mayoclinic.com/health/teens-health/CC00019

Your circadian rhythm changes in your teens, and too few schools are willing to work with this rather than against it. The results are sleep-deprived kids doing crap work in school and getting into car accidents like the one mentioned at the start of the HBR article.


I've collected some citations on the topic: http://www.gwern.net/education-is-not-about-learning#school-...

(Too bad that there doesn't seem to be any large authoritative collection of relevant material.)


This page, and your entire site in general, are very interesting. Thank you for taking the time to compile all of these links, I'll have to start sending them to my teachers :)


Well, it's not finished\; so if you are going to send'em to anyone, might I suggest you pick particularly relevant citations and just link those instead or something like that?

\ You obviously have no way of knowing this but my convention on gwern.net is anything in lower-caps is draft/incomplete, and so I don't link it unless it's extremely relevant (as it was in this case).


I've always thought it was ridiculous how early school starts. I had to get on the bus at 7:00 when I was in high school (before I started driving).


My earliest lessons in high school started at 6.55 before you even got on the bus. (But then, our schools in Germany are also at the bottom of coal mines..)


Wow, what's a normal time to start work in Germany?


The fact that sleep is not really valued in the "start up" community is a disgusting reflection of stupid people. SLEEP MORE.

A great resource is a book called `Lights Out! Sleep, Sugar, and Survival` by T.S. Wiley. They book eventually gets into some controversial territory without much data to support the claims, but the first half or so is largely about sleep and invaluable. There's a ton of references for the sleep section.

Tl;dr - get the book I mention and sleep. Human sleep cycles change with the seasons and you shouldn't be up too much outside the realm of daylight if possible.


> People in executive positions should set behavioral expectations and develop corporate sleep policies (...) It’s important to have a policy limiting scheduled work:ideally to no more than 12 hours a day, and exceptionally to no more than 16 consecutive hours.

This sentence seems like having been taken straight out of Marx's "Das Kapital", volume 1. I know this is the Harvard Business Review, a bastion of capitalism if there ever was one, but I gathered that business owners had smarten up a little in the last 150 years.

If they don't want to do it for the benefit of their employees, they should do it for the benefit of their family and business friends, because at some point or another some guy like Lenin or Trotsky could come again and ride the wave of discontent and the consequences would not be pretty.


This article was written for an audience of pointy-haired bosses, and so the meh scientific insights are followed by groundbreaking! executive! decisions!


Sleep certainly is important but things like "Educational programs about sleep, health, and safety should be mandatory" and "Additionally, companies should provide annual screening for sleep disorders in order to identify those who might be at risk" sound like well-intentioned wastes of time.


99% had a great video from Tony Schwartz just last week:

http://the99percent.com/videos/7110/Tony-Schwartz-The-Myths-...

Basically, he believes that mental exercises is just like physical exercise. During rest periods is when your capacity to do the exercise expands, not during the exercise itself.


Be ready for this when you have kids. Your performance will drop right off for the first 12-18 months or so for each child. They disrupt your sleep and make your day physically and mentally longer and more demanding. The number of crazy things I have done in the car through chronic sleep deprivation is scary. I love my kids to bits by the way :-)


This is completely true. I was focused completely on work when I was at work when I had no kids. My first kid brought that building completely down. I would sit at my desk staring at the screen not knowing that time was even passing. I finally had to force myself to sleep at lunch in my car in the parking lot so I could focus enough in the afternoon to not get fired. Luckily it passed and I started getting sleep after a while. It is amazing the stuff you think and do when highly sleep deprived.


As I sit here at nearly 3 AM trying to finish something for Monday that we over-promised a client...


I have to say, this is one of the goals were wanted to achieve when I made SleepBot (Android). Sleep is such a big issue especially for us programmers. We often not aware of how much sleep we actually lose everyday and how much performance it kills.


Seth Roberts' sleep experiments: http://escholarship.org/uc/item/2xc2h866#page-1


In corporations this is a classic prisoner's dilemma: If EVERYONE agreed that they'd be sane and work optimal 40-50 hour weeks, it'd be great. But someone always 'defects', going to 80+ hours in an effort to show off. The result is actually lower efficiency, but since most corporations do very little track efficiency and results (they track effort, instead), it doesn't get caught.


I think I must be doing my bit by making sure I don't ever go over my 40 hours, even if I work in teams where others or the manager works long hours.


I wish I could upvote that at least twice - once for the game-theory bit, and once for the tracking-effort bit. Those are both excellent insights.


Thanks. I actually run a company - I have to constantly watch people and make the effort/results separation. Folks never understand when they get a negative review "for putting in 12-hour days." See, it's because you spend 4 hours/day running in circles.


summary: get enough good sleep




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

Search: