Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

When Willpower Attacks

17 Post author: jimrandomh 03 October 2009 03:36AM

Less Wrong has held many discussions of willpower. All of them have focused on the cases where willpower fails, and its failure causes harm, such as procrastination, overeating and addiction. Collectively, we call these behaviors akrasia. Akrasia is any behavior that we believe is harmful, but do anyways due to a lack of willpower. Akrasia, however, represents only a small subset of the cases in which willpower fails, and focusing on it too much creates an availability bias that skews our perception of what willpower is, how it works and how much of it is desireable. To counter this bias, I present here some common special cases where strong willpower is harmful or even fatal.

Consider the human sleep cycle. By default, we settle into an equilibrium in which we sleep for about eight hours per day, starting at about the same time each evening. We can override the normal sleep schedule, and stay awake when we should be asleep, or sleep when we should be awake, by applying willpower. For this example, we'll confine the definition of willpower to just this one ability. Willpower is the ability to control sleep times; we'll leave the ability to work hard and resist cake for later.

We use sleep-willpower to arrange our hours strategically. However, we can only do so within certain constraints, which cannot be overriden by conscious choice. It isn't possible to refrain from sleeping entirely, because the amount of willpower required to stay awake increases with the amount sleep missed, until eventually it exceeds the amount of willpower available. Attempting to minimize sleep for a long enough period will eventually cause a shift to an alternative equilibrium, polyphasic sleep, in which all but the most important stage of sleep (REM) are skipped. It's harder to stay awake in some circumstances, like dark rooms and boring lectures, than in others. And it's impossible to sleep too much more than normal; the better rested we are, the harder it is to fall asleep. These restrictions are protective; when rats are artificially prevented from sleeping for too long, they die, and when humans use stimulants to stay awake too long, they die too. If a human had unlimited sleep-willpower, then they would risk serious injury every time they used it.

Next, consider diet-willpower. Humans have a system that decides what is good to eat, which manifests as cravings and taste preferences. Many people try to override this mechanism, usually for health reasons, replacing them with advice from a perceived authority. Consider a hypothetical person who decides to follow a highly restrictive diet, to lose weight. Unfortunately, most of the diet advice available is bad; diet advice available in the recent past was very bad; and some of the diet advice floating around is disastrously bad. Restrictive diets tend to have harmful deficiencies. If that deficiency is one of the ones that our biochemical diet manager knows how to handle, then following the diet will require willpower. The more severe the deficiency, the more willpower will be required. Try to follow a no-fat diet, and lapses will be the only thing keeping you alive.

In these examples, we have a mechanism which provides defaults, and a mechanism for overriding them. In general, we call the defaults "System 1", and the override mechanism "System 2"; and whatever influence System 2 has over System 1, we call willpower. System 1 represents a general policy, and System 2 adjusts it for the present circumstances. This pattern applies across many different and sometimes unrelated mechanisms, and we use the term "willpower" for all of them. The larger the adjustment, or the longer it is applied, the more willpower is required; and the different implementations of willpower - sleep-willpower, diet-willpower, work-willpower, etc - all require resources, such as glucose and focused attention, which are finite and fungible. When willpower fails, it usually "breaks" - that is, it fails all at once - rather than failing gradually. Deviating from innate or habitual behavior requires willpower. Willpower is reduced by pain, discomfort, and many types of biochemical imbalance. Sustained application of willpower may adjust habits and produce a new equilibrium, but if no such equilibrium exists, it will eventually fail and cause a reversion or overcorrection in the opposite direction.

Willpower lets us adjust our behavior to match our decisions, and willpower failures protect us from mistakes. Generals often successfully convince their soldiers that they should be courageous and always fight to the death, but when stressed enough, they'll break, desert or surrender, and survive. For soldiers on the losing side of a battle, excess willpower is fatal. Teenagers sometimes pledge to stay celibate. Their psychology is designed to make sure they don't keep that pledge. Married couples pledge monogamy, but women with infertile husbands and men with infertile wives tend to cheat. When stubborn people argue, one of them has to give up and end the conversation eventually. The sooner that happens, the less time they waste and the fewer black eyes they get. Workers may be convinced to keep a frantic pace, but sheer laziness will keep them from working hard enough to injure themselves.

The next time you see a willpower hack, remember that willpower isn't always a good thing. Sometimes, what we think of as akrasia is actually protecting us from harm.

Comments (77)

Comment author: Psychohistorian 03 October 2009 04:13:11AM *  18 points [-]

This is interesting, but I think it could essentially be summarized in two sentences:

"Willpower is good if used to fulfill a healthful or desirable goal, and bad if used to fulfill an unhealthful or undesirable one. In other words, willpower, in a strict sense, is like any basic tool - however sharp or strong it is, its value is determined by the tasks to which it is applied, and the context in which those tasks are completed."

Comment author: Bongo 03 October 2009 09:35:05AM *  12 points [-]

plus "Most people would kill themselves with unlimited willpower. Maybe willpower is limited so that one can't pursue bad goals too far."

Comment author: DanArmak 03 October 2009 09:40:22AM 5 points [-]

This is the theory saying that if people really ran their own lives, they would mostly run them into the ground. The power of our conscious selves seems to be severely limited for sound evolutionary reasons. It is restricted by emotions like fear and love, self-deception due to signaling, akrasia, conservatism and authority memes, and so on.

Comment author: Aurini 03 October 2009 10:42:55PM *  10 points [-]

If they actually developed a will-power drug, can you imagine how fast the legislatures would move to ban it? Within the first months you'd have dieters and athletes dying, college kids cramming for eighteen hours, and working class 'victims' who chose to slave away at their crummy jobs (which they're already doing with meth amphetamine addiction).

Still, it might wake people up to realize what a brain is exactly.

Comment author: Psychonaut 04 October 2009 12:59:07PM 4 points [-]

Arguably, clubbers taking ecstasy and dancing until they die of hyperthermia are doing exactly this.

From my experience, dopamine agonism can significantly improve my ability to exert willpower. That does not mean it allows for unbounded enhancement, however: psychoactives merely allow for intertemporal transfer of particular cognitive functions, due to downregulation of receptor sites. However, just like credit allows for intertemporal transfer of purchasing power, you can vastly increase your utility by shifting your cognitive function to optimal times.

Comment author: CannibalSmith 04 October 2009 05:12:37PM *  2 points [-]

Arguably, clubbers taking ecstasy and dancing until they die of hyperthermia are doing exactly this.

No, on ecstasy, you need willpower to stay still.

Comment author: pdf23ds 03 October 2009 10:56:06PM 2 points [-]
Comment author: gwern 06 October 2009 11:38:47PM *  1 point [-]

That'd make an interesting scenario, actually: if people had to live up to their stated ideals, how many would take the drug and do so (possibly destroying themselves), and how many would realize that their stated ideals were social fictions and come up with more honest ones?

It reminds me of Robin Hanson's little paper about social signals in the future, with computers revealing whether the person is being honest or not. The transition period might be rough, but humanity might be better off.

Comment author: cousin_it 03 October 2009 08:29:40AM 5 points [-]
  1. Zeal conquers all!

  2. Sometimes zeal conquers even understanding.

-- Kozma Prutkov

Comment author: randallsquared 05 October 2009 04:07:04AM 3 points [-]

And it's impossible to sleep too much more than normal; the better rested we are, the harder it is to fall asleep.

This isn't true, as far as I can see. I know several people who have fallen into the habit of sleeping much more than is normal for months or years, with no apparent physiological reason. Also, I notice that when I sleep in, I typically get up because I don't want to waste more of the day, and never seem to have trouble going back to sleep if I choose to do that. For someone sufficiently unmotivated to do anything, I think sleeping for most of each 24 hours would be possible.

Comment author: Carrie 07 October 2009 01:28:48PM 2 points [-]

I think that has less to do with willpower and more to do with the way that most of the developed world has been crazy on sleep-dep since the invention of the electric light.

Comment author: randallsquared 07 October 2009 03:49:22PM 1 point [-]

Hm. One point in favor of that possibility is that both of the people I know closely who've done this spent virtually all of their waking hours when it was dark out, during these periods.

Comment author: CronoDAS 07 October 2009 06:17:35PM *  0 points [-]

And the alarm clock. I'd like to go back in time and kill the evil genius that invented the alarm clock. I hate hate hate hate hate hate hate hate HATE them!

(And no doubt the evil genius's evil twin contributed the snooze button.)

Comment author: CronoDAS 07 October 2009 06:11:43PM 1 point [-]

For someone sufficiently unmotivated to do anything, I think sleeping for most of each 24 hours would be possible.

I've actually tried that, and it doesn't really work. First of all, sooner or later, I need to go to the bathroom, eat, and all those other things. Second, I generally just end up lying in bed, mostly awake, letting my mind wander freely thinking about nothing in particular instead of actually sleeping.

Comment author: gwern 05 October 2009 05:22:54PM *  0 points [-]

In The Discovery of France, I read of one region of France where all the peasants during winter simply holed up in their houses, bundled up together, eating very little, and just sleeping or dozing the days away. Outsiders were struck how very like hibernating bears they were.

Comment author: taw 03 October 2009 09:11:18AM 6 points [-]

Both cases of sleep and hunger are demonstrably untrue, for we can turn off those systems (with modafinil for sleep, and ECA stack for hunger), and there are few ill effects. These systems are horribly misregulated, and cause more problems than they save from.

We know that not sleeping for days, or sleeping significantly less per day than 8h, doesn't cause any serious problems in healthy adults. Modafinil completely turns off sleepiness, and modafinil users do not die in droves, or even get sick from that.

Sleepiness on the other hand forces you to sleep even when it's not necessary at all, by reducing your alertness, making it painful to keep awake etc. Sleepiness often kills people, for example by making them insufficiently alert while driving, far more often than it saves them from death due to sleep deprivation.

With appetite suppressants it's similar story. How many people die due to diets, and how many people die due to obesity?

These systems are completely broken.

Comment author: Vladimir_Nesov 03 October 2009 11:32:49AM 6 points [-]

We know that not sleeping for days, or sleeping significantly less per day than 8h, doesn't cause any serious problems in healthy adults.

Maybe if you do something simple. I experience serious mental impairment on the whole next day even if I sleep 7 hours (in the sense of not being able to get on with reading math or thinking as fast and productive/deep as after good sleep). This isn't about sleepiness, as I can be perfectly alert, just dumber (there's also coffee that helps with sleepiness but not with thought). This is probably too fine a point to have been adequately researched.

Comment author: wedrifid 04 October 2009 09:46:03PM 4 points [-]

There has been plenty of research, complete with assigning numbers to the expected IQ drop. From what I recall, acute sleep deprivation has been found to lower measured IQ for up to a week.

(This is without the sleep effects that are particular to modafinil, as opposed to other stimulants such as caffeine or amphetamine.)

Comment author: gwern 23 July 2011 09:16:05PM 1 point [-]

If people will forgive me not digging up the details likepage number & citation & metrics, I recall that the big USAF study on modafinil and sleep deprivation found that mental performance on modafinil did begin declining during and past the first night, it just declined less than the alternatives.

Comment author: anonym 04 October 2009 09:28:05PM 1 point [-]

I notice the same if I get an hour or two less than normal. It's extremely obvious when I'm doing anything that is very cognitively demanding, even if I wouldn't notice anything out of the normal other than a mild sense of malaise if I were watching TV or doing easy cognitive tasks.

Comment author: MendelSchmiedekamp 05 October 2009 04:19:06PM 0 points [-]

I have a similar result, except I've never experienced a stimulant effects from anything other than blood sugar I'm not certain I can discount sleepiness. Also, I suffer from a migraine condition which has a much more severe affect on my mental faculties on a day-to-day basis.

And since improper sleeping is one of my triggers - "Happiness is getting enough sleep." Not too much, not too little.

Comment author: DanArmak 03 October 2009 09:36:38AM 3 points [-]

We know that not sleeping for days, or sleeping significantly less per day than 8h, doesn't cause any serious problems in healthy adults. Modafinil completely turns off sleepiness, and modafinil users do not die in droves, or even get sick from that.

I believe that sleepiness during the waking hours isn't useful, but if you can significantly reduce sleeping hours indefinitely with no ill effects, that's very surprising to me. Can you provide pointers to a study testing this?

Different people vary in their need for sleep, and their sleepiness systems may not always be attuned to this variation, but if you took a large random sample and literally turned off sleepiness for all of them for a long period of time I would expect significant negative effects in many cases.

Sleepiness often kills people, for example by making them insufficiently alert while driving, far more often than it saves them from death due to sleep deprivation.

You can't compare the two directly, because almost everyone is sleepy now and then, while almost noone (as a percentage of the population) uses modafinil-grade sleep prevention. Being in danger of death due to voluntary sleep deprivation is a very rare condition.

Comment author: gwern 05 October 2009 05:53:34PM *  1 point [-]

Do we know of any actual cases of sleep-deprivation death?

I note that while rats will die of it, mice and pigeons won't; and Randy Gardner went 11 days without sleep, and I can't find anything about any long-term health problems (50 years later, he sounds perfectly hale & healthy in http://www.gelfmagazine.com/archives/sleeping_in.php - note also he probably could've gone more than 11 days if the descriptions of his press conference at the end as being 'lucid' are to be believed).

Reading about sleep deprivation studies and the 'micro-sleeps' that occur in many species, I wonder if the processes making up sleep might be analogous to garbage collection: you can have 'stop the world' GC schemes, or incremental ones

(The one human example I did find was 'fatal familial insomnia', but I don't think anything can be safely inferred from a rare genetic disease like that.)

Comment author: NancyLebovitz 11 April 2010 06:34:15PM 1 point [-]
Comment author: Cyan 11 April 2010 06:56:09PM *  2 points [-]

While the article is interesting (thanks for posting the link!), the disease doesn't appear to cause "actual cases of sleep-deprivation death".

Comment author: NancyLebovitz 11 April 2010 08:35:13PM 1 point [-]

It's hard to be sure-- the deaths could be simply caused by lack of sleep, or what's causing the inability to sleep could also be causing additional damage.

Comment author: pdf23ds 05 October 2009 06:25:24PM 1 point [-]

This reminds me of that sleep schedule where you sleep about 3 hours a day total in 20-30 minute increments. From what I've read, a bunch of people (mostly nerds) have tried it, but few have succeeded.

Comment author: gwern 05 October 2009 07:39:25PM *  3 points [-]

I gave it a try one summer. Toward the end, it did sort of work, but my general conclusion was that you were permanently on a lower level of mental functioning (which seems to accord with Stampi's results). The experiences I read online generally either didn't have a good test of mental functioning available, or were focused on creativity - which I figure is something that could well be boosted by the slight delirium/loss of inhibition one sometimes experiences...

Comment author: pdf23ds 05 October 2009 07:43:33PM *  1 point [-]

Hmm. I wonder how a cost/benefit analysis would work out. On the one hand, you have a lot more time to make up for the mental deficiency, but on the other, small differences in mental performance can add up really quickly. (Anyone who has spent a day chasing a bug that would have taken them five minutes had they just thought a little more clearly will understand this.)

Comment author: gwern 06 October 2009 12:12:19AM *  1 point [-]

It'd be difficult to estimate. Intelligence is valuable; even the most basic minimum wage job can be done better if you're more intelligent.* And then there's the schedule disruption - more than one polyphasic sleeper has cited that as a reason to go back to monophasic.

So you want to be someone with a flexible schedule & undemanding job. A good method for freelancers, I suppose, or students (eg. GPA correlates higher with 'conscientiousness' than IQ; consider http://medicalhypotheses.blogspot.com/2009/05/do-elite-us-colleges-choose-personality.html ); positions where consistency can be more valuable than peaks.

* claimed in Murray's infamous The Bell Curve; I have no particular reason to disblieve it

Comment author: Douglas_Knight 06 October 2009 12:37:07PM 0 points [-]

[polyphasic sleep] did sort of work, but my general conclusion was that you were permanently on a lower level of mental functioning ... [other sources] were focused on creativity

One of the arguments for polyphasic sleep is that some inventors, like Edison, did something like it. Were they trading off intelligence for creativity? Also, you get more time immediately after waking up, which is supposed to be a good time to work.

Comment author: gwern 06 October 2009 02:02:50PM 1 point [-]
Comment author: DanArmak 05 October 2009 06:25:02PM 1 point [-]

Do we know of any actual cases of sleep-deprivation death?

I couldn't find any. On the other hand, Wikipedia claims that total and indefinite sleep deprivation is "impossible" to achieve, possibly even under torture, due to microsleep and extreme tiredness enabling brief ordinary sleep in almost any circumstances. Other reported ill effects of SD might be linked to the cause of the sleep deprivation instead.

However, that doesn't answer the question of what might happen to an average human who was sleep-deprived by whatever method, as far as possible, for a really long period of time - months, not days. I expect there would be physiological or mental damage of some kind in (almost) everyone. That doesn't mean there isn't a way to negate those effects and do away with sleep one day - it's just a question of how narrowly we define the "consequences" of sleep deprivation vs. "removable side effects".

Comment author: loqi 04 October 2009 10:33:27PM 1 point [-]

We know that not sleeping for days, or sleeping significantly less per day than 8h, doesn't cause any serious problems in healthy adults.

...in the short term. I need more evidence than "they don't get sick or die right away" to believe that cranking a knob as central to our physiology as sleep doesn't deliver some unintended consequences further down the road. For example, possibly contributing to Alzheimer's.

Comment author: taw 05 October 2009 02:31:18AM 3 points [-]

By simple evolutionary argument, it's extremely unlikely that sleep would develop to any considerable degree to protect against very late age mental deterioration like Alzheimer's (Wikipedia says it's "generally diagnosed in people over 65 years of age") - we're trading 8 hours a day at reproductive age (very high cost) versus something that happens far post reproductive age, and until late stages is only a mild problem compared to other not-sleep-dependent deteriorations (benefit very close to zero).

You might want to accept far lower standards of evidence in general. There will never be enough long term medical studies in humans. No long term studies of which early adulthood activities contribute to Alzheimer's will be done until you reach age when it will be too late to act.

Comment author: loqi 05 October 2009 07:43:04AM 3 points [-]

By simple evolutionary argument, it's extremely unlikely that sleep would develop to any considerable degree to protect against very late age mental deterioration like Alzheimer's

Indeed. But that doesn't mean sleep can't protect against it anyway. Not every biological mechanism that protects against pathology X is an exclusive anti-X bullet. Also, the fact that Alzheimer's tends to set in after 65 isn't strong evidence against the claim "human sleep has evolved to delay Alzheimer's until after reproductive obsolescence".

You might want to accept far lower standards of evidence in general.

I was talking about belief, not action. You said:

We know that not sleeping for days, or sleeping significantly less per day than 8h, doesn't cause any serious problems in healthy adults.

I don't think we're so sure of this. When sufficiently compelled to act (e.g., to gain to more time by sleeping less), we may well decide to scrape together what applicable knowledge we have and take a calculated risk. This is distinct from knowing the risk is low: It takes much less evidence to change which action appears optimal in the low-knowledge scenario than the high-knowledge scenario.

Comment author: taw 05 October 2009 08:27:09AM 3 points [-]

Also, the fact that Alzheimer's tends to set in after 65 isn't strong evidence against the claim "human sleep has evolved to delay Alzheimer's until after reproductive obsolescence".

Risk of Alzheimer's is 0.3%/year past 65 years. Only 19% of people in 75–84 group get it, and it's still a few years before it becomes fatal. It's extremely far removed from reproductive age (and there might be various environmental factors present, I'm not even sure if wild humans could get Alzheimer's at all).

If sleep was mainly to protect against Alzheimer's, we would sleep quite a lot less, and have significant risk of Alzheimer's by age of 50 or so, just as happens with most other deteriorative diseases. Evolution never overprotects like that.

I don't think we're so sure of this.

We know there's virtually no short to mid term risk.

We know next to nothing about long term risk of pretty much anything, we can only take our best guess and hope for the best.

Comment author: Jayson_Virissimo 05 October 2009 08:42:18AM *  0 points [-]

How many hours on average do you sleep a week? Also, I know that the ECA stack makes losing fat much easier and raises your resting heart rate, but what evidence do you have that ECA stacks "take away hunger"?

Comment author: taw 05 October 2009 08:53:11AM 0 points [-]

I don't think amount of my sleeping diverges much from 8h/day, but it's very irregular. I don't take modafinil unless in special circumstances, but ask people who do so a lot.

With ECA I have personal experience, and you can ask pretty much anyone who tried it, or tried any amphetamines, which act in a very similar way. As far as I know it does not raise resting heart rate.

Pubmed has plenty of studies of ephedrine and weight loss; I don't have any particular links ready right now, but it shouldn't be hard to find them if you're interested.

Comment author: pdf23ds 05 October 2009 01:17:24PM 0 points [-]

I have tried many stimulants for my attention problems, including amphetamines, and all (except a norepinephrine reuptake inhibitor) raised my resting heart rate unacceptably, from 65-70 to 85-95. I think I might try ECA and report back.

Comment author: taw 05 October 2009 04:07:52PM *  0 points [-]

Amount of ephedrine in ECA stack is really pretty low. It won't have significant stimulant effects past the first couple of days, caffeine in the stack is probably more stimulating than ephedrine.

The stack is good exactly because it provides good appetite suppression to stimulation ratio. You can reduce your hunger by taking a lot of amphetamines, but stimulation will be huge before appetite suppression becomes significant.

And you seem to have some heart problems. In such case, maybe it's a better idea not to take even mildly stimulating drugs, and go for a willpower-based diet, preferably one with a pseudo-cultish background like paleo or Atkins to help you stay on track.

(I recommended ECA for appetite suppression and fat loss, I don't have any opinion on their use for attention deficit or any other problem).

Comment author: pdf23ds 05 October 2009 06:02:31PM 0 points [-]

I recommended ECA for appetite suppression and fat loss

That's what I was interested in it for.

Comment author: wedrifid 05 October 2009 01:26:59PM *  0 points [-]

Also, have you tried selegiline? Sulbutiamine? Nothing magical, but worth a play. Selegiline in particular I've found to boost motivation. That makes the constant effort of turning back to whatever I was supposed to be paying attention to less of a drain on willpower.

Comment author: wedrifid 05 October 2009 01:23:14PM *  0 points [-]

Try intense cardiovascular exercise for 50 minutes every day. It'll lower the heart rate, never mind the boost to cognition! Not entirely joking. Stimulants impact both my blood pressure and my heart rate. With exercise I can reduce my non-wired resting heart rate to 40 and blood pressure to borderline hypotension (when in a near meditative state and according to my monitor). That's two side effects down!

Comment author: pdf23ds 05 October 2009 01:28:10PM 1 point [-]

I did that for a while. Unfortunately it's incredibly mentally painful for me--intense boredom. Also, I have weird knee problems.

Comment author: Jonathan_Graehl 05 October 2009 08:08:16AM 0 points [-]

I feel like ephedrine makes me less efficient at doing the same work, which I guess is part of the attraction to dieters.

I'm not sure that ECA turns off hunger, either.

One thing to consider about the use of stimulants to increase (endurance) exercise performance (caffeine definitely helps me) - you could probably do the same if you felt your life depended on it - this article on [Jure Robic] suggests that extreme endurance athletes may only be better in that they can force their body to use all its reserves.

Comment author: taw 05 October 2009 08:35:23AM 0 points [-]

Big enough dose of ECA significantly reduces hunger on pretty much everyone. You can easily try it on yourself.

Professional athletes undergo years of training, special diets, various legal and illegal performance enhancement, are brainwashed into performing better, and are selected from people genetically predisposed into their sport. That's a long list of differences.

Things like ECA and modafinil work on an average healthy adult.

Also the most interesting athletic performance enhancing drugs don't work on CNS, but to enlarge muscle cells or increase their number (HGH, steroids), increase blood capacity to carry oxygen (EPA), reduce body fat (different steroids), or water (diuretics) etc. CNS-affecting drugs seem to be a minor issue.

Comment author: jimrandomh 03 October 2009 06:25:57PM 0 points [-]

Both cases of sleep and hunger are demonstrably untrue, for we can turn off those systems (with modafinil for sleep, and ECA stack for hunger), and there are few ill effects.

The fact that we can alter our psychology with drugs does not change what that psychology is. We can also eliminate or cause many emotions and perceptions, and will probably learn to do so with for a larger variety of cases and with less side effects as time goes on. These drugs are properly considered as an extension of willpower, not a counterexample to it.

These systems are horribly misregulated, and cause more problems than they save from.

It only seems that way because of availability bias; we only know these systems when they go awry.

Comment author: taw 03 October 2009 10:35:28PM 2 points [-]

Oh no, we can take marginal view. If sleepiness was slightly more severe, it would kill more people due to diminished attention. If it was slightly less severe, nobody would die due to sleep deprivation, and plenty of lives would be saved.

By this marginal argument, it's misregulated. It doesn't mean it would be useless at some different point.

Comment author: gwern 06 October 2009 11:30:54PM *  0 points [-]

Or we could apply Nick Bostrom's evolutionary heuristics and say that the systems were regulated fine for older environments, but not modern ones.

This is obviously true for food in 1st World countries, and plausibly true for sleep. (If calories are nearly free & electrical lighting universal, then the safe-place/low-energy rationale holds much less water.)

Comment author: anonym 04 October 2009 07:47:32PM 0 points [-]

These drugs are properly considered as an extension of willpower, not a counterexample to it.

These drugs don't extend willpower, they alter the system 1 defaults so that willpower is no longer needed, for there is nothing to be overcome.

Comment author: Kaj_Sotala 03 October 2009 05:15:47PM 0 points [-]

We know that not sleeping for days, or sleeping significantly less per day than 8h, doesn't cause any serious problems in healthy adults.

Does "no serious problems" include "no impairment in learning"? I've seen plenty of studies saying that learning, memory formation and skill acquisition is consolidated during sleep.

Comment author: wedrifid 03 October 2009 05:36:52PM 2 points [-]

The boost to alertness and general stimulatory properties more than compensate. Learning is improved over baseline (in non extreme sleep deprivation scenarios). Of course, some memory effects are easier to study than others. I don't recall what impact it has on skill acquisition.

Something I am more wary of is that forgetting may be impaired. The process of filtering through memories, sifting out the useful abstractions and discarding the crap is a fairly important part of sleep. I love having extremely retentive memory but I don't necessarily think it is healthy!

On the health side of things, taking Modafinil and cutting down on sleep doesn't make you sick on it's own if you are otherwise healthy. But it isn't without physical costs. Many users (including myself) note that you need to be in far more in touch with your own physiology. If you've turned the safety mechanism off and don't take care to monitor your health and force sleep when you need it you will wear yourself down over a couple of weeks. This means that you will be more vulnerable to the effects of stress and less able to recover from intense exercise.

Comment author: gwern 05 October 2009 05:26:32PM 0 points [-]

Many users (including myself) note that you need to be in far more in touch with your own physiology. If you've turned the safety mechanism off and don't take care to monitor your health and force sleep when you need it you will wear yourself down over a couple of weeks.

Thanks for the advice; I'd just started modafinil and thought I only needed to make sure I was eating enough. What about your health needs monitoring?

Comment author: wedrifid 05 October 2009 06:37:40PM 0 points [-]

Unless you are far more responsible than I you have experienced the effects of a period of sleep deprivation over multiple days. Now, forget the cognitive impairment. Forget the vulnerable mood and the constant urge to sleep. Moadifinil takes cares of those.

But there's other effects, more physiological, that are sometimes only noticeable after a few days of this self abuse. A somewhat paler face. Fatigued eyes (bloodshot or dry). A hint in the throat, almost a cough. Basically, whatever you usually get when you are run down. If you add more physical (too much exercise?) or emotional stressors with an unlucky exposure to a pathogen then watch out.

Just something to watch out for if you rely on higher doses over a period of weeks. Basically, just add 'sleep' and 'rest' to 'food' and it's fine. Easy to forget if you don't feel the need for any of them.

Comment author: gwern 05 October 2009 07:41:23PM *  0 points [-]

A hint in the throat, almost a cough.

You're right! I hadn't even noticed that I was starting to sound like Leonard Cohen.

Comment author: taw 03 October 2009 10:33:18PM 0 points [-]

I cannot answer to you, because research is so limited. And research is so limited, because it's non-prescription use, so nobody is really bothered (and funded) to research it, and also because all enhancement uses are unofficial, and shunned.

Another reason is that we don't really know that much about the processes of learning, and we cannot easily test that. Testing simple things like attention and short term memory is much easier, so it's done.

Comment author: pdf23ds 03 October 2009 10:45:57PM 2 points [-]

Testing long-term memory would probably be pretty easy, actually, using flash card programs like Supermemo. You just measure the retention of foreign vocabulary over time, and compare based on when it was first learned and whether you were sleeping regularly or not over that time, or using modafinil, or whatever. Of course, it's still much more tricky than testing short term memory.

Comment author: anonym 04 October 2009 09:02:09PM 0 points [-]

Testing learning is not difficult, and lots is known. There is no doubt that decreased sleep negatively affects the brain and learning and impairs existing abilities in normal human beings that aren't taking Modafinil. Modafinil research is barely out of its infancy though, which explains the relative lack of studies related to Modafinil-modified sleep regimes and learning.

I did a quick search and turned up Modafinil restores memory performance and neural activity impaired by sleep deprivation in mice, which identified a plausible physical correlate of impaired spatial working memory due to sleep deprivation and observed that Modafinil eliminated the impairment and restored normal neurological activity.

Another paper, Modafinil activates cortical and subcortical sites in the sleep-deprived state found that altered neural activity and cognitive impairment due to sleep deprivation were both improved by Modafinil on an n-back task (which taxes attention, working memory, and the executive system), but only when working on a moderate difficulty task -- not on either an easier task or a more difficult task (for those familiar with n-back, the easy/moderate/difficult levels were 1-back/2-back/3-back, resp.). That study also showed slower reaction times under sleep deprivation even with Modafinil, but they were greatly improved over sleep deprived without Modafinil, even if not at the level of non-sleep deprived with or without Modafinil.

(Modafinil: A Review of Neurochemical Actions and Effects on Cognition)[http://www.nature.com/npp/journal/v33/n7/abs/1301534a.html] looks like it might be an interesting read, since they reviewed all English language papers on Modafinil in pubmed, but I unfortunately don't have access. Maybe somebody else who does have access could summarize the main points.

Comment author: taw 05 October 2009 02:07:37AM 1 point [-]

If large portion of effects of "sleep deprivations" can be reversed by drugs, it begs a question if they are effects of lack of sleep, or more side effect of the systems trying to get you to sleep.

Let's take as a hypothesis that they are entirely of the latter kind, and modafinil and stimulants simply don't turn off the sleep forcing system completely. Is there any strong evidence against it?

It's trivial to point evidence against hypothesis that they're entirely based on sleep deprivation, as even partial reversibility disproves it.

I'm also surprised by many of the procedures, which tried modafinil and/or stimulants as a single dose taken only after a lot (24h or more, often 36h+) of being forcefully awake. The right way that everyone uses them is to use them before first effects of sleepiness happen, which would be no later than about ~14h after waking up, and then boosting regularly every now and then, so you never go into sleepy phase. I don't know if this affects the results, but intuitively it should make them less effective.

Comment author: wedrifid 04 October 2009 09:24:33PM 1 point [-]

Modafinil research is barely out of its infancy though

I'd go with 'neglected adolescent'.

Comment author: vinayak 20 October 2009 11:27:57PM 2 points [-]

I think there's a fundamental flaw in this post.

You're assuming that if we have unlimited willpower, we are actually going to use all of it. Willpower is the ability to do what you think is the most correct thing to do. If what you think is the correct thing to do is actually the correct thing to do, then doing it will, by the definition of correctness, be good. So if you do some "high level reasoning" and conclude that not sleeping for a week is the best thing for you to do and then you use your willpower to do it, it will be the best thing to do, just because you've done the correct analysis and have taken all costs into consideration (including the cost of bad health because of sleep deprivation).

It's always good to be able to do the thing that's the best thing for you to do. What's bad is to not be able to decide what's best for you. So we shouldn't blame willpower. We should blame the inability to take correct decisions.

Comment author: Morendil 03 October 2009 06:53:40PM 1 point [-]

There are more failure modes of willpower than just acquiring the power to resist urges bequeathed us for good reasons by evolution.

Obsessive-compulsive disorders sound like a good example to study; possibly of willpower taken too far. Most of us have to be reminded to wash our hands before eating, or (in the current period of paranoid fears of swine flu contagion) after sneezing. OCD sufferers don't need reminding, they must wash their hands several times a day, or check, double-check and triple-check that they have locked their doors before leaving home, and so on.

Ainslie's analysis of "personal rules" as one of the components of willpower might be relevant here; he mentions "fussiness" as one downside of willpower, and OCD is an extreme of fussiness.

Comment author: anonym 04 October 2009 07:41:51PM *  4 points [-]

I don't think OCD makes sense as willpower taken too far, since the compulsive behaviors are the defaults and the lack of ability to override those defaults via willpower is the problem. To the OCD sufferer, overriding those defaults requires superhuman levels of willpower they can't reliably summon and sustain.

Comment author: michaelkeenan 05 October 2009 02:58:56AM 1 point [-]

According to Aaron Haspel, "it is the incorruptible politician who is especially dangerous."

I didn't immediately see what he was getting at, but today a parallel occurred to me: a powerful, incorruptible politician is like a person with great willpower. Directing great resources to the wrong project can do a lot of harm, at individual and society-wide levels.

Comment author: MugaSofer 18 September 2012 10:51:02AM *  0 points [-]

when humans use stimulants to stay awake too long, they die

Source?

IIRC, problems caused by sleep deprivation - hallucination, paranoia etc - are entirely cleared up when the subject is allowed to sleep. They spend a longer-than-usual time asleep, then wake up fine. Wikipedia claims those rats were forced to walk whenever they began falling asleep, which seems like it could could have adverse effects.

Comment author: jimrandomh 03 October 2009 03:46:19AM 0 points [-]

My first draft of this article included a formal definition of willpower, which I cut out because it didn't really fit.

Since willpower is a purely mental construct, we have to define it relative to some psychological model. Unfortunately, we have many imprecise and conflicting models of how minds are structured, and they're probably all at least somewhat wrong. Rather than define willpower in terms of a model that will probably be disproven, we should define it in a way that can be applied to any psychological models - even models of animals, aliens or AIs. We also need to be careful to separate the definition from empirical observations, such as the observation that willpower tends to act like a fungible resource, tends to fail all at once rather than gradually, because these observations might not always hold.

For the sleep-willpower example, our model needs at least two parts: one which provides defaults, and one which overrides them. We'll call the defaults system 1, and we'll call the override system 2. Now consider a complete model of minds, with many parts that have various degrees of influence over each other. Use some centrality measure to designate one of these parts as the center. Most reasonable measures should put the center somewhere inside what we call "higher reasoning", since that's the one part of the human mind that can most directly influence every other part. Now take any pair of elements in the model which act in opposition to each other. Whichever is farther from the center, is system 1; whichever is closer to the center, is system 2; and whatever influence system 2 has over system 1, we call willpower. In general, we use "system 1" to mean "things that are far from the center", and "system 2" to mean "things that are close to the center". Because different models may place things differently, whenever we use the terms system 1 and system 2 we assert that all reasonable models should agree on which is which.

Comment author: atavist 06 October 2009 04:40:46AM 0 points [-]

One could also imagine an individual who aspires, through sheer dint of willpower, to become homo economicus. Whenever he senses the beginnings of a bout of empathy, guilt, shame, or love, he exercises top-down cognitive control to extinguish his emotional reactions and stay on course to maximizing his utility. It won't be long before his social life becomes impoverished. Akrasia in his case would be quite adaptive.

Comment author: gwern 19 October 2013 09:26:22PM *  3 points [-]

It won't be long before his social life becomes impoverished. Akrasia in his case would be quite adaptive.

Ah, but you forgot that this particular homo economicus has a utility function valuing only reproductive fitness. By suppressing emotions like empathy, guilt, shame, love, and ordinary creature comforts, he is able to make a fortune and pay/bribe sperm banks in countless countries to use exclusively his own sperm; by the time he dies heavily in debt (a poor decision by his creditors), his sperm has been used in tens of thousands of inseminations, almost all of which come to fruition, and he is officially the Most Reproductively Fit Man in the World. Akrasia in his case would be quite maladaptive.

Comment author: DanArmak 06 October 2009 12:52:21PM 0 points [-]

The force of love, guilt, shame, etc. are much stronger and more common than akrasia. It is rare for humans to want and to be able to disregard these feelings completely in the long run. I don't think akrasia functions as a social adaptation - these feelings are themselves the adaptation.

Not to mention the fact that by far most cases of akrasia hurt your social lfe: akrasia can prevent you from making new friends or meeting existing ones or indeed from doing anything at all, in the worst cases.