Hacker News new | comments | ask | show | jobs | submit login
Researchers suggest that R.E.M. sleep serves to warm the brain (nytimes.com)
291 points by dnetesn 8 months ago | hide | past | web | favorite | 163 comments

Just so you know, there's half a dozen of equally plausible hypotheses about REM function, and dozens more of more fantastical. Most amusing one yet is about synthesis of molecular chaperones that guard against aggregation of chains produced during NREM. But honestly we just don't know.

Personally I agree with Marcus Schmidt and his idea of "energy allocation function of sleep". Basically it says that distinct physiological states provide viable configurations of priorities for vital functions that allow for the lowest possible energy expenditure over the whole cycle. I.e. there's no single specific function of REM sleep, it hosts all functions that are too costly to prioritize during other states, and the same is true for NREM.

That means that biology uses the same idea of having "background jobs" and worker queues that get more relevant while the main function has a bit less to do: many IT systems will schedule some tasks during the night when there is not much going on, so they can prioritize them because otherwise they wouldn't get a chance during the day.

I wonder if there are universal laws that transcend biology and technology, all following the same rules that we haven't figured out yet.

Everything derives from the laws of physics (well as we understand it now at least; our understanding has and will change in the future) so its not that surprising that in aggregate they would result in similar systems.

I'm not saying its not fascinating though. It is incredibly fascinating, and this kind of research is exactly why we should be investing MORE not less in pure science.

>"this kind of research is exactly why we should be investing MORE not less in pure science."

I disagree. Look at the paper: https://www.cell.com/current-biology/fulltext/S0960-9822(18)...

Can you point out where they even report the temperature of the brain/brainstem? This looks like it is just wild speculation.

Agreed. Hopefully someone will so we can get some good data on this.

> many IT systems will schedule some tasks during the night when there is not much going on

Indeed, it’s natural that we do this but kind of funny when you consider that our machines now have their own circadian rhythms in tune with ours.

What might our IT infrastructure look like if humans evolved to sleep far less or not at all I wonder!

> What might our IT infrastructure look like if humans evolved to sleep far less or not at all I wonder!

Sounds like your typical developer or sysadmin to me, so probably not much different at all. ;)

> What might our IT infrastructure look like if humans evolved to sleep far less or not at all I wonder!

We probably would have started with multi-core systems much sooner.

Can you explain to someone without a graduate degree in biochemistry what about "synthesis of molecular chaperones that guard against aggregation of chains produced during NREM" is amusing?

You don't need a degree here.

It's not amusing in a bad way, it's not ridiculous. I mean to say that it's highly surprising and counterintuitive, like a punchline in a joke. Most theories about REM are focused on brain function, as is almost all research on it. Chaperone production, though, is a basic housekeeping routine. Saying "well, REM isn't about episodic memory consolidation or dreaming or wherever you might think, it's an epiphenomenon of a very low-level mechanism that prevents boiling alive" is almost as unorthodox as claiming that learning to code is about forearms' dexterity.

So the hypothesis is that mammals need REM sleep to prevent the brain from getting too cold while sleeping, but the hypothesis hasn't been tested yet.

If this turns out to be true, I wonder if it has implications for night terrors. It seems plausible that if the brain fails to warm enough through REM sleep, hormones are released to wake the brain up completely which lead to night terrors and the adrenaline rush feeling that accompanies them.

These are just hypotheses, but interesting nonetheless.

The hypothesis seems a little unlikely. If REM sleep was to warm up the brain you might think people wouldn't do it in hot climates but I don't think that's the case.

Maybe it's more like the "Google’s DeepMind AI gives robots the ability to dream" stuff that increased their neural nets rate of learning 10x https://www.extremetech.com/extreme/240163-googles-deepmind-... and "Rats Dream About the Places They Want to Explore" https://news.ycombinator.com/item?id=9791927 - about improved cognition.

Also the 'inceptionism' stuff looks so reminiscent of human dreams that it makes me suspect there is neural network stuff going on rather than heating. https://news.ycombinator.com/item?id=9736598

> If REM sleep was to warm up the brain you might think people wouldn't do it in hot climates but I don't think that's the case.

Also, fever dreams.

> If REM sleep was to warm up the brain you might think people wouldn't do it in hot climates but I don't think thats the case.

Isn't internal body temperature kept constant (or at least the body tries very hard to) irrespective of environment?

The body tries to keep the core temperature constant around the heart and internal organs and will divert blood there and let your feet get cold. I'm not sure about the brain but in a 40C climate say I doubt your head would get cold inside.

Depends on where you are. In a land climate, days can be searing hot while nights are freezing cold.

There's plenty of places in the tropics where it's hot during the day and night for months on end, and yet as far as I know there's no evidence of less REM sleep in the people living there. And even if there weren't, these conditions would be easy to replicate in a laboratory.

Another thought after sleeping on it - after a good nights sleep you are often clearer mentally and can see better ideas whereas after broken sleep you feel fuzzy headed which also could suggest REM sleep is about cognition - probably running what if scenarios in effect which is why if you interrupt it you get odd future scenarios as dreams. Its interesting that the networks can probably do that without conscious awareness.

There was also some research a while back where scientists theorized that sleep is also used to clean out the brain:


It could mean that constant sleep deprivation could increase the risk of Alzheimer's, but of course that's really difficult to test since you'd need a lifetime of sleep data for humans in their 60s/70s.

Also I think that paper was just about sleep in general, not specifically dreams/REM sleep.

I read an article long ago (I searched now but didn't find it) where a family had a rare genetic disorder that prevents sleep and gets worse with time. The afflicted family members essentially go into an early dementia in their fourties or early fifties that's fatal. I don't know how similar that is to Alzheimer's, but it seems to lend weight to the theory that lack of sleep could be a contributing cause of dementia.

Thank you, I think that's the one. It amazes me how whenever I mention an obscure article I once read on HN, someone comes along with a link.

Fatal Familial Insomnia.

It's a prior disease.


I have a pet theory, based on sensations I've experienced, that:

A. 'sleep cleaning the brain' is real

B. The things that are expelled from the brain are expelled through the sinuses

C. If you are sleep deprived enough, you can feel this process happening while you're awake

How would your brain dump something in your sinuses???

The blood from your brain (like all your blood) is cleaned by your kidneys and liver.


Are you confusing those with the sinuses in your nose? Two very different kinds of sinuses.



Lymphatic system does connect to the circulatory system, so can eventually be handled by the kidneys, liver, etc. But it's not necessarily a direct path, according to recent research.

The glymphatic system collects into the cephalospinal fluid, which is drained by the dural veinous sinuses through the Pacchionian/arachnoid granulations.


Could just be anxiety and tension headaches making you feel like your sinusitis is caused by sleep, rather than masked by it. It’s common enough.

I don't have sinusitis or tension headaches. When I've stayed up many hours into the night I get a particular sensation of vibration in my sinuses and a compulsion to breathe deeply through them.

well that's a scary thought, because --

my mom's side of the family has a continuously running history of early-onset Alzheimer, AND terrible sinus problems.

my sinuses are expelling so much goop on a constant basis that by the end of the day my teeth are covered in a layer of slime.

You may be allergic to something, and/or you could have an infection.

A 3x former coworker and friend had a sinus infection that wouldn't go away. Like 10 years. Had his tonsils removed, etc. He finally cut out everything in his diet, and eventually discovered that despite not being gluten sensitive, celiacs, crohns, IBS, etc., or allergic (he got the allergy scrapes too), not eating bread / barley / wheat was the thing. Whether it was bacterial, fungal, or 'not allergic unless I eat it' type thing, the answer was the same: change your diet, write everything down, see what happens.

Another family friend who had sinus issues and allergies for 30+ years ended up going keto. In the process of losing 20 pounds (in the last 4 months, so far), she also is no longer allergic to Norwegian spring/summer. Don't know if it was the sugar, starch, etc., but low/no carb basically changed her life.

Note: I'm not a medical doctor. The above are anecdotes. But changing your diet could seriously change your life. Give it a shot!

"change your diet, write everything down, see what happens"

if any reader wants to go this route, "whole 30" is a relatively well-known "elimination diet" that omits the vast majority of the things that cause food sensitivities in people. the idea is you do a 30 day purge with an allow-list of foods, and after that you can slowly reintroduce individual things and keep track of how you feel.

a benefit of using a well-known diet over any other one is that finding recipes etc is much easier when there's an entire subculture of "whole 30" food bloggers.

Or you can go to an immunotherapist for a prick test and be told precisely what you're allergic to (in addition to being given a recommendation for immunotherapy treatments) in a matter of minutes.

I went to an immunotherapist six times and they were unable to determine my allergies. After a whole year, I finally figured out on my own though an elimination diet that I'm allergic to eggs.

You may want to eat less dairy, I've found that milk in particular seems to increase the amount of mucous my sinuses and lungs produce.

I know a very successful banker who prides himself on sleeping 4 hours a night because he believes that is the optimal point where each additional hour no longer has any incremental benefit.

I have gone through periods of high and low amounts of sleep for both work and personal reasons.

Sleep has always been phrased in terms of time - ie get x hours of uninterrupted sleep to perform your best. But there has to be more to it than that.

I've stayed up all night and felt rejuvenated the next day. And I've had lazy Sundays where I've remained tired the entire day.

I might not have not seen the scientific study on it yet, but I believe the quality of your sleep has a lot to do with the nature of when and why you're sleeping as well as your mental state when doing so.

After giving up an enforced routine schedule and allowing my sleep to settle where it will, I have found my sleeping "schedule" to be largely irregular.

Sometimes I want to stay up all night and keep going the next day. Other times, I go to bed at 9 and get up early the next morning. Sometimes I only need a couple of naps. Other times, I want to keep on sleeping, which I allow myself to do.

I have slept for 20 hours straight before, with only short breaks to pee and drink water. I think this typically happens when I contract a cold or flu, and helps my body focus on fighting it off.

I feel much, much better than when I used to have to be at the office every weekday, even if I could come in at noon.

I presume you work remotely?

I've gradually given up using currency, and no longer need to work.

Well, you may as well have changed the laws of physics. Care to share how you managed this?

I sleep outside, occasionally couchsurfing. I scavenge my food and everything else. I'm not 100% currency-free, but close to it.

In the traditional medicine system of India (Ayurveda), there is an anecdote passed around that goes something like this: for each hour of sleep before midnight, count it as two hours. For each hour of sleep between midnight and sunrise, count it as one hour. Sleeping after sunrise counts for zero - it is useless.

And I've observed this to be true in my life. The most peaceful, restful sleep I experience is consistently between 10pm and 1am.

I fully agree with your dismissal of hours slept being the quantitative unit of sleep, and I'm similarly curious about what constitutes a healthy, effective sleep habit...

> Sleeping after sunrise counts for zero - it is useless.

I don't know about some people, but for me this is practically the silliest thing I've ever heard.

I need 8 hrs of sleep period, and the time of day has zero effect on my alertness. It can be 9pm-5am or 3am-12pm, it's exactly the same.

But if I get only 6 hrs of sleep? I'm miserable.

I can't even begin to imagine where the idea that the hour of the day matters would have come from.

And just think of nighttime workers. Obviously their daytime sleep counts...

> I can't even begin to imagine where the idea that the hour of the day matters would have come from.

The idea that sleeping in is lazy and self-indulgent while rising at dawn is virtuously ascetic?

> I can't even begin to imagine where the idea that the hour of the day matters would have come from.

Many bodily functions such as hormone production are affected by your circadian rhythm [1]. Ambient light also seems to have an effect on quality of sleep.

[1] https://www.nigms.nih.gov/Education/Pages/Factsheet_Circadia...

> I need 8 hrs of sleep period, and the time of day has zero effect on my alertness.

It matters a lot, especially if it’s 5 AM - 1 PM. In high school, I was an RF technician, testing the rollout of LTE in my area. We had to conduct drive tests at night, due to traffic concerns. Working at night and sleeping during the day seriously disrupts your natural circadian rhythm. The only reason we as humans don’t sleep immediately after sunset is due to the invention of artificial light.

I sleep during the day and work nights, so yeah, that's complete crap.

> And just think of nighttime workers. Obviously their daytime sleep counts...

I think the quote is more an aphorism than rigorous guide. The framers of the quote probably didn’t anticipate the night shift ;)

It's interesting to consider that the 9-5 workday is a very modern innovation, especially considering how long Ayurveda has been around.

> In the traditional medicine system of India (Ayurveda)

“Ayurveda medicine is considered pseudoscientific.”


You skipped the following sentence in your quote: "Other researchers consider it a protoscience, or trans-science system instead."

Early thought systems evolved from hundreds of generations of folk wisdom. There are significant enough gaps in modern scientific understanding that to discount these systems is foolish.

Note that the current discussion is set in such a gap.

From the god of the gaps to the pseudoscience of the gaps. The same 5 old tricks to justify superstition and folk belief.

I'd expect most traditional medicines to be considered pseudoscientic if they predate the invention of the scientific method.

That said, I'd take science-driven modern medicine over weird herbs and rhino horns any day of the week.

If you go to sleep at 4pm and wake up at midnight, does that count as 16 hours of sleep or zero? This system suffers from the reverse Gremlins problem.

Sounds like it's not really worth sleeping during summer above the Arctic Circle, then.

My grandparents are farmers in part of Northern Europe where currently it’s dark from 11pm - 3am (it’s never truly dark though, and if you look North in the middle of the night it is distinctly brighter). Their sleeping patterns do strongly follow that of the sun, in the winter they sleep more and in the summer they sleep a lot less.

For the work they do this is good though, in the summer there are a lot more things to do, whereas in winter when it’s -20c and the ground is covered in snow, there isn’t very much to do.

In a way I think they have it better than me, trying to go to sleep when it’s still a bright sunny day outside just feels weird.

Or it is possible that humans are maladapted to living in the artic circle.

Maybe by extension it’s not really worth living there ;)

Naps in the day have repeatedly been shown to be helpful and have positive health benefits. I’d say this anecdote is just like a lot of anecdotes passed down by humans and doesn’t match science.

The anecdote does, in some ways, match science: hormones in the body operate on a schedule that is derived from specific frequencies of ambient light, and therefore ultimately the position of the sun.

But there are definitely people who sleep right in the middle of the day, eg siestas in sunny places.

can confirm. taking my nap now, in fact!

If you follow the "normal" schedule of sleeping (i.e. you don't work nights or something), then this probably has more to do with the deeper stages of sleep being more common at the beginning of the sleeping session and REM sleep more common as sleep duration continues.

As a person who induces lucid dreams as a hobby, I sometimes manage to make use of those long lazy mornings on weekends to lucid dream. I've found it's very difficult to have a lucid dream in the first two or three sleep cycles. After that, it's much easier because the REM stage of sleep is longer. Not that I've found that having lucid dreams is easy, I get one day like that every month or two and a spontaneous one here or there in between.

My problem with ayurveda and other traditional stuff is people not realizing they were not your popular paracetamol tablets you could gulp down and be effective. These were wholesome lifestyles. So cherry picking parts may just not be effective for most people and gives traditional stuff a bad name.

I have benefitted immensely from yoga and parts of ayurveda and I laugh at people who dismiss them but I don't like them being cherry picked and tested which diminishes trust in them

Wait - what? You like parts of them but you don’t like people who like parts of them?

There are some humans who only need 5 hours of sleep a night. Others that need 8 or 10. A good indicator is during a long vacation, without an alarm, how long do you typically sleep? If I allow my body to wake up naturally, it's almost always after 7 hours.

If you go long enough without needing to be on a schedule, studies have found a lot of people move back into a rolling/rotating sleep cycle .. a few hours at a time throughout the day/night. I found myself doing this while I took a several month sabbatical from the workforce.

> There are some humans who only need 5 hours of sleep

Define "need". Also, citation needed.

Just because you can get away with it and live, doesn't mean it's the best move.

> citation needed

“In 2009, a woman came into Ying-Hui Fu’s lab at the University of California, San Francisco, complaining that she always woke up too early. At first, Fu thought the woman was an extreme morning lark – a person who goes to bed early and wakes early. However, the woman explained that she actually went to bed around midnight and woke at 4am feeling completely alert. It was the same for several members of her family, she said.

Fu and her colleagues compared the genome of different family members. They discovered a tiny mutation in a gene called DEC2 that was present in those who were short-sleepers, but not in members of the family who had normal length sleep, nor in 250 unrelated volunteers.

When the team bred mice to express this same mutation, the rodents also slept less but performed just as well as regular mice when given physical and cognitive tasks.“


And we have not started genetically engineering this mutation into humans, why exactly?

Unforeseen consequences is a reason I would not touch this. Something that correlates with one positive thing can also (and often does) have correlations to other negative things that greatly outweigh the positive. And sometimes the latter can be much more difficult to discern than the former.

For instance in the US today childhood injury and fatality rates have plummeted [1]. And this seems like an obviously good thing. And one might reasonably hypothesize that the increase in rules and regulations against letting children play independently, as well as the rise of helicopter parenting has played some substantial role in this. Since it's very easy to see the link there.

Yet at the same time other factors such such as childhood mental illness, obesity, and other major issues have sharply increased. Relative academic results are down and even IQ levels are declining. Are these linked to the behaviors that have driven the decrease in injuries in fatalities? Well that's a lot harder to say, though I think it's a very reasonable hypothesis that at least some of these issues are linked with the same behaviors that have reduced childhood injury/mortality rates.

This is a recurring pattern in many things. Our understanding of the causality of genetics right now is essentially 0 keeping in mind that saying gene A 'causes' effect B is no more a meaningful understanding of the reason for the causality there than saying rain clouds cause rain.

[1] - https://www.childtrends.org/?indicators=unintentional-injuri...

Right, that's why our ancestors didn't all evolve that mutation anyway. But our ancestral environment isn't the same as our current environment: just engineer more genes to paper over each successive disadvantage, recursively, until the only remaining disadvantage is "reduced energy efficiency". Because our current environment is more than efficient at supplying humans with energy.

There are some studies - I'll have to do some digging to find them - that show that even if you "feel rejuvenated" the next day, you will still not be performing at full cognitive capacity. I think this makes much anecdotal evidence invalid.

Melatonin is interesting, I occasionally take it when I can't fall asleep, it generally works although I try not to use it too often. Occasionally, I wake up in the middle of the night tho and got into the habit of taking a 3mg melatonin pill to get me back to sleep. More often than not, the resulting sleep, probably around 4 hours left me feeling absolutely amazing, I mean better rested than I have been in years it was that big of a difference.

Of course sometimes it doesn't have that effect, but I've managed to repeat it 2-3 times and I know for a fact that simply being awake in the middle of the night and going back to sleep again doesn't have that effect.

You might be interested in this link. https://www.gwern.net/Melatonin

Gwern claims:

1. 0.1mg is an effective dose.

2. It is probably beneficial to take it every night before sleep.

Bankers also think they add value to the world, so

If it's related to warmth, then why would rats have to make it up after days of deprivation? Overheating seems like it would be just as destructive.

I'd recommend the book "The Promise of Sleep" [0] for anyone that's interested in learning more about sleep. The author basically pioneered the sleep medicine field and was at the University of Chicago in 1951 when R.E.M. sleep was discovered.

[0]: https://www.amazon.com/Promise-Sleep-Medicine-Connection-Hap...

Thanks, purchased!

Sidenote: The book recommendations on HN are always top notch.

I would recommend reading Why We Sleep by Matthew Walker instead. [0]

The Promise of Sleep is an okay book, but it is almost 20 years old now and it shows.

I have read both books and it is remarkable how much more we learned in the time between them.

[0]: https://www.amazon.com/Why-We-Sleep-Unlocking-Dreams/dp/1501...

You know my performance might change imperceptibly with sleep, but the area where I know it makes a difference -- probably the most important area -- is how nice I am to fellow human beings.

I am a better co-worker, family member, client etc with sleep.

They should look at aggression in animals rather than how well they do tasks.

What is the connection between swimming and heat? How did they even arrive at this hypothesis? Why would rats need to catch up on heating their brains, days after missing REM sleep?

The facts are interesting, the conclusion doesn't make any sense at all to me.

It doesn't make any sense at all to me either. Because of the difference in specific heat and heat conductivity of water vs air, an animal that can sleep under-water (even whilst alternating the sleeping part of their brain), surly doesn't need to warm its brain while on sleeping on land.

(But I'm no expert.)

It could be that the cooling of the brain is dominated by convection through blood flow rather than conduction through the skull/fat/skin. If that's true, then the effect of water vs air would be on the whole body which still has to be maintained at proper body temperature so there would be no difference in brain temperature sleeping the same way in water or in air. If the body's temperature regulation is designed assuming that the brain is always generating too much of its own heat and doesn't need to be supplied with additional heat like the extremities do, then, it would still need to keep up that heat generation independently of the rate of heat loss to the environment.

It's weird to me too. But articles says it's because the brain alternates activity in hemispheres. It switches activity to each hemisphere, possibly as it gets too cool.

>These marine mammals have evolved a half-brain style of sleeping, perhaps as a way to remain alert enough to avoid predators and drowning.

>Because part of the brain is always active, it’s always warm. As a result, it never triggers R.E.M. sleep.

It's weird they don't mention the much more obvious fact that an air-breathing mammal might need to maintain some level of alertness when sleeping in water.

From the article: "These marine mammals have evolved a half-brain style of sleeping, perhaps as a way to remain alert enough to avoid predators and drowning."

Regarding the rats: the study has an explanation for that too. It says the rats were woken up thousands of times a day, and they are catching up on the lack of sleep rather than lack of REM.

The article says they have more REM after they miss out on it. The ratio is not the same as normal sleep.

Doesn’t make any sense to me. I don’t really think that people sleeping in warm countries have less R.E.M. sleep than people sleeping in cold countries...

The issue with your statement though is that people in colder climates adapt via technology to keep a comfortable environment at night (houses, blankets, etc). So it’s not the kind of comparison that you can make like that.

If anything, it could be the reverse, it's just too complicated to immediately see. As another counter-obviousness example, living in a cold environment helps you lose weight because your body has to generate more heat all the time. But in real life, people keep their bodies warm with clothes and houses and actually gain weight in winter because they eat more and don't go outside and exercise as much. Just because the weather's cold or hot, doesn't mean people are.

Sounds like pseudo science to me.

My first reaction was that it seems so inefficient to bring the entire computer online just to maintain its temperature - which could be controlled physically using blood. But couldn't we test this by artificially keeping the brain warm and seeing what sleep patterns emerge?

As an anecdote, I always have intense dreams with fever. I'm prone to kidney infections so I experience this often. My head is clearly very warm, my brain swells a little with the heat, and my dreams are clear and intense.

this would seem likely because you aren't sleeping well or deeply, hence able to recall the dreams.

I usually recall my dreams, fever or not. There is a marked difference between them.

> it seems so inefficient to bring the entire computer online just to maintain its temperature

In what sense is it inefficient? In terms of energy 100% is converted to heat anyway

Some of that should be converted into lowering entropy (building neural structures etc) so I don’t think 100% escapes as heat.

We're talking about biological evolution here. There's no design process, except by chance and fitness. Even some traits that might be better for an animal may not make it -- there's a whole chance and probably thing there.

There is a paper where some researchers look at biological processes vs computer software. Biological processes seem to have a lot of repetition between systems:


If heating is your goal, thermal inefficiency isn't really a problem.

There's also the issue of the blood-brain barrier. Might make normal thermal regulation insufficient. In any case, it seems like they intend to do further study to see if this hypothesis is really correct and if there is more to the story. Who knows--it might be that for some species it's not necessary, but left over as a vestigial behavior like a tailbone. That's pure conjecture on my part, excuse me.

Something I find interesting about this, if a little off-topic, is that the Ancient Egyptians had a misconception that the brain simply existed to cool the blood. When they mummified a body they threw that part in the trash. How far we've come.

"If heating is your goal, thermal inefficiency isn't really a problem." - I guess my objection is based on an analogy with human-made devices, which degrade the more you use them. Thermodynamically it's fine to power up a computer to produce heat, but the inefficiency I really meant is that every moment the computer is running is another moment when it can go wrong. Perhaps the brain isn't like that: perhaps the brain's activity doesn't change the mind's lifespan. Or, more prosaically, perhaps the mind does degrade faster if it's active more, but evolution didn't care as long as the meatbags kept pumping out their genes.

True, that's a good point. As far as evolution is concerned, if you have effectively raised offspring until you aren't fertile anymore then the warranty is up and all the time you get after that is a bonus. There isn't really much pressure to select for longevity after that period.

I've heard a lot of office buildings will leave every floor's fluorescent lights on in the winter to take a load off the overnight heating bills. I'm sure that lessens the lifespan of the bulbs and ballasts, but for people to adopt that practice it must still be economical in the final tally.

Who knows, I'd love to see more study on this because I think everybody in history has always wondered what the hell is up with dreams. "To keep your noggin warm" seems like a super anticlimactic purpose and I love it. It's so absurd and nihilistic it's funny, which kind of jives with nature's whole style.

> I love it. It's so absurd and nihilistic it's funny, which kind of jives with nature's whole style.

I couldn’t agree more and wouldn’t it be so indicitave of the scientific trend that ensmallens man?

From the center of the universe, the sun orbiting our land to the Big Bang and inflation theory where there is no center.

From gods chosen creation to common ancestors shared with apes.

Now even our dreams to be punctured?

Brilliant to think about!

Actually it's the thermal expansion/contraction cycles of turning on and off, as well as ablation of the electrodes on startup. So leaving it on is marginally better for bulb longevity.

My own understanding of this, as a layman who has been driven by curiosity to read a bunch of journal papers on the subject:

Consider the developmental paths of cell differentiation/specialization over the evolutionary history of animal species. Neural cells and muscle cells both have a common progenitor cell (i.e. a stem cell will differentiate into this cell, which then has one or the other as potential descendants.)

This sort of means that there is "code sharing" going on between the structure of nerve and muscle cells, moreso than between these and, say, liver cells. And, since animal muscle cells evolved before animal nerve cells, it's likely that it's directional—it's less that these are two "classes" with a common, well-factored "parent class"; and more like nerve cells are a "subclass" that has something closer to muscle cell functionality in its "parent class", and overrides that functionality heavily until it does what it wants.

Or, to put that another way: ignoring the particular way nerve tissue structures itself, and the receptors it grows, etc. Functionally, nerve tissue is, effectively, just muscle tissue that is balanced very precariously by its electrochemical environment to be in an always semi-conducting state, where the tissue as a whole will never go all-the-way-resistive (that's excitotoxicity, both in muscles and in nerves) or all-the-way-conductive (that's flexion in a muscle; but that's a seizure in a nerve. For the nerve-as-muscle, that's okay, it's built for that; but for the nerve-as-controller-of-functioning-elsewhere, that messes things up.)

Nerve cells rely on a lot of the code they've "inherited" from their muscle-cell "parent class" for their functioning. This includes the part where muscle tissue (and so nerve tissue) gets oxygenated by pumping its own blood through it [and clears waste by pumping its own lymph through it], by inducing subtle sinusoidal increases/decreases in electrical activity to cause increases/decreases in capillary diameter.

This is especially important for the brain, given that the blood-brain barrier prevents blood from being "driven" through the brain by the heart. The brain has to pump itself, and therefore has to induce electrical activity in itself.

For the brain-as-muscle, this electrical activity is useful. For the brain-as-neural-network, this electrical activity is—from what I understand—a form of interference to the signal the brain is trying to propagate/compute. The brain has evolved to deal well enough with this interference to make it not get in the way of accomplishing the goals the brain was evolved to accomplish; but, necessarily, evolution never bothers to go any further than that. So, as long as the effects are "harmless" to adaptive fitness, that interference is allowed to leak through as a signal interpreted by the brain-as-neural-network as thought, or perception, or what-have-you.

And thus, dreams.

Fantastic idea! It better explains rebound REM than "brain heater", while still explaining the lack of REM in aquatic mammals - half of the brain is always "pumping".

thats what i thought. seems easy enough to do. I'd like to see more evidence to support that theory, I'm rather sceptical.

i was remembered of that very rare condition where your brain cannot ever enter the deeper sleep states. these people usually die rather soon after onset of exhaustion. what has been confirmed through these people is that narcotics dont help at all. while you can ise it to make them 'sleep' long enough, it's not helping with the condition. so brain warming cannot be the main reason for REM, if it's one at all.

edit: fatal insomnia is what i was referring to.

> n fact, the earlier studies on R.E.M. deprivation might not have been as compelling as they once seemed.

> In those earlier studies, researchers kept animals from going into R.E.M. sleep by waking them up. “In some experiments, they wake up the animals a thousand times a day,” Dr. Siegel said.

So much science that I used to take for granted is garbage.

I think the problem is that society accepts results like this uncritically. I don't think this study was innately garbage, as waking the animals was the only way they could disturb rem sleep, but there should have been (maybe there were?) big caveats attached.

As a society we need to realize that studies are not "proof". The scientific process is more than just studies.

The actual papers probably did. News writeups are often much more sensationalized than the actual research abstract and conclusion. This PhD Comic describes it really well:


This article is actually fairly balanced in comparison to many of the other paper-based news write ups I've seen.

My favorite quote from a favored author, and former head librarian of the US Library of Congress:

"Imagination drew in bold strokes, instantly serving hopes and fears, while knowledge advanced by slow increments and contradictory witnesses."


-The Discoverers


It isn't the science itself, it's the conjecture built up around its popularization.

"We have two observations of X when Y"

"So Y causes X!"

"I didn't say that."

"Hey everyone! Y causes X, probably."

Quick, get your own Z, it helps prevent Y...

It is often valuable to question how you know what you know.

> Quick, everyone get your own Z, let's test whether it helps prevent Y... it might just work


The text in italics was about people trying to cash in on the media exaggeration by selling bogus products based on the initial false assumption.

That's how it is. Don't be discouraged or disillusioned. The beauty of science is not that it gets things right, it's that it is always a little less wrong than before.

I many cases, the science is not talking about real proofs (in the mathematical sense) but marely about assumptions based on some (or many) indicators. Such "science" should not be part of the real science. The real sciense should contain ONLY the things which are proved with real proofs!

But this is not possible because the scientific method is based on inductive, rather than deductive reasoning. Once you get into deductive proofs, you're not doing science anymore. You're doing maths.

If that were true we'd be paralyzed by solipsism and skepticism. Sometimes you have to make do with a working hypothesis.

You can't eat mathematical proofs.

But that's the essential point, that what you call working hypothesis are actually untested hypothesis. Using an untested hypothesis as a working hypothesis is bad science that may actually be counterproductive, unbeknownst to the users of the working hypothesis.

As opposed to doing nothing because of some kind of epistemological catch-22? How do you plan to start with a tested hypothesis, as opposed to an untested hypothesis? How did you test it before you conceived it?

It's not an ideal solution but it's a practical inevitability to make some assumptions to get things moving. The best of us come back and check the assumptions as soon as they are able to quantify them and perform an empirical experiment, and as a backup we have others independently verifying people's assumptions by trying to replicate experiments and seeing if the predictions line up.

Nobody has magic powers to divine truth from nothing. Science is a filter which gradually strains out the misinterpreted and the biased and the just-plain-wrong, but it still starts with subjective observation and reasoning. As does all human experience. The important part is that it moves away from that instead of stopping there.

You always depend on an untested hypotheses at some point. They are absolutely unavoidable.

Here's a very simple example:

- How much testing is sufficient? For example, as far as we can tell the laws of physics don't vary over time scales that we can directly observe. This has been true as far as we can tell.

- So far so good: the hypothesis that 'the laws of physics don't change' has been 'tested'. Next question: is that amount of testing sufficient?

- Well, that seems to be an untestable proposition; it's always philosophically possible that things get pulled out from under your feet, and no amount of testing up to time T lets you know with complete certainty what happens for t > T.

Go even one or two steps deeper, and you quickly find yourself resting on either unfounded, untested, or untestable assumptions.

That's diverging from what was being discussed and creating a strawman that you are now debunking. The context is the use of method-less science that uses assumptions based upon indicators as working hypothesis - not that there are always base assumptions but that the refutation of testing while holding assumptions up as tested is something that is bad science.

I thought what was being discussed was:

"The real sciense[sic] should contain ONLY the things which are proved with real proofs!"

As in mathematical proofs. You can't prove most things the way you can in mathematics, which is rock-solid through tautologies and self-consistent abstract models. When you are observing actual reality you have to make some assumptions.

What was being discussed was the reply in the context of my comment that had been replied to. I was not the original GP poster. Sorry, if I didn't make that clear.

You are describing the anti-intellectualism movement. We wouldn't have many of the conveniences of today with that type of thinking.

You expect a future where brain implants will be tiny computers that make you smarter, but perhaps instead you get a future where brain implants are tiny heating coils that make you wake up earlier.

I‘m pretty sure I won’t have any kind of brain device installed, no matter the upsides. No upgrades for that hardware!

I feel the opposite. As soon as it's possible to replace biological neurons with artificial one, even if they aren't perfect replicas, I'm in. I'm pretty sure the mind is an emergent phenomenon robust to slight variations in the lowest levels of hardware, and I'd like to live for billions of years. (And if the tech lets you process faster, it could feel like trillions.)

My personal theory was that dreaming allows humans to rest while staying somewhat alert to the environment (because outside phenomena tend to manifest inside dreams). The phased aspect of REM sleep allow groups of humans to sleep at night while guaranteeing that some members of the group will be in REM rather than deep sleep at any given time, so the group never becomes completely unconscious of the surrounding environment.

My couch scientist theory is that REM sleep is phase when brains reconnects and tunes quantum links between neuron groups (assuming brain is big quantum computer underneath), that why there are chaotic movements.

That's a pretty big assumption there. There's a huge difference between "exploits a quantum effects" (like classical computers today) and "is a quantum computer". What evidence leads you to suspect the latter?

My guess was that dreaming is just cross-talk while our brain consolidates and archives our memories from the day, and uses them to train for the next day.

Even if dreams arose as interference, given that evolution "tries" to capitalize on every possible opportunity, doesn't it seem likely that there's useful information in them? The ego controls the self all day in a process that discards lots of information; at night, other processes (perhaps that were ongoing in the day, or perhaps new ones) might be able to "speak" and be heard where in the waking state they would be ignored.

Hypothesis, not theory.

Yes, but....the usage you apparently objected to is how the word theory is used in the world, i.e. it's one of the meanings of the word. At least, I looked in a couple of dictionaries and they gave sentences just like the one you objected to, as examples of correct usage. Yes, in science, the word has a narrower meaning; I'm sure people on here know that.

REM cannot be strictly necessary because I don't get any REM sleep (or so little it's difficult to detect) due to severe sleep apnea (AHI=148). Three different sleep studies (using two unrelated labs+doctors) produced very similar results. The hypnogram was always a narrow rectangle of rapid oscillations (λ ~= 25-35 seconds) between NREM stages 1 and 2.

Obviously this has caused harm, but it's difficult to isolate how much of that harm (if any) is caused by lack of REM when the dips in oxygen saturation could easily be the causing most of the problems.

I think it's got more to do with maintenance, not temperature. From my own experience, lack of sleep leads quickly to "spinning my wheels" as my thought process becomes bogged down to the point of not being able to "finish a thought".

At some point the brain needs to clean out molecular by-products, replenish stores of needed neurotransmitters, I'm sure at some level there's some garbage collection going on, maybe mark and sweep everything out of short term memory that's been moved to long term. Maybe dreaming actively helps us to connect disparate thoughts and concepts into some kind of of aggregate.

Some of my best thinking comes easily after a great nights sleep...

If this is true, then why are there such extreme consequences to not sleeping for extended periods in mammals ?

Without REM sleep (deep sleep does not work by itself), you lose control of your emotions in 2 days at most. You go insane between 1 and 2 weeks (mania, depression, and likely quick switches between those 2), and your body will physically shut down and die a week or so after that. (note that in humans it hasn't been well established that death occurs, for obvious reasons. In mice, however, it's well studied and accepted. No sleep for ~10 days, half of the mice just die).

This seems to be a bit at odds with the idea that it's just a brain warming technique.



Natural question follows: can humans train ourselves to do the same as animals in water? Live a bi-phasic life where creative / analytical sides of brain each get to "sleep"?

Probably nobody knows, but I guess these animals have an special adaptation to make bi-phasic sleep possible. I guess that to evolve it you only need to keeping an isolated human population for a few million years were you kill at random some of the one that are sleeping, to encourage swallow sleeping and later bi-phasic sleep. Another possibility is trying a genetic modification, but it's impossible with the current technology and knowledge, perhaps in the future ...

If you want to have a peek, you can read about the Wada Test https://en.wikipedia.org/wiki/Wada_test . (And somehow related https://en.wikipedia.org/wiki/Split-brain )

It would depend on how this evolved in the water mammals researchers have found this in. Maybe some land mammals do the same thing, we just have preformed this type of experiment on them yet.

If it does appear in humans, it'd probably be best to study those with extreme insomnia or people who claim to have very short sleep patterns (<2 hours a night) or maybe those who are narcoleptic (the other end of that scale).

So we need a volunteer to put the electrodes on the hippopotamus?

The character Kraiklyn in the Iain M Banks novel Consider Phlebas has had enhancements to his brain to be able to do this.

The creative / analytical dichotomy of the brain is a myth I'm afraid.


Sure, but we're not "animals in water" as OP wrote? Perhaps reading the full sentence before commenting? Just an idea...

I read the entire sentence, but had not yet read the article, and had not had my morning coffee yet. After looking through the article, given the context it's pretty obvious they were probably referring to Amphibians.

The sentence is poorly worded, because when a human is in water, it's an animal in water. Through that lens alone it makes no sense.

It is not the same thing, but some people learn to switch between analytical and creative activity. I am many more could, if they tried hard enough.. Of course, there’s no recipe for how to learn this sort of thing which makes it a lot harder.

True. I go through phases like this. For two whole weeks I will write (creatively) and draw, and then I usually find there is a day or two when I can't seem to do much of anything useful besides menial tasks and watching stupid videos. Then I go through another two weeks of analytical stuff-- reading, catching up on work (programming and mathy stuff), and doing planning for my writing. I can control it, if necessary in a day. The urge to do one thing or another is an intense "mood" for me (which, like I said, normally has a two-week period). This indicates some kind of synchronization, so if I want to switch I use musical triggers: airy, melodic music for creativity, and beat-heavy, algorithmic music for analytical activity. I only need to listen to the music for a few minutes to switch, and of course it sometimes doesn't work but I don't know of any other triggers. Possibly going out for a short walk could do it too? Meditation?

I feel weird about this post. It feels like it's posted by people wanting to somehow give evidence or support to the idea that sleep is good and necessary. I feel weird about the fact that that is something that needs to be proved. Don't you trust in the signals your body is sending you? Do you really trust a model of average human behaviour more than your body; which is perfectly conditioned your precise enviourment and history?

To me this looks like evidence that seals don't enjoy sleeping in water very much. Not much else. Also sleep-deprived rats seem to consume more calories. Wow.

Have there been studies of sleep need, or sleep quality, vs temperature?

I wonder if the modern (American?) trend of running heating at night is affecting people's sleep?

As a Briton who moved to the USA a few years ago, I found the American habit of running heating overnight to be very odd - as I think most Europeans do. In Britain, we tend to run heating only during the day, unless it's extremely cold outside.

Do we know which animals experience the most R.E.M. sleep, or if humans display atypical R.E.M. sleep characteristics compared to other animals?

Not a direct answer, but immediate REM sleep is a symptom of narcolepsy. So the animal displaying the most REM sleep is probably a narcoleptic human.

It just makes you tired, no matter how much sleep you get.

I thought the received wisdom was that REM sleep was connected to memory. E.g. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3768102/

I've always found that I get exceedingly warm when I sleep. Whenever I fall asleep on a bus or plane, I eventually wake up drenched in sweat. It's not from bad dreams or night-terrors; my body temperature just increases. I now wonder if it's specific to REM.

Every night I have to throw a few blankets off of me in the early morning.

I read somewhere (?) that the body cools during sleep. Somehow one of the implications was that taking a shower before sleep can help you sleep faster.

The fact that a body is sweating indicates not necessarily that it is hotter than during the day; it could be that the body is trying to be cooler than it was during the day. (Of course, if you wake up feeling hot, it's probably because you're hot.)

Thermodynamically, it makes sense that being in a bed, with insulation against almost all of one's body, would (once the insulation gets warm) leave someone warmer than when they're sitting or standing -- at least if the ambient temperature is the same in the day and night.

Night sweats can also hint at a chronic infection or hypoglycemia. Can you exclude those?

I would be cautious about extrapolating seal metabolism to mammal metabolism. Diving mammals have many special adaptations to deal with high pressure, breathing, oxygen, nitrogen etc. Is an interesting history in any case.

My personal theory is that REM has many uses - one of which is giving the brain an opportunity to do reinforcement learning.

Is it just heating of brain required or blood flow? One crazy way to test would be to artificially heat the brain!

If you found that interesting you should read about why our bodies are at the temperature they are, 37C. The tl;dr is to fight off fungus.

Another clickbait title, it should really be “an unproven theory for why mammals need R.E.M. sleep based on seals”

Or even

“R.E.M. sleep might be a trait evolved to keep your brain warm at night”

Applications are open for YC Summer 2019

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