Evidence that birds sleep in mid-flight | Hacker News Reader

Evidence that birds sleep in mid-flight

nature.com

161 pointsMz75 comments

I'm reminded that we still don't know conclusively what purpose sleep serves[1]. We know about things unimaginably distant in scale or time -- like quarks and the Big Bang -- but we don't know the answer to something seemingly pedestrian.

[1] https://en.wikipedia.org/wiki/Sleep#Functions

It's meant to separate days. That way each day is a new day, and you have some sort of mental partition between each one.

meant by who?

Whoever you want.

What purpose does that serve? Why would it be more advantageous than a continuous time stream?

We don't have similar separation between morning & evening, for example. (though I've always been a fan of the siesta)

Its meant to serve physiological homeostasis in some way yet to be completely understood :)

as someone who has had a rotating schedule their entire life, I wonder what that would mean for my brain.

What do you mean by your "entire life"? Since you were an infant? I have non-24[1], but it didn't manifest until I was in my 20s.

[1] https://en.wikipedia.org/wiki/Non-24-hour_sleep-wake_disorde...

I think it is because of a few interrelated things:

1. Ill-posedness of the problems: what is sleep?, why do animals sleep?, etc. vs. the well-posed physics problems such as, what happened at the beginning of the universe?

2. Data sufficiency: there is too much variability across people and animals to have enough representative data marginalized on each variable for many biological systems in general, and sleep in particular

3. Complexity of the system: cosmological systems and even subatomic particles are less complex models than biological systems

On incentives for researchers, there is also the maxim that there are many unsolved and interesting problems to work on, but scientific judgement is in knowing which are solvable. Of course, the occasional maverick goes after the hardest, most ill-posed, most unsolvable problem. But the existence of such mavericks is not guaranteed for every problem, sleep biology included.

Brain is very active when you sleep, a lot is happening. Imagine sleep is just a mutex on big set of data on which computation needs to be made, compression, data reorganization, index optimizing. Data that you received that day is segregated , decisions are made which data to keep and which to delete. After that lock is freed and you wake up. If you don't sleep your system will run slower and slower over time and will timeout more and more queries until it will start to return corrupted data. You need to sleep to keep the system healthy.

Sure, but that's just pure speculation.

Sleep = vacuum freeze analyze

Garbage Collection

We need a concurrent GC.

It's called Red Bull. It gives you wings. Nature is now selling native advertising.

That is a theory. There are many other theories too. The OP's observation was that we don't really have a firm grasp on which theory or theories is/are correct, in the way we (believe) that we understand something like the nature of chemical interactions, or the limits of finite-state computation. Sleep, like many many other functions of the brain, not the least of which being its most recognizable function (to us), remains something speculated about without broad consensus. As per the fantastic quote in the linked wiki article,

"As far as I know, the only reason we need to sleep that is really, really solid is because we get sleepy."

A theory, sure. Another thing to remember about biological systems is that complex features like sleep rarely serve just one purpose. There will generally be one cause that was killing lots of members of a population which the feature was developed to mitigate, but other "reasons" may have grown over time as evolution "discovered" other ways the feature can help.

I think in general, good-sounding theories produced by the "speculative evolutionary biology" game are actually very often right -- but only describe one facet of a supremely messy system with boundaries that don't conform cleanly to the mental models we try to impose.

Sleep does more than one thing. One advantage is lower energy needs, so saying we don't know what it does is very misleading. We don't know Everything sleep does in Every situation, but the same is true of everything else.

A better way of posing it might be to say that we don't know why the stuff accomplished during sleep can't be accomplished during wakefulness. We do know that the purpose of sleep is to maintain pretty much everything.

>> We know about things unimaginably distant in scale or time -- like quarks and the Big Bang

Oh, we do?

That's funny, Afaik quarks still confuse physicists because they behave erratically and appear to be composed of other particles. And the big bang is simply an assumption that makes the cosmological model work well for our human brains, but it is not how the universe started specifically.

We don't know jack shit, that's the bottom line, and we like to pretend we do.

"We live in a culture of casual certitude. This has always been the case, no matter how often that certainty has failed. Though no generation believes there’s nothing left to learn, every generation unconsciously assumes that what has already been defined and accepted is (probably) pretty close to how reality will be viewed in perpetuity. And then, of course, time passes. Ideas shift. Opinions invert. What once seemed reasonable eventually becomes absurd, replaced by modern perspectives that feel even more irrefutable and secure—until, of course, they don’t."

http://www.goodreads.com/book/show/27068734-but-what-if-we-r...

Yes. We were, are and will always be living in "modern times"...

> That's funny, Afaik quarks still confuse physicists because they behave erratically and appear to be composed of other particles.

I'd really like a source for the composite quark claim.

Well, String Theory.

What evidence is there for quarks being composite?

Couldn't have given a worse answer.

1. String theory has no experimental backing.

2. It doesn't posit that currently fundamental particles become composite.

For a decent explanation see:

https://physics.stackexchange.com/questions/38767/relation-b...

You might consider the The Relativity of Wrong:

http://chem.tufts.edu/answersinscience/relativityofwrong.htm

Sure, maybe we know jack shit. But we know unfathomably more jack shit than in previous ages.

You might call this bias of similar scales: things that happen within our attention must have similarly brief explanations.

Scientists still don't understand the mechanisms behind lightning formation! Turbulent flow is also another great unsolved physics problem.

I looks like they're referencing this paper[1] which suggests that like dolphins, birds can sleep 1 brain hemisphere at a time. That's pretty interesting.

[1]http://www.nature.com/nature/journal/v397/n6718/full/397397a...

Ducks do that while sleeping. If you observe ducks sleeping near a pond you'll typically see them sleeping with one eye open and one shut. That way they can still stay alert enough to watch for predators.

I've had several pet birds and I've observed the same thing from them. I just assumed they were just pretending to sleep.

Your comment made me wonder how common it is with birds and it seems to be trait shared with all of them[1]. Pretty interesting. I asked my friend if her bird does that and it only seems to when it's not in a deep sleep yet, but eventually it will close both eyes.

[1] https://en.wikipedia.org/wiki/Unihemispheric_slow-wave_sleep

Gives a new meaning to the term, "sleep with one eye open at night." :)

Let me venture a guess: that expression comes exactly from observing animals keeping one eye open.

Haven't bothered to search for this, but do you know offhand if they also have a L-R hemisphere/body crossover?

All vertebrates have that, IIRC.

> On land, birds can switch from sleeping with both hemispheres simultaneously to sleeping with one hemisphere at a time in response to changing ecological demands17,18. During such unihemispheric slow wave sleep (USWS) birds keep the eye connected to the awake hemisphere open and directed toward potential threats.

This reminds me of the low-power sleep states of PCI (L0s, L1, etc.) or a multicore CPU

I wonder if it correlates with flocking behavior: More individuals makes it less risky to have a blind-side.

If a potential predator starts to circle them, would they switch sleeping sides without waking in order to follow it with their awake eye?

We really need to figure out what sleep is for. It seems important.

It functions as a regularizer in learning and reduces unnecessary energy use.

While those are benefits of sleep, I don't believe we consider them to be the reason we sleep.

https://www.nih.gov/news-events/news-releases/brain-may-flus...

This seems the most likely to me, purely based off the fact that going on without sleep gives the same feeling in the head as walking a long time gives me in the foot and legs.

So, the brain is an electro-chemical engine, and like other engines, it must be shut down to perform regular maintenance so the engine lasts longer, and is less prone to error conditions.

Not just the brain, but the whole body.

If I go without enough sleep, my hair and skin start getting greasy, my immune system is weakened, my muscles start feeling tired, etc.

Your hair and skin get greasy because you aren't bathing, not because you aren't sleeping.

You are incorrect, as you can trivially verify yourself with a few days’ first-hand observation.

Even if you don’t bathe at all for a few days, or just rinse yourself with water (skip the oil-stripping soap for this experiment), the greasiness of hair/skin fluctuates noticeably throughout the day, and increases substantially after a day or two of insufficient sleep. (That’s right, the greasiness will go up and then subsequently go down, without your doing anything special in between.)

[Side note: it’s pretty disappointing when people make confident sounding assertions about things they clearly don’t know anything about and didn’t bother to do the tiniest trivial bit of study. I find this happens quite commonly even in discussion venues where there’s ostensibly a cultural value placed on reasoning and evidence.]

I've been sleep deprived for extended periods before and never noticed the difference. Is there sample size > 1 evidence for this you're aware of?

You want to casually study myself while sleep-deprived? That sounds like a recipe for trustworthy results, I'm sure.

Of course I'm going to get more greasy after two days of not using soap. I don't see how that would verify anything that you are claiming.

I'm currently reading Where Song Began[1] which discusses this, and is super-interesting besides. Has an anecdote of WW1 pilots encountering flocks of dozing swallows at night.

1. https://www.amazon.com.au/Where-Song-Began-Australias-Change...

Just me or the fact that they managed to record the brain activity of birds in flight seems more interesting than the fact that birds sleep in flight?

I mean, the reason I clicked the link is because I wanted to learn HOW they recorded the brain activity.

I just never imagined EEG sensor + logger can be miniaturized to such a degree.

Is it possible for a human(for example a meditator) to learn to sleep with one eye open ?

Yes. My son used to sleep with one eye open. After serving in Vietnam, my father habitually napped with both eyes open.

A human can sleep with one or both eyes open, but as far as I know it's not at all related to brain activity. The person's brain is still fully asleep.

See: https://en.wikipedia.org/wiki/Lagophthalmos#Nocturnal

Your brain can still react to things when it's asleep though. You can be shaken awake, or shouted awake, or woken up by a bright light. It doesn't seem impossible that seeing something dangerous while asleep could wake you up.

Presumably this is exactly why my father learned to sleep with his eyes open while caught behind enemy lines in Vietnam for, as I understand it, a month or more.

I remember once waking up, when, for some reason, I had my eyes open. It was weird: all I saw was just shape; not shapes of anything in particular, but just pure essence of what shape really was.

Then, over a period of a few seconds, and in distinct chunks, it suddenly resolved into things I could actually identify, and I realise what I was looking at.

After the fact I put this down as the pattern-matching areas of my brain starting working rather more slowly than the rest of my visual cortex, so it took a while before they starting resolving what my eyes were seeing into meaningful concepts.

I have a variety of mild sleep disorders. I can recommend them as a hobby; they're fascinating. Admittedly, it helps that being hag-ridden is so familiar it's practically an old friend; the first few times were utterly terrifying[ * ]. But the weirdest of the lot is something I've dubbed false memory syndrome, where I'll remember something, while quite conscious, which never happened, and frequently in quite some detail. Just last night I had apparently written and drawn a comic book. That was better than the time I was on the run after having killed someone...

[ * ] Apparently there's a theory that this is where alien abduction myths come from.

> It was weird: all I saw was just shape; not shapes of anything in particular, but just pure essence of what shape really was.

Wonder if that's what it's like for a newborn.

https://en.wikipedia.org/wiki/Molyneux%27s_problem

There is some conjecture that babies have synesthesia too - so it would be a case of visually receiving sense shape imagery and the brain interpreting that as colours, forms, tastes, pressure, warmth, scents, and such too. Now that seems like enough to make you scream and seek solace in innate behaviours.

Yeah, that first part happens when you on occasion mentally wake up faster than your body does, which is then still sleep paralysed. It can be pretty scary the first time, as it feels as if there are shapes (ie. humanoids) lurking around you. Also when you try to move your brain remembers and plays the motion as if it's happening, but you're still paralysed so you keep resetting back to whatever position you're lying it. The same probably holds true for sight.

That seems to imply that the unconscious part of the brain still can use visual input while the 'mind' is sleeping (just like it can 'read' the output of your ears, so that it can wake you when your alarm clock rings)

Does anybody know whether that is true? Similarly, does the knee reflex work when humans are sleeping?

Edit: on the knee reflex, there's http://ajplegacy.physiology.org/content/68/2/345. Paywalled, but the knee reflex seems diminished/absent during sleep.

Edit 2: quite surprising, as the knee reflex is so fast that a nerve signal cannot travel to the brain and back within its response time (and of course, there's the spinal cat, that walks on a treadmill after its brain has been disconnected from its spinal cord (https://en.wikipedia.org/wiki/Spinal_locomotion))

I don't know, but I would suspect that doing the knee reflex thing with someone asleep would often cause them to wake up?

But, also, I think I heard that the signal doesn't get all the way up to the brain before the signal to kick or whatever starts going down? (I am unsure of this though.)

So, I would think that the brain being like it is when the person is asleep wouldn't stop the kick reflex thing, because the brain isn't even what causes the kick reflex thing?

So my guess is that they would kick and then likely wake up.

I'm not certain of this though. Please correct me if I am wrong.

Yes, the reflex is controlled by the "dorsal root" ganglion, at the spine.

Perhaps sleep is a full body affair.

Bright lights wake you up right? I'm not sure how complex a stimulus you can interpret while asleep, but it wouldn't surprise me. After all your brain doesn't turn off, it's just in a different mode of operation.

EDIT: To be fair, I wouldn't be surprised either if there was a dedicated circuit that sensed change in luminous energy that bypassed V1. I'd google but I'm on limited wireless right now.

Here's an article that's informative, but doesn't quite answer the question: http://phenomena.nationalgeographic.com/2014/03/13/how-light... It talks about the blue light pathway that David Berson discovered in intrinsically photosensitive ganglion cells. This was surprising because, while this was before my time, I think most people thought that it was the rods and cones that were photosensitive, containing proteins called opsins that transduce light into neural signal. Rods and cones can be considered input cells, ganglion cells sit above the retinal circuitry can can be through of as output cells that perform computations from their inputs and pipe that information down the optic nerve, which eventually leads to primary visual cortex located at the back of your skull. Though there are plenty of opportunities for side branching minor paths.

Can split brain patients do that? https://en.wikipedia.org/wiki/Split-brain

Kind of OT, but does anyone happen to know if a person with a severed corpus callosum can sleep with one hemisphere at a time?

It would make sense that birds do fall asleep while flying, as they are sometimes prone to fly into buildings or windows that were obviously visible.

wow, birds can auto-pilot.

if two halves of the brain can work independently, wouldn't there be a sync problem! I wonder how the brain handles it.

Sometimes I can hear myself lightly snoring while asleep. Its kind of weird

I wonder if they have that "I'm falling" dream and then wake up.

I think that's actually unique to primates (who do/did sleep in trees).

Personally, I always have trouble sleeping while flying.

Reply on news.ycombinator.com