I mean literally not sleeping for the rest of their life, not like giraffes that only require short rests, or dolphins that can sleep with half of their brain at a time, or like otters and sharks that sleep walk.

An external solution like drinking caffeine or using other drugs is out of the question.

I don't mind, however, if one of the creature's organs can produce caffeine or another chemical that allows the creature to never sleep, but please elaborate on how it would work instead of merely hand-waving it.

(I use science fiction tag because it may not possible for science-based answer, but much appreciate if it really possible in real science)

what organs need or developed for a life biological creature to never need to sleep?

  • 4
    $\begingroup$ Its not about organs or chemicals, but about radically overhauling a body plan that requires regular sleep or risks breaking down. You can't get there from here; you'll need to design some new things that never needed sleep in the first place. $\endgroup$ – Starfish Prime Sep 12 at 13:42
  • 30
    $\begingroup$ The brains of vertebrates need sleep, or else they malfunction with very serious effects. We don't know why exactly the brains of vertebrates need sleep; all we know is that they do, and that nothing we tried can replace sleep. $\endgroup$ – AlexP Sep 12 at 13:54
  • 11
    $\begingroup$ We barely understand sleep. What we do know is that during (deep) sleep your body goes into regeneration mode. This is important as people with fibromyalgia cant enter deep sleep properly and these people dont heal the damage from normal life as well. Each day micro damage to muscles, tendons and other tissues that naturally happen over the course of a day are only partially healed. Do that a full lifetime and you have permanent pain. We also have a fair idea its needed for processing things you've learned/experienced that day. So we dont know how to remove sleep without compensating for this. $\endgroup$ – Demigan Sep 13 at 5:27
  • 24
    $\begingroup$ One might say the solution is a no-brainer ... ;) $\endgroup$ – Hagen von Eitzen Sep 13 at 6:59
  • 6
    $\begingroup$ You can literally stop sleeping for the rest of your life right now, but with the caveat that "rest of your life" will be massively foreshortened.. $\endgroup$ – mccdyl001 Sep 13 at 9:59

15 Answers 15


Polycephaly is a medical condition, in which an animal has multiple heads. This is a special type of conjoined twin. 2 is the most common form, but there is at least 1 case of 3 heads documented (a turtle).

In many cases, each head is capable of sleeping independently of the other, allowing your Biological Creature to function without turpor.

In some cases, each head controls a different part of the body (e.g. Abby and Bittany Hensel, born 1990, each control one hand and one leg. Interestingly, if one of them develops stomach ache, the other twin experiences the sensation instead).

In other cases, both heads are able to control all-or-most of the body - however, when they disagree on an action, this can cause the body to "lock up" and paralyse.

What you require is a situation where both heads can control the whole body, and sleep independently.

  • 14
    $\begingroup$ Isn’t this just a more complex version of the already vetoed dolphin brain? $\endgroup$ – Joe Bloggs Sep 13 at 7:59
  • 10
    $\begingroup$ @JoeBloggs The reason that the dolphin brain was vetoed was because the dolphin undergoes turpor - while not completely asleep, it becomes slow, sluggish, and unable to react or make complex decisions - essentially stuck on "autopilot", and for longer than most mammals sleep (e.g. half the sleep, for double the duration?). The aim here (by having 2 full brains instead) is to allow the resulting creature to avoid that situation, and the dangers resulting from it. If you consider the Dolphin Method analogous to being heavily drunk, then this is more like wearing an eyepatch and 1 earplug. $\endgroup$ – Chronocidal Sep 13 at 8:16
  • 3
    $\begingroup$ @Chronocidal except a creature with two brains would likely have much more advanced cognitive functions, and disabling one of the two brains would comparatively have the same cognitive deficit as disabling half of one. Your CPU has multiple cores and it can work with all but one disabled, but it's obviously unable to perform the same amount of work while doing so. The half-sleep dolphin thing would be a single core cpu undergoing thermal throttling. In both cases the overall computational performance is drastically reduced. $\endgroup$ – Demonblack Sep 13 at 13:04
  • 3
    $\begingroup$ You can get around that by saying that your creature evolved to only ever use one brain at a time, so it can't run them in parallel, but that would be highly dubious from an evolutionary standpoint. Once you've incurred the cost to build and maintain a second brain, you want to utilize it to its fullest potential. $\endgroup$ – Demonblack Sep 13 at 13:06
  • 7
    $\begingroup$ I guess though that in that scenario you could argue it's not so much that the creature is "sleeping" when one of the brains is off, but more that it has engaged turbo core when both are on. Everything is relative... $\endgroup$ – Demonblack Sep 13 at 13:51

Just use an alien.

If it has a large brain you can't get around sleep, every organism we know of with a brain that is enlarged (complex at all) sleeps and will die if denied it long enough. There is just no escaping sleep. As long as you use earth life you have no options. All life on earth with brains share a common ancestor, so sleep is likely due to something about how neurons are structured, whatever we get from sleep many be just a quirk of how neurons evolved. We know brains need sleep but we don't know why we need it, which leaves you an out.

If your lifeform evolved on a different planet they may not need sleep, you can literally just say it has drastically different neurochemistry/biochemistry and does not need sleep and no one can argue since we don't know why earth life needs it. It does sound like a handwave, but any answer will just be a handwave since we as t he scientific community don't know what is so vital about sleep.

  • 27
    $\begingroup$ @LiJun sorry there just isn't an answer, we really don't know why brains need sleep, there are a lot of ideas, but nothing solid. We don't even know everything the brain does while asleep. We know they need it but not why. Any answer you give will be by definition handwavy. $\endgroup$ – John Sep 12 at 15:05
  • 4
    $\begingroup$ @LiJun : Handwavy (to one degree or another) is probably the best you can get on this one unfortunately. $\endgroup$ – Pelinore Sep 12 at 15:32
  • 7
    $\begingroup$ @LiJun You're assuming someone knows why we need sleep. Fact is, we don't. We just know complex brains need it because sleep deprivation kills them after about 11 days. That's all we got so far. There is actually reported cases of people that don't need sleep but it has never been figured out why they can live so long without sleep. $\endgroup$ – Nelson Sep 13 at 2:21
  • 4
    $\begingroup$ @Nelson I would not call those cases they are claims that once investigated turn out not to be true. in short they don't live without sleep they are either lying or delusional. Herpinforstnace slept every day he just did it in a reclining chair instead of a bed, but insisted that did not count as sleep. $\endgroup$ – John Sep 13 at 3:32
  • 1
    $\begingroup$ Interesting. Well, the bottom line we we just don't know how sleep works. People who don't need sleep are super duper rare (much rarer than any other form of medical condition), or actually don't exist. Sleep deprivation also causes huge amount of problems. After couple days it is outright hallucination and crazy mental problems. $\endgroup$ – Nelson Sep 13 at 5:40

While multi-brain and multi-hemisphere answers have been posed, they all result in some loose of function while one part of the cognitive system is down. This seems against what the OP is looking for since he does not want dolphin style torpor.

However, computer scientists have already devised a number of systems called RAID controllers to be used by servers to replicate data across multiple hard drives in case of a disk failure and to allow for scheduled maintenance without downtime. When you consider sleep as analogous to server maintenance, it becomes obvious that a brain could use a similar system to overcome the need for sleep with a somewhat larger brain that functions as 2 identical brains. By replicating functionality, when one brain is ready to sleep, it wakes up the other brain and duplicates it's memories of the day to it, and then goes to sleep. In this fashion, you would not notice a performance drop-off or change in personality because both brains are doing the same things with the same hardware and the same information, just at different times of day.

So what does this look like on a brain? During the day you are bombarded by more more information than could be stored in your entire brain, but we still have long term memories that go back for years because our hippocampus processes all of this information and reduces it to a few trivial bits of information that it writes to long-term memory. Now imagine two brains connected by a new organ at the hippocampus. Let's call this new organ your raid ganglion. Your raid ganglion will work like a cache, a small and efficient place designed to store the memories you've acquired in a day. Each time your hippocampus writes a long term memory to the active brain, it also writes the instructions for creating that memory to the inactive brain in the raid ganglion. The raid ganglion retains these instructions until the sleeping brain wakes up (so as not to disturb the sleeping operations), and then dumps all the memories all at once into the second brain. The the raid ganglion then goes through its own sort of sleep cycle to purge its memories while both brains are active recording things in stereo. Once the raid ganglion is purged, the first brain can go to sleep and the second brain can begin using it to store its memories of the day.

As various comments and other answers have pointed out, there are several possible concerns here that require elaboration:

1. Sleep is poorly understood

The nature of sleep is generally irrelevant for this answer. We do not need to understand much about how sleep works to understand that it is an "offline process" where the brain is doing something other than what it does while you are awake. Whether it is a time to sort memories, replenish neurotransmitters, repair damage, etc. is inconsequential. As long as it is an isolated enough system to do what the brain is designed to do while sleeping, it can do that while the second brain takes over.

2. Consciousness is poorly understood

This is a more existential question than anything. To the outside observer, two brains attached to the same body with all the same capabilities, memories, skills, endocrine system, etc. would be indistinguishably the same person. If you ask either brain the same question, they will generally come to the same answer as long as they are syncing up their memories every switch over, and neither of them is significantly damaged.

From the inside, you would technically only "be you" half of the time, but you would always receive all of the memories and experiences as the other you. When "you" are in control, you really would not be able to tell the difference between when were awake, and when you just remember being awake. You might experience something like deja vu twice a day for those moments when both minds are trading off, but the human mind is full of filters designed to merge multiple inputs into a single stream of consciousness; so, a species designed to function this way would likely have a mechanism for making the switch over feel seamless the same way we can have 1 eye open, then open our 2nd eye, then close our first eye and feel like we are experiencing a continuous view of what we are looking at.

3. Removing sleep makes our bodies less optimised

This becomes a matter of perspective that we can again reference how RAID servers work to gain more insight. RAID servers always need more hardware than an equally powerful desktop computer. It needs more disk space, more wiring, more data processing, etc. A RAID server can cost several times as much as an equivalently powerful PC, so why do we use them at all? The answer is because they are optimized for something other than cost. A RAID server can maintain much better uptime and information preservation than a traditional computer. If part of a RAID brain gets damaged, it can potentially heal the damaged part of the brain using the duplicate structure to fully restore the lost memories. If the organism needs to think around the clock more than it needs to be able to be motile, then the bigger, heavier, brain would be more fit. We as humans use the same optimisations that were relevant to our ancestors, but after a few million years of civilization, or a few rounds of genetic engineering, humans could find RAID brains to be optimal in our environment where our greatest threats are the ravages of age and mental competition.

4. RAID brains are just a less optimal version of having two independant brains

It is true that you will need more neural structures operate a RAID brain than two independent brains, but this again comes down to what you are optimising for. The point here is not just no sleep, but no sleep without loss of function. The problem with two brains and one body is that both brains will have different experiences based when they have been awake. So, they develop different fears, values, skills, etc. If one decides to flee and one decides to fight, then neither strategy will work and your creature will die. If a RAID brained animal is threatened, during a time both brains are active, they will both generally react in unison instead of against each other because they have the same experiences.

As for what happens when only one brain is awake, both animals will be able to act in a single minded fashion, but the brain that does not share experiences will be unable to use what it's learned during one time of day to function at the other. This means the animal will often have to learn the same thing twice. Not only does this waste a lot of time much like sleeping does, but it exposes the organism to twice as much danger. If a one-brained or RAID brained animal does something stupid that almost kills it, then it learns from its near death experience and never does that again. If a 2 brained animal does something like that at night, then during the day it does the same thing, it has twice as many chances to die.

So, a 2 brained animal is in many ways less optimal than a 1 brained or RAID brained animal.

  • 1
    $\begingroup$ I think this is specious answer; as detailed in other answers, we don't understand the mechanism by which sleep renews the brain, so there's no guarantee that having a second brain would address the issue. In addition, we don't know how consciousness works, so the facile approach of "just run consciousness.exe on a second brain" is also lacking any sort of support of the sort the OP asked for. $\endgroup$ – jdunlop Sep 12 at 22:43
  • 2
    $\begingroup$ @jdunlop "We don't understand the mechanism by which sleep renews the brain" sounds more like a frame challenge than a critique of this particular answer since it applies to all previous answers as well. You may consider posting a frame challenge answer outlining evidence that the process of sleep is not well enough understood to answer the question as asked. $\endgroup$ – Nosajimiki Sep 12 at 23:39
  • 1
    $\begingroup$ This is just a less efficient version of having two brains, (it takes more mass to fit two intermingled brains in the same space because dendrites have to weave through each other) and unlike two brains does not allow some of the know things that happens during sleep, like cilia streaming, which disrupt the brain as a whole when they occur. $\endgroup$ – John Sep 13 at 18:44
  • 1
    $\begingroup$ @jdunlop : If we're talking about the ability to bio-engineer on this level of sophistication, then I'd imagine whatever posited super-civilization can do this will have already figured out all the functions sleep performs in the brain long in advance and hence can figure out how to redistribute that accordingly. We already know hemispheric sleep is a real thing right here on Earth, and this would not be too different. $\endgroup$ – The_Sympathizer Sep 14 at 2:49
  • 1
    $\begingroup$ en.wikipedia.org/wiki/Th%C3%A1i_Ng%E1%BB%8Dc $\endgroup$ – Aadam Sep 14 at 5:20

You don't need new organs you just need a few tweaks to the way the brain operates & Dolphins have already solved this one for you, obviously they sink if they stop swimming & if they sink they drown so they can't sleep, well they do, but they sleep one hemisphere of their brain at a time.

There's some evidence that dolphins can display slightly different personalities depending on which half (or none) of their brain is sleeping, which is interesting but perhaps not relevant :)

Ah! but on a re-read I see you don't want that option.

Well there's also a commercially available drug in the US that stops you needing any sleep, there was a documentary I saw on it a few years ago.

I can't remember what either that drug or the documentary was called but you might find this article helpful.

Sleep is largely a function of biochemical process in the brain so simply manipulating the genes to adjust the Adenosine receptors to prevent uptake of this neurotransmitter or reduce or else prevent production of it might do the trick.

A downside is that sleep is an integral part of transferring short term memory to long term memory & as far as we know no organism that uses long term memory can do without it indefinitely.

I don't know 'why' the brain needs sleep to transfer short term memory to long term but as sleep is a such a vulnerable state to be in there must be strong evolutionary pressure against any alternative to the dolphins for it not to already have appeared.

Long & the short is the dolphin answer seems the best option to get the 'appearance' of what you ask for despite the fact you remove that option in your question.

But there's no reason to stick to the two hemisphere plan of the the Dolphins (presumably shared by whales).

You could split the brain into four or more with overlapping sleep patterns such that each part (eventually) shares sleep time with each other part, you can use that to plausibly say long term memory is duplicated to all parts of the brain during this time to avoid any long term variation in memory & personality during different parts of their sleep pattern.

  • 1
    $\begingroup$ any external solution like a drug is not an answer (cit.) $\endgroup$ – L.Dutch Sep 12 at 14:02
  • 2
    $\begingroup$ the issue is dolphins do go into torpor and for much longer than a mammal of the same sizes sleeps. $\endgroup$ – John Sep 12 at 14:47
  • 5
    $\begingroup$ @Pelinore "Just as well" is patently wrong. "can get by" is closer to the truth, and that's only true with humans who have a huge highly plastic brain plus large external support systems. dolphins with one side sleeping are almost completely unreactive, with only the bare minimum of functions happening. You are also wrong about how long a dolphin sleeps for, each half of the brain sleeps for about 4 hours, for around 8 hours of total torpor.. $\endgroup$ – John Sep 12 at 15:00
  • 2
    $\begingroup$ @Pelinore Eventually recover normal semblance is not the same as no change, the can eventually converse normally, but there is loss of functionality. from your own source there is a great quote." the [patient] adapts the task to fit his brain more than he adapts his brain to fit the task." and the next sentence" Neuroplasticity after hemispherectomy does not imply complete regain of previous functioning, but rather the ability to adapt to the current abilities of the brain in such a way that the individual may still function, however differently the new way of functioning is." $\endgroup$ – John Sep 12 at 15:47
  • 2
    $\begingroup$ @John the question allows for an alien or novel creature, and so the argument that humans can't function normally seems misplaced. This would be something that has the opportunity to not only modify the tasks to fit the brain, but also evolve away the major problems thus incurred. It wouldn't behave quite like a human or any other known mammal would, but that's well within the premise of the question. $\endgroup$ – sfink Sep 13 at 15:47

Brain with three hemispheres

The state in dolphins and some birds that could keep some functions while sleeping is achieved by working with only one of the hemispheres of their brains while in "sleep state" the other part is resting.

If you add an extra hemisphere that would support the "awake" part with their normal functions you could cycle between them to achieve a "non-sleep state" during most part of the life of the creature that you are designing.

Edit: The correct definition would be a Trientsphere divided brain. @jdunlop

  • 5
    $\begingroup$ This would be a Trientsphere by definition, rather than a hemisphere. But are you suggesting that two trientspheres would be active at any one given time, with each trientsphere taking a third of the day to sleep, or that the new component would always be active, supporting the active bit? Because then the question of "when does the new lobe sleep?" arises. $\endgroup$ – jdunlop Sep 12 at 22:49
  • 2
    $\begingroup$ @jdunlop the idea is that always there is atleast one of the Trientsphere in resting state and the cognitive load is shared between the awake parts. Thank you for the new word I will edit the answer. $\endgroup$ – Tridam Sep 13 at 14:30
  • 1
    $\begingroup$ Perfect. Given the sleep requirements in dolphins (8hrs sleep per hemisphere, I believe, for 8hrs with both awake) a trientsphere-style brain would fit perfectly to have just two parts active at any time when on a 24hr schedule... functionally they'd be always awake with full (intended) capacity. Its also easier to imagine since its a fairly slight modification from what we already see. Not that growing an extra brain hemisphere is easy, mind, but the rest of the setup is in place should an oddball mutation start to happen (...how much harder is extra brains than extra toes? lots, yes?) $\endgroup$ – Megha Sep 19 at 5:12

First, you need to understand why we need to sleep. Sleep is to pay a debt with interests Let's put it in this way: Imagine that exists a creature just like us, but does not requires to sleep (later we see this). The body and mind will have some features, like move X kg/hr, and to solve Y problems each hour. That will be limited by available energy, matter, and complexity (among other variables). It does not matter which value X and Y are, every individual will have some limit, and the species will have some median limit.

Now, imagine that her/his life (or perpetuate genes) requires, somedays, to move twice the mass (2*X kg/hr) in one hour, and to solve twice problems just one hour a day, but the amount of matter, mass and biological complexity can't grow that much.

Which can be an evolutionary path? Sleep. Sleep is the counterface of overpowering (just like cooling could a requirement to use a machine over normal limits).

Examples (not exactly this way, just some simplifications):

  • When you rest, your body accumulates creatine in your muscles. It's a reserve of energy that does not requires oxygen to be consumed (so, it's not directed limited by circulation or respiration frequency). If the body does not did this, you'll need to oversize everything (and the cost sometimes is exponential), to do the same (but you could do it all day long).
  • When you think, every activated neuron in your brain drops neurotransmitters to next neurons in the "gap". The next neuron need to detect the change of neurotransmitter concentration. Until you eliminate the remaining neurotransmitter, the same neuron can't be activated and have effect (like pushing a button before releasing it, it's just one big push). The brain has mechanisms to allow "repush" but they generate more and more wastes. If you remove this, you think slower. Or maybe you'll need a really complex and big brain.

So, to not sleep, your creatures needs to don't have a big evolutionary pressure on be awake than "overpowering". It will be slower both physicaly and mentally, or: bigger, with much hunger, a more complicated "waste cleaning" system for the brain, a more complex circulatory system for the muscles, a more complex "cooling" system (more energy consumed, more heat), a more complex "memory system" for the brain, etc.

  • 1
    $\begingroup$ creatine regeneration and accumulation does no happen only during sleep, nor is it terribly vital, there is no reason all of it could not happen while awake. $\endgroup$ – John Sep 13 at 18:50
  • 1
    $\begingroup$ Not, it was an example, also I did not say while sleep,but REST as a general concept (while it could be accumulated even while aerobic activity, during anaerobic it can be lesser than consumed, so in total you're not accumulating). $\endgroup$ – ESL Sep 16 at 1:07

I've seen people saying that vertebrates need sleep. Well, not only them. Cephalopods also experience brain activity similar to REM sleep. Insects sleep also. For example, fruit flies have been found to take rests at certain periods in their circadian rhythms, staying still for 2 hours or something, exhibit very poor or near zero response to stimuli.

This suggests sleep or sleep-like states as part of an animal's circadian rhythm. The state leaves animals vulnerable, so there must be a very damned powerful evolutionary pressure for animals with significantly different nervous systems and family branch pedigrees to evolve such a trait (possibly convergent.)

Now, let us understand what sleep is.

Honestly we still do not know. All we know is that it is a regenerative process that requires the creature to go into a torpor-like state.

It also seems that the drive for this regeneration is not based on cognitive abilities. Fruit flies aren't necessarily the brainiest of creatures.

Heck, cnidarians sleep, and it appears the C. Elegans nematode with only 504 neurons also sleep (and can be induced to sleep or wake up by modulating oxygen levels.)

So it seems to be that sleep, where it occurs it is tied to a circadian rhythm. And it could be an evolutionary trait that allows animal to somehow control when to go into torpor/hibernation (as opposed to wait for external stimuli to do so.)

So, for an animal (or alien animal analog capable of locomotion and with a "nervous system" capable of acting, reacting and pro-acting to external stimuli) capable of going without sleep, it might have to have a sort of branched/parallel nervous system.

I'd say only one is active at any given time. That is, they are not "on" at the same time. These systems will need to be integrated with (also separate) endocrine systems that produce the necessary hormonal analogs to regulate organ activity and tissue recovery.

That is, the body can recover while it is "on" with at least one nervous system as a hot backup (and the other one in a cold backup state, sleep.)

I have a hard time imagining such a biological system not needing to be in a state of torpor, to recharge and accumulate energy stores or flush metabolic byproducts, for instance.

Now, if we relax the requirement and allow the creature to enter a state of near torpor or body relaxation, then it is easier to imagine a creature that can remain neurologically away (with parallel or hierarchical nervous systems where one system remains "on" and one or more remain "off".)

Other than that, I don't see how (not with Earth biology, no.)


I think there's a lot of wiggle room here, so you might need to clarify your question.

For example, what is "sleep"? Would it be ok to be responsive to stimuli, but incapable of conscious thought? That would require only a small amount of additional cells to handle not sleeping (or none at all, given that we will be awoken by a strong enough stimulus -- again, it depends on definitions), plus some changes to allow faster recovery from dormant mode.

You reject the dolphin solution, but it's not clear to me what aspect doesn't fit your criteria. Do you want identical behavior throughout the day? If each hemisphere had its own short-term memory, but a shared long-term memory that could only be properly consolidated and organized during low periods of mental activity, would that be adequate?

Identical behavior throughout the day is very different from any known organism with a nontrivial brain, by the way. We just don't notice the extreme differences from moment to moment. Try doing something intensely mental for a long period. We might have 8 hour work days, but nobody does a full 8 hours of intense mental work every day. Even consciousness is patchy; most of the time, we aren't truly consciously aware or making decisions. We don't need to be; there aren't that many decisions to be made. It's more like vision, where we think we can see a nearly 180° field of view, but in actuality our brain is stitching it together from small regions of high acuity surrounded by a huge region that is extremely low resolution. Your creature mostly just needs to fool itself into thinking it's awake all of the time.

If you really wanted to dig into it, you could come up with mechanisms for doing maintenance and consolidation that can operate during brief slices of downtime. Perhaps have a shadow network right alongside the main neural network that computes the "rested" state, and quickly swaps it in during periods of inactivity. (Activity would inhibit the modifications from the shadow network.)

But that requires a lot of speculation about mechanisms that we don't understand in the first place. How different from just waving your hands and saying, "It just works, m'kay?" do you need?

  • $\begingroup$ as i already say, literaly never sleep for the rest of their life,never rest like giraffe, not sleep walk either, also iam not reject dolphin solution, it was already one of the exeption from begining i make the question, some answerer just jump to make that dolphin idea. but your answer seems like the giraffe type though. but yeah if possible i want a believable reason not just "it just work dont ask" thing. $\endgroup$ – Li Jun Sep 13 at 16:48

There is a huge difference between a brain that has evolved to depend on sleep and a brain with an evolutionary history that deals with the brain's needs in other ways. During sleep the body goes into a different modus, and there are a lot of things happening in the glans, brain and biochemistry. Take clams for instance; there are some species that are often exposed to tidal waters, and are exposed to air for hours when the tide is low. They have practically no brain, but you could call this inactive time for sleeping, because they have become dependent on this daily rhythm, even if we should submerge them in water 24 hours a day.

Could a complex brain that don't need sleep evolve? Possibly:


"Many fishes, however, seem not to sleep. Pelagic species such as tunas and some sharks never stop swimming. One theory suggests that during sleep, sensory information (predominantly visual) gathered during the day is processed to form memories. Fishes that swim constantly in blue oceanic waters receive little 'unusual' visual input and require less 'memory-processing time' and thus need no sleep. This is supported by studies on several species of blind fishes that live in caves. These sightless fishes do not sleep."

If this theory is correct, sleep is the brain's way of "digesting" information that have been accumulating during the waking hours. The brain of an intelligent species would therefore need a brain that is able to digest information fast enough for it not to build up. Other species are only active at certain times of day, and when they are first inactive, why not do something useful while you wait? So they evolve a sleeping pattern, even if they don't have to worry about that much visual inputs.

For a sleep dependent species, like a human, to be healthy and function optimally, they can't go too long without sleep. So no magic pills. The only thing I can imagine would work in a science fiction setting, except from artificial selection over many, many generations to breed a new kind of human that don't need sleep, is a highly advanced gene therapy. If you could rewrite large portions of your DNA, which would modify your brain, glands, biochemistry and habits, I guess it could be done. (Alternatively you could also inject some serum from a sparkling vampire.)


Though we don't know exactly why we need sleep, we do know that there are biological processes that occur during sleep that aid in things like memory. And things like cleanup and maintenance, I want to say? So, theoretically, I suppose if there was a way to cause those biological processes to occur during wakefulness, that would do it. If those processes are only semi-compatible with wakefulness, maybe you get a half-sleep state. And maybe because of the brain's architecture, in real life it's not possible to run the wake and sleep processes simultaneously. But maybe off sci-fi land, someone can create a way for them to occur together. For example, nanobots that perform the sleep processes to keep your brain healthy but do it in such a way that they don't interfere with waking activity. Or maybe they do interfere in some interesting way that is a plot mechanic.

Another sci-fi option might be neural implants of some kind. Could you replace the part of your brain that needs sleep with a piece of hardware that did not? For example, if your brain needs sleep primarily to maintain memories, what if you replaced that biological memory storage with a man-made memory storage unit of some kind? Maybe you'd need less sleep then. But I'd guess your brain probably needs sleep for a host of things and that isolating them into replaceable components isn't so simple. But this is sci-fi land...

[EDIT] Wanted to add that a single chemical is not going to do the trick, because at the most it can just prevent sleep. It can't replace the necessary biological processes that your brain needs and sleep provides. These processes are probably very complicated, so...nanobots?



Sleep is a natural stress reliever. In humans, and even in animals, you can expect a lack of sleep to lead to a situation of Hypertension. So, any sentient animal species that doesn't sleep at all has to be really stressed, which means its lifespan will be short lived.

Next, not sleeping is akin to an extreme case of insomnia which is biologically evolved in your case.

Since insomnia is a common occurence in humans, you can look into conditions that can trigger it/ break the regular sleep cycle, and factors that can keep it broken.

For example: In below, the bodily chemicals/ hormones work in tandem to affect the regular sleep in humans:

Exposure to light stimulates a nerve pathway from the retina in the eye to an area in the brain called the hypothalamus. There, a special center called the suprachiasmatic nucleus (SCN) initiates signals to other parts of the brain that control hormones, body temperature and other functions that play a role in making us feel sleepy or wide awake.

Melatonin is a natural hormone made by your body's pineal (pih-knee-uhl) gland. This is a pea-sized gland located just above the middle of the brain. During the day the pineal is inactive. When the sun goes down and darkness occurs, the pineal is "turned on" by the SCN and begins to actively produce melatonin, which is released into the blood. Usually, this occurs around 9 pm. As a result, melatonin levels in the blood rise sharply and you begin to feel less alert.

So, all you need to do is establish interference in a similar system/ remove the hormone, and your fictional species can go sleepless.

Next, give this species a short life, and an early reproductive age, where the stress they have on themselves doesn't affect their reproductive abilities by much yet.

  • 1
    $\begingroup$ That's OK for a hand-wavium explanation of sleep but a likely-factium version indicates that sleep is an essential process in its own right. At least akin to 'garbage collection' in a computer system where resources are cleaned up and data compacted. In the case of the brain, compacted ina different form of storage. $\endgroup$ – Russell McMahon Sep 18 at 2:55
  • 1
    $\begingroup$ @RussellMcMahon yes, that is true. The longest a human managed to go sleep less is 11 days, which is why I mention the [Fiction] on top of it :) $\endgroup$ – mu 無 Sep 20 at 16:01

REM sleep cycles in the human brain can last from about 10 minutes each to about an hour each.

A person is typically not using most of their brain at any given moment. Therefore its logical that most of the brain (the unused parts) could be asleep at any given time without loss of function.

Divide the brain into multple lobes (the more the better) that can each nap for 10 to 30 minutes at a time. Whenever some lobe is not being used it goes to sleep. It wakes up on-demand if some stimulus tells it to. This could be an external stimulus or an internal one generated by another lobe. This is not much different than the human brain waking up from sleep when you hear or feel something.

There would be multiple lobes that redundantly perform each function.

For example, memories are recorded redundantly in multiple lobes. There are enough memory lobes so that at least one copy of any memory is available at any given time.

The only part of the nervous-system that can't take a break is the brain stem which aggregates the signals from the active lobes. But this is already true in the human brain because we always need our heart and lungs to work anyways.

Other benefits to the organism would be that failure of any one section of the brain would not completely eliminate a function (only reduce the available processing power a bit).

If each lobe needs to rest at least 1/3 of the time (like a human brain sleeping 8 out of every 24 hours), then the overall size of the brain would likely be 1.5 times as large as a normal human to have the same number of neurons awake at the same time.


To solve sleep dependency You need a much larger brain with duplicated brain pathways as well as two separate hormonal balances.

You would achieve that by having two different hormonal responses depending on a neural pathway. One is active when hormones are "high", one is active when it's "low".The problem is in the "switching" when you could hallucinate which, while not a debiliating state, can be rather confusing and unpleasant, especially while driving.

Long term and short term memory would also need a phase-shift solution with the brain being able to communicate with a part of the concepts and memories and worked out patterns at one time. So you could be a good programmer for 12h and a good welder for another 12 (example).

Schizophrenia would also be a substantial issue, where basically everyone suffers from a mild version of it because the small differences of not-perfectly replicated pathways could lead to reality dissociation.


Because on Earth day and night are so very different, most creatures have evolved to be well-adapted to either daytime-living or nighttime-living. During the time a creature is not well adapted for it, seems to be advantageous that it conserve its energy and remain mostly inactive, and this seems to be the behavior that has evolved.

For complex life to evolve that does not go periodically inactive, you thus need to take away the regularly occurring sub-optimal time periods. A planet without a sun, but a more active internal heating system, could perhaps work. Or a planet that does not turn around its own axis and so has day and night determined geographically, that is, there are regions where it is always night and other regions where it is always day.


Sleep is mostly required for highly intelligent lifeforms or lifeforms with complex memory functions as far as I know (bacteria don't sleep). This is different from physical rest, which is for conserving energy.

As such, a life form with two independent brains, that take turns, would probably not need to sleep, since each brain sleeps half the time in turn, despite the creature itself being awake. Both brains need to have identical functions, and the only problem is sharing/syncing of memories between the two. Otherwise you will end up with a race with dissociative personality disorder.

As for chemicals to prevent sleeping, that's just overdrafting. The brain will eventually shutdown, maybe even permanently.

  • 2
    $\begingroup$ Any animal with a central nervous system experiences some form of sleep or sleep-like state. But, plants and some bacteria do have circadian rhythms where their behaviours change between light and dark, which is analogous to sleep in some ways. We're not sure why bacteria do this, though with plants it's likely tied to photosynthesis. In short, sleep (or something like it) seems to be very deeply tied into our genetic history indeed. $\endgroup$ – Benjamin Wandio Sep 13 at 8:49

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.