Would any creature ever need to evolve a second (or multiple) brain(s)? If so, under what conditions, why, and what implications would it have on the creature's intelligence?

Note that while an octopus' arms are independent from its brain, the animal itself doesn't have more than one brain.

Also note that it was discovered that the stegosaurus didn't actually have nervous tissue in its tail section, and as such fails the 2 brain minimum requirement.

In order to qualify as having multiple brains, the brains must be able to work as individual unit(s), while at the same time being able to cooperate to focus on more intensive tasks. If the creature were to lose one brain, the other brain(s) must be able to assume full control of the creatures body, such that it would be able to function (almost) as efficiently as before. In other words, you can't have a brain that deals with just movement while another one deals with just regulating bodily functions. The brains must be able to "swap jobs" as necessary.

I'm not looking for the design of a creature that has 2 brains; however, if the answer so requires the design in order to make sense, feel free. I'm more for looking specifically at the "how this evolution would make sense".

The environment: doesn't matter; it can be in a jungle, the sea, the land, the skies, whatever you see fit, but assume an earth like planet and its current environmental conditions (without the humans).

Predators and prey of said creature: Whatever makes sense. Perhaps the creature needs a second brain to rotate sleeping functions in order to avoid predators (not unlike the dolphin, with Unihemispheric slow-wave sleep that rotate sleep - however, still one brain, so it fails the requirements)

Note: Not a duplicate of Creature with a non-centralized brain because even though the requirements are similar, I'm asking for more than one centralized brain while the other is asking for no centralized brain.

Note: while it's very true what most answers have pointed out regarding the inefficiency and improbability of this occurring, the question does in fact ask for the conditions of said mutation occurring, and not the plausibility of of the mutation. My is not about how likely it is, but how it could happen. Please do consider coming up with an environment in which said mutation would occur instead of telling me why it won't occur (at this point, I already know it's highly unlikely).

  • 8
    $\begingroup$ What is the difference between "rotating hemispheres" like a dolphin has and "two brains"? Is it distance between the "halves"? $\endgroup$ – Erik Apr 20 '15 at 19:42
  • 5
    $\begingroup$ could you define what you mean by 'brain'. There are creatures, like octopuses, which have parts of their body act independently like a brain. Humans, and all species, have a brain with large redundancies, you could argue that we have two brains in our head, main and backup if the main gets damaged. If neither of these count then at what point does it stop being 'extra processing power' and when does it become a second brain? and darn it Erik beat me to pretty much the same point as I wrote :P $\endgroup$ – dsollen Apr 20 '15 at 19:42
  • 1
    $\begingroup$ The Stegosaurus had been thought to have had 2 brains for a long time (And it's still an actual theory although it's been slightly dis-proven with later discoveries) you can read up on it here: en.wikipedia.org/wiki/Stegosaurus#.22Second_brain.22 $\endgroup$ – Spacemonkey Apr 20 '15 at 19:45
  • 2
    $\begingroup$ @Aify humans can function at full capacity after a rod was run through their brain, severly damaging half of it. We have effective 'backups' in our brain (sort of, it's more complex then that) which allows us to recover from significant brain damage. It's effectively what your asking for, other then being located in one spot. That one spot is the most defended of our body, and thus the best place to store an 'extra' brain without it being damaged. $\endgroup$ – dsollen Apr 20 '15 at 20:03
  • 2
    $\begingroup$ @Aify: I think the consciousness thing is the crux. Consider the brain as two separate brains, connected by a high-bandwidth communication channel (the corpus callosum), in the same way that our two lungs are connected by the trachea &c. If the channel is wide enough, the two brains integrate into a single personality; if not, they become distinct individuals. So if you have say the dinosaur with a brain of roughly equal capacity (IOW not just a peripheral processor) at each end, then it would develop two personalities, with obvious anti-survival potential. $\endgroup$ – jamesqf Apr 22 '15 at 5:22

30 Answers 30


Sleep? The body part needing sleep most is the brain. Now imagine a world where it's too dangerous to go to sleep, maybe one with two suns or placed in a crowded part of the universe, so that there's always enough light. Two brains might evolve to do a shift work.

  • 15
    $\begingroup$ Brilliant in its simplicity $\endgroup$ – Attackfarm Apr 21 '15 at 2:05
  • 2
    $\begingroup$ Saturn Rukh by Robert L. Forward. There's no surface, the ecosystem is made up of floaters and fliers--and the latter are two-brained because they have to keep flying no matter what. (Yeah, his evolutionary biology wasn't up to his storytelling. I don't think any of his ETs could have evolved.) $\endgroup$ – Loren Pechtel Apr 21 '15 at 4:31
  • 3
    $\begingroup$ I'm pretty sure we know that without sleep you will die (scholar.google.com/…). It obviously does some important stuff whether or not we understand it. $\endgroup$ – thanby Apr 21 '15 at 13:34
  • 46
    $\begingroup$ Notice that Dolphins have solved the sleep problem by having their brain halves sleep independently. $\endgroup$ – Cephalopod Apr 21 '15 at 14:18
  • 7
    $\begingroup$ Dolphin-style half-brain sleep might be how it starts. If they're less functional with half the brain asleep, and they get into danger frequently enough that this reduction is a risk, then it's possible that the two halves would evolve to compensate more effectively for the other half being asleep so that they don't always have to wake up and lose the benefit of sleeping. As the impact on function decreases, there's less to stop them from always having one half sleeping, to save energy, and waking up both halves only when they need the brainpower. At that point it's effectively two brains. $\endgroup$ – anaximander Apr 23 '15 at 10:09

A creature is unlikely to have a fully redundant brain, but could well evolve to have one or more brains that control different parts of the creature. There's two possible reasons that this could happen:

The animal is really big.

As many others have mentioned here, there was a theory, now generally considered inaccurate, that some dinosaurs had a second brain due to their extreme size. While this may not be true of dinosaurs, it makes sense in principle for extremely large animals. A creature which is big enough for nerve impulses to take a prohibitively long time to travel the length of its body could evolve a second brain to reduce this lag time.

The animal evolves from multiple animals/is a colonial organism.

We do see some animals on Earth that have evolved a single body comprised of multiple organisms. The most common example of this sort of a creature is the Portugese man o' war:

enter image description here

In a man o' war, the floats, tentacles, and reproductive bits are all different organisms, known as zooids, which are all structurally similar to other simple organisms, but grow in a giant jellyfish-shaped colony in which different organisms perform different functions in a similar manner to the internal organs of complex creatures like beavers and humans. The man o' war, of course, has no brains what so ever, but we can imagine a similar creature which is comprised of an agglomeration of simple creatures with brains.

During its evolution, the creature would probably need to exist in its environment in such a way that the different organisms that make it up could behave relatively independently of one another. Perhaps some have evolved to pull the animal towards food, some have evolved to eat whatever is nearby, and others have evolved to defend the rest. All creatures would share nutrients.

In such a multi-brained multi-organism creature, neural interconnectivity would probably evolve as well, as knowing (on some level) what everyone else was thinking would be incredibly useful. The though process of the creature would likely be entirely alien to us, since it would resemble a distributed computing architecture, rather than the single high speed processor that evolution has endowed humans with. (Which is actually massively parallel as well, but in a different way.)

  • 17
    $\begingroup$ Arguably humans have a "second brain" for reasins similar to the reasons you give under "the animal is really big". The spinal cord conducts basic logic and can make its own decisions. a key examples is the response to touching something hot and flinching away; if you waited for the brain to decide to pull the hand away you would be badly burned but the spinal cord "decides" quicker $\endgroup$ – Richard Tingle Apr 21 '15 at 5:17
  • 1
    $\begingroup$ @RichardTingle I think that's a very important point - it appears the OP wants a reason why logical thinking tissue would exist in disparate parts of the body. $\endgroup$ – Adam Davis Apr 21 '15 at 15:34
  • $\begingroup$ "imagine a similar creature which is comprised of an agglomeration of simple creatures with brains" : An episode of Star Trek The Animated Series featured a character that was a colony being. Unbeknownst to the good guys, parts of him could go off and do things independantly. He was "passing" as a biped through configuration and concealing clothing. $\endgroup$ – JDługosz Apr 22 '15 at 6:52
  • 2
    $\begingroup$ The colony organism brings up another interesting option: breeding. What if one way to overcome the dangers inherent in the environment is to split off a body part that includes a fully-educated brain? This species would have no children, only adults immediately able to survive. And if they are successful enough, they grow large enough to split and have not-babies. $\endgroup$ – Peter S. Sep 24 '15 at 12:44
  • 1
    $\begingroup$ Arguably it might be possible that communities of people could be viewed as a multi-brained organism, eg. a village, takes in nutrients, removes waste (garbage dump), might reproduce (Greek city states establishing colonies) can expand, although it's probably a stretch to imagine it like one. $\endgroup$ – Ethaba Feb 16 '18 at 9:06

The brain is not redundant in the way that man-made machines are redundant. It is far too expensive for that. Rather, it is plastic, meaning that it can adapt to changes (including damage). However, all damage to a brain means reduced capacity somewhere. The only questions are: "What have you lost?" and "Was it important?"

The human brain is only 2% of the body by mass, but consumes 20% of the metabolic energy. That makes it literally 10x more expensive to operate than the average tissue. No creature would out-compete another with a 20% penalty to energy usage. Thus, having more brain must be useful, and significantly so. While humans do have one of the highest "encephalization quotients", it is only modestly larger than other animals, yet we have taken over the planet. A little brain goes a long way, apparently.

Having 2 distinct brains seems very unlikely to occur spontaneously in a single creature, because of the efficiency of having a single unified brain. But sometimes biology screws up and a creature is born with 2 heads. Most of the time, this defect is so detrimental that the creature doesn't even reach maturity. It may be the case that in a world with a certain kind of hunting environment, having two heads which can look/focus in different directions more than compensates for the cost of having to operate 2 brains (especially if the brains are "cheaper" than ours). The more complex the environment, the more computational ability will be rewarded.

Just finding predators or prey in camouflage may not qualify. And Newtonian mechanics appears to be mostly tractable. But what if there existed a planet with strong geomagnetic anomalies, and this caused objects to move in complex ways that we would not recognize? Trying to play baseball on such a world may prove to be past our abilities, even after hundreds of years of adaptation. Or maybe the planet has quickly moving gravitational anomalies, like very dense chunks moving chaotically within the mantle which cause balance the shift quickly over time?

Obviously, the idea is that the environment must put computational demands on the creature which justify having more brainpower. It's difficult for us to imagine these conditions, because we live on a world that we have already mastered, and with only 1 brain at that.

  • 5
    $\begingroup$ Of course 2 heads and 4 eyes doesn't necessarily require 2 brains (after all I have 2 arms and my 1 brain controls both just fine $\endgroup$ – Richard Tingle Apr 21 '15 at 21:04
  • $\begingroup$ What if the brain can cold sleep? We can't do that now because we'd die without special medical intervention, but if there were a second brain ... $\endgroup$ – Joshua Nov 30 '15 at 19:36
  • 2
    $\begingroup$ "A little brain goes a long way" - I know too many people for whom this adage holds true. $\endgroup$ – Stephen Jun 28 '16 at 4:32
  • $\begingroup$ Siamese twins. If you were somewhere where this was a real advantage, evolution would favour them. Your characters could then tinker with genetic material to make them on purpose, following some plan. $\endgroup$ – RedSonja May 16 '17 at 8:48

And now for a completely different approach: We have egg layers and live births, lets hypothesize a creature that reproduces by fission instead. They have sex with the exchange of genetic information (although they would lack the concept of male and female) and a new creature would be formed. However, it would grow to maturity as part of the parent. Imagine a starfish that actually consisted of 5 subcreatures that were semi-fused. It wouldn't make a baby starfish, it would become a 6-armed starfish. When that arm was mature it would become a 7-armed starfish. At some point it would have enough arms and would split into two starfish.

Each component would have its own brain, they would work as equals, any brain capable of commanding the others based on the urgency of the situation. (Thus if arm 1 sees a predator about to eat it that overrules arm 2 seeing food.)

This would almost certainly be suboptimal but the evolutionary chasm to escape this would be almost certainly uncrossable, if it developed (and the pattern would be laid down long before brains entered the picture) it would persist unless the entire kingdom died out.

  • $\begingroup$ See salps for something like that. $\endgroup$ – JDługosz Apr 22 '15 at 6:46

Parting from the attempts at pessimism, I would suggest that a second neural mass could easily evolve to handle sensory input from a second (or more) location for some perfectly reasonable environmental factor. At first thought, I would suggest a creature that walks long, narrow tunnels. A second predatory species (or multiple) might hunt that species in those tunnels. The prey might easily develop sensory organs for the "front" and "back", and "brains" to handle both. Given the right circumstances, these brains might develop towards higher cognitive functioning simultaneously, eventually resulting in sentience.

That, however, is just an example. Any environment requiring multiple distinct sensory locations could result in similar evolutionary tracks. Multiple locations needing sensory organs -> favoring neural masses to handle reaction speed -> raising of complexity of those neural masses.

The end result depends heavily on the environment. In the above example, the physical environment and nature of the predators. They could be very similar, or one could be specialized for one task or methodology while another for another task or methodology. In the above example of the push me pull you-esque creature, maybe the front brain evolved for hunting food and results in a sentience good at engineering, math, the arts, etc. The back brain might be the sentry brain, and have evolved for predator awareness resulting in a sentience good at "fighting" or athletics or vision and spatial-awareness.

As a note, your own octopus example is a good reference for this concept, and was the first example of which I thought when I read the question title. You say that the octopus' secondary neural areas aren't brains, but that doesn't mean they wouldn't evolve further if the need was present. A pressing evolutionary need to advance those neural masses in the arms, and the octopus might one day have a sentient head and sentient arms. Sentience is simply the evolution of incredible neural complexity.


As mentioned in the comments, some dinosaurs had spaces in their skeletons to house what might have been second brains. As the theory goes, these beasts were so large that the distance alone was enough to cause issues with reaction time as the nerve impulses had to travel down to the leg/tail.

Likewise, there would be the problem of sensory input. If something bit the dinosaur's tail, there would be a relatively long delay before that got to the brain and processed into input. Simple things like walking require that sort of sensory input -> motor control feedback loop, and such large beasts might've run into issues due to nerve latency.

One other point about the theory is about blood pressure. Giraffes have a very specialized circulatory system that helps them actually get their blood up to their brains and keep it at relatively consistent pressure as the animal raises the head up and down. By having only half the brain up there, the demands on bloodflow would lessen, placing less demands on the circulatory system. The thought was that beasts like the Brontosaurus would have had similar bloodflow challenges, making the dual brain approach a potential solution to that.

  • $\begingroup$ So, 2 smaller brains instead of 1 larger brain - this is an interesting concept. $\endgroup$ – Aify Apr 20 '15 at 20:34
  • $\begingroup$ wasnt the myelin sheath evolved to solve that ? en.wikipedia.org/wiki/Myelin $\endgroup$ – Jorge Aldo Apr 21 '15 at 1:07

I don't think there would ever be a reason for a fully redundant brain in sapient creatures. Redundancy is good, but our brains already have it. You could increase the redundancy in your brain, but ultimately it makes more sense to simply add more redundancy in an already existing brain then to create a second brain. Your skull is designed to protect your brain well, no other part of your body is. Why create a second location to protect your second brain, as well as re-wiring all the neural 'controls' to a new brain location, when you could simply add more redundancy to the one you already have protected and wired.

However, there are two approaches which get somewhere close

1) the creature can function after being dissected. Think of starfish, when they break off a limb that limb can grow back to a fully starfish. If a creature was symmetrical, with one brain in each 'half' of their body, then breaking their body in half would allow two functional versions that could both grow. This could even be their method of asexual reproduction. However, it's hard to imagine a creature like this being sapient, or even highly intelligent. This trick only works because the star fish 'brain' is not much of a brain. The more complex the mental functions the less viable this method is. In addition if the creature developed entirely by asexual reproduction by splitting in half it will never grow advanced enough to be intelligent, you need sexual reproduction of some form to have the sort of evolution that can develop complex creatures like sapient.

2) non-redundant brains. Think of your octopus example, where you have limbs that function separately. Perhaps two brains exist, but with different roles. One brain controls higher order thought and one brain controls certain advanced motor functions. They would be separated either to increase reaction speed (the closer to the object controlled the better), or because they function in separate manner. However, the creature may have evolved a separate 'redundancy' approach where, if one brain fails to function, the other brain will grow to control the functionality of the first brain.

For this second approach to be viable though the creature would have to be limited in functionality after loosing one of his brains. Each brain is specialized for specific functions, and it's quite unlikely that they will preform the functions of the other brain equally as well; if they were redundant one would evolve away due to it's high energy cost! Each brain may be able to have limited ability to take on the role of the other brain, but that ability is likely no where near as good as what the creature could do with two brains.

For example perhaps a creature has a main brain, and a second smaller 'brain' for controlling motor functions of a particularly complex appendage. Perhaps the original main brain use to control the functions that the sub-brain now controls, but a new brain evolved due to the extra control it provided. If the brain controlling the appendage fails the main brain may still be able to control those appendages perhaps retaining the abilities it possessed from before the sub-brain evolved. However, it will not be able to perform the role as well as the main brain could have. Now instead of complex and effective control they can manage a haphazard control of their limbs, enough to not be fully crippled but clearly worse off then they were before. It's unlikely that their second brain could take over higher level thinking if the main brain failed though; higher order thought is much harder.

However, as a creature grows more intelligent these approaches become less viable. The brain becomes a greater and greater expense, too expensive to be able to keep a backup 'just in case'. In addition the need to defend the brain from accidental damage becomes likewise greater, meaning necessitating a 'skull' or equivalent to defend it.

Also, any higher level learning would almost have to be limited to a single brain, trying to duplicate 'learned' behavior from one brain to another would be nearly impossible. Thus only one brain could 'learn' a given task, and losing one brain would lose anything you learned. Simple creatures do little learning, relying on instincts primarily, and thus this is not an issue; but it is a major problem with sapient species, or even species as advanced as your common pet or...really any mammal.

  • $\begingroup$ This is an interesting answer, especially regarding point 2; however, why do the roles of the brains have to be separated? For example: sometimes B1 controls left appendage x, while B2 controls right appendage y, and then for unknown reason R, B1 controls y and B2 controls x. Given enough redundancy in each separate brain instead of a while second redundant brain, this approach would satisfy the requirements of having 2 brains while able to take over control of the others functions in the event of single brain failure. $\endgroup$ – Aify Apr 20 '15 at 20:40
  • 1
    $\begingroup$ @Aify because of the later points. seperate brains are expensive, you need a good reason for them. If it's possible for either brain to be just as good at the other brain's role then you have a massive expense for redundancy that is rarely needed. It only makes sense if each brain is specialize to be better suited at what it controls; if there isn't a advantage to specialize then evolution would eventually move all function into one brain to save resources. $\endgroup$ – dsollen Apr 21 '15 at 2:29
  • $\begingroup$ Dully noted, but instead of considering the implausibility, of said mutation, the question is really asking "what conditions are needed for this mutation to form" $\endgroup$ – Aify Apr 21 '15 at 16:28

While most of the upthread posts point out the objections, I will try to hand wave a semi plausible way for this to occur.

If a creature with multiple limb sets evolved for whatever reason to be very large and long, it would make sense for multiple nodes to develop in the nervous system for the control of limbs or other bodily functions. For the most part this would not lead to sentience, or even multiple "brains" but rather a sort of distributed nervous system with controlling nodes similar to the hind brain spaced at strategic locations. This isn't a huge issue in terms of protection, since if we think of these as being auxiliary controls to ensure fast and efficient use of the limbs, the nodes will probably be protected by structures similar to the pelvis where they will be close to the parts they control.

Assuming there is some evolutionary pressure to achieve intelligence and then sentience, one of the nodes will become a full fledged brain (presumably the one closest to the sense organs), while the rest of the nodes may increase in complexity to control fine motor skills needed to manipulate the environment, and ultimately tools and technology.

The creature may resemble a centaur with several very complex structures at each set of limbs that serve to anchor the limbs, provide attachment points for the muscles and protect the controlling nodes at each limb set.

If you want to take this further and have the nodes become more "sentient", then they will need to process more information individually and collectively. We will need to come up with a more alien body plan which involves each node being responsible for dealing with sensory inputs, locomotion, feeding or other complex functions. Perhaps the best example might be the truly alien initial reconstructions of Hallucigenia, which had the animal balanced on a series of "stilts" and tubular protrusions rising from the back which were theorized to be "mouths" or manipulators of some sort. A creature like that with a row of sensory organs rising from the back would need some pretty sophisticated processing power at each organ, and could, in time, evolve a set of fairly primitive "brains"; each one individually not very smart, but perhaps able to run all of them in parallel to deal with complex problems in the environment.


So multiple brains may be possible, but they require a very convoluted evolutionary history to even approach.

  • $\begingroup$ Huh. Instead of an alien body with lots of sensory input, maybe each brain node would become more complex if there was a greater risk of losing any of them - a high rate of injuries, or it can regenerate if cut between nodes. This would slowly evolve towards complexity and redundancy in all the nodes, instead of only specializing one, since that would let the creature function better if any was damaged. By the time one specialized enough to be a full-brain, the others wouldn't be far behind... and it might be possible then to have pressures encourage independence in the subordinate nodes. $\endgroup$ – Megha Jul 3 '16 at 20:58
  • $\begingroup$ Your Hallucigenia is an old interpretation which is upside down and missing half the legs. It did not walk on its spines. en.wikipedia.org/wiki/Hallucigenia $\endgroup$ – CJ Dennis Aug 10 '19 at 4:37
  • $\begingroup$ You will note I said these were the initial reconstructions of Hallucigenia. I used this as a means of modeling a truly alien creature and how they might come to have multiple brains $\endgroup$ – Thucydides Aug 10 '19 at 10:50

Well, a leech apparently has 32 brains. There is one for each segment. This might be different from the idea of having many centralized brains, but I think any neuroscientist would agree that humans have two brains. Motor skills and various senses are split between the left and right, and the output of this processing is then rectified via the connection between the two (the corpus callosum).

You might argue that such a connection between the two hemispheres would define that as one brain, but if you physically split this connection, the brains get by just fine without it. There are some abnormalities, to be sure, but otherwise the human can get by rather normally (see split-brain patients--an operation sometimes undertaken to treat severe epilepsy). The two brains then "communicate" by observing what the body ends up doing, not through actual neurological processing -- this is how conjoined twins (aka "Siamese" twins) can get along, despite having clearly two brains (and two heads).

The reasons for this redundancy are largely speculative. One could argue that we have redundancy for the sake of redundancy; if one side is damaged, the other might be available to take over. That's somewhat unlikely. It would mean that there was a point when evolution tried one-brained pre-vertebrates and two-brained pre-vertebrates, and one-brained vertebrates were not selected because they died en masse compared to the two-brained variety. But really, the one-brained variety that was around is still around, and in large numbers. Vertebrates just evolved rather symmetrically, and two brains were more consistent with that physical structure. Vertebrates did not "win" -- there are way more invertebrates than vertebrates.

So, the main reason for multiple brains might well be because body parts started to become physically separated or duplicated. They most likely stuck around because the separated/duplicated parts were useful, not because the redundant brains offered a clear advantage. (Remember, extra brain material takes a lot of energy to support. Under the strength-via-redundancy theory, this energy trade-off would have to be outweighed by the rather rare occurrence when exactly one side of the head was damaged and the animal still survived to reproduce.)

To support multiple brains in evolution, you have to answer the "how" more than the "why." It's more about how the reproductive process could have evolved these multiple brains, and then you just have to show why this evolution did not kill off the animal. If there were an advantage, all the better, but it's not actually necessary. It just has to be survivable.

Finally, if I were to hypothesize a likely scenario for multiple brains, I would define an organism which reproduced with many children in the womb that tended to not fully divide, leaving a conjoined organism. By chance, the conjoined version of this organism ended up with an advantage because, for instance, it could observe its environment in many different directions. It would have to have a locomotive ability that happened to work very well in a loosely coupled way. (Human conjoined twins get by, but they would usually be at a disadvantage, even given the hyper-vigilance afforded by multiple sets of brains/eyes.) Or maybe they would be prized as sentries in a communal society that also had non-conjoined versions of the organism.

I'd just like to add one more thing. Many responses point out that a single brain would generally be simpler and more advantageous, therefore the multi-brain is unlikely and a bad idea. I agree with the original poster, though. It really doesn't matter how unlikely it is. The fact is, evolution often creates very strange, seemingly ridiculous creatures. It's a matter of whether that creature can just survive long enough to not go extinct rapidly. And, as in the case of the leech and probably a good number of other organisms, a separated brain system might actually yield simplicity when the added complexity of wiring up body segments is factored in. Moreover, there are environments that are not particularly resource constrained and where predators are few. Look at animals like pandas and sloths -- slow, gentle creatures that certainly would not have survived in the jungles of Africa, but do ok in their particular jungle environments. In those sorts of environments, there is not a lot of pressure to kill off non-optimized species.

The point is that the evolutionary selection criterion is actually a very complex function which cannot really be predicted by a simple rule of thumb. So, just use your imagination, and you'll come up with an evolutionary environment that would yield just about any sort of fantastical creature you could think of, as long as it physically makes sense in the end.

  • 1
    $\begingroup$ Great point about the already existing redundancy of the human (mammal?) brain. Besides it's not just split brain but also downright hemispherectomy: you can shut an entire hemisphere down and things still (mostly) work. $\endgroup$ – Lloeki Apr 23 '15 at 7:28

For a second, completely redundant brain to make any sense, I would think that the animal would have to be living in some sort of environment where destruction of one brain would be a likely and common occurrence.

Evolution is all about efficiency.

Why don't human beings have two brains right now? If evolution was not about efficiency, then after the first human was born with a functional second brain, any benefits/advantages that second brain will make this specimen stronger and more desirable than puny single-brained humans. This, in turn, makes two brained humans more desirable, more likely to reproduce. Eventually, all future humans would have two brains right?

Not right

So maybe having a secondary, redundant brain might not give humans any extra benefits other than making it possible (barring blood/fluid loss and infection) for us to survive catastrophic damage to our heads. If Joe Two-brains gets shot in the head, he can still live and function normally.

Added Complexity

But what if Joe Two-Brains never gets shot in the head in his lifetime? What if he existed in a time where guns were not even invented yet? And it became increasingly unlikely that getting shot/crushed/dropped on his head was a possibility? Then, what you would consider an advantage would be somewhat just wasteful. What if he lived in a time/place where food/nourishment was exceedingly difficult to get a hold of; so now that having two brains to provide sufficient nourishment for was now actually a liability to his survival?

The animal

So I think the second brain would have to evolve out of necessity. So if this creature was on a small planet with a very thin atmosphere; where small pea sized meteorites came crashing down on it. Even if the likelihood that during it's lifetime, that one hole was punched through the animal right in one of it's brains. It had a secondary brain to take over and keep it alive while it's first brain healed. Therefor, the planet has pea-sized meteorites crashing down, frequently enough that they might strike the animal's brain on any given time, but infrequently enough that the chances that one brain gets struck and the second brain gets struck before the first one heals occurs.

I would like to think it's a two headed shark or something.


The title and body of your question are slightly at odds with each other, but a high rate of birth defects may be all that is necessary for a creature to "evolve" two brains. If you start with a creature that regularly gives birth to multiple offspring, and the environmental conditions are such that there is a high incidence of birth defects resulting conjoined offspring (or perhaps more specifically, two-headed offspring), one might conclude that the creature "evolved" to have multiple brains.

Although one of the brains is often more dominant, some cases of human conjoined twins have met your requirements for being able to work individually, as well as assume full control. It isn't clear to me why you distinguish cooperation from working individually--to me that doesn't necessarily seem like a direct function of the brain itself, but rather along the lines of a social adaptation.


Interesting question. Let's explore the premise of the question itself a little more deeply first:

"In order to qualify as having multiple brains, the brains must be able to work as individual unit(s) while at the same time being able to cooperate to focus on more intensive tasks."

What does "working as individual units" mean? For example, we have different portions of our brains that are constantly working independently of each other. For example, like you pointed out, different parts of our brain that deal with movement/speech/sight. However, these are very specialized portions of our brain that lack any flexibility to "cooperate on more intensive tasks" outside their domain.

The portion of our brain that has this sort of non-specialized general-purpose nature, is our consciousness. What does it mean for our consciousness to branch off and "work as individual units"? That would imply being able to have multiple streams of thought. For example, one stream of thought that is fully concentrated on read this post, while another stream of thought is fully concentrated on watching Breaking Bad.

Even though many of us think we're capable of this, our brains cannot handle any form of true parallel-processing. Whenever we try to "multitask", our brain is actually time-multiplexing between these different tasks, on a single stream of thought.

There is no evolutionary free lunch, and multiple-streams-of-thought will only come about if it leads to an evolutionary advantage. Ie, we will need an environment that rewards cerebral activity, but not too much of it in any one concentration. Ie, thinking about 2 different things with half-the-processing-power has to be more beneficial than thinking about 1 single thing, with full focus. This can only happen in an environment featuring an abundance of shallow-cerebral-content, such that it benefits from multiple streams of thought.

For most of natural history, this has never been the case. Cerebral activity has rarely been a great evolutionary advantage. And even when it did, the amount of cerebral content available at any one point in time, is limited such that a single stream-of-thought has always been sufficient.

Even in today's world, I would argue that there's little benefit from multiple streams of thought. Even though there's so much content out there, you can still only be in one place at a time, and you only have one pair of eyes to see with. Given that this limits you to a single input stream of information, a single stream of thought is sufficient to process this.

And even though there's so much content out there, there's economies of scale involved in choosing which content to consume. Most human progress has come from deep thinkers who focused on one problem, devoted it their full focus, and achieved breakthroughs that way. Even in today's world, which revolves much more around "multitasking," time-multiplexing between all the different things you'd like to do is sufficient to achieve our goals. Very rarely do we ever benefit from thinking about 2 different things simultaneously, with half the processing-power given to each. Hence why this abundance of shallow-cerebral-content requirement is completely foreign to anything in our world thus far.

Let's look at the next requirement:

"If the creature were to lose one brain, the other brain(s) must be able to assume full control of the creatures body, such that it would be able to function (almost) as efficiently as before. In other words, you can't have a brain that deals with just movement while another one deals with just regulating bodily functions. The brains must be able to "swap jobs" as necessary."

This requirement is going to be tricky. Imagine if you're running a dual-core computer and one of the cores stops working. Would you still expect the 1-remaining-core to perform almost-as-efficiently as your previous dual-core-setup?

If you're using your computer to its full potential, this will definitely not be the case. Losing half your processing power will leave you with a computer that is much more sluggish and slow, even though it's still fully functional.

But if you're not using your computer to its full potential, this could be true. Imagine if the only thing you're using your computer for, is playing solitaire. Going from 2-cores to 1-core wouldn't affect your computer performance in any practical way. The question is: Why would you pay for 2-cores if all you need is 1? From an evolutionary perspective, brains are very energy intensive. Having 2 brains when you only need 1, will put you at a severe caloric disadvantage. To make up for this, you will need an environment that is either:

a) Abundant in energy/nutrition needed to develop/power 2 brains


b) Very hostile and injury prone, such that backup organs significantly improve survival rates

Either/both of these will be needed to justify the added cost involved in paying for redundancy in your brain.

To summarize, if you want 2 brains that are capable of sustaining 2 different streams of thought in parallel, you will need an environment featuring an abundance of shallow-cerebral-content, concentrated within a single physical locations.

And if you want 2 brains that can deal with failures/injuries with minimal side-effects, you will need an environment featuring an abundance of energy/nutrition, and/or very high injury rates.

  • $\begingroup$ The sleep thing: redundantbunits allows a duity cycle in normal cases, and both brought to bear as an advanced feature. If one were lost, the animal would seemingly shut down and be unresponsive to stimula half the time, putting it at a severe handicap. $\endgroup$ – JDługosz Apr 22 '15 at 6:40
  • $\begingroup$ The caveat I would add regarding sleep, is that we don't fully understand why we need sleep, or even what happens while we're sleeping. Depending on the answer to this question, having 2 brains might not help with this problem at all. To give a simple analogy, if you were running a laptop with 2 processors, would you be able to reboot-the-OS while the 2nd processor is still actively running software applications? Of course not. Similarly, having 2 brains capable of full multi-tasking, might still require the same sleep habits and shut-down that we have today. $\endgroup$ – RvPr Apr 30 '15 at 15:05

Since evolution is more of an accidental thing that happened to work than a goal-driven mechanism. Your way of asking is a bit strange.

However, given the fact that dolphins in the wild already have a sleeping system where half their brain sleeps while the other half is awake, it is not unimaginable that at some point some dolphins will evolve to have more separated brain halves and eventually even complete separate brains.

As to the environment, looking again at dolphins, I would say the most likely environment is one where the creature cannot (or can hardly) hide from predators and/or needs to actively move to survive (like dolphins need to surface for air).

Theoretically speaking it is plausible that humans would at some point develop a similar system due to the continuous drive for more efficiency. However, it seems to me that, the intermediate stage of 2 brain halves functioning in turn would not make for very intelligent/efficient human beings, thus lowering their chances of survival (although in modern society they would most likely still survive).


There are already some great answers here but at your suggestion I am submitting my own.

To restate what was said in the comments, evolution only happens out of necessity, and only very strong necessity at that. As many have mentioned, brains are (by our Earth standards) a very expensive investment, so natural selection would only breed a second one if there was a very, very good reason, critical to the organism's survival, to do so. The mother of evolution is survival, and of all survival reasons, nothing prunes weakness like pure environmental hazard.


Someone has already mentioned brain-killing physical forces like crashing micro-meteoroids, so I'll focus on the more sinister threats... those of a biological nature. Imagine a creature with a complex neurological system to control either a large body, or complex senses/musculature. It doesn't have two brains (yet) but it does have an extra "neural node" to help coordinate its activity and awareness. Now imagine a virus or bacteria that evolves to feed on brain tissue. It doesn't necessarily like neural tissue, but uses it as a means to "find" the brain after it enters the body. As more creatures start to die from the disease, those with mutations of extra neural tissue in the neural node (not the same as brain tissue) tend to survive to some extent because they can continue slightly more complex rudimentary functions even after brain death (Edit: A mass-extinction of all those without this mutation, leading to the leftover "zombie-like" creatures still able to breed, would speed up the process significantly). Over time this evolves into a second brain, which is biologically different from the first brain but allows the creature to survive even if the disease kills the original brain. Over time the creatures could find a way to combat the disease and save the original brain, and thus you end up with a species that has two distinct brains, each more-or-less capable of controlling the whole body. The differences between the two could create some really interesting opportunities for cultural and "personal" development, plenty of ritual or religion, persecutions, etc to really spice up the world you're designing.

More Creative

A more fantasy-oriented (or soft science depending on how you spin it) approach might be having a second "brain" evolve as a response to the environment slowly being introduced to a dangerous energy that's simply impossible for a "normal" brain to deal with. It could cause unusual chemical reactions in things it touches, and some quirk of evolutionary mutation would make people with a certain chemical (or magical) composition able to handle it slightly better, due to how their bodies process the reaction. That process could start out as a fairly "mechanical" means of barely surviving the reaction, but better survival would be obtained through further mutations that allowed their bodies to exert some control over reaction. Rinse and repeat several times and you could end up with a second "brain" devoted entirely to controlling parts of the body that enable it to handle this destructive force and turn it into an advantage. Such a "brain" would certainly work much differently that a normal one, so as it develops into a sort-of intelligence (out of necessity, mind you... perhaps the environment is constantly changing and requires an intelligence to keep up with it, and the original brain can be killed by the reactions so this one would need the ability to take over) it would have a completely different "voice" than the original brain, and a whole different perspective on priorities for the creature.

Edit: Per your comments on the original question, both of these scenarios could also result in the separate "brain" not being a separate consciousness, but an extension of the first. But at that point it's not so much a separate brain as it is a distributed one. I guess. We're so far into the theoretical realm we don't really have a clear definition for what we're discussing, so you could do it however you want at that point.

Further Edit: Consider also that they could be separate but telepathically linked so they act more-or-less as one entity. Depends on how "out-there" you want to get.


I think Chimerism is the best bet as a starting point. Brains are restricted in size by the calories they consume. The path of least resistance is to evolve more brain. The only reason we have bigger brains than apes is because we have more to learn, and the things we learn get us more food. A big part of what aids in survival are our socialization skills and language. I think, for a creature, with what we view as consciousness, to evolve a second brain, it would have to be the result of some sort of merger of fertilized eggs - eggs of an already conscious species. You'd never evolve a redundant conscious brain one gene at a time, this is something that would require whole genome duplication, like mono-zygotic twins that fused.

Then you'd need to isolate a population of primates that were genetically inclined to this sort of merger. But I'm not sure that such mergers are genetically linked, so it sort of falls apart there.

At that point you'd have two-brained, one-body primates. And dicephalic twins do seem to share control of organs and motor reflexes, sometimes they're even in conflict over control of motor reflexes.

Lets say shared thoughts and non-verbal communication is a driver. Certainly cooperative motor control would be a benefit. That connection would occur in the axons and grey matter of what, hopefully, is a shared spinal column.

To suppose that we have one mind is incorrect I think. People with schizophrenia demonstrate an inability to recognize their own self-generated thoughts as self-generated. When you ask yourself a question, it's pretty obvious that you have multiple minds. You can pharmaceutically induce the symptoms of schizophrenia by slowing areas of our neural network that analyze other distinct parts of our neural network.

So for a few hundred thousand years you'd likely have a two-headed primate that shared muscle memory, probably basic sensory perceptions since pain and sensations originate largely from the nervous system, eventually, perhaps non-verbal communication, shared vision and perhaps eventually, some merger of consciousness and with that some form of redundancy. Well, meeting your criteria of redundant - I wouldn't call consciousness redundant, so I think the question is ultimately self-defeating.


Possibility 1: Very, very, very long body.

Upon reading just the title of the question, I immediately thought of Dinosaurs, some of which, such as Diplodocus and other sauropods were so large that some scientists had speculated that they might have had two brains (disproven for stegosaurus), one in their heads, and another lower down, in their hips.


This is so that the lower part of the body could be controlled independently, and reflexes coordinated at a faster rate, I read once that if you dropped a rock on a Sauropods tail you'd have enough time to run away before the Dinosaur's brain registered the event!

Whilst in the real world, these cases for two brains have been either disproven or seem unlikely, there is no reason why other, alien organisms may not have evolved two brains because of similar constraints.

Possibility 2: managing sleep, staying awake.

Another reason for multiple brains, as previously stated, would be to manage sleep:

Sleep? The body part needing sleep most is the brain. Now imagine a world where it's too dangerous to go to sleep, maybe one with two suns or placed in a crowded part of the universe, so that there's always enough light. Two brains might evolve to do a shift work.

However, I now point you to Dolphins which manage a similar sleep-problem by only sleeping with half their brain at once, an ability called Unihemispheric slow-wave sleep.


Whilst this sleep-pattern eliminates this need for two-brained beings on earth, there is no reason why other alien organisms wouldn't evolve two brains to conquer this problem:

Possibility 3: Improve coordination and awareness.

If multiple organisms came together to produce one organism (previous examples included the portugese man o' war and Siamese twins) then these organisms would have to be coordinated by multiple "brains."


But what conditions would cause hybrid organisms like these to become populous?

Well, the Portugese man o' war already benefits from being a siphonophore, it's almost like a normal eukaryote, all these "cells" coordinating to create an organism is actually quite efficient.

For beings like Siamese twins to evolve naturally there would have to be some strange environment, perhaps two-heads are necessary to spot dangers and predators, or extra limbs need to be coordinated independently and quickly, a set of limbs per brain.

These conditions could be imposed by faster predators, or coordinating pack animals which approach from multiple directions. Or, say the ground was unfit for living on, a species of ape could evolve which used two sets of limbs to maneuver the perilous branches of the tress above, trees which move, or hide ambushing hunters.


I've looked over the current answers and have not seen the Angler Fish mentioned yet. The female angler fish lives in the deep ocean and is best known for her bioluminescent lure with which she 'lures' her prey to her. The male on the other hand is only a tiny fraction of her size. His only purpose in life is to mate with her by biting into her and being absorbed into her body. Over time, he melds into her flesh and acts more like an extra body part to provide sperm when she lays eggs and is fed by the nutrients that flow to him. Many females have been documented to have several mates.

What I see in this example is the melding of two organisms into one. A brain would be supplied by both. One sex has the dominant body and, in a fantastical setting, one could use this as a basis for creating such a creature as described in the original question. Obviously, there would be a time period whereby the newly acquired brain would have to get wired in and the female would desire some kind of defenses to a potentially parasitic invasion of her body in such a creature. Just as much, the male side would want something of its own to combat the defenses.

Continuing this thought process, the question should be asked whether the two would have to be of the same species. There are many examples of multiple species creating a super colony that acts as a single unit and many species that have evolved to be solely dependent upon another for (sometimes mutual) survival where the extinction of one would cause the extinction of the other.

Overall, if you go with the route of the second brain coming from another creature and being wired in, your possibilities may be endless.


What are the conditions in which a creature would evolve more than one brain?

The rest of the question makes it clear that you're not actually asking for multiple brains, but brain/thinking tissue located in disparate parts of the body.

We already have multiple brains that take control of the body. For instance, observe your breathing for a few breaths. Merely by mentioning it, you are now breathing using your brain. It will take awhile to switch control back to your autonomous system. You can't do that with your heart. Touch something painful and your muscle retract before your conscious self receives the pain impulse. This is due to the spinal cord - it can actually make some decisions on its own.

Further, you specify that the brains must be able to be independent, assuming full control of the creature if/when necessary.

This essentially involves two separate consciousnesses, but both must pass through a single "routing" system that gives control to one or the other, and routes signals to both. Alternately, the body must develop complete, separate nervous systems and the fighting for control happens at the muscle level.

Honestly, none of this sounds like a reasonable mutation that would increase the organism's ability to procreate, or survive. If this mutation provided positive evolutionary pressure in any way, we'd see a lot more animals born as conjoined twins.

In essence, anything that can be accomplished by your desired two brain outcome, can also be accomplished with a single brain that has multiple units - like our existing brain. Look, you're breathing manually again. There is no reason to separate the parts of the brain and move them to other parts of the body. It increases risk and complexity, with no benefit.


I completely agree with most of these answers - we already have 'multiple brains' which is why we sometimes feel conflicted about things (I want to eat AND sleep). And the biological conditions to evolve 2 brains seem fairly implausible. There is, however, a biological condition called Polycephaly where an organism ends up with 2 heads.

While both minds are distinct, there is little to suggest that their experiences would be unique enough to provide the same variance in perspective that we expect when talking to a physically distinct person, so the whole two heads are better than one seems unlikely. Most of these organisms die quickly compared to their expected lifetimes.

The conditions for such a creature to become a distinct species might involve a cooperative species becoming so successful that enough of these 2 headed creatures lived at one time to create a breeding population (in love, we're looking for something distinctive that resonates - what resonates more than having 2 heads?). With humans, this is unlikely because we surgically separate twins, but many species have had periods of wild success where the typical breeding pressures change. It seems wildly implausible, but if there are advantages to having 2 of something else found on our heads, perhaps such a creature would be successful...


We already have a second brain.

Gut Feelings–the "Second Brain" in Our Gastrointestinal Systems - Scientific American By Justin Sonnenburg and Erica Sonnenburg | May 1, 2015

The enteric nervous system is often referred to as our body’s second brain. There are hundreds of million of neurons connecting the brain to the enteric nervous system, the part of the nervous system that is tasked with controlling the gastrointestinal system. This vast web of connections monitors the entire digestive tract from the esophagus to the anus. The enteric nervous system is so extensive that it can operate as an independent entity without input from our central nervous system, although they are in regular communication.

  • $\begingroup$ I don't think you can see it, but there was already another answer that was almost exactly the same as what you just posted. It mentions and links the Enteric Nervous system as well. It's been since converted to a comment by a mod. $\endgroup$ – Aify Jul 27 '15 at 23:31
  • $\begingroup$ I was the one whose answer was converted to comment by some pedantic mod who said my answer didn't answer the question. The question was "What are the conditions in which a creature would evolve more than one brain?" If we already have a second brain, then that implicitly answers the question. $\endgroup$ – dan-gph Oct 4 '16 at 8:14

In Saturn Ruch, each side (top and lower surfaces) had a stalk with a head and hands, used for grooming and maintaining that side. The characters notice two such eye-stalks and say that you can't have eyes without a brain immediately behind: certainly on a carrier-sized animal the nerve impulses are two slow.

So each half has a separate brain and is a distinct individual. This is due to its size and the shape making it advantageous to have a stalk on each surface rather than one that can reach all over.

Since the ruch must remain on active alert, one head controls the body in the day and the other at night, with different evolved behavior.

Some birds have unihemispheric sleep: that could further evolve into more independent functioning, as opposed to our situation which combines consciousness into one individual with one "thread".


The answer is ridiculously obvious and the title gives it away; What are the conditions in which a creature would evolve more than one brain? Well, in conditions where a creature has more than one body.

The Man o' War is a type of creature that is not one, but instead multiple creatures, all of which come together to form a single species. An advanced version of this multi-bodied concept is the anthill. While the creature does have multiple brains, one could argue that a multiple bodied creature is even its own creature.

  • $\begingroup$ The man o war is already covered by the second most voted answer in high detail. Also, your answer fails to meet the requirements to my brain requirements - specifically: If the creature were to lose one brain, the other brain(s) must be able to assume full control of the creatures body, such that it would be able to function (almost) as efficiently as before. In other words, you can't have a brain that deals with just movement while another one deals with just regulating bodily functions. The brains must be able to "swap jobs" as necessary. An anthill does not qualify as a single creature. $\endgroup$ – Aify Jul 15 '16 at 5:56

I would like to point out that there are several two brained vertebrates in captivity that you could arrange to observe, and even at least one two brained vertebrate you could arrange to meet and talk with, Abby and Brittany Hensel (born 1990).

The problem with 2-brained conjoined twins is that they are not genetically 2-headed. The descendants of any would not inherit genes for being 2-headed. But there could exist species that are genetically a lot more likely to have conjoined twins than most Earth species. And if there is some survival advantage to being 2-headed it could become more and more common in the population. Perhaps being 2 headed and 2 brained might in some cases improve survival rates in a social species.

Perhaps in a futuristic biologically advanced society 2-headed and 2-brained species might be artificially created. A future emperor might be inspired by heraldry to have 2-headed eagles created, for example.

If most Earth vertebrates have right and left brain hemispheres that sometimes operate independently, some evolutionary process might develop multiple brains out of such a situation.

For example, there are species on Earth that are possibly semi intelligent or even fully intelligent that have heads much larger than their brains. An Elephant's brain is larger than a human brain but it's head is much vaster and includes empty spaces to save weight. A hypothetical elephant species evolving a much larger brain might do so by having one hemisphere grow an extension out of the brain case and into the hollow regions. Eventually that entire hemisphere might be in the hollow regions of the skull and the other hemisphere might grow to fill the brain case and anatomists might consider that species might has two brains.

Similarly many cetacean species have brains larger than humans that occupy tiny parts of their heads. Possibly they might evolve even larger brains by having one hemisphere grow out of the brain case into another region of the head while the other hemisphere grows to fill the brain case. And they might seem to have 2 brains.


Many of the proposed answers here claim that efficiency is a requirement for an evolutionary path to be manifested & maintained. That's not at all true. What makes it seem true is that the more competitive (that is to say, if things are inefficient they die or fail to reproduce) an environment is, the more likely inefficiencies will incur a cost that results in failure of the line to procreate and thus evolve more (or maintain functionality.)

So scenario 1 is: The environment in which the subject species evolves is essentially non-competitive for that species. 1a: Subject species procreates as a community, rather than as individuals..that is to say that there is no mate selection process. 1b: Environment is cold, species is cold-blooded, in this way the high metabolic requirements of 'normal' brain matter become a boon quite aside from it's nature as problem-solving tools. (Why: Whilst 1b can help explain why it would not necessarily be a net loss to the organism, we can also say that See 4)

Scenario 2: Subject species' brain matter has evolved in such a way as that it grows/reproduces to fit available space. Another species evolves that as part of the life-cycle burrows into the flesh of subject. Net result: on occasion members of the subject species will form tendrils of brain matter back along the path the parasite has burrowed, where said burrowing has created cavities the reactive-brain has the local mass to take on a greater role as part of the consciousness. (Why: Of course in this scenario we might say that the growth mechanism can be a reasonable response to the parasite, if it regularly occurs that said parasite destroys brain matter, the evolution of a process to stimulate additional neural matter makes sense.)

Scenario 3: As scenario 2, but some part of the parasite's biology is close enough to the hosts that it triggers the formation of brain matter around it(once inert) by the hosts physiology.

Scenario 4: Feedback. Simply put, when a human is growing it's cells are instructed as to what to be, and by and large they do as told. If our theoretical species evolved in an environment in which damage regularly occurred which interfered with a sensory organs ability to communicate with the rest of the body resulting in eventual death, we might possibly see a scenario in which when the feedback loop from brain<>sense organ was interrupted, the sense organ had the functionality to stimulate the local(or secondary) creation of neural matter (that linked into the general nervous system). If what damage had caused this growth was eventually repaired by other functionality, the subject would have it's central neural mass + dedicated neural mass) // Then we have Unihemispherical Slow - Whatsitcalled, wherein certain creatures have developed a capacity to sleep with half a brain at a time.

This shows us that brains don't need to be so alien to extend their ability to operate as two parts than in the fashion they already do among humans.

But in the end we have an easier solution, it seems to me.. it's insects that become sapient, likely in a slightly cooler environment but with denser concentrations of oxygen in the atmosphere, reducing load on heat exchangers and increasing the efficiency of tramsdermal respiration.


Another option would be a massive, largely or entirely immobile planktiverous organism, if the feeding mechanism is passive but food source is abundant. The issue changes from "this uses too much energy and so too much cooling" to "this needs a much more efficient nutrient delivery system than a terrestrial mammal of the same eq. This being the case, where nutrient collection is distributed across the body, so must digestion be or your species ends up using as much energy moving food from 'mouths' to 'gut' as is gained from the food (in a sufficiently large organism.) If we accept this as true, it then also becomes an energy saver to have the energy derived from local digestion to local uses and so we justify the evolution of multiple brains...at least in part.

Of what use they'd be to an immobile planktiverous organism I'm not sure :)

  • $\begingroup$ It seems you've really thought this out. While I can understand your thoughts I'm afraid this answer may not be clear to all users. $\endgroup$ – Zxyrra Feb 12 '17 at 22:46
  • $\begingroup$ Thanks for the consideration of the first sentence, but I'm afraid I just thought it out whilst typing. Perhaps I need to evolve a secondary neural mass that enhances my ability to communicate concepts. I suppose the simplest way I could refine it is? Instead of or as an enhancement of the ability of man to repair damaged tissue, an organism might evolve the ability to create replacement tissue in another location. If that original damage was then repaired 'traditionally' your organism would have a secondary brain, going forward. $\endgroup$ – mensenisevirem Feb 12 '17 at 23:04

Let's run with the starfish idea and build a hydra. Like a big reptile with many heads

A hydra - the scaly reptile kind A lot of these posts are approaching the problem backwards, and answering "how could an existing single-brained creature develop an additional brain" and that's an impossible evolutionary path - neural tissue is just too expensive and a semi-evolved brain would not be useful alongside their existing fully-evolved one.

We need to instead answer the question "under what circumstances would a non-brained creature evolve multiple brains" and the various starfish posts above spotted the solution - the answer is radial symmetry. Or basically any body layout in which it would make sense to have the sensing/processing apparatus (It's best to imagine the brain as a thing that evolves from the eyes - it is efficient to colocate them and they are made from the same type of tissue) located on an extremity, and not along a line of symmetry.

This seems quite unlikely to develop with bilateral symmetry, as the sensing/processing apparatus can at once be an extremity (the terminal end of the symmetry line) and place itself equidistant from the two halves of the organism, which feels like an efficient organism.

But this isn't so with the starfish... it would have to be very TALL for the centre to be a sensible place for a head... and it already has quite a lot of sensory apparatus on the tips of the arms, so this makes the tips of the arms a strong candidate for the development of additional processing tissue.

EVOLUTIONARY BONUS - evolution tends to flow more smoothly if it doesn't need to develop multiple systems in parallel... if each incremental new change is separately useful to the organism. And we get that with the starfish - it is already a radially redundant organism, consisting of identical modules. This means for this creature to evolve multiple brains it need simply continue to add processing tissue to the tips of its arms incrementally. It doesn't NEED to develop complex systems for coordinating between the brains - at least not right away! Whilst the ability to coordinate between brains is obviously useful; each arm being separately clever is also independently useful. This gives us a good evolutionary ramp, we can add intelligence incrementally and coordination will follow in it's wake.

We're still a squishy thing though

Alright so now what? We don't want a squishy boring sea creature. We want a cool interesting alien/monster. So where does evolution go next?

Well, now that the arms are specializing themselves into sensing/processing devices ("heads"), maybe we want to do more with the central body where our gut and reproductive organs live? Lets extend this mass at the base, and develop a "foot" like snails have, so we can bring our "arms" up off the ground and allow them to continue their specialization away from being a locomotion device and further into being our sensing/processing apparatus (much like how humans specialized our forelimbs away from locomotion and into manipulators).

A Sea Anemone/Hydra/thing

So now we have something that looks a bit like a sea Anemone... we've got a foot for locomotion and organs we don't want extra copies of, and we've got a set of "braintacles" that house our sensing, processing and manipulator apparatus.

Let's step back. Where are we in the evolutionary history of the world? We're right near the start. The Ediacara. Bilateral symmetry has not achieved dominance and large carnivores do not exist yet - all life is squishy sea creatures feeding off plants and microbes in the water column or stuck to surfaces. They were simpler times, when evolution could try all sorts of wacky experiments without worrying about armour or teeth or speed, because there was nothing out there trying to eat you unless you were plankton.

Weird sea creatures

Lets add a new feature for thriving in this environment. Most sea creatures (and we assume most Ediacarans) reproduce by spawning and and thus have a plankton stage... which as we mentioned a moment ago, is the only way to get eaten in this era.

So let's use our radial redundancy as an advantage to skip the planktonic stage.

Our organism now has two reproduction modes - sexual (spawning and creating planktonic larvae) and division (no sexual recombination, but we skip the planktonic stage and thus skip predation during the Ediacara). For coolness, let's imagine that each of our "braintacles" can seperate off from the main trunk of the organism, becoming a free-swimming snake-like creature.

So our lifecycle is plankton -> gets bigger to become braintacle -> fully matures to develop a trunk (housing a bigger gut and sexual organs) -> and then can both develop additional braintacles (making the organism more advanced and also allowing future reproduction by division).

This is a strong basic model. We now have a creature that can add additional sensory/processing capacity as it ages (the Elders are wise!), an individuals genome is likely to survive injury (if the trunk is damaged irreparably all braintacles can separate, disperse, and generate new trunks). And because it has 2 different reproduction modes that are optimized for different ecological conditions, it should be robust to major ecological upheaval.

Which is important, because the end of the Ediacara is coming.

The Predator Apocalypse

There's always a bigger fish! Atleast post-Ediacara The first macro-predators emerge, and before "food chains" can properly become a thing almost everything just gets eaten. Almost all body-types are completely wiped out. But not our creature! Whilst they haven't evolved combat abilities yet (there was no reason to in the Ediacara!) they are well equipped to escape combat. Infact escaping is a reproduction event for our creature. So the predators come, trunks die and braintacles disperse... some of the braintacles survive long enough to grow a new trunk and spawn and... hallelujah. Because of the sudden predator-apocalypse, there are not many filter feeders left to eat our planktonic stage, and there is not much competition for food. Our creature has a population explosion, and achieves the population numbers it needs to evolve combat abilities.

And so the Ediacara ended, and all life that remained had teeth, claws, shells, bones, and other hard things for battle/predation. But our little many-brained creatures survived it, it's one of the fighty monster things in the food chain now. Infact, as one of the few creatures to survive the Ediacaran extinction event, it now diversifies and fills many ecological niches. It is one of the two dominant bodyplans of this world. It is the hydra bodyplan... and the only other major bodyplan is that of the first predator.

Fast forward to the Jurassic period of planet Hydra.

Our canonical example of the hydra bodyplan stalks the forest. It is a large many headed reptile. Each head sits atop a snakelike-neck, which in times of severe injury can separate from the main trunk of the body and become an independent snakelike organism. But the heads do not separate by choice. For when they are together they are smarter, they can pool their intellectual and sensory resources to better understand their environment. They can draw upon each others memories, most valuable of which is the oldest head, their leader and mother, the one which originally grew this trunk. And what a trunk! The claws! the musculature! The large digestive tract and sexual organs! In their Hydra form they are a lord of the forest, a powerful combat machine with prodigious intellectual abilities.

And should they fall from grace they shall become snakes. Separate. Weaker and dumber... but nonetheless whole. As snakes they can flee and hide, and grow. To one day become a Hydra lord once more.

And that is good... for a shadow crosses the sky... that which hunts the hydra is coming...

Hydra stalks the forest

p.s. apologies for the necromancy on a 5 year old post... I just... I just love mad things like this...


If there was natural or unnatural evolutionary pressure for Siamese Twins to survive by some far fetched scenario that would meet your criteria.

An entire planet or continent somewhere in the cosmos populated by Siamese Twins.

Title of the book: The Siamese Gamble Subtitle: Two Heads are Better Than One

  • 1
    $\begingroup$ Yes, Siamese twins could be considered a single creature with multiple brains, but this doesn't answer the question of what conditions would be necessary for it to become common. $\endgroup$ – Rob Watts Apr 22 '15 at 18:59
  • $\begingroup$ Sexual selection. They have twice the number of penises, and that really turns women on. They're natural politicians because each head can promise a different thing to their constituency without being dishonest. I mean, there's a million fantasy reasons why Siamese twins would outperform normal people evolutionarily, until they become the norm . $\endgroup$ – Salmoncrusher Sep 4 '16 at 4:46

Taking a page out of nature, we could think of a Mantis' abdominal ganglion as a secondary brain being evolved. Perhaps, over time, a situation could arise where a secondary brain is necessary to reproduce.

  • $\begingroup$ Yes, but what would the situation be? $\endgroup$ – Aify Apr 22 '15 at 20:11
  • $\begingroup$ That is the situation. The secondary "sex" brain confers a reproductive advantage and in your scenario perhaps the more complex this secondary brain is, confers a greater advantage. $\endgroup$ – Biff MaGriff Apr 22 '15 at 21:13
  • $\begingroup$ That doesn't answer why it confers a reproductive advantage. What are the environmental factors resulting in those with secondary brains being superior to the single brained counterpart? $\endgroup$ – Aify Apr 22 '15 at 21:24
  • $\begingroup$ In the mantis' case it allows them to continue having sex after decapitation. $\endgroup$ – Biff MaGriff Apr 24 '15 at 14:49

This seems to be a question of why not one big brain.

If a brain gets too big and too busy it'll get hot. To control the heat you have to expose more of the brain to more coolant (or colder coolant, which I take to be impractical). Stuffing more pipework in there spreads things out and so signals have further to travel. This slows things down.

To mitigate that intrusion, brain work has to be localised by reducing the dependency on connections to other lobes. You just take the process of compartmentalisation and deeper intrusion of the cooling system until you've pushed lobes so far apart that they fit the description of "multiple brains".

It's not just heat, of course. It's fuel and waste too. In fact, it appears that sleep is necessary to give the brain a regular deep clean. Partitioning allows that maintenance cycle to be done piecewise.

The supposition is that there are significant pressures to think fast on simple problems, to stay cool, and to stay awake. This probably comes about in an environment of competition with other creatures that are similarly capable.

Rather than try to imagine a world with predators and prey and stars and satellites that force this condition, I'm just going to say it's a secondary sex characteristic.


I just don't see this coming up. You are essentially asking for a creature with a completely redundant brain. (This based on the creature having to manage effectively the same with one functioning brain or two; this requires one of the brains to basically not be doing anything, otherwise you'd quickly notice a difference between one and two)

In addition, the brain is the control centre for the whole creature; having two brains would mean conflicting commands are being sent out if the two brains disagree. And since brains are constantly changing, they would quickly start running out of sync.

On the other hand; if you have a true redundant brain, you are spending a lot of energy to grow and maintain a complex structure that does nothing. And if anything ever happens to your primary brain, the redundant brain will most likely not even be prepared to take over because it's essentially untrained (not having been fed any impulses until then)

Not to mention that it'd be hard to trigger the redundant brain to take over from the primary one, since a trigger of that sort will generally be sent out by... the brain.

  • $\begingroup$ plus your brain is already located in the most secure spot in your body. To make another brain means making another skull to defend it. $\endgroup$ – dsollen Apr 20 '15 at 20:05
  • $\begingroup$ The question is, why is redundancy assumed? What if the brains weren't truly redundant, and both did the same things, but at random intervals? eg: sometimes B1 controls left appendage x, while B2 controls right appendage y, and then for unknown reason R, B1 controls y and B2 controls x. The second brain doesn't have to be an empty backup at all. $\endgroup$ – Aify Apr 20 '15 at 20:37
  • $\begingroup$ Because you requested that losing one of the two brains would have it acting (almost) as effeciently as before. That implies one of the two brains isn't really doing anything useful, otherwise losing one is going to be very noticable. Even in the case of interval swap, you will notice the loss of a brain because the other will overwork itself after a while. If that doesn't happen, you have redundancy again, except now you have a creature with a split personality as well. Which probably isn't very competitive. $\endgroup$ – Erik Apr 20 '15 at 20:40
  • $\begingroup$ I don't fully understand the logic behind your conclusion. It was mentioned in another answer that the current brain already has a lot of redundancy - why would using that redundancy to take over the workload of a second brain upon failure result in overworking of the brain? $\endgroup$ – Aify Apr 20 '15 at 20:52
  • $\begingroup$ If it's redundant, it wouldn't result in overworking. Your previous comment made a case for brains that were not redundant; thóse would get overworked. If a brain only works half of the time but it could also work all of the time if need be, then half of that brain is redundant. (That's what redundant means, after all. If losing it does not harm you, it was redundant. If losing it does harm you, you're not working at full efficiency anymore. It's one or the other.) $\endgroup$ – Erik Apr 20 '15 at 20:57

Upper Limit on the size of the head

Technically humans have two brains: One in the head, another(smaller, much less powerful) in the stomach. The second brain basically regulates function of the digestive organs, or something. Maybe it can be thought of as an outpost of the first brain, a small outcropping of nerves that need to be "where the action is," i.e. a field operation. Now imagine that evolution decides humans need to get smarter, but the head is already big enough to cause childbirth difficulties(this is the case). The solution is that the second brain gets bigger, and is used for "multi-threaded computation."

Here would be an example of Multi-threaded computation:

Such tasks would comprise situations when the primary brain can give the secondary brain a simple problem and a time allotment, and the secondary brain can solve it all on its own, without using the incredible faculties of the primary brain. An example is counting, or arithmetic: It's very simple to do(theoretically at least), and it does not require very much real-time data transfer between the primary and secondary brains. Basically, the primary brain sends some simple information(what's the prime factorization of 23856293698273325739248573029876129384762983462138741241234612394872649876219834?), where the number of nerve impulses needed to send the problem is relatively small compared to the complexity of solving the problem, and then the secondary brain works on it, and works on it, and works on it, and then sends back: The answer is 2*3*5*7*11 etc. etc., which is also a relatively small number of nerve pulses relative to the complexity of the problem.

Now, imagine that the secondary brain gets bigger and more powerful as the need arises until it's fair to call it an actual brain as opposed to just a nerve cluster in the stomach.

Disclaimer -- When you assume human brains are basically just squishy computers, it will be much easier to think about them, and also a whole lot more incorrect.

  • $\begingroup$ Evolution doesn't "decide" anything. Roughly, evolution is the unguided response to changing demands on a lifeform, most often by changes in the lifeform's environment, selecting among random mutations. That's also why evolution often hits upon local maxima rather than a global maxima; it's possible that there's an objectively "better" way to do something, but getting there would require an initial reduction in fitness, which is heavily selected against. $\endgroup$ – user Feb 12 '17 at 21:01
  • $\begingroup$ @MichaelKjörling, I think the question of whether evolution can "decide" anything is better suited for a philosophy or semantics discussion. I can fairly say that "The stock market is very intelligent," while still acknowledging that it is an unguided process. $\endgroup$ – Salmoncrusher Feb 15 '17 at 1:37

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.