No Brainer.

For decades now, I have been haunted by the grainy, black-and-white x-ray of a human skull.

It is alive but empty, with a cavernous fluid-filled space where the brain should be. A thin layer of brain tissue lines that cavity like an amniotic sac. The image hails from a 1980 review article in Science: Roger Lewin, the author, reports that the patient in question had “virtually no brain”. But that’s not what scared me; hydrocephalus is nothing new, and it takes more to creep out this ex-biologist than a picture of Ventricles Gone Wild.

The stuff of nightmares. (From Oliviera et al 2012)

The stuff of nightmares. (From Oliveira et al 2012)

What scared me was the fact that this virtually brain-free patient had an IQ of 126.

He had a first-class honors degree in mathematics. He presented normally along all social and cognitive axes. He didn’t even realize there was anything wrong with him until he went to the doctor for some unrelated malady, only to be referred to a specialist because his head seemed a bit too large.

It happens occasionally. Someone grows up to become a construction worker or a schoolteacher, before learning that they should have been a rutabaga instead. Lewin’s paper reports that one out of ten hydrocephalus cases are so extreme that cerebrospinal fluid fills 95% of the cranium. Anyone whose brain fits into the remaining 5% should be nothing short of vegetative; yet apparently, fully half have IQs over 100. (Why, here’s another example from 2007; and yet another.) Let’s call them VNBs, or “Virtual No-Brainers”.

The paper is titled “Is Your Brain Really Necessary?”, and it seems to contradict pretty much everything we think we know about neurobiology. This Forsdyke guy over in Biological Theory argues that such cases open the possibility that the brain might utilize some kind of extracorporeal storage, which sounds awfully woo both to me and to the anonymous neuroskeptic over at Discovery.com; but even Neuroskeptic, while dismissing Forsdyke’s wilder speculations, doesn’t really argue with the neurological facts on the ground. (I myself haven’t yet had a chance to more than glance at the Forsdyke paper, which might warrant its own post if it turns out to be sufficiently substantive. If not, I’ll probably just pretend it is and incorporate it into Omniscience.)

On a somewhat less peer-reviewed note, VNBs also get routinely trotted out by religious nut jobs who cite them as evidence that a God-given soul must be doing all those things the uppity scientists keep attributing to the brain. Every now and then I see them linking to an off-hand reference I made way back in 2007 (apparently rifters.com is the only place to find Lewin’s paper online without having to pay a wall) and I roll my eyes.

And yet, 126 IQ. Virtually no brain. In my darkest moments of doubt, I wondered if they might be right.

So on and off for the past twenty years, I’ve lain awake at night wondering how a brain the size of a poodle’s could kick my ass at advanced mathematics. I’ve wondered if these miracle freaks might actually have the same brain mass as the rest of us, but squeezed into a smaller, high-density volume by the pressure of all that cerebrospinal fluid (apparently the answer is: no). While I was writing Blindsight— having learned that cortical modules in the brains of autistic savants are relatively underconnected, forcing each to become more efficient— I wondered if some kind of network-isolation effect might be in play.

Now, it turns out the answer to that is: Maybe.

Three decades after Lewin’s paper, we have “Revisiting hydrocephalus as a model to study brain resilience” by de Oliveira et al. (actually published in 2012, although I didn’t read it until last spring). It’s a “Mini Review Article”: only four pages, no new methodologies or original findings— just a bit of background, a hypothesis, a brief “Discussion” and a conclusion calling for further research. In fact, it’s not so much a review as a challenge to the neuro community to get off its ass and study this fascinating phenomenon— so that soon, hopefully, there’ll be enough new research out there warrant a real review.

The authors advocate research into “Computational models such as the small-world and scale-free network”— networks whose nodes are clustered into highly-interconnected “cliques”, while the cliques themselves are more sparsely connected one to another. De Oliveira et al suggest that they hold the secret to the resilience of the hydrocephalic brain. Such networks result in “higher dynamical complexity, lower wiring costs, and resilience to tissue insults.” This also seems reminiscent of those isolated hyper-efficient modules of autistic savants, which is unlikely to be a coincidence: networks from social to genetic to neural have all been described as “small-world”. (You might wonder— as I did— why de Oliveira et al. would credit such networks for the normal intelligence of some hydrocephalics when the same configuration is presumably ubiquitous in vegetative and normal brains as well. I can only assume they meant to suggest that small-world networking is especially well-developed among high-functioning hydrocephalics.) (In all honesty, it’s not the best-written paper I’ve ever read. Which seems to be kind of a trend on the ‘crawl lately.)

The point, though, is that under the right conditions, brain damage may paradoxically result in brain enhancement. Small-world, scale-free networking— focused, intensified, overclockedmight turbocharge a fragment of a brain into acting like the whole thing.

Can you imagine what would happen if we applied that trick to a normal brain?

If you’ve read Echopraxia, you’ll remember the Bicameral Order: the way they used tailored cancer genes to build extra connections in their brains, the way they linked whole brains together into a hive mind that could rewrite the laws of physics in an afternoon. It was mostly bullshit, of course: neurological speculation, stretched eight unpredictable decades into the future for the sake of a story.

But maybe the reality is simpler than the fiction. Maybe you don’t have to tweak genes or interface brains with computers to make the next great leap in cognitive evolution. Right now, right here in the real world, the cognitive function of brain tissue can be boosted— without engineering, without augmentation— by literal orders of magnitude. All it takes, apparently, is the right kind of stress. And if the neuroscience community heeds de Oliveira et al‘s clarion call, we may soon know how to apply that stress to order. The singularity might be a lot closer than we think.

Also a lot squishier.

Wouldn’t it be awesome if things turned out to be that easy?

This entry was written by Peter Watts , posted on Tuesday July 28 2015at 07:07 am , filed under blindsight, neuro, Omniscience, sentience/cognition . Bookmark the permalink . Post a comment below or leave a trackback: Trackback URL.

80 Responses to “No Brainer.”

  1. There is a TED talk floating around that focuses on neuron density in brain tissue. In mice it is pretty low and as you move through mammals to the primates the density of neurons get higher in the tissue. Humans are very dense I wonder what the density in VBN’s is?
    https://www.ted.com/talks/suzana_herculano_houzel_what_is_so_special_about_the_human_brain?language=en

  2. What if that stress is something we already apply, to a smaller degree, by the process we call “education”? If so, the methodology of education could be greatly enhanced, too.

  3. Also, could we apply this to other animals, maybe? We’ll finally get to talk to cats!

  4. Guy with an empty skull goes around walking and talking and doing derivatives…

    Man I give up, sign me up for the soul train. I mean, what more proof do I need? :)

  5. Adam Etzion:
    Also, could we apply this to other animals, maybe? We’ll finally get to talk to cats!

    I talk to cats all the time.

    Be careful what you wish for.

  6. So the question becomes: hydrocephaly is apparently not all *that* uncommon; the stress and damage which causes it is thus also not that rare. Regular hydrocephaly might take away as much as the stress/damage/resilience response gives, leaving them net normal, but if there is such a response, then it should sometimes be accidentally triggered on its own, in the same way that other repair processes (like bone growth) sometimes get triggered when they shouldn’t, making for a few people who get all the upside and none of the downside.

    If net brain performance could be boosted by one order of magnitude, or even just doubled (compare humans with ~100b neurons to creatures with 50b neurons), they would be as gods unto men. Either by accident or by evolution, they should exist. As Fermi might say: where are they?

  7. Surface area. Smaller heads can’t make up the surface area the way the swollen heads of hydrocephalus cases can. Maybe the stuff in the middle is a redundancy against being dropped on our heads.

  8. Holographic brains don’t need all the bits to maintain or develop function?

  9. gwern,

    If they have any sense – and they would – then they’re hiding!

  10. BMT,

    Normal brains still have some free space and can be a bit denser. Plus, even if we assume the volume doesn’t change in the lucky freaks, they should *still* be way smarter on a neuron by neuron basis; neuronal efficiency varies considerably from person to person (myelination comes to mind) and other structural things like white matter density seem to predict intelligence in the brain imaging studies, so…

  11. These folks’ brains are rewiring for normality. There is no blueprint for supra-normal. “Overclocking” normal brains for superior performance is completely speculative fiction at this point in time.

  12. Could you correct the name of the author? It’s Oliveira not Oliviera.

  13. Oh wow. I did a presentation on this very topic in 2007. 😀 John Lorber’s work was pretty wild.

    https://drive.google.com/file/d/0BwI95WRbQAHbMGFjYzQ1YjYtOTNjMy00ZTQyLWIyZjAtNjZiZWQ0NzMzZWM0/view?usp=sharing

    The most fun I had with it was statistical derivation of how many “Brainless among us!” there would be given the population size of my home city. (I’m probably off by an order of magnitude when it comes to the drivers.)

  14. gwern:

    If net brain performance could be boosted by one order of magnitude, or even just doubled (compare humans with ~100b neurons to creatures with 50b neurons), they would be as gods unto men. Either by accident or by evolution, they should exist. As Fermi might say: where are they?

    Winning Nobel prizes for creating quantum electrodynamics, and the like, I expect.

  15. In the first link, the chap has an IQ of 75; in the second link ‘Sharon Parker’ actually has a normal brain volume, it’s just oddly distributed. It would’t be surprising if many of the older cases were similar to Sharon’s – brain mostly all there, but not where a 1980s scan would find it.

    On the other hand it won’t be long before things without any brain at all regularly beat 126 on IQ tests:

    http://observer.com/2015/06/artificially-intelligent-computer-outperforms-humans-on-iq-test/

  16. gwern: Either by accident or by evolution, they should exist. As Fermi might say: where are they?

    Maybe they’re among us, and don’t realize it themselves— some of the documented cases were well into adulthood before discovering that they had extra storage space in the attic, I bet a lot of other VNBs go peacefully through their whole lives without ever getting a brain scan.

    Or perhaps they do know, and are smart enough to stay hidden…

    BMT: Maybe the stuff in the middle is a redundancy against being dropped on our heads.

    Apparently the stuff in the middle is largely white matter, not grey. Trunk lines, module-connecting circuits.

    Bil: Holographic brains don’t need all the bits to maintain or develop function?

    Yeah, but they’d lose an awful lot of resolution.

    Are holographic brains still a thing? I thought they’d been discredited.

    Hrothgar: “Overclocking” normal brains for superior performance is completely speculative fiction at this point in time.

    Well of course. Awesome speculative fiction, potentially.

    André Esteves:
    Could you correct the name of the author? It’s Oliveira not Oliviera.

    Oops. Thanks. Fixed.

    Patrick “Bahumat” Rochefort: Oh wow. I did a presentation on this very topic in 2007.

    Wow. WOW. I didn’t know that about the rhombencephalon. Maybe these guys are pattern-matching, nonconscious p-zombies!

  17. Simon: On the other hand it won’t be long before things without any brain at all regularly beat 126 on IQ tests:

    http://observer.com/2015/06/artificially-intelligent-computer-outperforms-humans-on-iq-test/

    Uh oh. That didn’t take nearly as long as I’d hoped.

  18. Simon,

    This prosaic explanations seems plausible, no?

    Never mind mental super-powers; if you could just run a normally functioning brain with a fraction of the cells, why would there not be strong selection pressure for hydrocephaly? Brains are metabolically expensive.

  19. Phil Koop: if you could just run a normally functioning brain with a fraction of the cells, why would there not be strong selection pressure for hydrocephaly? Brains are metabolically expensive.

    Depends on a couple of things. If the malady’s congenital, then the net benefit to the small fraction of hydrocephalics who are functional VNBs would have to outweigh the deleterious effects imposed on the greater hydrocephalic population (effects that include a shitload of nasty diseases in addition to the usual impacts of retardation, blindness, and convulsions inflicted directly by the condition, according to Wikipedia).

    And of course, if the malady’s developmental, then there’s no gene to spread.

  20. First of, do we really know there are no cognitive defecits? Something that might not show up on an test for g, whatever that is, but maybe on some test for spatial orientation, motor coordination, attention shifting or episodic memory? Knowing some mathematicians, deficits in those are not necessarily detrimental for mathematics. It might also be interesting how many social interactions they have, since there is some correlation between the size of some brain structures and group size in primates, e.g. Dunbar’s number.

    Second of, it might be the parts least affected are the ones at the surface, e.g. the cortex, so maybe it’s just the deeper areas missing. So viewing etc. usually just work normal, but the deeper areas that just play in with blinsight etc. are missing. Seems still somewhat strange, but evolution is somewhat conservative, and if e.g. said older structures are needed in early development, but not later on, selection would keep them.

    Third of, usually half your brain is occupied with shouting the other half down, so maybe it’s an reallocation of resources.

    And last but not least, maybe it’s not so much neuron volume but surface area that is important for cognition, which might not be that much changed in those subjects.

    Still somewhat strange, though…

  21. And of course, if the malady’s developmental, then there’s no gene to spread.

    The point isn’t that hydrocephaly itself would be selected for, rather that if all that brain matter isn’t actually needed, then why are so many humans spending 20% of their energy on a mostly useless organ – not to mention the selection pressure of mothers not dying in childbirth that would come from smaller heads. There are so many negative effects of the larger brain/head that there must have been some immense selection pressure for more brain matter to even get to this point. That pressure seems to be strongly at odds with the claim that a human can be normal with only 5% of the tissue.

  22. Did some computational science (AKA a quick Python script) on volumes of normal brains compared to these “shell” brains.

    Head circumference in cm seems to be the most common measurement used, with Canadian adults averaging 55-57cm according to Wikipedia.

    If the “shell” brain is 1cm thick, then at 65cm circumference it would be roughly the same volume as a 43cm circumference normal brain. These are both outside the average range, but are they excessively so? Are there no examples of people with below average brain volumes having an IQ of 126?

    Has anyone tried measuring the reflexes of VNBs rather than IQ? Would sending signals around the circumference rather than through the interior be measurably slower?

  23. “Are there no examples of people with below average brain volumes having an IQ of 126?”

    Guinness lists (or used to) Anatole France as having the smallest documented brain for a human.

  24. Most of our brain is simply spares, basically. Spares are really really important. Stuff breaks a lot.

    My grandmother was an old yogi, trained from a fairly young age. She had a stroke and lost the ability to speak. Part of her brain (one of the language centers) was blacked out, permanently gone, never lit up again. She taught herself how to speak again, using a different part of her brain. Had another stroke, lost the ability to speak again, permanently lost another section of her brain. She taught herself how to speak again, using a different part of her brain.

    The standard “map” of what is processed in what part of the brain is just an average. For any given feature, there’s a fairly large percentage of people who are wired completely differently — a different part of the brain performs the function which Broca’s area performs in average people. Et cetera. This is just known science. Brain tissue is exceptionally plastic, and can be used for a lot of purposes.

    Having spares seems really really useful.

  25. Also, remember that old claim that people only used a small fraction of their brains? The missing statement there was “at any one time” — it all gets used, generally, but not all at *once*. Perhaps you can’t overclock the brain because it’ll overheat. Maybe you can run everything on a small, tight brain network (with sufficient cooling). But arguably it’s better to only wake up a few smaller sections at a time and let them cool down in between thoughts, which is roughly what we seem to actually do. Do these functional hydrocephalics die younger than average?

  26. The point, though, is that under the right conditions, brain damage may paradoxically result in brain enhancement. Small-world, scale-free networking— focused, intensified, overclocked— might turbocharge a fragment of a brain into acting like the whole thing.

    Vernor Vinge’s _A Deepness in the Sky_ had something very similar with the Focused. The wikipedia page has more details.

  27. What I find interesting is that in the Lewin paper you can still see gyri and sulci, while in other disorders that also present with enlarged ventricles (like classical lissencephaly) those are absent. Naturally, the absence of gyri and sulci is part of lissencephaly as a condition, but one has to wonder.

    I hope the patient has a hippocampus……somewhere. *shudder*

  28. Kurzweil’s *How to Create a Mind* does talk about redundant clusters as a form of backup, especially for memory. Still have a hard time imagining overclocking being so darn effective.

  29. The point, though, is that under the right conditions, brain damage may paradoxically result in brain enhancement. Small-world, scale-free networking— focused, intensified, overclocked— might turbocharge a fragment of a brain into acting like the whole thing.
    Can you imagine what would happen if we applied that trick to a normal brain?

    Allegedly, that would be the case with Jason Padgett. Brain damage he suffered in an assault enhanced his ability to do high level math and physics intuitively.

    http://www.livescience.com/45349-brain-injury-turns-man-into-math-genius.html

  30. This is a brilliant and fascinating blog post. I have never heard of this phenomenon and it seems to have so many avenues of inquiry built in to it. The plasticity of the brain, of neurons, is nothing short of amazing. Wow.

  31. My best friend of 40 years happens to have a large portion of his brain “missing” due to a traumatic head injury. He has since earned his 2nd degree in “special education”. Perhaps there really are no roadblocks for the determined.

  32. His brain didn’t rewire for normality, it rewired for its genetic potential. If his brain was at its normal size his iq would be 126. When he lost most or part of the brain due to an illness or injury remaining part worked better to reach the genetic potential. So if you want to overclock normal brains for superior performance you have to engineer the source: the dna.

  33. eriko,

    And some humans are more dense than others.

  34. There is a lot of nerve tissue in the body other than that in the head. Could these nerves be involved in picking up the slack? I’ve seen articles suggesting that spiders and octopi use most of their body as a brain.

  35. Hugh:
    Has anyone tried measuring the reflexes of VNBs rather than IQ? Would sending signals around the circumference rather than through the interior be measurably slower?

    I wouldn’t care so much about reflexes, of which quite a few are in the brain stem or even the spinal chord. It also depends somewhat which structures are spared; as already mentioned, it seems like the hindbrain (rhombencephalon) is often relatively unaffected, which might indicate they use some phylogenetically older structures for apei, err, lizarding human behaviour. Though being myself somewhat on the autism spectrum [1], I wonder how effective that would be. BTW, they would not necessarily be P zombies, since there would be quite some other differences beside self consciousness. Hm, remind me to do a WADA test to know what it’s like to switch of brain areas…

    Also, if the damage goes from the in- to the outside, it could be the outer areas are less affected. Now humans differ from chimps not just in brain volume, but also in the, err, invagination of the brain tissue,

    https://commons.wikimedia.org/wiki/File:Human_and_chimp_brain.png

    which might indicate whatever makes us superior in some (but not all) areas (AFAIR chimps have better memory, inter alias) has more to do with surface area than with volume. Maybe associative networks work best in 2D or limited 3D, or whatever. And just getting rid of much of the deeper structures might not damage those that much, but that would raise the question why evolution kept the underlying structures in the first place. So it would be quite interesting to search for subtle differences, e.g. more concrete thinking in some areas.

    OTOH, it might be that the surface areas are more damaged, though that would raise the opposite question, e.g. why the deep sulci.

    Also, maybe the kind of work the brain does is subject to some strange work requirements,

    https://en.wikipedia.org/wiki/Time_complexity

    and it takes relatively little to get work done relatively well, e.g. touching your nose. to get a somewhat more precise result, though, e.g. touching the tip of your nose, which is maybe 10 times as precise, you might not need 10 times the time or number of processors, but 100 times. So big brain differences might not mean that big differences in absolute abilities, though the relative difference might be vital in some situations, e.g. stalking some prey.

    And last but not least, we don’t know what VNB would be like with full brains. Maybe our IQ of 128 guy would be 3 sigmas more out with a full set up cups.

    [1] Though I guess I have graduated from high functioning autism with ADHD to ADHD with autistic traits. 😉

  36. Trottelreiner,

    And dammit, I’m getting old, the term for the enlarged surface area was “gyrification”, so strike “invagination”:

    https://en.wikipedia.org/wiki/Gyrification

    Funny thing, in my inner ear I hear “invagination” with a heavy Upper Saxon accent. Don’t ask, traumatic university experiences…

  37. […] by Arkanj3l [link] […]

  38. If the explanation is really that simple, “all” it takes is to find out what triggers the mechanisms that create all these connections, and inject the same kind of hormones into a normal brain. But when you have problems learning math or a new language, could this task really become easier with more connections? Instead of asking why it is easy for some people to learn these things, maybe one could also ask why it is so difficult for others who have a perfectly healthy, normal brain.

    An old myth is that we only use 10% of our brain. But how efficient could it become if the neurons were more connected?

    Stuart Hameroff and Roger Penrose have been arguing for long now that the important stuff happens in the microtubule, and the neurons and glial cells are just specialized cells that connects all microtubule processes in each individual cell with each other. Can’t say if they are right, but it does sound interesting.
    When you look at a tiny nematode finding its way around in a pond, and looks at a large ciliate about the same size as the nematode, has the multicellular worm mental qualities that place it higher than the protist? They can sense the environment, orient themselves, try other solutions when they meet obstacles, speed up to chase a prey or escape danger, mate and show similar behavior in general. Are the nematodes in any way better than the similar sized protists in this? Faster reactions, better memory or something else? I don’t know, but a study could show if the worms have any advantages over the protists thanks to their neurons.

  39. First of all, 95% volume loss doesn’t mean 95% brain loss. As we know, the vast majority of brain cells within the cerebrum is in the cortex. The cortex is clearly intact in this man with hydracephalus. The rest of the space is white matter that composes of axon fibers. What this case probably demonstrates is that the axon fibers in the white matter can condense tremendously under the pressure of hydrocephalus and still function. Clearly knowledge is known to be stored in the brain as you can lose it by suffering brain trauma or stroke. For example, Dr. Taylor regressed into an infant as she suffered stroke on her left hemisphere as all the memories of her past became inaccessible during the stroke.
    http://www.ted.com/talks/jill_bolte_taylor_s_powerful_stroke_of_insight?language=en

  40. Uhhh…is anyone really thinking that we’d see the same computational intelligence had this patient been born this way?

    Hydrocephalus is degenerative. You’ve got to look at the rate of degeneration vs neurological rewiring. Basically it comes down to one main function:

    Neuroplasticity. Both synapse & structural.

    Small world/scale-free networks only work if there’s data density. I think many/most operate on some cellular automata variant (would need to go back & look). It appears to be an optimization function of space. The brain will use space more optimally if it has less to work with. BUT it has to stay ahead of the curve. I’d bet that rate of degeneration, combined with prior brain function are the two main variables at play here.

  41. […] Is your brain really necessary (and why is there a maths genius who only has 5% of one?) […]

  42. Eric: The point isn’t that hydrocephaly itself would be selected for, rather that if all that brain matter isn’t actually needed, then why are so many humans spending 20% of their energy on a mostly useless organ – not to mention the selection pressure of mothers not dying in childbirth that would come from smaller heads. There are so many negative effects of the larger brain/head that there must have been some immense selection pressure for more brain matter to even get to this point. That pressure seems to be strongly at odds with the claim that a human can be normal with only 5% of the tissue.

    The code for biological growth is incremental. Ie. you grow structure B, then you have code that grows structure C by making the required changes on/to B. Which means it’s almost impossible to get rid of old, objectively redundant structures and code.
    So the most likely explanation to me is that most of the brain (by volume) is useless.

  43. Mike,

    To play devil’s advocate for the mammal in gasform folks, what if the brain is just a receiver, and the stroke just shuts down part of the antenna array or some other part of the damn radio?

    Not that I believ that, but I had some fun with born-agains…

  44. There’s another interesting case worth mentioning: a woman without a cerebellum with mostly normal motor function (mild impairments; can still walk and talk):

    http://www.wired.com/2014/09/24-woman-discovers-born-without-key-brain-region-cerebellum/

    It is claimed that the cerebellum contains about 50% of the neurons in the brain; and the reports on this case seem to suggest she is missing all of them. The doctors believe the rest of her brain learned to compensate for the missing brain region.

  45. If we can boost the cognitive function of brain tissue with the right kind of stress, then we’d probably want to try it on chimpanzees before humans, which would result in intelligent chimpanzees.

    Planet of the Apes anyone?

  46. Peter

    I seriously love your writing and this blog. It is always though provoking a filled with new or just shown in completely new light ideas to the brim.

    As a fan of yours, I would like to ask you a question. Do you have any plans on writing wider ( and longer! ) popular science work.

    Coming back to topic of your post. Could it be possible knowing those stressors to enhance certain functions of brain or certain parts of brain?

  47. […] Gilead’s Greed That Kills Clallam’s measles outbreak price tag comes in at $223,223 Your online advice is rubbish if I can’t read your CV!!! Science: D.C. Is Literally Sinking No Brainer. […]

  48. I’ve wondered this with regards to sensory deficits and high IQ. Myopia correlates with high intelligence. Is compensation a factor in enhancing cognition? High IQ + low vestibular function is another thing I’ve noticed occurring together (anecdotally.) Potentially interesting.

  49. Very interesting. However, there are a few issues that many people forget about.

    First, consider Moravec’s paradox: what seems easy is hard, and what seems hard is easy. In principle ‘higher math’ could be done with very little brain tissue – it just seems hard to most of us, because it’s not what our brains evolved for (like using a jumbo jet to travel down the street to the grocery store). If this patient grew up with this condition he might have a ‘savant like’ efficient wiring for math.

    The other issue is that we evolved to compete and win, but there are diminishing returns. For example: suppose that in order to run 5% faster, you need 50% more muscle mass (I made this up). In the wild the person that runs 5% faster eats and the other person does not eat and dies. However, they can still run at 95% of the speed with half the muscle mass. From that perspective it might seem like 50% of your muscles are useless, but not really.

    Consider vision. Each eye has about 1 million nerve fibers connecting it to the rest of the brain. However, this is massively overkill for most tasks. You can lose almost all of your retinal nerve fibers through glaucoma and still read, drive a car etc. That doesn’t mean those million fibers are useless. In a winner-take-all competition, the person with a million fibers per eye will likely beat out the person with only 100,000 fibers per eye. But the person with 100,000 fibers would still appear to function almost normally for non-competitive simple functional tasks.

    I mean, compare a 2 megapixel camera with a 25 megapixel camera. You can still see basic objects with the 2 megapixel one, right? But if you are hawk searching for rabbits, you want the 25 megapixel equivalent.

    It’s not that most of the brain is useless. it’s that the human brain is a bleeding-edge combat system designed for maximal performance. It can still work pretty good with only a fraction of the neurons, though, as long as the damage is not all at once and doesn’t cut the major connecting paths or the vital subcortical nuclei.

  50. Oh sorry, just had another idea.

    It does suggest that if we were willing to settle for slightly (maybe just SLIGHTLY) lower levels of performance in things like visual acuity and coordination etc., we might be able to free up maybe half the cerebral cortex or more for other tasks, for which we presently have no ability at all. Probably hard/impossible to do once you are grown up, but maybe during development? Force the normal brain functions into a restricted space – the person likely won’t grow up to be a major league baseball player but they will still be able to do normal tasks (drive, read, etc.) fine. Now trick the remaining cortex into doing something new…

  51. SI2K,

    SI2K- while Myopia does seems to correlate with high intelligence, the current work on it seems to make it that the myopia is brought on by the attributes of high intelligence i.e. lots of indoor close up work and reading.. This is gaining more strength as the current epidemic of myopia seen in the Pacific Rim is taken into account, and there is some very cool work being done on causing, and not causing, myopia. It may turn out that the stereotype of the geeks reading in their basements do indeed cause myopia.

  52. SI2K:
    I’ve wondered this with regards to sensory deficits and high IQ. Myopia correlates with high intelligence. Is compensation a factor in enhancing cognition?

    With myopia the thing is that if your eye is oversized, then you’ll be myopic. Plenty of brain size genes are going to affect the eyeball as well, so it may simply be a side effect from having too many genes that increase the head size, in one individual.

    (Other possibility is that myopia is the result of eye adapting to reading, which higher IQ people do more of).

  53. “myopia is the result of eye adapting to reading, which higher IQ people do more of”

    It’s sortta like this. Only not as simple. But for now, I’ve had very smart people advise that if you have young children make sure they spent an hour a two a day outside playing. In my day all the smart kids did not pay football or baseball- we stayed in and read. Now it looks like science is starting to show this actually may be responsible. With young people always indoors reading or on their computers, myopia is epidemic. Taiwan recently wads unable to find anyone that could qualify for fighter pilot school due to bad eyesight, and it’s been proven to not be genetic.

  54. Adam Etzion,

    The cats would still ignore us. ^_^

  55. gwern:
    If net brain performance could be boosted by one order of magnitude, or even just doubled (compare humans with ~100b neurons to creatures with 50b neurons), they would be as gods unto men. Either by accident or by evolution, they should exist. As Fermi might say: where are they?

    Maybe it’s that guy who does 2 man years of programming in 1 year (and can solve slightly harder programming problems).

    There’s huge variability in programmer performance, some programmers are an order of magnitude better than average.

    I don’t think there’s anything dramatically awesome going on with that hydrocephaly example, though. It is the white matter is affected, and you can make all those cables go around and be thinner at the expense of minor increase in propagation delays. And you can probably lose a lot of connectivity with the effects such as being less accurate at locating an itch with your right hand on your left side.

    Picture a slowly enlarging plastic bubble in a well maintained supercomputer room. The cables now go around the bubble, the machines are all squished to the walls, the airflow is complete shit. But none of that matters for a task that is too short term for the supercomputer to overheat.

  56. Ohh and don’t forget that the skull is significantly enlarged in hydrocephaly, so picture that the supercomputer room itself got blown larger. Devil is in the details – one guy has normal skull volume, hydrocephaly and IQ of 75 , another guy has larger head, normal sized cortex, and the IQ of 126.

    Mix those two up and it’ll be a lot more mind blowing than it really is…

  57. I’ve did a little reading on hydrocephalus and it’s as I remember: in general, serious mental impairment, likely death if untreated, etc etc. Overwhelmingly not compensated for.

    The 126 IQ guy may be a measurement error; the white collar guy from 2007 got IQ of 75 (verbal 84 and performance 70) and Sharon Parker the nurse with IQ of 113, well,

    For Sharon, the most important discovery the American doctors made concerned the size of her brain. Something that only the very latest technology allowed them to measure. Having been told, all her life, that she has only 10-15% of a normal brain the result came as a welcome surprise, but it was something of a shock for Dave. While the volume of Dave’s brain was 1300cc Sharon took great delight in telling him that hers was 2300cc.

    The IQ 70 guy, we don’t know how much impairment we should have expected given that he got close to normal brain surface area, and his number of neurons may be perfectly in line with high IQ.

    Most of the neurons present in adult human brain were once packed into a much smaller volume (most neurogenesis ceases by the third trimester. The head is tiny then!). The extra volume that is added is astrocytes, myelin, and other non neuron “stuff” which fills most of your skull. (Not extra connections though – connections are pruned in childhood, like chipping off marble to make a statue).

    It is highly plausible that an IQ test does not fully assess performance of all that non neuron machinery in your head. Myelin is crucial for fast reaction times, glia is crucial for providing brain with nutrients and removing spent chemicals. Impairment in the latter would most affect long term endurance.

  58. […] No Brainer […]

  59. […] Some number of hydrocephalics (people with fluid replacing up to 95% of their brain) are completely unaffected and in fact undetected until they get head scanned for other reasons.  Leading explanations include “isolation makes remaining area strong” and “souls”. […]

  60. BMT,

    Phil Koop,

    And expensive to younger siblings coming into existence, thanks to maternal mortality.

  61. OTOH, extracorporeality echoes Penrose quite a bit, doesn’t it?

  62. Trottelreiner: It might also be interesting how many social interactions they have, since there is some correlation between the size of some brain structures and group size in primates, e.g. Dunbar’s number.

    Don’t know the details, but the guy Lorber reported was said to be pretty normal, socially. Some of the more recent cases, too.

    Eric: The point isn’t that hydrocephaly itself would be selected for, rather that if all that brain matter isn’t actually needed, then why are so many humans spending 20% of their energy on a mostly useless organ – not to mention the selection pressure of mothers not dying in childbirth that would come from smaller heads. There are so many negative effects of the larger brain/head that there must have been some immense selection pressure for more brain matter to even get to this point. That pressure seems to be strongly at odds with the claim that a human can be normal with only 5% of the tissue.

    Yeah, these are all really good points. All I can suggest, off the top of my head, is that the strength of the selection doesn’t matter unless there’s relevant variation for the selection to act upon. The developmental recipe for growing a brain would have to experience some kind of a mutation to permit going down a different road; maybe the right dice roll never came along.

    Maybe we could induce it now, though; in which case, the New PostHumans would all be microcephalics. Wonder how many Kurzweillians would be eager to jump on board that/i> train…

    Hugh: Would sending signals around the circumference rather than through the interior be measurably slower?

    Might be more complex than that. White matter is better suited for long-distance connectivity than your average grey neuron, and apparently that’s most of what gets lost; so a different kind of circuitry may have a tougher time taking up the slack in principle. Given that we see evidence that such underconnectedness results in greater functionality at the local scale, I’m guessing that might be a more likely result here.

  63. Nathanael: Do these functional hydrocephalics die younger than average?

    Good question.

    Steel Phoenix: There is a lot of nerve tissue in the body other than that in the head. Could these nerves be involved in picking up the slack? I’ve seen articles suggesting that spiders and octopi use most of their body as a brain.

    I’d be really surprised if this were the case. I don’t know about spiders, but the reason octopi brains have outsourced so much of their cognition is because all those incredibly dextrous prehensile suckers need control at the local level. There’s even been some suggestion that individual octopus arms have their own personalities as a result. But there’s no comparable feature of the human body to justify a distributed intelligence. (Although they do say that the neural net runing around our GI tracts have the neural complexity of a cat brain…)

  64. Sibylle,

    A UCLA Newsroom press release details the findings of life scientists from UCLA and Australia showing that when the brain’s primary “learning center” is damaged, parts of the prefrontal cortex take over. The findings could help in the development of new treatments for Alzheimer’s disease, stroke and other conditions involving damage to the brain.

    Click here to read the press release.

  65. […] Peter Watts: "[Roger] Lewin’s paper reports that one out of ten hydrocephalus cases are so extreme that cerebro… […]

  66. […] It happens occasionally. Someone grows up to become a construction worker or a schoolteacher, before learning that they should have been a rutabaga instead. Lewin’s paper reports that one out of ten hydrocephalus cases are so extreme that cerebrospinal fluid fills 95% of the cranium. Anyone whose brain fits into the remaining 5% should be nothing short of vegetative; yet apparently, fully half have IQs over 100.— Peter Watts […]

  67. Peter Watts:

    Wow. WOW. I didn’t know that about the rhombencephalon. Maybe these guys are pattern-matching, nonconscious p-zombies!

    Well, there you go with your zombies. Now that you’ve explored hyperactive weaponized zombies, why not hypersocial economized zombies?

    I mean really, so many jobs out there just waiting to be filled by customer service drones… 😉

  68. Grey matter is neurons, white matter is axons. My guess is the grey matter is normal, or at least not severely compromised. The axons may be just squashed against the grey matter. It’s also possible the axons are less numerous. Axon deficit would hamper variation of firing rates. It would then seem “cliques” could be an important adaptation. I guess this may turn out to be a lot less mysterious than it appears at first glance.

  69. Peter Watts:

    Wow. WOW. I didn’t know that about the rhombencephalon. Maybe these guys are pattern-matching, nonconscious p-zombies!

    I once spent a few days wondering if maybe I was just a pattern-matching nonconscious p-zombie and came to the conclusion that if I was, I’d never know it and would probably deny if it anyone accused me. Then I pondered a bit farther and decided that if I was, it could explain a lot about me, and also decided that if I was a p-zombie I’d probably rather not know since there probably wouldn’t be anything anyone could do about it that didn’t involve violence or sequestration.

    But back to the search for relevance: I had often wondered about the source of the old saw “people don’t even use but about 10 or 20 percent of their brain”. Reading this article and the comments, it looks like maybe 80 to 90 percent of the “normal” brain mass is missing from these well-adapted survivors of hydrocephalus. I don’t know how far back in the history of neurology/medicine this phenomenon has been observed, but the old saw seems to fit, so to speak, but it seems to perhaps be an unreasonable generalization from extreme examples. One does wonder what might be the function of all of the “extraneous” tissue without which these people seem to be able to function on a level considered to be at least normal.

    I would propose that the “extraneous” tissues are a reserve against trauma. There are some classics of the field such as various head-trauma cases in military medicine, where someone pretty much loses half a brain and has a miraculous recovery months or years later, regaining most of their previous capacities practically overnight (a year later), but now they are left-handed instead of right-handed. There are the folks who get half of their brain anaesthetized and we see the emergence of different personalities which disappear as the anaesthesia subsides. There are the fairly rare cases of people who get a bang on the head and they’re perfectly functional aside from not knowing who they are or much else that’s personal to their own lives. The astonishing thing is that despite trauma that we might expect to permanently cripple, recovery may be more common than not, for some values of “complete” in recovery.

    I can’t think of many ways to test this theory which wouldn’t be totally out of ethical bounds, but perhaps however well these people may seem to live normal and competent lives having lost a great deal of brain mass to hydrocephalus, they might have almost no hope of recovery following additional brain trauma, because there’s just no tissue there to be re-purposed by plasticity of the brain. Leaving out of the discussion that these people are in their condition as the result of a technological intervention, we might think that there might be some evolutionary value in having all of the capability of a “normal” using only 20 percent of the brain mass, but in the pre-technical world in which we all evolved, it might be far more valuable to have a reserve left idle against the likelihood of needing it for trauma recovery. In a social animal such as primates, which also might be thought more likely than most to suffer brain traumas as a result of falling out of trees, such a reserve might actually be likely to get a chance to be used, and might likely be conserved in the gene pool if recovered specimens went on to successfully reproduce. That normal function is possible after technical intervention in hydrocephalus, that’s just serendipity. But if you were trying to write a story about specially bred/engineered hominids designed for “normal” mental function in extreme environments (military etc) you would probably have to throw in some bit about any specimen’s fellows instantly abandoning their members who had any significant cranial damage, as almost incapable of recovery to anything like “normal”.

    I guess a good question here might be “but how could we use imaging techniques to discern in healthy brains, what might be “daily use” brain tissues, from “idle in reserve” tissues? Maybe there’s some indicator that tuning to task or function hasn’t occurred yet? Or are there papers that definitively rule out this sort of thing?

  70. http://johnhawks.net/weblog/reviews/brain/development/ten_percent_brain_myth_2007.html

  71. Nestor:
    http://johnhawks.net/weblog/reviews/brain/development/ten_percent_brain_myth_2007.html

    Beautiful. This is just about exactly what I was looking for… thanks very much and thanks even more for that site in general. Regards,

  72. Yeah, almost decided not to watch Lucy because they used that urban legend in the ad. Probably not fair since writer/director may have no control over ads.

  73. Trottelreiner: Also, if the damage goes from the in- to the outside, it could be the outer areas are less affected.

    I have vague memories of a documentary dealing with Lorber’s case (I think it might be up on Youtube or something), in which they did eighties or nineties-era MRI on a VNB (although back then, they were still calling it NMR, having not yet pussied under to antinuclear paranoia). IIRC, the outer layers were not relatively unchanged; the brain parts that lit up during sensory processing, motor commands, etc, did not necessarily correspond to the same areas in normal brains. The remodelled brain seemed to just allocate functions to whatever substrate was available, without too much respect for legacy.

    Biff: But when you have problems learning math or a new language, could this task really become easier with more connections?

    My understanding is that learning involves a reduction in the number of neural connections. Newborn babies apparently start with a headfull of randomly-wired mush, and the redundant connections get pruned away as meaningful experience accumulates.

    Biff: Stuart Hameroff and Roger Penrose have been arguing for long now that the important stuff happens in the microtubule, and the neurons and glial cells are just specialized cells that connects all microtubule processes in each individual cell with each other. Can’t say if they are right, but it does sound interesting.

    Most experts do say they’re wrong, actually. I myself have nowhere near the expertise to pass judgment, but there’s a part of me reserving judgment; their model does seem to explain some interesting things, and if nothing else, I think Penrose is right when he claims that we won’t truly understand consciousness without coming up with a new kind of physics.

    sadpenguin: The code for biological growth is incremental. Ie. you grow structure B, then you have code that grows structure C by making the required changes on/to B. Which means it’s almost impossible to get rid of old, objectively redundant structures and code.
    So the most likely explanation to me is that most of the brain (by volume) is useless.

    I buy the first part, not necessarily the second.

    It’s definitely true that genes don’t program specific synaptic connections; all genes do is say “Glial cells, migrate now! Glial cells, >\stop now! Dendrites ready three two one, proliferate! (The famous shorthand is that genes are not a blueprint, but a recipe.)

    Don’t know if that leads to the conclusion that most of the brain s useless, though. The brain is a big glucose hog; there’d be strong selection pressure to reduce that cost if it didn’t carry it’s weight somehow. Of course, no matter how strong the selection pressure, nothing happens until you get a favorable mutation that lets you start down that road— but there’s so much brainular variability throughout the animal kingdom that I gotta think such a mutation would have happened by now.

    Trottelreiner: To play devil’s advocate for the mammal in gasform folks, what if the brain is just a receiver, and the stroke just shuts down part of the antenna array or some other part of the damn radio?

    Actually, I think that might be one of the claims Penrose and Hameroff are making..

  74. Merkator: As a fan of yours, I would like to ask you a question. Do you have any plans on writing wider ( and longer! ) popular science work.

    No plans, no. I’ve considered it now and then— given my biology background and my prose chops, I think I’d be pretty good at it— but I’ve never pursued it as a serious option. Someone once offered to put me in touch with someone over at the Ottawa Citizen WRT pitching a regular column there, but I never took them up on it.

    I’ve done the occasional nonfiction piece (that recent Aeon essay, a couple of commissioned true-story bits on my brush with nec fasc); those paid well, and I’d gladly do more. But the pay for regular columns seems to be shit.

    Quite honestly, I’d love to get paid for the science pieces I stick right here on the ‘crawl— I spend a lot of time researching and writing the fucking things, and I frequently resent the fact that I spend so much time on something that I don’t get paid for when I should be writing novels. But without the science parts, the ‘crawl would just deteriorate into one of those me-Me-ME! blogs o’Authorial Narcissism that already infest the internet. So the only alternative is to not blog at all, and then I just drop off the scope.

    Merkator: Could it be possible knowing those stressors to enhance certain functions of brain or certain parts of brain?

    Well, that was the potential breakthrough I kinda left hanging at the end of the piece…

  75. Timothy Gawne: It’s not that most of the brain is useless. it’s that the human brain is a bleeding-edge combat system designed for maximal performance. It can still work pretty good with only a fraction of the neurons, though, as long as the damage is not all at once and doesn’t cut the major connecting paths or the vital subcortical nuclei.

    That’s a really good point.

    Ann Gunning: Click here to read the press release.

    Um, that wasn’t a link.

  76. Peter

    I think, there are tons of people, who would find you posts interesting.
    If you would make this a little more popular ( posting on some bigger social pages, think reddit and similar ones ) would give you needed popularity.

    If not among fans, maybe across science geeks ( I am both ).

    Some short commentary video would be great too.

  77. Peter Watts:
    Um, that wasn’t a link.

    Gunning was likely linking to this release.

    The release is likely referring to this article in PNAS v110n24, with the full text a tab over from the abstract.

  78. Peter Watts:
    Actually, I think that might be one of the claims Penrose and Hameroff are making..

    Well, there are also Eccles and Popper, no idea who was first.
    Not that I believe that is true, it just makes for some nice trolling (in the positive, Stephen Colbertesque sense) in discussion.

  79. Eric,

    The point isn’t that hydrocephaly itself would be selected for, rather that if all that brain matter isn’t actually needed, then why are so many humans spending 20% of their energy on a mostly useless organ – not to mention the selection pressure of mothers not dying in childbirth that would come from smaller heads. There are so many negative effects of the larger brain/head that there must have been some immense selection pressure for more brain matter to even get to this point. That pressure seems to be strongly at odds with the claim that a human can be normal with only 5% of the tissue.

    I won’t speculate on the childbirth factor, but as for the energy usage, well, have there been any studies of the metabolism of VNBs? Maybe it costs just as much energy to run a brain at 126-IQ-level processing power regardless of how much gray matter you’re running it on?

  80. Peter Watts: No plans, no. I’ve considered it now and then— given my biology background and my prose chops, I think I’d be pretty good at it— but I’ve never pursued it as a serious option. Someone once offered to put me in touch with someone over at the Ottawa Citizen WRT pitching a regular column there, but I never took them up on it.

    I’ve done the occasional nonfiction piece (that recent Aeon essay, a couple of commissioned true-story bits on my brush with nec fasc); those paid well, and I’d gladly do more. But the pay for regular columns seems to be shit.

    Quite honestly, I’d love to get paid for the science pieces I stick right here on the ‘crawl— I spend a lot of time researching and writing the fucking things, and I frequently resent the fact that I spend so much time on something that I don’t get paid for when I should be writing novels. But without the science parts, the ‘crawl would just deteriorate into one of those me-Me-ME! blogs o’Authorial Narcissism that already infest the internet. So the only alternative is to not blog at all, and then I just drop off the scope.

    Well, that was the potential breakthrough I kinda left hanging at the end of the piece…

    Patreon is (as suggested earlier in this thread) just for this kind of thing, but with the caveat that (I would guess) you might need a quite a reader base to get enough people to bother to sign up. A couple of bucks a head from the usual commenters don’t add up to a whole lot.

    The modern media landscape(TM) has a quite heavy bias against well researched content, bang per buck (or click per buck) just doesn’t scale. Not that I’d like the old landscape back, in the old landscape I wouldn’t have this post at all :).