Quantum Brains

OK, I should be preparing a talk, but it is late and my mind is wandering, so it’s not going to happen tonight.  Instead, I’ll pose this puzzler:  If quantum computers are more efficient than classical ones then why didn’t our brains evolve to take advantage of quantum information processing?

I have a vague recollection of seeing this question on a physics blog somewhere before, and it does have a family resemblance to Scott’s infamous post, albeit a more politically correct version.

There are a number of assumptions behind this question:

  • Evolution usually does a very efficient job of coming up with information processing devices.  As evidence for this note that the best algorithms we have for some tasks simply imitiate nature, e.g. neural networks, simulated annealing, etc.
  • Some functions of the brain, such as the ability to solve math problems, are best understood by regarding the brain as a kind of computer.  Note that we don’t need to say that the brain is merely a computer, only that it can be regarded as such for understanding some of its functions, i.e. we don’t need to get into a big philosophical debate about conciousness and artificial intelligence.
  • Further, in these respects the brain is a classical computer and not a quantum one.  It certainly seems that the information processing function of neurons can be understood in classical terms, i.e. neural networks again.  There is a small minority of experts who believe that quantum mechanics plays an essential role in the information processing functions of the brain for whom my question is nonsense.

Here are all the possible explanations I can think of.

  • The set of problems in BQP, but not in P does not include anything that would have conferred a significant survival advantage for our ancestors.  Admittedly, efficient factoring could be useful for surviving high-school math class, as well as for cracking codes, but this wouldn’t have mattered so much to cave-people.  This would be disappointing, although not devastating, news for people trying to come up with new quantum algorithms.
  • There is some big problem with building a stable quantum computer of any appreciable size, and so present day experimentalists will eventually run into the same problems that nature did.
  • Dumb luck.  Evolution tends to find local minima in the landscape of all possible species.  Having a quantum brain is indeed a lower minimum than our current classical brain, but we never got a big enough hit to get over the mountain separating that solution from ourselves.

The first two explanations seem like the most interesting ones.  If the third explanation wasn’t a possibility then there would have to be a tradeoff between the amount of progress possible in developing quantum algorithms and the amount possible in actually building a quantum computer.  Given that much quantum computing funding is predicated on the idea that massive progress is possible in both areas, I’d say we should thank Darwin for dumb luck!

Creative Commons License
Quantum Brains by Matthew Leifer, unless otherwise expressly stated, is licensed under a Creative Commons Attribution-Noncommercial 3.0 Unported License.

16 Responses to Quantum Brains

  1. Ralph Hartley

    Well, the most obvious reason would be number 2, but without the non sequitur second sentence. Something like:

    There is some big problem with building a stable quantum computer of any appreciable size out of biological materials in solution at body temperature , and so present day experimentalists will never run into the same problems that nature did (but may run into other problems).

    Even if it were possible to build, the the cost could easily overwhelm the benefit. Most animals just don’t need to factor numbers that badly, and most have a very limited budget for computation.

    The phrase “Dumb Luck” seriously mis characterizes the third reason. Evolution has rules that have nothing to do with “Dumb Luck”. In order for an adaptation to evolve, it has to be reachable by a path, and every step along that path has to be an improvement locally.

    I don’t see how this question as significantly different from “why don’t any large animals have wheels?” Faster locomotion would surely be a bigger win for most animals than anything a quantum computer could do for them.

  2. Robin Blume-Kohout

    I like the ponder, but I think two of your points are weak.

    First, Evolution usually does a very efficient job of coming up with information processing devices…the best algorithms we have for some tasks simply imitiate nature, e.g. neural networks, simulated annealing, etc. Annealing isn’t a product of evolution. A stronger statement might be “Nature usually does…”, but that doesn’t tie in to your thesis as well. Furthermore, for a lot of information processing tasks, no natural/evolutionary solution is very good. Sorting in O(n*log[n])? Fourier Transform in O(n*log[n])? Fast integer multiplication? I submit that the IP tasks where technology hasn’t massively outperformed nature are the exception, not the rule.

    Second, There is some big problem with building a stable quantum computer…experimentalists will eventually run into the same problems that nature did. What about wheels, or propellers? Rotary motion is a great solution to a common problem, but wheels and propellors have never evolved. Dawkins’ argument about the eye shows that evolution can get to amazing places, but there are lots of isolated global minima (involving, e.g. axles and bearings) that it doesn’t get to. I submit that coherent information processing, like axles and bearings, is separated from the rest of the computational landscape by quite a high potential barrier. Such a barrier can be crossed rather easily by design, though not by a local random walk.

    So, if our brains are BPP machines, it’s good evidence against intelligent design. Surely God would have upgraded us to BQP, no?

    P.S. Why can’t I preview my comment?

  3. OK, almost all “why didn’t X evolve?” questions are extremely silly and this is no exception, but they do get people going. Please take into account the fact that it was a night thought.

    Ralph Hartley said, “Most animals just don’t need to factor numbers that badly, and most have a very limited budget for computation.”

    I should comment here that we don’t know for sure that factoring is the only useful thing that quantum computers can do. Part of the point was to infer that they shouldn’t be able to do anything that confers significant survival advantage, which would be a far stronger result than any quantum complexity has ever managed to obtain. However, I admit that the entire premise is deeply flawed.

    Robin said, “Annealing isn’t a product of evolution. A stronger statement might be “Nature usually does…”, but that doesn’t tie in to your thesis as well.”

    Granted, but evolution itself could be thought of as a kind of annealing. I guess that’s still irrelevant to my argument, but I just felt like saying it.

    Robin said, “P.S. Why can’t I preview my comment?”

    No idea.

  4. Torbjörn Larsson

    Interesting way of formulating the question.

    Evidently I don’t know enough about quantum computers, because I don’t understand why Tegmark’s paper on decoherence time in brains and their classical behavior wouldn’t preclude QC here.

    If it doesn’t, I would think that the tasks involved doesn’t benefit from QC. There are a lot of possible systems that doesn’t seem to be used by evolution, like the mentioned wheels for large animals and radio waves (but electric fields and magnetism) et cetera. A few animals roll for locomotion, at least part time, but it is usually not practical. Useful systems have been reinvented a number of times, eyes upwards 40 times or so IIRC, wings several times et cetera. (Granted, for the eye photosensitive chemistry and cells have been kept.)

    “Rotary motion is a great solution to a common problem, but wheels and propellors have never evolved.
    Dawkins’ argument about the eye shows that evolution can get to amazing places, but there are lots of isolated global minima (involving, e.g. axles and bearings) that it doesn’t get to.”

    Not being a biologist I haven’t read Dawkins, but I have heard that he is leaning towards selection as the useful mechanism. Other theories are things like neutral drift, where may populations change without selective pressures. So contingency can have a lock-in effect (hard to get from legs to wheels) but smaller fitness gaps can presumably be traversed.

    Oh, and I assume this is discussing macroscopic animals. Flagella can be propellers and have axles and bearings. (Btw, our cells have them too. For example, it is flagella that helps establish left-right asymmetries in the fetus by setting up a gradient flow over surfaces when spinning in a set direction forcing a tilt. People who have deficient flagella have a 50 % chance to end up with the heart on the right side.) And IIRC small animals or protists may look and move like a wheel or rather spinning ball in water – it is practical on their scale.

  5. Personally, I think it is pretty clear that having a quantum computing-enabled brain would not have conferred any significant reproductive advantage to our ancestors, simply because evolution does not seem to have selected that strongly even for classical computing ability. Judging by my own performance, my ancestors clearly did not gain a big advantage by being able to keep pi’s, i’s, minus signs and factors of 2 straight in a long and messy calculation. And even for simple arithmetic, our brain is not very good at handling large numbers (larger than say the typical size of a group of social primates) with high precision, and it is even worse if it has to deal with several numbers at once. So if evolution did not select strongly for even these very simple classical capabilities, why should it select for an exponential speed-up in factoring ability, or anything else that a quantum computer might conceivably be better at than a classical one? Indeed, evolution does not seem to select very strongly for overly brainy creatures at all, and our own existence could be seen as a fluke. After all, we are the only species left of a large tree of brainy hominins, and genetic evidence suggests that even we nearly went extinct several times in relatively recent geological history. Now, after we evolved technology that is sophisticated enough to more or less guarantee our survival against anything other than that technology itself, our brains are a spectacular advantage, despite consuming a ridiculously large portion of our calorie intake. But we probably had to clear a pretty big threshold to get to that point.

  6. Robin Blume-Kohout

    Matt,

    Let me hasten to encourage late-night thoughts! We lurkers in the blogosphere like nothing better than a half-baked comment that lets us demonstrate that we, too, have something to say. Tightly reasoned and elegant posts are disappointing, because they leave nothing for us vultures.

    Or, to put it another way, late-night thoughts often provide the nucleus for lots of discussion.

    Anyway, in other news… it would be rather interesting to characterize the computational problems whose solutions do arguably provide a fitness advantage. Or, if you like, the fitness advantages that could be seen as solutions to computational problems. Did T-Rex evolve a stomach that more efficiently solves the napsack problem? Okay, that’s a really dumb example…

    Torbjorn (sorry, don’t have an umlaut here), do you have examples of microscopic axles? As far as I know, while flagella do make use of rotational motion, it is indirect — akin to a baseball player’s or stone slinger’s rotational motion, which is induced by coordinated linear muscle action. More like an undulation (fish-tail) that’s modulated rotationally. But I’d love to see an actual axle or bearing example! (Having the whole animal spin, like a armadillo rolling down a hill, is rather a different matter).

    Matt, w/respect to previewing… there’s a button here that “Submit Comment”, and I’m used to seeing one next to it that says “Preview Comment”. It’s rather nice for making sure I didn’t forget a trailing HTML tag or something. Does this option not exist on your blog, or is my browser missing it?

  7. Torbjörn Larsson

    “evolution does not seem to have selected that strongly even for classical computing ability”

    But the methods of the brain doesn’t have a direct bearing on math ability. In fact, our representation of the environment is incredibly sparse, which probably explain why we have problems with this new task.

    Evolution is an ongoing process, so if you wait a couple of thousand years, who knows what we can do? :-)

    Robin:

    “an umlaut”

    Thanks. Technically it is a diaeresis in Swedish, it is a genuine letter. (It can be done with HTML: “& # 246 ;” remove quotation marks and spaces. Testing: Torbjörn.)

    “As far as I know, while flagella do make use of rotational motion, it is indirect — akin to a baseball player’s or stone slinger’s rotational motion, which is induced by coordinated linear muscle action.”

    I don’t know about those, but it could be – apparently there are a huge number of variants. But this has been discussed a lot because creationists use them as examples of cellular machines. (Though they mistake it for an outboard engine, when in fact it is an inboard engine.)

    Here is a post that links to a really nifty youtube animation of bacterial flagella evolution and workings: http://www.pandasthumb.org/archives/2006/12/ode_to_the_flag.html . It shows the bearings and the rotor engine, driven by ATP. Note: It isn’t entirely accurate.

    And here is the (accurate) source material for the animation, including drawings of different flagella and TEM pictures: http://www.talkdesign.org/faqs/flagellum.html . It discusses some of the variants like you mention, starting from the archetypical membrane secretion system, over sticking, pushing, and wiggling flagella to rotational. I believe the article developes an evolutionary model over this sequence, btw.

  8. “Does this option not exist on your blog, or is my browser missing it?”

    It’s probably something to do with the settings, so I’ll look into it. This being a Quantum Foundations blog, I’m worried about the crank contingent taking over the comments, so I’ve got just about every filtering option set as high as possible. You’re doing well if you managed to get a comment to appear at all!

  9. It is the ‘Algernon law’, named after Algernon, the mouse from the novel ‘Flowers for Algernon’, written by Daniel Keyes Moran. Any simple major enhancement to human intelligence is a net evolutionary _disadvantage_. Human intelligence is (supposed to be) at its optimal level. Not maximal, but optimal for our _survival_.

  10. > In order for an adaptation to evolve, it has to be reachable by a path, and every step along that path has to be an improvement locally.

    Not necessarily, there can be some non-immediately-improving steps which are still kept in the genome and later harnessed together with some improvement.

  11. Jan: Your arguments were actually nicely addressed in the original blog post (specifically, explanation 1).

  12. So what would the classical world look like to a quantum observer?

  13. Anonymous, there is a paper by Albert about ‘quantum automata’ and their behaviour. But also this amazing quote …
    “Once, at the afternoon tea, in the Institute [Copenhagen] E.Teller tried to explain to Bohr
    why he thought Bohr was wrong in thinking that the historical set-up of classical concepts would forever dominate our way of expressing our sense experience. Bohr listened with closed eyes and finally only said: “Oh, I understand. You might as well say that we are not sitting here, drinking tea, but that we are just dreaming all that.”
    -‘The Copenhagen Interpretation’, by C.F.von Weizsaecker, in ‘Quantum Theory and Beyond’, Ted Bastin ed., Cambridge U.P., 1971.

  14. I argue that in nature robustness is very important. The reason nature never evolved wheels is because wheels are not guaranteed to work in all possible conditions: in snow, in water, up mountains, in forests etc. My view is that evolution is capable of pretty much anything we have done but much of what we have done is very nonrobust so evolution does not tend to favor it. The brain is extremely robust. You can cut in in half, remove parts of it, bang on it, change temperature etc and it still keeps functioning. Most of our technology is not capable of that. Robustness is important because evolution is trying to create survival machines and the environment is constantly changing.

  15. Pingback: Philosophia Naturalis #8 « {metadatta}

  16. Pingback: Test Post « metadatta

Leave a Reply