Slashdot Mirror


A Savant Explains His Abilities

numLocked writes "Of the few hundred autistic savants in the world, none have been able to explain their incredible mental abilities. Until now, that is. It seems that Daniel Tammet, a mathematical savant who holds the record for the most digits of pi recited from memory, is able to explain exactly how he intuits answers to mathematical problems. Tammet is quite articulate and speaks seven languages, including one he invented. The Guardian is running an article about his amazing abilities."

10 of 930 comments (clear)

  1. Crypto by koreaman · · Score: 5, Interesting

    I don't really know a lot about autistic savants or encryption technologies, so this may sound idiotic, but if these guys can so easily factor large numbers why don't they have them working for NSA breaking public-key encryption?

  2. Does not Compute! by ParadoxicalPostulate · · Score: 5, Interesting


    "When I multiply numbers together, I see two shapes. The image starts to change and evolve, and a third shape emerges. That's the answer. It's mental imagery. It's like maths without having to think."

    I don't understand. There is nothing intrinsic in the number 2 and the number 5 that will tell you what they will equal when they are multiplied.

    The way we arrive at the solution is extrinsic, namely in the form of the operator (multiplication in this instance).

    But if it's extrinsic, I don't understand what the author of the article means by "instinct" and "shapes" and that sort of thing. As far as I can understand, the only explanation would be the ability to compute those operations at much higher speed, then any "non-savant."

    If that's the case, then, theoretically, would there not be a limit associated with the physical properties of the nervous system that would cap out at a certain number of such operations per unit time? So theoretically might we not be able to test such a thing by running him through a long list of operations? That'll let us know if he's really just making those calculations really, really fast, or if he really is viewing the mathematics in such a fundamentally different way (something I find rather unsettling).

    Then again, how would we design such a test? I fear that the number of operations we can demand his brain to perform per unit time will be limited by his powers of cognition (i.e. by the time he reads/hears all the stuff he needs to hear, we'll already be beyond that critical operating time interval).

    Eh, I think I come off as somewhat difficult to understand. Oh well, I wanted to make sure my question appeared in the main thread of discussion (rather than being posted after most people have moved on).

  3. The brain of a savant by Space_Soldier · · Score: 5, Interesting

    "He can't drive a car, wire a plug, or tell right from left."

    Is it possible that knowing how to drive a car, wire a plug, tell right from left, and other banal things that we do require a ton of processing power? Since he cannot do these things, all that processing power goes to processing numbers and memorising words.

    It we would be cool if on a math test we cold forget our ability to drive cars and concentrate on processing numbers.

  4. It's not intelligence in any conventional sense... by Caspian · · Score: 5, Interesting

    It's an unusual form of brain damage. Look at how he describes the way he does sums; he doesn't think about it consciously at all. He just sees two shapes morphing into another shape, which to him represents a number. He then simply recites the number out loud. On the conscious level, there is no "calculating" involved whatsoever. It's all done for him by the deep recesses of his brain, without him lifting a metaphorical finger.

    I would say that this isn't any sort of "intelligence" in any conventional sense; it's simply that his damaged brain has given him the ability to access "hidden" subroutines of the neural wiring we all have.

    For instance, it's no secret that the human brain can do maths in real-time with frightening speed. Just walking involves real-time feats of calculus that would choke a calculator. The problem is that it's all subconscious. Well, in Tammet's case, that "subroutine"-- which is supposed to be wholly subconscious-- now has a window into his conscious mind, expressed through pictures.

    This is fascinating, but arguably it's no form of intelligence. At least, not in any conventional sense of "intelligence".

    Mind you, I fully understand what it's like to be able to do something without mentally "lifting a finger". It's the way I've always been with language. I first spoke at age one, and I've been able to write and speak at an "adult" level since early childhood. My grammatical skills are quite high, but if you asked me to diagram a sentence, I'd choke. I usually can't describe why I know that a certain sentence structure is "right" or "wrong", since I can't consciously describe many of the rules of language.

    I suppose this fellow is much the same way with the pictures in his head. He's described to us how he (as in the conscious entity known as Tammet) does sums: He just sits back and his brain feeds him the answer without any conscious sort of calculation. However, he hasn't described to us how his brain does the work, which is the really interesting question.

    --
    With spending like this, exactly what are "conservatives" conserving?
  5. Re:Savantism by Monkelectric · · Score: 5, Interesting
    Obsession. I think thats the common thread through all of these things... My cousin is a "high functioning" autistic. He has a (crappy) job, and his superpower seems to be memory. He remembers *everything*, he is obsessed with movies and remembers where he bougt each one, for how much, what else he was considering buying, and sometimes even what was on the shelf next to it!

    Sometimes he'll get obsessed with a particular person -- when its me for instance, he will send me several emails *per minute* until whatever it is about him passes.

    Id hate to think of where he would be without the memory though, its clear he doesn't really understand the interactions between people, or emotions. He sent me a picture of himself with some of the budweiser girls (he met them at a promo thing), and he's got this mean scowl on his face in the picture. He was horribly excited about the whole thing and he waited days and days for the photo, but simply doesn't *know* to smile. He can *remember* the thousands of little things that his family has told him over the years, and usually remembers a short phrase that tells him what to do, "My grandfather said when somebody gets real mad the best thing to do is let them cool off for a bit and then go talk to them." And he does that thing.

    --

    Religion is a gateway psychosis. -- Dave Foley

  6. Re:What is mathematical genius by SleepyHappyDoc · · Score: 5, Interesting

    My goodness, what is good enough for you? The fact that he can do this, despite the fact he can't tell right from left, is the story. He's not the latest new processor or kernel, he's a human being with a severe disability. For a lot of disabled people, standing upright is an amazing feat (and for many it's beyond them). As a person who suffers from a severe mental disability myself, I am darn impressed.

    --
    Stasis is death. Embrace change.
  7. Re:What is mathematical genius by rkmath · · Score: 5, Interesting

    His ability to multiply numbers quickly, or test for primes quickly (not sure if he does this), or factor large numbers quickly (never does an article about a math idiot-savant talk about this - a problem that is *hard* by all known algorithms - but that is another story) does not say anything interesting for mathematics. It is interesting purely from the viewpoint of understanding how the human brain works.

    And if we are on the topic of raw computing ability - and we decide that computing ability _is_ interesting - could we *please* have them try computations in a more general number field? Could we *please* have them solve problems that we can't yet solve efficiently by any known algorithm? (And, could someone also study how fast this guy computes factorisations as a funtion of the input size? Fr instance, could we find out how fast his brain's process works - O(n) ? O(log(n))? This question could at least be answered experimentally.

  8. (Temporarily) turning people into savants by FleaPlus · · Score: 5, Interesting

    A year or two ago the New York Times had a neat article titled Savant for a Day about research by Prof. Allan Snyder. Basically, he uses a technique called transcranial magnetic stimulation (TMS) to temporarily induce savant-like symptoms in volunteers. The journalist writing the story also acted as a volunteer, and experienced greatly-increased drawing ability while the device was turned on.

    From the article:

    As remarkable as the cat-drawing lesson was, it was just a hint of Snyder's work and its implications for the study of cognition. He has used TMS dozens of times on university students, measuring its effect on their ability to draw, to proofread and to perform difficult mathematical functions like identifying prime numbers by sight. Hooked up to the machine, 40 percent of test subjects exhibited extraordinary, and newfound, mental skills. That Snyder was able to induce these remarkable feats in a controlled, repeatable experiment is more than just a great party trick; it's a breakthrough that may lead to a revolution in the way we understand the limits of our own intelligence -- and the functioning of the human brain in general.

    Snyder's work began with a curiosity about autism. Though there is little consensus about what causes this baffling -- and increasingly common -- disorder, it seems safe to say that autistic people share certain qualities: they tend to be rigid, mechanical and emotionally dissociated. They manifest what autism's great ''discoverer,'' Leo Kanner, called ''an anxiously obsessive desire for the preservation of sameness.'' And they tend to interpret information in a hyperliteral way, using ''a kind of language which does not seem intended to serve interpersonal communication.'' ...

    In a 1999 paper called ''Is Integer Arithmetic Fundamental to Mental Processing? The Mind's Secret Arithmetic,'' Snyder and D. John Mitchell considered the example of an autistic infant, whose mind ''is not concept driven. . . . In our view such a mind can tap into lower level details not readily available to introspection by normal individuals.'' These children, they wrote, seem ''to be aware of information in some raw or interim state prior to it being formed into the 'ultimate picture.''' Most astonishing, they went on, ''the mental machinery for performing lightning fast integer arithmetic calculations could be within us all.''

    And so Snyder turned to TMS, in an attempt, as he says, ''to enhance the brain by shutting off certain parts of it.''

  9. a really interesting field by urdine · · Score: 5, Interesting
    How memory works is woefully understudied. You'd think we'd know more about this by now.

    When you get down to it, though, we do most of our "thinking" in sounds or visuals. Everything else is translation. For instance, LANGUAGE is incredibly complex, but we can do it with ease since our brain has such an amazing "processing chip" for sorting sounds. Reading is simply converting things to sounds (or visuals - when you "remember" a quote you will normally either remember it by sound or by a visual memory of the words.)

    Even math is, at it's root, visual for all of us. Take 2 + 2 = 4. There is cold memorization of the result, but if you were learning math for the first time, you would break it down to:

    || + || = ||||

    ie. a visual representation, or counting fingers etc. The reason many people have so much trouble with math is they end up doing too much cold memorization - the brain remembers associatively, so this doesn't work well (but it explains why mneumonic devices DO work well). Unfortunately, that's how they teach it.

    I tend to believe that we have an amazing ability to remember sound and sight (makes sense from an evolutionary standpoint) but we're NOT hard drives and "cold memorization" just doesn't work. By knocking out some part of the brain, the brain is forced to take in math through the visual/sound process, inventing a network of logic that handles all the work in the subconscious.

  10. Re:Resume Puzzle by jd · · Score: 5, Interesting
    What do you expect from me? I'm autistic myself. (Well, the "formal" diagnosis is Aspergers, but I consider that to be on the Autistic Spectrum.)


    It's not unusual for people with Aspergers to have trouble recognizing the "correctness" of behaviour, facial expressions, etc. Sure, it's not universal, either, but it's definitely not a rarity.


    Do I care? No. But, then, I'm not built to care about things like that. This isn't an "I can't help it", because that implies it's wrong to be anything other than a highly socially-aware, socially-structured individual.


    I am Autistic. I don't see it as anything to be pitied, or even delighted in. It's just a word that describes how some aspect of the chemistry in my brain differs from the "norm". It is a description, not a definition.


    Idiot Savant is the same thing. It is just a description, no different from "hot", "yellow" or "crispy". Someone might get angry with the words, finding them offensive. That's not my problem. How you choose to understand words is entirely up to you.


    True, I could be better understood, if I spoke in a language closer to your own. But if I want to be understood by fellow autists, I go join the autistic channel on an IRC network specially set up for such folk. Here, I expect to be understood by geeks, who know how to dereference the pointers of obscure and arcane language.


    Well, they must. When karma was still counted in points on Slashdot, my score was in the thousands. Someone out there must have understood me. :) Well, at least modded me up for being humerously incomprehensible, at least. :)

    --
    It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)