Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
×
Biotech Science

Scientists Begin Mapping the Brain 129

Raindance writes "A team at the University of Utah has unveiled a system to map and digitize brain tissue — thus fulfilling one of the long-standing holy grails of neuroscience and enabling for the first time in-depth analysis of how mammalian neural networks function. So far, maps for the entire retina and related neural networks have been released; no ETA on a full-brain digital reconstruction yet. (One of the lead authors hangs out here on Slashdot.)"
This discussion has been archived. No new comments can be posted.

Scientists Begin Mapping the Brain

Comments Filter:
  • Now... (Score:2, Interesting)

    by Daswolfen ( 1277224 )

    Maybe I can finally get that direct neural interface I have always wanted :)

  • I can see the inevitable result now. They'll get the whole thing mapped, petabytes of data, the position and connection of every neuron to thousands of other neurons will be known... and they'll be left to say, "now what?"

    This may be of use for diseases, but the greatest use - understanding consciousness - is still well beyond simply mapping the brain.
    • Then you can start all over again and map my brain.
    • Re: (Score:1, Interesting)

      by Anonymous Coward

      Once they have it mapped, they can warehouse the data, and come along in a few hundred years and run it in a simulation.

      Seriously.

      There's certain schools of thought that subscribe to the view that if you can save a complete state of the brain, there's no reason why, with exponentially increasing compute, you couldn't come along later and start it running again...

      If you are going to start talking about the inevitable result, you need to think about patterning consciousness, and some big philosophical issues.

      • by brian0918 ( 638904 ) <brian0918@gma[ ]com ['il.' in gap]> on Tuesday April 07, 2009 @04:27PM (#27495071)

        There's certain schools of thought that subscribe to the view that if you can save a complete state of the brain, there's no reason why, with exponentially increasing compute, you couldn't come along later and start it running again

        That would be determinism, and would require precise measurements for future reproducibility - the further into the future you go, the more precise they would have to be to remain accurate to reality. And it in no way accounts for stimuli, the very things of which one is conscious. So maybe you could restart a simulation, and it could last a short time, but it would not remain accurate long, and would not be able to handle changes in input.

        • by Cyberax ( 705495 )

          But digital revolution taught us that precise duplication of information is easy. If this information is digital in its nature.

          It might very well be that each neuron can be accurately characterized by several numbers (I'm not including connectivity information here). In this case it might be possible to accurately emulate brain.

          • Actually it was the inaccuracies in copying digital information that lead to the initial forays into chaos theory. Some people were doing a weather simulation and stopped it. They printed up the information and let it run. Later they re-entered the information and let it run again. This time the results were different. It turned out to be related to rounding. The computer was displaying only three decimal places, but was calculating much more. That miniscule difference between what was calculated and
        • No it wouldn't. He's not talking about re-running the simulation and it outputting the exact same thought patterns etc. as the original brain, he's jus talking about running a simulation of a brain from a stored model that would behave similarly to the original brain (as I read it anyway).

          What do you mean by it not "remaining accurate long" and "not handling changes in input"? If it's a brain (simulated or real) then it changes its internal state in some structured way while it handles input, that's what
      • Re: (Score:3, Insightful)

        by MrEricSir ( 398214 )

        Exponentially increasing != infinite

        This is a MASSIVE mistake that sci-fi (aka syfy) writers make all the time when talking about computer power. Just because you have a lot, and you will have a lot more later, doesn't mean you have enough to compute everything.

        • Actually speaking purely hypothetically it is possible to not only compute everything... but everything x2. Although it depends on your definition of everything. If there is somesort of subspace it's entirely possible you could put a computer in an alternate universe which makes our own look microscopic. Based on scale if there was a way to use this alter-dimensional computer you could theoretically build a system larger than our universe.

        • by Kjella ( 173770 )

          Of course not computing power to do everything. But you know that:

          1) The brain would be a lower bound - even though we don't understand it, we know it's possible. Maybe we can't do it with little pieces of silicon but at worst we could build organic computers of the same primitives our minds are made up of. Unless the brain got magic soul bits we should be able to duplicate anything it does with enough effort.

          2) We know evolution doesn't find the optimal solution, just whatever evolutionary path works. We s

          • In terms of pure bit rate of calculations, we should have commodity desktop computers capable of outperforming our own brains within a decade. This paper [transhumanist.com] (from 1997, but I doubt human brains have changed much since then) estimates our brainpower at 100 million MIPS, or 10^14 calculations per second. By comparison a Radeon HD4870 x2 graphics card is 2.4 TFlops (2.4 million MIPS at 1 flop/instruction), or roughly 1/5th of a human brain.
            • In terms of pure bit rate of calculations, we should have commodity desktop computers capable of outperforming our own brains within a decade. This paper [transhumanist.com] (from 1997, but I doubt human brains have changed much since then) estimates our brainpower at 100 million MIPS, or 10^14 calculations per second. By comparison a Radeon HD4870 x2 graphics card is 2.4 TFlops (2.4 million MIPS at 1 flop/instruction), or roughly 1/5th of a human brain.

              OK, so a brain is worth five graphics cards.

              But then, the brain only needs about 20 Watt. With five graphics cards, that would be 4 Watt per graphics card.

              • Graphics cards don't have to be able to run on instant noodles and kebabs. If I could put a human-class AI on my desk I wouldn't care if it used a kilowatt, it'd still be cheaper than paying a human to do whatever job it is that I set it to.
                • can you accurately emulate changes in bloodflow, reaction to allergins and so forth that offer some diminished work and others increased motivation? The brain is only part of the equation--is a pure "brain" sufficient to emulate an effective human worker? (I am reading That Hideous Strength at the moment and have my questions on the rationality of such a project anyway)
      • Plus every year the scanning tech will get better and better, maybe someday they will beable to do scanning to backup people regualrly like we backup our songs on our iPod.

        • Worked for the Asgard.
      • I'm amazed no-one's mentioned Greg Egan's Permutation City yet - this is pretty much the concept of the novel. What happens when you can save a whole person onto a computer and let them loose in a virtual reality? And what happens if the original of the copy is still alive and controlling the copy's ability to interact with its surroundings?

        Great novel.

    • by thepotoo ( 829391 ) <thepotoospam@yah[ ]com ['oo.' in gap]> on Tuesday April 07, 2009 @04:26PM (#27495053)

      Bullshit. A complete map of a brain of someone with and someone without gene XYZ will tell us about the role played by gene XYZ without the ethical or temporal problems associated with creating an XYZ knockout. A neural network running a simulation of a human brain would be a Turing-complete strong AI. Throw a evolutionary algorithm onto this, and you can start looking at where different types of selective breeding could take humans, or the long-term effects drugs could have on personality.

      That's off the top of my head; there'll be a million and one uses for this eventually (ever wanted to live forever inside a computer?). Besides, this is in the preliminary stages, they are still doing stuff like classifying synapses by hand. By the time this is workable, we may already know what consciousness is.

      • by MozeeToby ( 1163751 ) on Tuesday April 07, 2009 @04:39PM (#27495291)

        A neural network running a simulation of a human brain would be a Turing-complete strong AI

        I understand what you're saying here but there are a lot of non-trivial hurdles to get over, even assuming you can accurately scan and simulate the brain.

        First, the brain also includes a lot of chemical transmitters which we really don't understand the function of yet. You would have to include them in your model as well, including the ones that don't originate in the brain.

        Second, you have to interpret the simulated neuron firing into something that you can actually understand. It's pretty pointless for your brain-on-a-hard-drive to be saying 'hello' if you can't understand what it's saying. An accurate simulation of my mind would have all the neurons that control my breathing, lips, tongue, and vocal cords firing like crazy, but good luck figuring out what I'm saying.

        Third, you have to be able to supply meaningful input into the brain. You're essentially talking about submitting a consciousness (assuming your simulation is 100% accurate) to the most horrible sensory deprivation imaginable. In order for your research to be useful, you would have to supply it realistic input (including feedback based on it's output) otherwise the brain would change drastically just from that.

        • by FLoWCTRL ( 20442 )

          All good points; it's a big project. Mapping is an important, but not the final step.

          The sensory deprivation aspect raises an interesting ethical problem -- if the simulated brain is in fact 100% accurate, then wouldn't running it without normal sensory input be the same as torturing a sentient person?

          I suspect that part of creating a functional full-brain simulation will have to involve it's being embedded in a robot (or biological body) which can supply the expected sensory environment.

        • Third, you have to be able to supply meaningful input into the brain. You're essentially talking about submitting a consciousness (assuming your simulation is 100% accurate) to the most horrible sensory deprivation imaginable. In order for your research to be useful, you would have to supply it realistic input (including feedback based on it's output) otherwise the brain would change drastically just from that.

          I dunno.

          I would find this very useful because in order to do an experiment just like this but with

        • Compare with mapping the genome. Does sending a long string of A, C, G, and T into space give an alien recipient to ability to re-construct a human body, much less a person to teach them our culture? No. But affordable gene sequencing still has dozens of applications.

          Here, I'll throw out a juicy one. The pattern of connections in your brain may very well capture all of your memories. You know those people paying big bucks to be cryogenically frozen? This is a big step towards re-animating them - ev

      • Re: (Score:1, Interesting)

        by Anonymous Coward

        Problem is, you'll never live inside a computer.
        A copy of you could, of course. One can well imagine his brain mapped and 'ran' as a neural network in a computer. Nevertheless, you are still alive, and there are simply two creatures (is a human brain ran in a computer a human or a computer?) that, for a short period of time, thought exackly the same. Then, due to different stimulus and perhaps some imprecision in copying, your thought processes would slowly diverge, and the machine-you would think different

        • by FLoWCTRL ( 20442 )

          You can copy yourself to a computer and then die, but that doesnt sound so cool anymore does it.

          Actually, it's still pretty cool. :)

        • You lack imagination. Imagine a cyberspace aspect of yourself, permanently linked to the organic brain. The *mind* itself would be spread across both the organic and inorganic components. When the organic bits wore out, the consciousness might notice, cobble up another brain out of the leftover bits (using the same molecules if it was sentimental), and download the memories at deathtime (and since) back into the organic matrix and go on it's merry way. Reincarnation or Extreme medical treatment? Life after
        • Yes, the machine copy and original would quickly diverge. Also, the machine copy would not be affected by changes in hormones, blood sugar, disease, etc. But then, the you that you are today is different from the you that you were yesterday too. And if the machine me gets an unlimited internet connection and no requirement to work for a living, I'm sure he would be quite happy to download porn all day... unlike the real me. Knowing me, we'd probably quickly grow to hate each other.

          On the other hand, if tha
          • by koro666 ( 947362 )

            Without hormones, the "machine you" would feel no need to watch porn.

            Same goes for any "survival instinct" you might have, of which "love" is a derivate. So at most, watching your grandchildren would be a pointless bore.

            • Yes and no. Without hormones, your basic instinctual drives would be nowhere near as strong. But you would probably still do things out of force of habit. I take care of my family now not because I enjoy it, but because of strongly held beliefs that I should do so. Take away the instincts, and a reasoned belief about my purpose in life still remains.
      • by FLoWCTRL ( 20442 )
        A neural network running a simulation of a human brain would be a Turing-complete strong AI.

        Actually, it would need to be severely dumbed-down in order to pass the Turing Test. Electronic circuitry operates millions of times faster than the electro-chemical circuitry of the brain, plus it could have instant access to vast databases of information, and no human has that much and that accurate memory.

      • Agreed, GP is insane if he thinks this is useless. Replacement bodyparts early on we could replace eyes. Later on maybe we could extend the brain (add external storage). But this is useful in getting people to accept artificial limbs. We could build the matrix even if it is shitty at first. Understanding our brains is understanding ourselves. This never hurts! We could look at how information is routed and exploit it. We have some things like mnemonics. But with greater specifics we can get ourselves to thi
      • I would expect the fine detail of the neural pathways to be created more by environment than by genetics. Also, it is time-variant, and not just run on electrical impulses of neurons -- there are lots of chemicals that also affect brain function. You need more than just a map to simulate a brain. Although this is a good step in that direction. Like functional MRI, this is mostly just a tool to facilitate further understanding.
        • That is quite correct (especially the bit about environmental regulation, so long as we're talking about vertebrate brains), however, neurotransmitters largely just regulate when and the extent of the action potential running down the axon. It should be possible to simulate everything needed with nothing but a detailed understanding of the systems biology of the neurons; IE a map.

          I can't seem to find a cite for this in my bookmarks, but I know it's been peer-reviewed before.

      • >>A complete map of a brain of someone with and someone without gene XYZ will tell us about the role played by gene XYZ without the ethical or temporal problems...

        Uh, no. Brains are plastic entities. A person who has gone to music school will have a much larger map for dealing with processing sound than a cultureless American. (Oh, wait, I'm a cultureless American.) In science, controls are essential for telling what changing variable X does, and unless you can control a person's behavior (thus raisin

      • Just because the brain map now run on silicon doesn't mean it doesn't have rights. Treating it as a test subject instead of a person is a sure way to wake up skynet or the cylons or that thing from "I have no mouth and I must scream"

        • by koro666 ( 947362 )

          You stupid hippie. Don't you realize that "rights" are just a mere social contract and nothing else?

          They are not intrisic, and what we have created, we can destroy if we want to. That includes synthetic brains.

      • No, no, no, no. That's so short sited. You have to think of Real Uses for mapping your brain. Things like, your wife's birthday, where you put your keys last, what your girl friend said about some useless piece of information that she feels deeply about. I don't know about World Peace, but peace in my home would be like Shangrala.

    • by interkin3tic ( 1469267 ) on Tuesday April 07, 2009 @04:32PM (#27495159)

      This may be of use for diseases, but the greatest use - understanding consciousness - is still well beyond simply mapping the brain

      A variation on the sentiment "Why bother investigating this, it's beyond our understanding and is useless," which has been posed at some point to every serious scientific inquiry. "Why study fungus, you can't do anything useful with it!" is probably something Flemming heard right before he discovered penicillin.

      Fortunately it's often wrong. In this case, it seems to me that knowing the map of a brain could have some real tangible uses

      -Understanding the sequence of wiring a brain, we know some things about the order in which brain cells develop, and we know unconnected neurons die, but beyond that I'm not sure we know anything. Is there an organization to how the brain initializes itself? Could this be one thing that goes wrong in, say, autism?

      -Better understanding of the interconnectivity of different regions of the brain. Obvious uses there for dealing with lesions to the brain, if you learn from this study that one of the areas damaged is highly connected to a distant part of the brain, you might want to watch out for effects on that other part of the brain

      -Helping us understand how or if new neurons generated in adulthood integrate into the already existing, quite complicated network

      -altering something and seeing how that affects the brain map, to study the plasticity of the brain and possibly learn how to learn better

      That's just the ones I could think up, there are undoubtedly more reasons one of the authors could fill you in on, and there are probably even more uses that even they haven't thought of, that some other researcher will.

      Anyway, since when did science ever need to have a clear use in mind before we did something? If you're anything like the typical slashdotter, you don't bother asking "What good will that do" when discussions of "Let's land on the moon or mars" come up. I would argue that this is clearly more useful than that, but that doesn't matter, it's not about knowing all that we can gain from an endeavor in advance.

    • by geekoid ( 135745 )

      Nope.
      Consciouness is just another part to figure out, and mapping the brain will start to pull it together to the point where we will be able to determine consciousness, not just understand it.

      For example, most people related 'who they are' or knowing where the are in the universal as a critical part of consciousness.

      Already we know which part of the brains tells you where you end and everything else begins. a critical piece to understanding consciousness.

      We pretty much know where most the pieces the we th

    • It's easy. They'll just map the brain of developing embryos at every stage, and see how it all goes together. 2 cells, 4 cells, 8 cells -- how complicated can it get?

      This isn't rocket science.

  • So a while back I was chatting to a maths guy over coffee and we started wondering about the brain. He figured that the average number of synapses between any two purkinje [wikipedia.org] cells was just over three - now that would seem to be pretty interconnected, to the extent that cortical differentiation really isn't all that.
    • That's interesting. Computer network models seem to say that with equal-weight connections, if you have, on average, fewer than two connections per node, the network quickly reaches a static state ("death"), whereas if there are more than an average of two connections per node, the network thrashes around in chaos. At 2 per node, "complex" behavior emerges.

      We know that many cells are massively more connected in the cerebrum (on the order of 10^4 connections per node), but obviously dif

      • Well, AFAIK there are theories that the brain is a chaotic system. But chaos is not mere randomness; you have things like strange attractors there.

        • Right... in a way. Chris Langton's phrase, popularized by Waldrop, is the edge of chaos http://www.romanpoet.org/223/Langton.EdgeOfChaos.pdf [romanpoet.org], meaning that the place in dynamical phase space where things get interesting is in Wolfram's Class IV of complexity. It's not really _chaos_ as that term is used by most people (or, rather, "chaotic" may perhaps be a less precise term than should be used for the systems which show long-term cohesion of some sort, while still being flexible enough to be "interesting").

  • In 500 mm turn right at the Hippocampus, and you have reached your destination.

    Seriously, this is pretty cool. Genomics and protenomics are cool, but you have to map the brain to understand it...
    • by robkill ( 259732 )

      In 500 mm turn right at the Hippocampus, and you have reached your destination.

      I knew I should have taken that left turn at the pituitary. (said in a Bugs Bunny voice of course)

  • It looks like this method, as others, divide the brain into slices for mapping. Doesn't the slicing of soft tissue make it difficult or impossible to determine the exact point of connection between slices? I imagine it like dividing a plate of spaghetti, and then trying to determine which noodles were connected to which just by looking at their new positions, whereas their previous positions were determined, in part, by the connections themselves, and the slicing process has introduced entropy.

    Are the bra

    • Re: (Score:3, Informative)

      I suspect that information is lost during the slicing; but that there are really no better alternatives, so you take the tradeoff, even if you don't like it.

      Optical microscopy, in combination with appropriately chosen staining techniques, offers good resolution and a large number of viewing choices(depending on what you stain for) at relatively low cost; but isn't all that useful for anything but thin slices or surfaces. Getting results equivalent to optical microscopy out of big solid lumps, by MRI or Xr
    • Your observation is really no different than e.g. numeric integration vs symbolic. The whole reason digital systems work is because with a high enough sample rate, they are a good enough approximation for the purpose. You start with something continuous (a blob of matter - a brain) and want a symbolic representation (a neural network) with discrete elements, so at some point, yes, you have to start discretizing things, but it doesn't mean you are discarding any of the desired structure.
  • Didn't google already do that? With streetview? I'm sure I heard a car drive through my cortex the other night.

  • One word. (Score:3, Insightful)

    by Sybert42 ( 1309493 ) * on Tuesday April 07, 2009 @05:09PM (#27495781) Journal

    Singularity

  • Comment removed (Score:3, Insightful)

    by account_deleted ( 4530225 ) on Tuesday April 07, 2009 @05:24PM (#27496003)
    Comment removed based on user account deletion
    • I have a question for the neuroscientists however... what's so critically important about this work, to demand the enormous resources being sunk into this?

      That's a puzzling question. What would you rather have the neuroscientists working on, if not neuroscience?

    • Come on, now you're talking like a politician.

      I have a question for the neuroscientists however... what's so critically important about this work, to demand the enormous resources being sunk into this?

      You seriously can't think of anything useful that could come out of this? Possibilities for artificial intelligence, medicine, origin of life.........if you can't see that, you really ARE like a politician.

      And exactly how much do you think is being spent on this, anyway? What kind of 'enormous resources?' The entire budget for the national science foundation is $6 billion. A much better allocation of resources than bailing out incompetent bankers, if you a

      • Comment removed based on user account deletion
        • HAHA cancer gets i'm sure thousands of times the budget this gets. Giving it another extra .01% of funding wont change shit. This also has more exciting implications than cancer immunity.
        • Why not do both?
        • by rts008 ( 812749 ) on Tuesday April 07, 2009 @10:47PM (#27498959) Journal

          It seems everyone is too busy being flippant or dismissive towards you, so I will attempt to give you an answer.

          1. Neuroscience is considered important in the medical field. The nervous system is important for almost every other bodily function if for no other reason than it is the control and communication system of the body.

          2. the brain is one of the least understood of our organs, and arguably, one of the more important ones. Anything we can learn about it helps a lot at this early stage.

          3. Scientists/med researchers are people too, and since people have diverse interests and passions, so do the scientists. They gravitate to fields that hold an interest/importance to them. Freedom of choice, etc....there is no pool of researchers and scientists that are assigned fields of study by some group/organization.

          4. Because it's there. This is a central drive inherent in humans...to 'boldly go where no man has gone before', and can be attributed to many reasons to do so.
          Curiosity, exploration for the thrill or ego(I was first!!!!), need to contribute/help, revenge/righting a perceived wrong...

          Yeah, this is all just basic stuff, but can be easily overlooked or taken for granted. On one side(funding) you have special interests, on the other you have researchers with special interests. They have a habit of finding each other.

          It gets to be easy to sit back and wonder 'why this and not that' from the outside. Maybe this will help:
          (I'm not asking for an answer, just giving food for thought, but it's okay to answer!)

          What do you do for a living? What got you into that, and why is it important to you? If not important to you(other than to make a living), then what would you want/like to do? Apply those answers to your question, and you may have an answer.(not trying to be an ass, but it's not an 'easy/one answer' question.

          On a more personal note, I'm all for neuroscience to blast forward. At my age, my mind is in the best shape of any of my other 'parts', and I would love to be able to go into a body shop and have my brain transferred(by some means) to a new body. :-)

        • "there are many equivalent and roughly as-challenging problem areas out there, and there seems to be a great deal of work being done on neuroscience and related fields, with breakthroughs announced on a daily basis. What's driving the interest, time and money being devoted into neuroscience, as opposed to, say, cancer immunology.... ?

          A key fact about science is that the ultimate utility of any line of work is impossible to predict. Sometimes the fields that people expect to yield big dividends have modes

    • I have a question for the neuroscientists however... what's so critically important about this work, to demand the enormous resources being sunk into this?

      Well, I have a brain, as do most people I know - I'm going to go out on a limb and say there may be some long-term benefits for medicine.

    • by HTH NE1 ( 675604 )

      I have a question for the neuroscientists however... what's so critically important about this work, to demand the enormous resources being sunk into this?

      Because we're there.

    • I have a question for the neuroscientists however... what's so critically important about this work, to demand the enormous resources being sunk into this?

      Hey, Skynet ain't gonna get built by itself.

    • I have a question for the neuroscientists however... what's so critically important about this work, to demand the enormous resources being sunk into this?

      I think the people who could answer that the best are those with or have family or friends with a neurodegenerative disorder [wikipedia.org].

      AKA Alzheimer's or Parkison's disease.

  • Do people remember all of the full "usable" gene mappings and correlations that were going to come out of the human genome project? We had smart genes, violence genes, political genes, blah blah blah, even though most of the genes were filler/junk lines of "code", which basically just meant we didn't know WHAT the hell they did. There was even a span of time when bio-med and genetic engineering firms were scrambling to patent various genes and their effects. It all turned out to be mostly a big joke, and
  • Dammit! Now my conductive tinfoil hat will backfire!
  • The scientists working on this problem suddenly realized they had found something! All were shocked at what their results were showing... Finally, one scientist, almost afraid to face the implications of this great discovery, broke the silence and spoke to his fellow researchers...

    "This is... astounding... ...Are you thinking what I'm thinking?"

    To which one of his colleagues prompty replied:

    "I think so, but where are we going to find a duck and a rubber hose at this late hour?"

  • This brain mapping would be great if they mapped an autistic brain and compared it to a 'normal' brain.

    I want to know if there is a difference in 'architecture' between autistic and normal brains.
  • then the much bigger task of mapping the interconnections can begin. That should not be more than one or two orders of magnitude harder.
  • Map this!
  • I'd hope that the "one of the lead authors" (isn't there only one lead author?) would correct some of the misstatements in the summary.

    The system digitizes the map, not the brain.

    Isn't "digitizes" a gratuitous inclusion by now?

    Mapping can produce a static picture of a dynamic system. The brain works by changing its structure. Thus, the "functioning" they can discern will be extremely limited. Actual functioning could be mapped by remapping the same brain after learning and such, if the analysis is sufficien

Math is like love -- a simple idea but it can get complicated. -- R. Drabek

Working...