Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Does your brain outpower a computer?

  1. Aug 16, 2006 #1
    I've been hearing that our very brains have more computing power than the highest of computers. IS this true. I mean what do we d o that's all that complex with out brains. I know consciously we don't have very much math ability but unconsciously we do. By the way, which mental activities use the most working of our mind?

    I also hear that there are more conversations from the left and right side of your brain than there are from North America to Europe per day. I wonder if we use much power as we sleep. Seeing that computers have millions and millions of transistors on one microchip and process extremely hard programs I don't see what we do that's so fantastic with out minds.
  2. jcsd
  3. Aug 17, 2006 #2


    User Avatar
    Gold Member

    Line, I don't know whether or not you've ever looked into neurology from the physical rather than psychological standpoint. The brain consists of something on the order of 200,000,000,000 neurons, each interfaced with up to a thousand others. When you start working out how many data points that can entail, you'll realize that the most complex artificial computer imaginable can't come close to replicating our thought processes. As a simple answer to your question, you thought to ask the question; a computer wouldn't because it can't think.
  4. Aug 17, 2006 #3


    User Avatar
    Science Advisor

    Well, but how much of the brain is actually devoted to thinking and how much is devoted to sensory or physical processing? In my opinion there is no particular reason why a powerful computer, programmed correctly, couldn't think.
    http://www.transhumanist.com/volume1/moravec.htm is an interesting article on the subject.
  5. Aug 17, 2006 #4


    User Avatar

    Staff: Mentor

    Computers/processors must be programmed whereas the brain seems to be able to develop and understanding without external programming, although brains require sources of sensory stimulation, which originate externally.

    Brains have intelligence, computers have 'artificial intelligence', or AI. Humans just haven't figured out how the brain develops intelligence, or the ability to think, so that the process can be rendered in a microchip.

    In general, what brains cannot do is perform complex computations which computers can do, but only with instructions.

  6. Aug 17, 2006 #5


    User Avatar
    Gold Member

    The problem with that question is that it isn't answerable at this stage of scientific understanding. No one knows what thinking is, or how we do it. There is intense research underway to determine that, but it's not tremendously successful. Bear in mind as well that the external signal processing is part of the thinking process. The preliminary 'filtering' is subconscious, but anything of interest is instantly flagged and sent to the cerebral cortex for attention. That's still 'thought', as evidenced by the fact that you can programme it to respond to something like a baby whimpering that would ordinarily be unnoticed, and to ignore something like a train going by.
    I agree that at some point a computer will be able to think, possibly even the same way that a human does if analogue circuits are used, but not until we know how to create it.
  7. Aug 17, 2006 #6


    User Avatar
    Science Advisor

    It's true that we are probably many years from a thinking computer, but we do already have programs that can learn--albeit often in simple ways--on their own. People also must be "programmed" to learn, by their genes.
    Last edited: Aug 17, 2006
  8. Aug 20, 2006 #7
    Thinking requires you to be conscious. COmputers far as I know can never be conscious. What we do is think, what computers do is proccess. Do we even use any brainpower while thinking> If it were just neurons it would be like the autonomic part of our brain, the part that controls heart function,temperature,and breathing while we are sleep. But there's something differnt in the conscious part, something that makes us...............aware. WHat I'm trying to ilude to is ,is it our brain that's doing the thinking when we think or our minds? Science ahsn't proved the existence of the consciousness. If you're still perplexed it's the same concept in QUantam Pgysics as the spirit or soul is in religion.

    Iw we had no consciouss it would be just liek we are when we're asleep...unconsciouss. BUt we are aware, alert and awake. I'm not sure if the nrain does any thinking at all or just translates our thoughts into electrical signals. Then come sthe hard question...........where are our thoughts,where do they originate from? Are they juststarted off by neurons or some consciousness linked to our in our brains.

    One sign that something is consciouss is freewill. We humans can do whatever we want to. SUre we've got instinct and natures calls but we don't have to do something if we don't want to. COmputers however live on a set plan, a set of rules called a program, no matter what it does it has to follow them. It's not thinking but just following rules and instructions. And just liek what happens to a thought where does sensory input go. We see it go from our eyes,ears, nose,tonguemand skin up our nerves to our brain centers. But where does it go after that? Does it just stop at the last neuron or go to a unexplored place. The place called the consciousness.

    COnsciousness can never be on a computer. Until we we can do that computers will never truely think. So for calculating it may be the unconsciouss part that's fdoing it.
  9. Aug 20, 2006 #8


    User Avatar
    Gold Member

    Unfortunately, the end of this is going to be a philisophical discussion rather than a scientific one. The opinion of how thought works, rather than the explanation, is purely subjective. I do not believe in anything that hints of supernaturalism, including a mind that's separate from the brain. To me, thinking is strictly an electrochemical process. How the neurons interact to produce consciousness is what current research is concentrating on, and that's the approach than my interest lies in.
  10. Aug 20, 2006 #9


    User Avatar

    my $0.02:

    this device called a "transputer" is more like a neuron connected to a bunch of other similar devices and represents more closely the architecture of the brain. not sure they'll ever make transputers small enough to connect 100 billion of them with 100 trillion different connections (to put it on par with the human brain), but i kinda doubt it.


    anyway, to compare crudely, it's like we have a trillion or so bytes of memory but a slow CPU (about 40 Hz in our prime, about 17 years old) in our head. compared to silicon computers with far less memory and a far faster CPU. so, as a consequence, the silicon computer will likely evalutate heuristic functions that require a lot of computations and relatively less memory, and we just "look it up". it's sorta like the silicon guys use a sorta Taylor or Maclauren series to evaluate [itex] \sin(x) [/itex] while our fleshy computer will just do a table lookup. slow CPUs with lotsa memory will optimally do table lookup over crunching numbers, but fast CPUs with much less memory will crunch the numbers instead of wasting a lot of space on a bunch of huge tables.

    so the 17 year olds mind works a lot faster (about 40 simple compare-and-conditional-branch instructions per second in our stream of consciousness, there are a lot more parallel instructions per second happening a the neural level) but has a smaller database than the middle-aged person with a slower CPU but lots of experience that can still be looked up pretty fast.

    about consciousness, it's been speculated (i dunno if it's true or not) that consciousness in higher species was evolved as a consequence of the need (or advantage) of integrating all of the different stimuli together with all of the different motor control. individuals of a proto-species that would better integrate all of this had a natural advantage regarding Natural Selection than individuals that had only reflexive reactions to stimuli. integrating all of this made for better decisions than mindlessly doing knee-jerk reactions and those protoplasms that could integrate and see a bigger picture of their existence and environment had a better chance of surviving long enough to do the horizontal bop and sire those who eventually sired people who could dream up microprocessors, space shuttles, and Barbie dolls. this "integration" of all of the various stimuli and motor control might very well be what our consciousness is.

    we are already designing computers to do self-modifying code, a crude form of cognition. might be scary someday, perhaps in this century, if one of those AI machines does enough self-modifying code and becomes sentient. it might never happen. i dunno.
    Last edited: Aug 20, 2006
  11. Aug 20, 2006 #10
    There are many estimates of the brain's computational capacity based on work investigating individual regions. For example, Hans Moravec, after calculating the computational capacity of the retina and its systems, estimates the computational capacity of the entire brain at 10^14 instructions per second. Lloyd Watts, after working with the ears, also came to an estimate of 10^14 calculations per second. After studying the cerebellum, researchers from the University of Texas estimated 10^15 cps.

    Ray Kurzweil outlines these studies in his latest book, The Singularity is Near, and opts for a more "conservative" estimate of 10^16 cps.

    Right now, IBM's Blue Gene/L provides 3.6 x 10^14 cps, greater than the lowest estimates above.

    Of course, this is all numbers. It doesn't take into account the functional differences between a computer and a brain. However, in terms of raw power, brains and computers appear evenly matched at this point in time.

    Savor it while it lasts.
  12. Aug 20, 2006 #11


    User Avatar
    Science Advisor
    Homework Helper

    Massive parallel beats clock spead.
    Still waiting to see a computer run up and down stairs.

    That and a few other things before I worry too much about consciousness.
  13. Aug 21, 2006 #12
    To the original post, Line: IMO our brain has computers beat in processing power due to the massive amount of connections neurons make. I think our brain makes up for a slow processing speed (if you will bear with the non technical explanations, that's all I have, sorry) by having these millions of parallel connections.
    I think in computer parlance (as a programming friend of mine put it) it means that while we have a slower clock speed, we have a deeper decision tree.
  14. Aug 21, 2006 #13
    BUt what do we use it all for?
  15. Aug 22, 2006 #14
    THat's still being researched by scientists from multiple fields. It's an ongoing discovery.
    BTW, Don't forget to check out the Biology, Mind and Brain Chemistry forum on this site.
    You can also get into some pretty heavy stuff in the Metaphysics & Epistemology section under the Philosophy forum too, if that's more of what you're steering towards which is my guess from post #7.
  16. Aug 22, 2006 #15
    Simple robotic contraptions have been able to do that for a while now. A slinky already accomplishes half of that task. Bipedal robots are jogging already, so emulating the energy inefficient human form is not far off either.

    Machines were flying upstairs before they were ever walking them. :smile:

    You shouldn't ever have to worry. Any fear of the fact that artificial life will be conscious is simply a prejudice.
    Last edited: Aug 22, 2006
  17. Aug 22, 2006 #16


    User Avatar
    Science Advisor
    Homework Helper

    I rather ment to limit my comment to bipedal robots.:tongue:

    And I might point out that the human is hardly inefficient. It's only been in the last year or two that any walking machine has been built that even remotely approaches human efficiency.

    Nothing to do with fear. Just don't see it happening anytime soon.
  18. Aug 23, 2006 #17


    User Avatar
    Gold Member

    Although it's not bipedal, here's an interesting walking robot. I can't help thinking of it as two guys in a horse costume. Patience is required; it's a slow site, and the sound is very irritating. :biggrin:

  19. Jan 13, 2009 #18


    User Avatar

    Our brains are far more powerful than any computer. No computer is more powerful than a Turing machine (a sequential symbol manipulating device [imagine a head that reads states or symbols off a tape ).

    Quantum Computers can break this law/thesis, but they haven't yet, because quantum computers have been simulated by a Turing Machine.

    Now I am not sure if our brains are as powerful as a Turing Machine, but I seriously doubt you can simulate a human brain on a Turing Machine. I personally think it will be an incomputable problem, because I do not think our brain works in the same way a Turing machine does.

    Take note: A computer is stupid, no computer ever thought of gravity. A computer is just a Mathematical Machine (look up combinatorics). No computer ever innovated or aspired. It isn't capable of individual thought (yet).

    Infact there is no such thing as Strong A.I (human intelligence or exceeded), I believe currently there is only Weak ( I have yet to research this though). Weak or Real A.I. according to my lecturer is just an approximation technique to run exponential time program (such as Traveling salesman problem) optimized to polynomial time.

    To directly answer your question: With all this technology, information, biology and mathemathics, we still as a human race do not fully understand how the brain really and truly functions. Therefore at present it is impossible to simulate on any machine. Therefore my honest opinion is that our Brain is more powerful and I base my assumption on the fact that we cannot simulate our brain on a Turing Machine. Also Strong A.I. doesn't exist.
    Computers cannot think, therefore our intelligence is far greater and if so I suppose more powerful.

    Might I also add a Researcher is at present is trying to decode the Brain language, I read this on the BBC and watched a video (Human V2 by horizon). Maybe you will find this interesting and might answer your questions better.
  20. Jan 13, 2009 #19
    Does your brain overpower a computer?
    This is the question.

    It has been stated that the brain does huge amounts of processing, well we cannot even imagine to what extent it's real power is. Do we even have enough brainpower to calculate this?

    If the brain overpowers a computer would we need computers? so this question might be the answer.

    The bottom line is: Computers overpower the brain in some situations, and of course the brain overpowers computers in other situations.

    if we compare the evolution time involved (millions vs hundreds), we will soon understand it's only a matter of time until computers overpower brain.
    Last edited: Jan 13, 2009
  21. Jan 13, 2009 #20


    User Avatar

    Since, according to the link below, the human brian consumes 40W or less I would say yes, a computer can out power the brain. But as one would suspect, it really depends on the particular computer and particular brain.

Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?

Similar Discussions: Does your brain outpower a computer?
  1. Brain signals (Replies: 1)