Brains vs. Computers

topic posted Mon, March 24, 2008 - 3:12 PM by  VoodooChild
An informative and readable article about some of the important differences between brains and digital computers, and the mistakes we can make when we take the computer metaphor too literally.

scienceblogs.com/developin...e_a_co.php

"Although the brain-computer metaphor has served cognitive psychology well, research in cognitive neuroscience has revealed many important differences between brains and computers. Appreciating these differences may be crucial to understanding the mechanisms of neural information processing, and ultimately for the creation of artificial intelligence."
posted by:
VoodooChild
SF Bay Area
  • Re: Brains vs. Computers

    Mon, March 24, 2008 - 5:27 PM
    Hey, looks interesting. Thanks. No time at the moment to read it through, but soon.

    Generally speaking:

    It occurs to me that any metaphor which compares an essentially mechanical, designed system to a biological, evolved system is bound to be fundamentally flawed.

    A designed system is preplanned and relatively efficient. (If you do your job right.) It's teleological in a basic sense.

    An evolved system is NOT designed/preplanned but rather *selected for* in the glacially slow and inefficient manner of evolution, although given the timescales available to life on the evolutionary scale, and the "willingness" of evolution to have many, many mistakes (mutations) fail to survive as long as some do survive, the inefficiencies don't matter much. It's not teleological in the slightest bit. There's no looking ahead, just selection pressures in the present time of the organism.

    No sane designer would or could go about making systems the way evolution can afford to. Living beings sometimes *look" designed, but that's a mistake of our perceptions and our conceptual limitations.

    If we do succeed in creating AI we may decide to come up with different implementations than Nature did, to the degree that that's even possible. (I say "we" loosely; ain't gonna be me!) That's one of the open questions - how much can function be separated from structure? Early, naive AI professionals assumed, using a hardware/software metaphor, that function could be entirely separated, just as a program can be compiled for different hardwares. Now that view is widely criticized.
    • Re: Brains vs. Computers

      Mon, March 24, 2008 - 6:01 PM
      Good points, Kai!

      <"That's one of the open questions - how much can function be separated from structure? Early, naive AI professionals assumed, using a hardware/software metaphor, that function could be entirely separated, just as a program can be compiled for different hardwares.">

      Yes, the idea that function can be separated from structure has been used an an excuse for psychologists and armchair philosophers to ignore cognitive neuroscience. But by now it has become evident that *how* neural networks do what they do holds important keys to understanding why they have the capacities and limitations they have.

      Beyond the question of structure vs. function looms the larger point that we're really not all that clear on what the functions are to begin with. We have some hazy grasp of what abilities brains have, formed largely from folk psychology, perhaps tempered with a little real science, but when we really try to separate and characterize what the functions are, we run into problems immediately. A simple folk psychological category like 'memory' breaks down into sub-categories like 'long-term memory' and 'short term memory', 'declarative memory' and 'procedural memory', memory for faces, memory for numbers, memory of what happened 2 minutes ago versus 2 seconds ago. We have lesion patients with very specific memory impairments in certain functions, while the rest of their memory remains normal.

      At this point we really don't know how to draw up categories that "cut nature at its joints" and we won't know until we've done a lot more gritty anatomical work, and probably a lot more studies of damaged individuals. Not only does it seem mistaken to try to separate structure from function - even assuming that we understand the functions to begin with seems misguided.
      • Re: Brains vs. Computers

        Tue, March 25, 2008 - 4:40 AM
        Voodoo - "Yes, the idea that function can be separated from structure has been used an an excuse for psychologists and armchair philosophers to ignore cognitive neuroscience."

        Actually, I'd say it's more a case of ignoring any neuroscience/neurobiology that doesn't support a pet theory really rather than ignoring all of neuroscience (which is all somewhat ironic since the scientific method is designed to try to address confirmation bias). It hasn't been very long since our ability to look inside the brain - to scan it and measure/pinpoint activity - has remapped our understanding of both the brain and the mind. There are a lot of theories - whole academic careers in philosophy, cultural theory and psychology/psychiatry - based upon theories about nature and nurture that no longer hold water. A prime example of this (and something that's still happening because of the resistance to accepting new understandings) is in the area of brain and gender and the biological underpinnings of gender and sexual identities. Doctors who performed "corrective" gender reassignment surgeries on children under the theory that one could be "made" into one gender or the other now either have to acknowledge that their actions, even if motivated by compassion and a desire to heal, actually created more suffering by imposing an inappropriate gender on a child. Some choose to cling to old ideas for a number of reasons - there's the obvious avoidance of guilt/personal responsibility but there are also potential personal biases that come into the equation. It's not surprising that some chose denial if they're so heavily invested in a previous paradigm rather than new understandings. It's not that surprising that armchair philosophers also cling to theories based in shoddy metaphors, particularly when they're sure they've personally figured out the answer to life, the universe and everything (since most of the armchair computer/brain analogies are used to support essentially religious ideas with a bit of technology involved to make it all seem scientific and avant-garde).
    • There was an article on "biomimetics" in NAtl Geographic in Feb or April 2008 issue that
      exposed similar problems, that, for example nature uses an incredibly complex system
      (7 layers of different materials) to make a gecko's feet stick
      to a surface, that engineered systems would be much simpler but more task specific.
      • Yeah - although it's important to remember that there can be a big difference between "designed" and (merely) engineered. (I think of "engineered" as "partly designed, i.e., in terms of inner function but not necessarily in terms of use, and rarely of usability.") UI work is full of examples of this. (see anything by Don Norman; The Inmates Are Running The Asylum, by Alan Cooper; Bringing Design to Software, edited by Terry Winograd; The Trouble with Computers: Usefulness, Usability and Productivity, by Thomas K. Landauer; and for a short conceptual overview of the subject, The Software Design Manifesto, by Mitch Kapor: hci.stanford.edu/bds/1-kapor.html - old, but still relevant.)

        In a way, Nature solves function, use *and* usability simultaneously, satisficing workable, not necessarily optimal, solutions; otherwise, creatures would die. If a lion's ability to stalk prey were "designed" like Microsoft Word, it would starve to death. Though we wouldn't want to (and can't) adopt the slow evolutionary way of making systems, there is a lesson to learn from Nature in any case about balancing various pressures. (Speed vs. accuracy, etc. Many examples in biology.)
  • Re: Brains vs. Computers

    Tue, March 25, 2008 - 6:35 AM
    "A good metaphor is something even the police should keep an eye on." - G.C. Lichtenberg

    The opening quote says it all really. We like metaphors because they help make something alien and complex understandable (something that we don't yet have the knowledge and language to understand becomes at least vaguely comprehensible to us), and then people assume that they can extend the metaphor as if it was actually a practical description (and that they understand the thing being alluded to by metaphor because they understand - at least sort of - how a computer works). All very useful for "programming" (or "reprogramming") followers of faith based pseudoscience belief systems (or shearing the sheep, as grifters supposedly like to say...speaking of metaphors).
  • Re: Brains vs. Computers

    Thu, April 3, 2008 - 1:06 AM
    Metaphor is pretty much what we do to understand process... this is like that... nuclei and electrons are like solar systems... photons are like waves and particles. The brain used to be like a big pneumatic engine with pressures, drives, forces. Sometimes it is worth extending the metaphor to see if it reveals new insight, but the mapping between the metaphor and that which it stands for would have to be pretty damned isomorphic for that to hold, and generally in more complex systems we simply a) don't know that that is the case or b) as in the case of the brain, know that it isn't!

    Interesting that our metaphors for the brain tend to reflect out most complex technologies at the time (probably with a timelag to allow for popularisation of that technology).

    I quite like weather as a metaphor for neural behaviour. Changes in state at one location cause cascading changes throughout the entire system and knowledge of state at time a only gives very limited predictive ability :-) (PS... I am aware of the inherent limitations of the metaphor!! :-)

    Anyone who makes the leap from this is/might be like this, to this is that, needs their metaphor examined if you ask me ;-p
    • Re: Evolved systems vs. Computers

      Fri, April 4, 2008 - 10:10 AM
      Just to return a moment to the evolution vs. design aspect of the brain ≠ computer argument, I realize that my statement

      "No sane designer would or could go about making systems the way evolution can afford to"

      is not entirely accurate - there has in fact been an attempt to develop at least a limited architecture (albeit only within software) in an evolutionary manner, which might produce results which fit the definition of cognition as discussed here.

      That would be Karl Sims' experiments with genetic algorithms for the purpose of making "artificial life": letting loose a bunch of competing algorithms to see what comes out the other end, to put it crudely. (I've met Karl, and he's quite sane.) The "time scale" for this particular form of "evolution" can be shrunk radically when the units involved are merely nanosecond processor cycles.

      en.wikipedia.org/wiki/Karl_Sims

      But I question whether one can eliminate any trace of teleology from such a process, since the programmer who sets up the initial conditions can't help but be a designer in some sense.

      re metaphors as a foundation of conceptualizing: Taleisin, have you read George Lakoff's work?
      • Re: Evolved systems vs. Computers

        Mon, April 7, 2008 - 4:00 AM
        "re metaphors as a foundation of conceptualizing: Taleisin, have you read George Lakoff's work?"

        No, but you are the second person to mention him to me in a short period of time in relation to metaphor... would you recommend, and would you have any suggestions as to what to start with?
        • Re: Evolved systems vs. Computers

          Mon, April 7, 2008 - 5:04 AM
          "Metaphors We Live By" by Lakoff is wonderful.

          In the political realm, "Don't Think of an Elephant" is an interesting read. (In addition to his work at Berkeley, he is a co-founder and senior fellow of the Rockridge Institute, a progressive think tank.)
  • Re: Brains vs. Computers

    Sat, April 5, 2008 - 12:20 AM
    All,

    >important differences between brains and digital computers

    The ARE NO commensurate similarity between the functional architecture of biological oppositional processing of natural neuro-structures of advanced consciosness organic 'mechina' and that of advanced electro-programmatic computational engines, ie 'digital computers. Even John von Neuman, (defacto 'father' of commonly used architecture digital computers) predicted (~ca 1925) that self-replicating autonomous evolutional computational engines would never equal even the simplest of oscillon wavelet neuro-structures, which in practice has proven to be accurate. Today, the most heavily funded AI research is that of the military domain for artificial consciousness weapon-systems and that course of study takes two basic architectural directions, --fast backward propagating neuro-alorythmic-networks (software running in super-computers) and that of the highly classified artficial silicon cortex research...eg, high-density-integration artficial (synthetic) solid-state neural self-organizing 'black-box' synapsie.

    However, there are other 'synthetic' architectures which exhibit sophisticated quantum oscillon wavelet transforms, which are analog in architecture and which exhibit advanced consciousness function such as self-awareness and machine telepathy . The 'former' Soviet Union is roughly 30 years advanced over that of current Western technology. For a glimpse of this type technology, see my friend, Dr. Ron Blue's AI website and read his articles regarding simulated 'synthetic' brain processing and that of his quantum consciousness research. Dr. Blue is a quantum neuro-cyberneticist / professor of psychology resident at a Pennsulvania .edu school.

    www.u2ai.us/

    Around 2001, Ron posted in the SETI-L list, his active SETI experiment, wherein, he had attached a CCD camera staring at a photographic image of Earth, (quantum-entangled) to an analog-quantum non-digital 'computer'. Chomp on that for awhile.

    Love - tron
    employed military electronics engineer

    ================

    Brains vs. Computers
    topic posted

    Mon, March 24, 2008 - 3:12 PM by VoodooChild

    An informative and readable article about some of the important differences between brains and digital computers, and the mistakes we can make when we take the computer metaphor too literally.

    scienceblogs.com/developin...e_a_co.php

    "Although the brain-computer metaphor has served cognitive psychology well, research in cognitive neuroscience has revealed many important differences between brains and computers. Appreciating these differences may be crucial to understanding the mechanisms of neural information processing, and ultimately for the creation of artificial intelligence."
    • Re: Brains vs. Computers

      Sat, April 5, 2008 - 10:38 AM
      "the highly classified artficial silicon cortex research...eg, high-density-integration artficial (synthetic) solid-state neural self-organizing 'black-box' synapsie" - *not* so classified, I guess, if you're posting about it. Sounds like the "bio-neural gel packs" Treknobabble from ST: Voyager.

      "...and machine telepathy"...? This really undermines what's otherwise a fairly interesting post, and makes me suspicious of the other terms: "quantum oscillons", etc.

      New Age Crapola Alert! "Telepathy" is unproven in humans; how would you then attribute it to machines? (No, that's a rhetorical question: don't bother answering with another jargon-filled "I'm privy to special knowledge" post.)

      I'm quite aware of the Soviets' *parapsychological* research (a decent account in novel form is [Esalen Institute's] Michael Murphy's book An End to Ordinary History), and willing to grant they may also have done some legit AI research, but the latter may be suspect, given how much credence they gave the parapsych stuff.

      Along with "machine telepathy", it all strikes me as just as pseudo-scientific as the CIA's infamous staring-at-goats shenanigans... It just reeks of lack of rigour born of Cold War desperation, on both sides. As does the claim that they're 30 years ahead of us in AI research: 1950s missile gap, anyone? Transparently self-interested. (BTW, someone must've written a book by now on the Über-silliness of the Cold War psychic arms race. Of course, New Agers usually seize upon accounts of governments joining them in unscientific mindsets as proof of such bullshit's legitimacy. No need for mythical evil dentists implanting CIA radio transmitters in your teeth when you can imagine that they're "reading your thoughts"!) )

      I don't expect any institution with an agenda less broad than pure research, especially an agenda as narrow as the US (or any) military's (produce weapons in a relatively short time frame), and the involvement of so much money, in the context of the diminishing funds in non-affiliated academia, that people will cut ethico-scientific corners to get the research grants, not to mention tell non-scientist managers what they want to hear, to come up with truly rigorous AI/cogsci results, even leaving out the history of flirtation with New Age Crapola. The results will be tainted. Being well-funded in that context may mean quicker results, but it doesn't guarantee, and even works against, rigorous ones.
      • Re: Brains vs. Computers

        Sat, April 5, 2008 - 9:53 PM
        lol.

        how on earth do people get into the habit of talking like this?

        it's like they're parroting the overall "feel" of scientific rhetoric, but where is the "content" coming from?




        "The ARE NO commensurate similarity between the functional architecture of biological oppositional processing of natural neuro-structures of advanced consciosness organic 'mechina' and that of advanced electro-programmatic computational engines, ie 'digital computers. Even John von Neuman, (defacto 'father' of commonly used architecture digital computers) predicted (~ca 1925) that self-replicating autonomous evolutional computational engines would never equal even the simplest of oscillon wavelet neuro-structures, which in practice has proven to be accurate. Today, the most heavily funded AI research is that of the military domain for artificial consciousness weapon-systems and that course of study takes two basic architectural directions, --fast backward propagating neuro-alorythmic-networks (software running in super-computers) and that of the highly classified artficial silicon cortex research...eg, high-density-integration artficial (synthetic) solid-state neural self-organizing 'black-box' synapsie.

        However, there are other 'synthetic' architectures which exhibit sophisticated quantum oscillon wavelet transforms, which are analog in architecture and which exhibit advanced consciousness function such as self-awareness and machine telepathy . The 'former' Soviet Union is roughly 30 years advanced over that of current Western technology. For a glimpse of this type technology, see my friend, Dr. Ron Blue's AI website and read his articles regarding simulated 'synthetic' brain processing and that of his quantum consciousness research. Dr. Blue is a quantum neuro-cyberneticist / professor of psychology resident at a Pennsulvania .edu school. "
        • Re: Brains vs. Computers

          Sun, April 6, 2008 - 1:47 PM
          matthew,

          >how on earth do people get into the habit of talking like this?

          Have you perchance had oppotunity to read anypapers written by my acquaintence, Dr. Jack Sarfatti? (Formerly a guest member of the CalTech/Jet Propulsion Laboratory facilities e-mail list, --he was quietly removed as his posts were considered by NASA mgmt as being too 'far-out-there' for most of the forum's scientists.) Jack, as you may know was a child protoge' of Dr. Boehm. Jack, while a tad self-centered, speaks a completely different language, it IS English, however the similarity ends there. Jack oncentold me that when he was a "child of five, ET called me on my plastic red toy elephone, and we discussed 'things'. I asked them why they called me, they told me, '"you are the most conscious person on Earth'". Jack as you know was recently awarded a $24.6 million Dollar research grant by the US National Science Foundation, for his work in artficial-quantum-consciousness studies.

          >it's like they're parroting the overall "feel" of scientific rhetoric, but where is the "content" coming from?

          To whom do you refer to as 'they'? "Content" is sourced from the scientific quantum-physics community. Working scientists. I read the papers. Typically in JBIS or the JSE. This stuff isn't published on the sci-fi channel or in the orthodoxia dogmatic religious science magazines to which most folks proscribe. Like safe SETI, they will never detect an exo-civilization who are using radio-waves to communicate, --subluminal and, shucks, everyone knows radio only works well inside a planetary ionosphere, true?

          Love - tron

          ===================

          Re: Brains vs. Computers
          Sat, April 5, 2008 - 9:53 PM

          lol.

          how on earth do people get into the habit of talking like this?

          it's like they're parroting the overall "feel" of scientific rhetoric, but where is the "content" coming from?
          • Sarfatti

            Mon, April 7, 2008 - 11:17 AM
            A few physicists unfortunately go off the deep end, and Sarfatti is one of them. His
            later stuff is not taken seriously by mainstream physicists...
      • Re: Brains vs. Computers

        Sun, April 6, 2008 - 12:19 AM
        Hey,

        I really didn't intend to get into a pissing match, rather just share some of the lesser but well founded research being conducted. I find these items to be interestng topics and desire to share them with 'friends'. When the pain becomes too great to answer poorly researched replies, I'll just disappear, --I have better things expend my time on aside from educating people who don't know very much outside of reading popular mechanics, the wallstreet journal high-school level physics books.

        >This really undermines what's otherwise a fairly interesting post,

        I take the compliment, thank you. It is the result of 15 years of research I have been conducting in support of a technical instrumentation treatise I am preparing...a book.

        However I take it you didn't visit Dr. Blue's web site which I cited?

        Again: www.u2ai.us/


        >and makes me suspicious of the other terms: "quantum oscillons", etc.

        Don't take my word for it, research the subject yourself. Also, checkout CalTech's J. Smarrs silicon cortex efforts. Different collaborating labs around the world are developing different parts of the Human neuro-strctures. I first became aware of this research back in 1990.

        >*not* so classified, I guess, if you're posting about it.

        I can see you have little or no contact with military engineering domain and apparently you don't introspect about the missing' sciences in the engineering PD, either.

        Can you elucidate for 'us' ANY hidden sciences which aren't referenced in the PD, but would naturally be a standard course of technical evolution? I have found several, not conspiracies, just subjects of interest in military weapons development.

        >New Age Crapola Alert!

        Hmm?

        >"Telepathy" is unproven in humans;

        You been living under a rock, these past thirty years? Try to do a little research on the subject before making statements as though you are an authority on the subject. I can cite over thirty papers wriiten by well known scientists without even working hard, how about you citing one recent counter AC paper written by anyone (Ph.D. or Masters level) except by Randie the entertainment stage magician skeptic. Show me yours and I'll show you mine, heh, mine are in the PD.

        >how would you then attribute it to machines? (No, that's a rhetorical question:

        A fair question which you immediately nullify in your next statement.

        How do we think? Conscious machines are nascently telepathic, -- 'we' Humans are 'machines'. Our whole body function, it is taught in primate anthropology (one of my courses of study) is for the acquisition, storage, protection and energy management of 'fuel' for our cognitive prcessor..also known as brain.

        >don't bother answering with another jargon-filled "I'm privy to special knowledge" post.)

        I'm glad to inform you the only reason the 'privy' information I shared with this 'tribe', is 'privy' is because I got off my fat butt and worked for it
        and you didn't. Get off yours and do some studying. No you won't find this type information in Popular-Science.

        >I'm quite aware of the Soviets' *parapsychological* research (a decent account in novel form is [Esalen Institute's]

        Yeah? I know Andreie Zelkovski, (formerly of the Moscow Institute for Advanced Studies), he is currently teaching at Tenn.edu, one of numerous American schools aside from Princeton, Stanford and MIT which also have funded psionics labs. You know about this stuff right? Its based in what?

        >Michael Murphy's book An End to Ordinary History), and willing to grant they may also have done some legit AI research, but the latter may >be suspect, given how much credence they gave the parapsych stuff.

        While it is a good book in my opinion, you are citing a book minimally related to the subject, yes there is some brain psychology in brains artificial or other wise and some cursory mention of OLD AI research. However, you illustrate that you don't know about the psionic subject and challenge my post logically supported by your knowledgebase. Psionics is the science of applied artificial consciousness and the control of specific nascent behavioral functions such as 'remote' sensory AC perception. How kewl would it be if one could presponse detect local intention to trigger an IED keyed with coordinate Doppler? A typical military application.

        Here are a few of the researchers currently (in PD) conducting ongoing private and military funded psionic science. You all can see for yourselves. Drs Kenneth Ring (parapsychologist, prof of psychology, formerly of UoA.edu, funded by billionaire Robert Bigelow), Robert Jahn (former Dir of PEARLab Princeton U.edu), Eric Davis (US Navy Research Labs physicist), Dean Radin (US Military psionics researcher, physicst, Applied Cognitive Research Lab), Jessica Utts (quantum cyberneticist, physicist), Hal Puthoff (quantum physicist, Dir SRI, & under US Military DARPA contract reverse engineered Soviet psi-spy program and developed US Military Remote-Viewing program), Robert Monroe (prof psychology & former Dir of Monroe Institute, former researcher for Esalen), Roger Nelson (prof psychology, prof physics, Dir of Princeton PEARLab, dir of GCP noosphere psi sensor array & developer of MindSong telepathy / ganzfield sensor using Brian Josephson uReg) to name a few of the better known psionic scientists / researchers. There are many others, these are just off the top of my head, so-to-speak.

        >it all strikes me as just as pseudo-scientific

        Good, one can always wiggle out by saying, 'seems' or 'in my opinion' (ok to admit ignorance, one can learn out of it, stupidity is a different matter 'tho). I have found that lots of scientists are bad at science, and are just as guilty of 'pseudo-science' as any lay person, and there is an old dictum first accredited to Boehm, "seems significant change only occurs when all the experts grow old and die, leaving the halls of science empty for new thought".

        >As does the claim that they're 30 years ahead of us in AI research: 1950s missile gap, anyone?

        Actually you missed the nuance, its not so much AI that they are more advanced in, but psionics. You might ask what is psionics? Well, however illogical this sounds, it is the science of applied psychic function in machine intelligence.

        >(BTW, someone must've written a book by now on the Über-silliness of the Cold War psychic arms race.

        I haven't seen any books written elucidating the 'silliness' of the psi-spying, however, there are literally dozens of books written by retired US Military sargents, captains, majors, colonels, warrant Officers and a general. These people (you can match rank with the names, not going to do it all for you), were participating members of "Project StarGate" under auspices of DIA & NSA.

        >Of course, New Agers usually seize upon accounts of governments joining them in unscientific mindsets as proof of such bullshit's >legitimacy.

        Many folks believe that the so-called 'new-agers' are typically love-light-and-clueless, unfortunately it doesn't stop with them alone, ignorance can be found anywhere in any class, level of education or ranking, eg anywhere. Relying on logic without sufficient current knowledge-base is just as naive'.

        >No need for mythical evil dentists implanting CIA radio transmitters in your teeth when you can imagine that they're "reading your >thoughts"!) )

        Seems you've been reading too much MK-Ultra pseudo-science-babble. And telepathy isn't range-power-limited or exhibit temporal latency which transversal electromagnetic energy emission exhibit.

        <<I don't expect any institution with an agenda less broad than pure research, especially an agenda as narrow as the US (or any) military's (produce weapons in a relatively short time frame), and the involvement of so much money, in the context of the diminishing funds in non-affiliated academia, that people will cut ethico-scientific corners to get the research grants, not to mention tell non-scientist managers what they want to hear, to come up with truly rigorous AI/cogsci results, even leaving out the history of flirtation with New Age Crapola. The results will be tainted. Being well-funded in that context may mean quicker results, but it doesn't guarantee, and even works against, rigorous ones. >>

        I am hesitant to inform you however, recent world history doesn't match your speculation. Get off your easy chair and do some studying, it will surprise you, I think. I speculate, that you are a curious fellow and that you, possibly, enjoy reading interestng subject matter. Psionics is a very interesting (and dangerous, in my opinion) science. It could be a world killer, and there is much supporting data which suggests that all of the 'civilized' techinical countries are deeply engaged in military psionic application development, and soon to be available limited (embedded) commercial psionic applications.

        I have learned from personal and certain aggrevious experience that one really shouldn't make wild verbal assumptions in the 'Net PD, one never knows in whose presence one is babbling.

        Love - tron

        ======================

        Re: Brains vs. Computers
        Sat, April 5, 2008 - 10:38 AM
        "the highly classified artficial silicon cortex research...eg, high-density-integration artficial (synthetic) solid-state neural self-organizing 'black-box' synapsie" - *not* so classified, I guess, if you're posting about it. Sounds like the "bio-neural gel packs" Treknobabble from ST: Voyager.

        "...and machine telepathy"...? This really undermines what's otherwise a fairly interesting post, and makes me suspicious of the other terms: "quantum oscillons", etc.

        New Age Crapola Alert! "Telepathy" is unproven in humans; how would you then attribute it to machines? (No, that's a rhetorical question: don't bother answering with another jargon-filled "I'm privy to special knowledge" post.)

        I'm quite aware of the Soviets' *parapsychological* research (a decent account in novel form is [Esalen Institute's] Michael Murphy's book An End to Ordinary History), and willing to grant they may also have done some legit AI research, but the latter may be suspect, given how much credence they gave the parapsych stuff.

        Along with "machine telepathy", it all strikes me as just as pseudo-scientific as the CIA's infamous staring-at-goats shenanigans... It just reeks of lack of rigour born of Cold War desperation, on both sides. As does the claim that they're 30 years ahead of us in AI research: 1950s missile gap, anyone? Transparently self-interested. (BTW, someone must've written a book by now on the Über-silliness of the Cold War psychic arms race. Of course, New Agers usually seize upon accounts of governments joining them in unscientific mindsets as proof of such bullshit's legitimacy. No need for mythical evil dentists implanting CIA radio transmitters in your teeth when you can imagine that they're "reading your thoughts"!) )

        I don't expect any institution with an agenda less broad than pure research, especially an agenda as narrow as the US (or any) military's (produce weapons in a relatively short time frame), and the involvement of so much money, in the context of the diminishing funds in non-affiliated academia, that people will cut ethico-scientific corners to get the research grants, not to mention tell non-scientist managers what they want to hear, to come up with truly rigorous AI/cogsci results, even leaving out the history of flirtation with New Age Crapola. The results will be tainted. Being well-funded in that context may mean quicker results, but it doesn't guarantee, and even works against, rigorous ones.
        • Re: Brains vs. Computers

          Sun, April 6, 2008 - 12:05 PM
          looks like an unfortunate combination of some technical training and way too much acid.
          • Re: Brains vs. Computers

            Sun, April 6, 2008 - 2:24 PM
            matthew,

            >blah blah blah
            >looks like an unfortunate combination of some technical training...

            I can take a hint, you guys just wanna sit around and impress each other with 1950's science, thats fun too, its all good. I've given you all plenty of good leads, all you have to do to illuminate that growing darkness between your ears is expend time reading some current physics. An interesting place to 'hang-out' on the 'Net...is browsing the LANL.gov abstract server. Those papers are the unclassified stuff which isn't regulated or restricted by ITAR. It may give you a little glimpse of the portent of classified physics which isn't in the PD.

            >... and way too much acid.
            Organic chemistry was never my cup'o tea, --in youth was too busy grad school to play. In talking with folks, seems might have been fun!?

            Love - tron
            pedigree 'babbler' as denoted by my new 'frndz'
            • Re: Brains vs. Computers

              Sun, April 6, 2008 - 4:25 PM
              Howdy guys,

              <moderator hat on>

              Please be careful to remain more or less respectful when disagreements get heated. Thank you!

              <moderator hat off>
              • Re: Brains vs. Computers

                Fri, May 2, 2008 - 6:53 PM
                And perhaps staying on subject?

                In a long conversation with an AI enthusiast form Carnegie Mellon, two of the major points were the problems of content/structure and processor/memory. In the AI electronic realm these are separate items; in the brain they are one and the same.

                To answer the query "Give me Clint Eastwood's phone number." AI has to look to see if the number is known. The brain does not need to perform a search - even if the answer is "I forgot the number."
                • Re: Brains vs. Computers

                  Mon, May 5, 2008 - 4:07 AM
                  • This is the maximum depth. Additional responses will not be threaded.

                    Re: Brains vs. Computers

                    Tue, May 20, 2008 - 6:51 PM
                    Cool article..
                    to me the most interesting point is that there is no hardware/software distinction within the brain. Hardware without software is useless, software without hardware in some sense doesn't even exist. The brains wetware though is one and the same with the brain itself.
                    There is alot of interesting fields of AI as far as doing things that our brains are not good at....pattern recognition on huge data sets for one..the whole "waking up your Dell like Dr Frankenstein" though is just an absurdity.