Monday, September 28, 2009

The Speed of Thought

Computers are becoming faster and more powerful all the time and those improvements have been mainly due to better hardware. Future improvements, however, may well rely increasingly on better architecture and software. One reason why this seems likely is that the human brain, with its very different architecture, dramatically out performs computers in performing various tasks (such as perceiving an object in a complex visual scene). If computers are to match the brain's performance, they likely will need to exploit features of the brain's design.

In some regards the brain's hardware is far beyond that of a computer. Its "wires," for instance, would go to the moon if stretched out. That's about 10,000 times longer than all the wire in a typical microprocessor chip. Similarly, the brain has about a hundred million million synapses (the connecting junctions between neurons) which is many thousands times more than the number of transistors in the microprocessor.

But perhaps even more amazing is the brain's architecture and software. For instance, whereas computers are based on the Von Neumann architecture which separates computation and memory, the brain combines these elements in ways not well understood.

Also, whereas computer components are highly predictable (deterministic) the brain's synapses are much less predictable (probabilistic). Any given signal may have only a 20% chance of successfully crossing the synapse and this is compensated with substantial redundancy. Aside from tremendous fault tolerance, this allows the brain to rapidly increase signals by modifying the probability of synapse crossing.

Such differences provide a profoundly different, and probably more powerful, computing architecture compared to the microprocessor. As one paper explains:

Because the brain is not bound by the Von Neumann architecture, exactly what a particular neural circuit computes can be modified on the fly without reference to other circuits (as when we shift our focus of attention from one thing to another) and can also remember things for a lifetime (how to ride a bicycle).

Exploiting the brain's architecture, however, will not be easy. For instance, we need to better understand how to use parallel computers:

The problem with emulating the brain’s massive parallelism, however, is that we are not even close to being able to use the increased hardware power efficiently; how to program parallel computers is a very active subject now in computer science.

But even the parallel programming problem is only the beginning. The far more difficult problem is to divine just what information processing and computations are carried out in the brain's neural circuits:

we believe the problem is not computer power and ability to program parallel machines, but rather our nearly total ignorance about what computations are actually carried out by the brain. Our view is that computers will never equal our best abilities until we can understand the brain’s design principles and the mathematical operations employed by neural circuits well enough to build machines that incorporate them.

If you find the brain's design astonishing, consider this even more astonishing fact: evolutionists say it all just evolved. The ability of the brain to develop from scratch, and its design and operation, they say, all arose from mutant forms. Amazing.

Two thousand years ago the Epicureans explained that the organization in nature arose from nothing more than the swerving motions of atoms. How naive. Evolution has moved us far beyond such a silly idea. Religion drives science and it matters.


  1. "Religion drives science and it matters."

    Floating at the end, with no support.

  2. I think you're confusing two types of complexity here. The first is that even deterministic systems can produce complex results - take a look at cellular automata to see how starting from simple rules you end up with complex results. The brain's neurology is just as predictable and deterministic as any other mechanical system.

    It's the network, the interconnections and the weighting's that are the source of the complexity. This second source of complexity is the primary difference between a computer circuit and the brain.

    You should also look into Chaos theory, another science, that will explain how starting conditions can affect outcomes in also unpredictable ways, despite the underlying mechanics are predictable.

    Emergent systems is another set of theories you should also educate yourself with.

  3. Seems just like another "argument from incredulity". Yawn.

    The Brain, like, it's so big, and so complex, and so full of lots and lots of neurons and other important stuff! We're NEVER, ever going to understand it! The ONLY explanation must be that You-Know-Who did it! Must be!

    (It's a real shame You-Know-Who doesn't just spill the beans and tell us exactly how, when, and what He, I mean You-Know-Who did it, it sure would clear up a lot of problems! Why doesn't You-Know-Who speak up, after all I thought You-Know-Who was quite chatty - even wrote a couple of books or two, right? )

  4. On the strong AI problem, if the materialist evolutionists were right, strong Artifical Intelligence should be achievable and it once was believed to be around the corner. Yet strong AI proponents have hit a brick wall, the qualitative differences between computer software and the mind, an even exponential growth in the data crunching and data storage and data transmission capabiliites of computers to the nth degree does nothing to address these qualitiative differences between computers and mind. Some decades ago, I remember back in the '80s, the naive enthusiasm of the strong AI techies, brilliant as they were, convinced that Asimov type robot intelligences would be developed, that it was just a question of time. A non-materialist approach, dare I say it, even (unfashionable) vitalist approaches to living systems and evolution always put a damper on strong AI ever becoming a reality, but naturally among many scientists and techies, non-materialist and frankly more scientifically rigorous criticisms of strong AI were simply old fashioned religious taboos motivated by those scared stiff of science and technology supplanting the specialness of man and nature with its own technological golems.

    Yet decades later we are as far away from strong AI as we were in the 70s, and this fact appears to have been grudgingly conceded by many in the AI community by the very fact that they are no longer given to the science fiction enthusiasms of the past in their pronouncements and predictions of future developments in AI. As far as strong AI goes - non-materialists have won this one.

    The definitive book on the subject setting forth the most rigorous mulitple scientific criticisms of strong AI, and why it would never be achievable remains the famed mathematician Roger Penrose's 'The Emperor's New Mind', published about twenty years ago. His predictions remain confirmed twenty years later despite the massive advances in computing in this time period. Of course Penrose has put forward a non-materialist/dualist theory of mind with Stuart Hamerhoff, so his criticisms of strong AI as one of the latest pursuits of Science's Naked Emperors (hence the title of his book) is compatible with his dualist outlook on the mind and brain and his opposition to strong AI ever being anything other than science fiction. He has always considered simplistic materialist monist notions of the human mind and brain to be scientifically unsound.

  5. Lawrence:
    You mistaken failures as a reason to stop. Science, values failures as much as its successes because they provide constraints against paths that we need not waste time in exploring.

  6. The brain's neurology is just as predictable and deterministic as any other mechanical system.

    This view is not about accurate knowledge/science, it's about supporting a mechanistic view of things even in the face of what can be observed or known. It's curious though, because if scientific knowledge is defined mechanistically as an unbroken chain of events that extends into the past and all other knowledge is excluded then is anything really open to study and knowledge at present? At any rate, for those interested in actual knowledge based on the pursuit of the truth instead of mechanical knowledge based on a philosophy that we already know to be false I'd suggest: (The Quantum Brain by Jeffrey Satinover) as a start on the subject.

    I'll agree with your mechanical philosophy and false the false "tick tock" mythology of evolutionary Progress rooted in a Blind Watchmaker in your case. And if we seek an unbroken chain of "mechanisms" which we imagine to "explain" the brain events which cause the text that you've written here then it would seem that it is caused mainly by natural selection operating on the reproductive organs of ancient ape-like creatures. And if we adhere to Darwinism then imagining things about the past can be treated as the equivalent of scientific evidence.

    Emergent systems is another set of theories you should also educate yourself with.

    The only thing that is emerging from the past naturally are the brain events which cause you to mistakenly think that you are a sentient or intelligent being. On your own terms you are a stupid subject that is the product of blind, unintelligent events in the past and so your sight and supposed knowledge and science are an illusion. Murmuring the term science over and over again as if it is a sentient being that whispers words of knowledge in your ear doesn't change anything, the mechanical philosophy of the Blind Watchmaker that you are promoting against science and empirical evidence has left you blind, ignorant and stupid.

  7. You mistaken failures as a reason to stop. Science, values failures as much as its successes...

    Science is not a sentient being that values anything. Knowledge is known by sentient beings who value, observe things and make judgments about progress based on what they value.

    Creation myths based on imagining things about evolutionary progress in the past have little to do with actual progress. In fact, historically they have had much to do with the absence of what we now know to be progress.

    Ironically mechanistic progress is actually generally created by engineers based on principles of intelligent design. Although actual science and technology of that sort (as opposed to Darwinian creation myths rooted in a mechanistic philosophy) does not ensure progress as we know it. The same knowledge can be used to create nuclear bombs or nuclear energy. It is ironic though, that Darwinists often seem to think that imagining mythological narratives of natural or mechanistic progression often based on nothing more than imaginary events in the past somehow have something to do with actual progress or science as we know it.

  8. Statements in the article like: “every year or so we find they [computers] can do something – like optical character recognition or speech recognition – that we never imagined they would be able to do” gives credit to computers when it should be attributed to persons who developed the software and hardware that make those things possible.

    Computers that have been combined with software by persons, do arithmetic, play chess or recognize speech. Computer hardware and software without human involvement can do nothing.

    The brain can’t do these things either. Brains don’t think, or see or hear or interpret languages. We may use our brain in the process but that’s light years from saying that the brain is thinking about designing a bridge, classroom curriculum or an F-35 JSF. The fact is, persons do. We need to stop seeing ourselves as “neural equipment.”

    A person is an immaterial substance not a property-thing and physical entities are not all there are. Neurons dying every second should be an indication that thoughts (e.g., of memories), desires, sensations, beliefs and volition are states of consciousness, not states of the brain itself. It’s hard to see how a brain (much less a computer) could retain all of my childhood memories and be able to determine which are happy ones and which are painful ones. The molecular particles that make up my brain can’t, but I have no problems doing so. We can remain the same person through time precisely because our personhood is not grounded in the meat that makes up the brain. So I may use my brain but I am not my brain.

    Consciousness, as a simple uncomposed substance, is the only way “enormous quantity of data can be handled at the same time” and give us that unified information field and the ability to “extract objects from a visual scene […] so rapidly and effortlessly” (vision btw continues to exist even with eyes shut). Otherwise the visual field and all the other input would be so fragmented it would be impossible to do anything, from writing a letter to flying an F-35 JSF at low altitude in the middle of New York.

    Could consciousness operate at speeds of C² (the speed of light²) and generate energy?

    Seems that way a designer could act freely (say when implementing a design) and create matter using M=E/C², essentially freezing the energy produced into matter during the volition act. Could we be using those energy bursts as signals (as some sort of morse code) in the brain?