«   |   »

Information storage in the brain

by Neil Rickert

The problem of information storage is raised by Cornelius Hunter in a post at UD and at his own blog.  I’m not quite sure why Cornelius posted that.  He often posts arguments for ID or arguments critical of evolution.  But he fails to connect this particular post with his ideas on evolution and ID.  But never mind.  It’s something to comment on, because my response illustrates my disagreement with the conventional wisdom.

Cornelius poses the issue with: “The problem is that how the brain could store information long-term has been something of a mystery.”

My reaction – as best I can tell, the brain doesn’t store information at all.  So there is no mystery.

Suppose I hear a tornado alert on the radio.  I might react by becoming more alert to the weather conditions outside.  That can be thought of as reconfiguring things.  And that reconfiguration can be said to be a kind of memory.  But, as best I can tell, there would never be a need to actually store the received information (the alert).

The idea of storing information comes from the way that we use computers.  Perhaps it is implicit in the conventional view that knowledge is justified true belief.  I disagree with that view of knowledge, and I disagree with the information processing view of what the brain is doing.  My example of how we react to a tornado warning illustrates why I disagree.

Posted on March 26, 2012 at 17:15 (UTC) in cognition, philosophy   |  RSS feed

Tags:

65 Comments to “Information storage in the brain”

    
  1. Jeffrey Shallit
    March 26, 2012 at 22:07 (UTC)

    I think this is absurd, but I’ll bite. If my brain doesn’t store information, then how can I recite the digits of pi to 30 decimal places?

    Like

    • Neil Rickert
      March 27, 2012 at 01:56 (UTC)

      I did not say that there isn’t memory. My point is that the way we get memory is not at all like the storing of information that we see in information processing systems.

      It probably required some practice for you to get that 30 decimal places of \pi right.

      Like

  2. Jeffrey Shallit
    April 5, 2012 at 10:50 (UTC)

    If that was your point, maybe you should just have said that.

    But why is it even interesting. Of course human memory is different from computer memory, since human memory is biologically based. That doesn’t mean that principles from computer science won’t be useful in thinking about human memory. And there’s also no reason to think that human memory is fundamentally different from computer memory, in the sense that it gives humans some extra computational edge over digital computers.

    Like

  3. Lage
    April 10, 2012 at 15:19 (UTC)

    I agree with Jeffrey on this one. Neil, you did specifically say that as far as you can tell “…the brain doesn’t store information at all.” It clearly does, or we wouldn’t ever have a background reference of temporal reality. Our short-term memory is basically a bio/chemical-based RAM, with our long-term memory more like a bio/chemical-based ROM (if I have to analogize it to a computer information storage system), but it is still a type of memory nevertheless. If our DNA stores information even more efficiently than a computer (a complex quaternary nucleo-based system rather than a less dense binary system), there’s no reason why our brains can’t do the same thing using action potential inhibitors and catalysts, and many other chemical forms of information storage. Every memory that you have is all a result of some type of information storage utilizing neurons/chemicals. The difference is that our memory has been created through trial and error over years of slow-paced evolution through the mechanisms of natural selection, the baldwin effect and others, whereas computer memory was “intelligently designed” by engineers starting from simple punch card logic all the way to solid-state transistors.

    We must remember that there are no original ideas anymore. Every idea, including “computer information storage” is just a modified version of something previously present (a mimetic representation of human memory which existed long before). I believe there’s a misconception that memory or “information storage” are terms exclusive to computers when they are anything but. Computers we’re a way of perfecting large amounts of memory (information) storage with an easy input / output system that the brain couldn’t accomplish as easily. When I memorized over 530 digits of PI, I memorized them in strings of 10 digits (that was my ‘bit’ method of organizing such a large string). It may not be memorized with the fecundity, fidelity, and longevity that a computer could accomplish, but it is done with a system created through natural selection, rather than direct engineering (like a computer).

    Like

    • Neil Rickert
      April 10, 2012 at 15:39 (UTC)

      It clearly does, or we wouldn’t ever have a background reference of temporal reality.

      That does not follow at all.

      To be more specific, I disagree with “the brain is a computer” idea. There’s no evidence for it as best I can tell. By now we have 60 years of research into to AI, and there is very little to show for it.

      Like

      • Lage
        April 10, 2012 at 15:56 (UTC)

        Then explain how we can have a background reference of temporal reality without information storage? I’ll explain my reasoning here so you can reply. Think of “me” right now. That is, the being that you think of as “Neil”. What runs through your mind? At the very least, it is a truncated feeling of your entire “being” which is constantly changing temporally (compare the “me” at 15 years old versus today). What you may think of as “me” is the current set of ideas, principles, likes, dislikes, philosophy, present physical and emotional state (pain, comfort, etc.). Every one of those things that you lump together are all bits of information that you’ve stored over time through individual experiences. They all form the “background reference of temporal reality” (I decided to quote myself there). Without that information, you would be a blank slate, and you would not have any idea of what “me” is. Thus, information storage is the simplest explanation for your background reference of your temporal reality. Does that make sense? Or is there something I need to explain in more detail?
        -Peace & Love

        Like

        • Neil Rickert
          April 10, 2012 at 18:45 (UTC)

          Every one of those things that you lump together are all bits of information that you’ve stored over time through individual experiences.

          Sure. And Adam and Eve really did eat the fruit of the magic tree as advised by the talking snake. And Noah really did build a boat to save the entire biosphere from a catastrophic world wide flood.

          That we can dream up plausible sounding stories does not make them true. We are supposed to be finding out how it really works.

          Thus, information storage is the simplest explanation for your background reference of your temporal reality.

          If you are a designer, thinking about how to design, maybe that sounds the simplest. If you are a creationist, you might be thinking about design. But I prefer to think about what might have evolved, and from that point of view it seems unlikely.

          Like

          • Lage
            April 10, 2012 at 20:09 (UTC)

            I’m not sure what your getting at with your “Adam and Eve” fable. “That we can dream up plausible sounding stories does not make them true”. Are you joking here? You really think that Adam and Eve is a plausible story? I don’t think so, as it makes more sense when we look at the discoveries in Biology, Geology, etc., that all life has evolved over time and human beings are included in that chain of evolution. On the other hand, my reasoning for how your background reference for temporal reality exists IS plausible, as it uses common sense, and evidence within the fields of psychology, etc. You still haven’t yet given an argument to refute my position, so let’s make this a two-way exchange.

            As for information storage being “the simplest explanation for your background reference of your temporal reality”, let’s forget that for a moment and go back to the main point I made regarding information and this aforementioned reference. The way you think right now is a result of numerous pieces of information coalescing into one subjective reality. If you suffer brain trauma or have a piece of the brain removed, scientists have shown that the feeling of “me” changes. Obviously this is because information has changed — plain and simple. Either it has reorganized, been reinterpreted, or it has been lost. You wouldn’t feel as you do with a temporal (time-dependent) reality if it wasn’t for all of the experiences that effected your psyche or view of “me” in some way. It’s really not that difficult to see why.

            Like

          • Lage
            April 10, 2012 at 20:55 (UTC)

            When I said “Thus, information storage is the simplest explanation for your background reference of your temporal reality”, you said “I prefer to think about what might have evolved, and from that point of view it seems unlikely…”.

            I disagree. In order for something to evolve through the mechanism of natural selection, only one requirement is needed: overall the traits of an organism as a whole have to be GOOD ENOUGH for that organism to survive long enough to reproduce. This means that they don’t ALL have to be good or even advantageous, as long as there are at least some traits that allow the organism to survive long enough to reproduce within the current environment.

            How can something evolve to store information? Well there’s a huge example of this that I mentioned earlier — an example that was the main requirement for life itself to evolve, DNA. DNA is nothing but stored information with every nucleo-base (whether it’s Cytosine, Guanine, Adenine, or Thymine/Uracil) having a conjugate base to form a base pair, and eventually sequencing to form amino acids, and eventually protein growth. This is all because DNA is a set of instructions for the growth and life of an organism. It is the clearest example of information storage in all of nature and it was accomplished through natural selection. Memory in the brain went through the same rigorous scrutiny.

            It’s easy to realize that brains that can hold onto information (more specifically, useful information), are going to be the most likely to survive because that organism is able to learn new advantageous behaviors/reponses more easily than an organism with a brain that can’t store information well (i.e. smaller brains, dysfunctional brains, etc.). An organism that couldn’t store any information would have a poor chance of survival unless it had some other defense mechanism (being poisonous, etc.) to counter that disadvantage. This is the case for some plants, etc., that have no brains. They survive through other means. On the other hand, once we enter the realm of brain-containing organisms, it’s completely different. Now behaviors can be learned that are no longer instinctual (like the sucking reflex of a newborn baby), rather we can solve problems like how to crack open a coconut, learn which foods are poisonous and to avoid them in the future by STORING the memory (what the dangerous food looked like, it’s taste, it’s effect, etc.). It’s very advantageous for a brain to learn to store and transmit information, and really that is a brain’s main purpose. If it is a brain’s main purpose to store and transmit information (unless you can share an alternative main fundamental purpose), then it would also make sense that this very storage is also responsible for our background reference of temporal reality (i.e. “me” or “consciousness”) as it is a subjective experience caused by the brain (according to the medical community).

            Like

          • Neil Rickert
            April 10, 2012 at 21:40 (UTC)

            You still haven’t yet given an argument to refute my position, so let’s make this a two-way exchange.

            Your claim, in effect, is that a store of meaningless abstract symbols provides a background reference to reality. I’m not sure what there is to refute.

            Like

          • Lage
            April 10, 2012 at 22:02 (UTC)

            Neil,

            No, you have missed the point here entirely. It isn’t just the storage of “meaningless abstract symbols” that provide a background reference to reality. It’s the combination of brain wiring which already exists (the blank slate due to genetic instructions which has been wired well over many years of evolution), and the information received (which the brain gives context to) as well as HOW the information is stored (i.e. how information relates to other information to build even more complex context and intra/inter-information data –), which creates this background reference to reality. We have consciousness. We are not just file cabinets that store meaningless symbols. Part of the equation is WHAT those symbols are stored in, HOW they are stored, and how parts of that information RELATE to other parts (i.e. pattern recognition).
            With all of these parts fitting together, now the symbols have context and meaning and thus provide that background reference. Without the information from our past stored, there is no background reference other than a pre-wired blank slate with no experience (i.e. sub-brain consciousness, such as cellular consciousness, etc.). In that last case, with no information stored, there would be no symbols which is what all of our thought and reality is based on.

            “You still haven’t yet given an argument to refute my position, so let’s make this a two-way exchange.”

            -Peace and Love

            Like

          • Neil Rickert
            April 10, 2012 at 22:19 (UTC)

            Well there’s a huge example of this that I mentioned earlier — an example that was the main requirement for life itself to evolve, DNA. DNA is nothing but stored information with every nucleo-base (whether it’s Cytosine, Guanine, Adenine, or Thymine/Uracil) having a conjugate base to form a base pair, and eventually sequencing to form amino acids, and eventually protein growth.

            Personally, I don’t think of DNA as stored information. Rather, I think of it as a template for making proteins.

            J.J. Gibson’s theory of perception is based on the idea of templates (he calls them transducers) for recognizing things in the world. And Gibson also argued against the idea of stored information. Making and tuning a template is different from the kind of store operation that we do with computers.

            Like

          • Lage
            April 11, 2012 at 14:17 (UTC)

            Regardless of what you may think, DNA is definitely stored information. It is something that has fecundity, fidelity and longevity (just like any other information). If you look at the DNA over time (assuming it is around body temperature), it does not change much at all other than minor mutations. It is preserved (stored) with very stable chemical bonds. The DNA may act like a template for protein formation, but it is no different than a set of instructions (information) to assemble a new office chair. The instructions are your template for putting together the chair, but it doesn’t mean that the instructions fail to be information. You see?

            -Peace and love

            Like

          • Neil Rickert
            April 10, 2012 at 22:35 (UTC)

            We are not just file cabinets that store meaningless symbols.

            But that is pretty much what information storage (as done in computers) can achieve.

            I’m glad you recognize that there is something missing. But shouldn’t we try to fill in the gaps of what is missing, before we jump to the conclusion that there is any actual information storage going on?

            Like

          • Lage
            April 11, 2012 at 14:22 (UTC)

            I don’t think that we can get around the information storage property. I think it is inevitable because information storage and transmission is such a basic requirement for matter and motion. I do agree with you, in that I think that there are many things still unknown, like the MECHANISM of storage (different from computers, etc., which people are constantly trying to compare to) as well as the interplay between some bits of information with others. There is definitely a lot more going on than what we see inside a computer (in terms of memory storage), which is why computers are such a bad analogy (at least it’s a very incomplete analogy).

            Like

      • Lage
        April 10, 2012 at 16:32 (UTC)

        Neil,
        Regarding your comment on “the brain is a computer”, the brain is certainly a computer. The misconception is that if anyone says that “the brain is a computer”, all they need to mean to be correct, is to show that the brain in fact computes. It does compute. Every time you “make a decision” your brain is computing your likes, dislikes, current emotional states, risk vs. reward, future consequences (accomplished by the pre-frontal cortex), etc., to make a decision. Every time you “crunch numbers” in your head, you are computing. So yes, one attribute of the brain is that it is in fact a computer. The issue is that people mistakenly think that this claim implies that the brain is EXACTLY like a “traditional” computer in every way. No, it does not have binary code (but it does have a way to code for storing and retreiving memories), and no it does not use solid state transistors, etc. (instead it uses action potentials to act as transistor gates at the brain’s synapses, with sodium and potassium ions used instead of electrons). This makes no difference in being able to call the brain a type of computer. The brain is definitely a computer in many ways (just not all). Another obvious difference is that the brain does not store memories with the fecundity, fidelity, longevity and speed of retrieval that a traditional computer can (if kept maintained), but the trade off is that the brain stores much more complex multi-dimensional information that a traditional computer can’t do. One synapse is like a micro-processor in itself, and it is doing what it’s doing based on primarily DNA instructions, secondarily, the presence of certain enzymes and chemicals that allow protein formation and breakdown, and tertiarily the action potential instructions given by drug receptors’ status on those synapses (allowing the flow of Chlorine, Sodium, Potassium, etc.) to make the next action possible. There are many more analogies that liken the brain to a type of computer; it has sensory inputs, as well as information/action outputs (nervous system, somatosensory, etc.).

        The trouble with AI research is that they are trying to mimic something that took millions of years to develop through careful trial and error (evolution/selection) — as well, they are trying to make their AI technologies silicon-based rather than carbon-based (like our brains are). It would be more of a shock if they were able to do this without serious difficulties.

        To say that: Due to the fact that AI researchers haven’t been able to make a traditional computer mimic the brain to some arbitrary standard (i.e. Turing Test, etc.) within 60 years, thus the brain must not be a computer is ridiculous and doesn’t follow. One synapse can be the equivalent of over 1000 transistors, and we have trillions of synapses! On top of that, our brains are very good at parallel processing and pattern recognition, due to the exotic interconnections present between neurons. We simply don’t have the technology to mimic such a feat, but it doesn’t mean we won’t be able to in the future. The point is, what AI experts are trying to accomplish is very difficult because they are trying to make something from a top-down approach. They don’t know how the brain works completely, yet they are trying to make a replica of it using a binary system computer. It’s a semi-blind reverse engineering project, and that’s the way it goes with that method. If we ever know how a brain functions down to the atomic level and have the ability to manipulate atoms at the level and speed required, we’ll have no problem replicating a human brain once that’s possible.

        -Peace and Love

        Like

        • Neil Rickert
          April 10, 2012 at 18:48 (UTC)

          Every time you “make a decision” your brain is computing your likes, dislikes, current emotional states, risk vs. reward, future consequences (accomplished by the pre-frontal cortex), etc., to make a decision.

          That seems extraordinarily unlikely to me.

          Like

          • Lage
            April 10, 2012 at 20:18 (UTC)

            Neil,

            As extraordinarily unlikely as it may seem, what other explanations do you have? Just think of something as simple as making a choice between going to work or playin’ hookey (which I don’t believe is a real choice since we dont’ have classical free will, but that’s another debate for another time). Now ask yourself these questions:
            What goes through your mind when making this choice?
            What existing opinions/thoughts may influence this choice (consciously or unconsciously)?
            What physically is happening in your brain as you are trying to make the choice?

            Let’s look at the first question, that is, what goes through your mind when making this choice?
            Well, I’m guessing you will tabulate a risk vs. reward list (pro’s and con’s of course). You may realize that you can relax if you stay home, but you may lose out on wages needed for future expenses or risk losing your job. On the other hand, you may consider that your supervisor won’t care (depending on what you’ve learned from their expectations), etc. The point is, unless you go through life not calculating anything (completely random, which is not feasible at all to survive), then you are calculating your decisions based on previously acquired information. If not, then provide an alternative explanation. I really don’t see why you think this is extraordinarily unlikely. If anything, NOT calculating decisions based on current acquired opinions and experiences is unlikely because it is not valuable for survival and natural selection would have filtered out those types of creatures long ago. We make decisions based on conditioning and learned experiences, not through some form of magic or random guessing — plain and simple.

            Like

          • Neil Rickert
            April 10, 2012 at 21:52 (UTC)

            As extraordinarily unlikely as it may seem, what other explanations do you have?

            That’s the kind of argument I hear from creationists. “I can’t think of anything else, so it must be true.”

            Just think of something as simple as making a choice between going to work or playin’ hookey (which I don’t believe is a real choice since we dont’ have classical free will, but that’s another debate for another time).

            I never played hookey, so I don’t have experience with that particular question.

            The ability to make decisions is important. I agree with that. You seem to be jumping to the conclusion that making decisions requires computation. I just did some computations, using a spread sheet. After I had the result of the computations, I still had to make the decision. The computation doesn’t make a decision for me.

            Like

          • Lage
            April 10, 2012 at 22:12 (UTC)

            Neil,

            You said:
            “That’s the kind of argument I hear from creationists. “I can’t think of anything else, so it must be true.” ”

            Have you ever heard of Occam’s razor? It is a good guide when you have little to work with. In the case of creationists, that has nothing to do with this argument. I suggest staying on topic, rather than changing the subject or attempting to invalidate my position by comparing mey argument to creationist arguments (of all things) — it’s ridiculous.
            I could say the same thing about evolution: “It seems extraordinarily unlikely to me, creationism is much more likely”.
            Which is why it’s better to analyze our arguments or present alternatives. If no alternatives exist, then Occam’s razor applies.

            And yes, making decisions does require computation. Give me one specific example when it doesn’t require it (and it can’t be a random choice to a decision, otherwise there is no decision made, rather it is made for you by chance).
            I guarantee you that every non-random decision requires computation. Humor me with an example and I’ll demonstrate how computation was used.

            -Peace and Love

            Like

          • Neil Rickert
            April 10, 2012 at 22:40 (UTC)

            And yes, making decisions does require computation. Give me one specific example when it doesn’t require it (and it can’t be a random choice to a decision, otherwise there is no decision made, rather it is made for you by chance).

            It seems to me that a thermostat (the old fashioned analog type) is making decisions to turn the heat on or off. But it is not doing any computation. It is just doing measurement.

            Like

          • Lage
            April 11, 2012 at 14:25 (UTC)

            “It seems to me that a thermostat (the old fashioned analog type) is making decisions to turn the heat on or off. But it is not doing any computation. It is just doing measurement.”

            It is computing, because it has to perform a choice (turn on or off) after taking the measurement. If it was only doing measurement, then it would be a thermometer, not a thermostat. A thermostat uses logic and makes a decision. IF the temperature is too low, THEN turn on the heat source. IF the temperature is too high, THEN turn off the heat source. It is definitely making a decision there, because it has two choice at any moment in time.

            Like

          • Lage
            April 11, 2012 at 14:32 (UTC)

            Also to be clear, “computing” only means to determine an amount or number. The thermostat is determining what the temperature is (albeit through a mechanical method in the case of analog), which satisfies the definition. Even if an analog thermostat were not technically computing (if we further refined the definition for our purpose), how does it compare to the brain? Are you saying that you think that your brain is analogous to an analog thermostat? If that’s what you’re saying, how so? Regardless, the argument that spurred this was because you said “You seem to be jumping to the conclusion that making decisions requires computation.” When did I jump to this conclusion?

            Like

  4. Lage
    April 10, 2012 at 15:49 (UTC)

    I want to add that generally speaking, “information” is a broad term that really encompasses everything. Getting all the way down to thermodynamics, one can argue than a quantum state is the smallest non-arbitrary bit for information storage. Every thing that exists can be described by sets of quantums states, including memory, memes, objects, anything. The tough part is actually determining those quantum states, as well as drawing an arbitrary subjective boundary to describe a “memory” and nothing else. The human body and mind are both subjective, even structurally so, as we are constantly exchanging atoms with the environment around us every second of our lives. Calling myself “Lage” is just a truncated way of describing an arbitrary set of several octillion atoms. There really is no “me”, or “you”, as objectivity doesn’t exist; rather we are all one with the universe and a part of a single whole. Once we accept this, then we can choose convenient labels for things including “information”, and we can describe it in various ways and at different levels of complexity (numbers, letters, codes, DNA, on/off, ideas, etc.). So, to return to my main point — due to the broadness in how we define “information”, it is most definitely being stored in our brains in many different ways (including storing episodic, semantic, iconic, echoic, etc., information through the use of memory). Neural network patterns are formed in the brain, and these paths change and/or become weakened/reinforced for memories’ storage. Think of episodic memory (some moment in your past) which you may remember like “it was yesterday”. You are clearly retreiving something that is stored in your “cabinet upstairs” — what else could it be other than information?

    Like

    • Neil Rickert
      April 10, 2012 at 18:51 (UTC)

      I want to add that generally speaking, “information” is a broad term that really encompasses everything.

      If everything is information, then the word “information” becomes meaningless.

      I prefer a more restricted meaning for “information”, roughly along the lines suggested by Shannon and Weaver.

      Like

      • Lage
        April 10, 2012 at 20:25 (UTC)

        It doesn’t lose it’s primary meaning however, and that is broad but still meaningful. It just means that our existence and experience is describable by basic fundamental units (bits) of information. How we choose to define information further is where it gains more meaning. The reason why I even mentioned this was to illustrate that the concept of “information” is not limited to a bunch of 1’s and 0’s in a traditional computer. Your initial comment stating that “the brain doesn’t store information at all” inclined me to explain this, as you must believe that information has some exclusive meaning outside the realm of brain mechanics. I mentioned the various uses of the term “information” to prove that this is simply not the case. Information, even if we confine the definition to be non-fundamental, could mean: ideas, thoughts, feelings, sensations, images, sounds, etc. All of those things are forms of information, and if we assume that our brains are the sources of our thoughts then the brain does store that information (otherwise, they are stored non-locally, and human brains are just receivers which is much less likely based on what we have learned regarding electromagnetic wave theory).

        Like

        • Neil Rickert
          April 10, 2012 at 22:09 (UTC)

          It just means that our existence and experience is describable by basic fundamental units (bits) of information.

          But it isn’t. All that is describable by a bit of information, is whether an abstract value is a binary 0 or a binary 1.

          How we choose to define information further is where it gains more meaning.

          But maybe when the meaning part is worked out, it will turn out that there is no need to store those bits in long term memory. Until you understand how the brain deals with issues of meaning, you have no basis for thinking about whether it is useful to store information.

          The reason why I even mentioned this was to illustrate that the concept of “information” is not limited to a bunch of 1′s and 0′s in a traditional computer.

          If you are depending on a different meaning of “information” then what does it even mean to say that information is stored? When the auto-mechanic gives my car a tune-up, I would not describe that as storing information. When the piano tuner adjusts the piano, I would not call that storing information. It seems to me that those kinds of tuning operations are closer to what the brain is doing than are information storing operations.

          Like

          • Lage
            April 11, 2012 at 14:12 (UTC)

            Neil,

            “But it isn’t. All that is describable by a bit of information, is whether an abstract value is a binary 0 or a binary 1.”

            Unfortunately, everything in the universe is ultimately abstract due to our epistemological indeterminism — that is, our lack of ability to truly know anything objectively about a system as we are a part of that very system and the system (if we consider it one object) can not know itself, thus we only have subjectivity in our universe. The closest that we can get to objectivity is abstract values and that is something we have to accept.
            Once we accept this reality, if we want to continue to discuss theory or otherwise, we have to accept that everything is ultimately abstract. It’s the degree of abstractness that we arbitrarily grade based on conventions, theory, semantics, etc. So yes, while they may be abstract values, there isn’t anything that exists that isn’t abstract due to epistemological as well as ontological indeterminism.
            Once this indeterminacy is accepted, we can now describe things that we experience in many ways (as best we can, since we can’t describe things fully due to the aforementioned indeterminism) even if using abstract values (since there aren’t any truly non-abstract values that exist).

            Also, the value doesn’t have to be a 0 or a 1, that is only the limit of a binary system. Who says that fundamental values have to be one of two possible values? You are making that assumption with no justification for doing so, other than that traditional computers, and some philosophical conventions employ this system (with 0/1 corresponding to Off/On states).
            DNA, for example, has a quaternary system for storing genetic information (4 nucleobases used), so there are an infinite number of systems to describe information (not just binary, albeit binary being the most commonly used as it is the simplest even though it is the least dense).
            In terms of quantum mechanics, to make matters more complicated, a quantum bit of information (Qubit) can be a 0, or a 1, or a superposition of both 0 AND 1 (three possible states, even with binary values: 0 and 1).

            “But maybe when the meaning part is worked out, it will turn out that there is no need to store those bits in long term memory. Until you understand how the brain deals with issues of meaning, you have no basis for thinking about whether it is useful to store information.”

            That may be true that it will turn out that there is no need to store bits. But first of all, your initial comment: “as best I can tell, the brain doesn’t store information at all” implies that you have strong reason to believe that it doesn’t store information. What reason is that exactly? I haven’t heard any rationale or theory from you as to how memory works if no information storage is used. What is your proposed alternative?
            Second, just because there is no need for something doesn’t mean that it’s not the way it is. We seem to have no need for an appendix, yet we have one anyways (sometimes ready to burst). It doesn’t mean that it didn’t serve a function in the past, nor does it mean that it shouldn’t exist. “No need” for something allows people to consciously engineer things better, but our brains weren’t consciously engineered — rather they are a product of evolution through natural selection. There are plenty of things that are the way they are, even though they don’t need to be that way (alternative possibilities to allow some property to exist). Natural selection is very sloppy because it doesn’t consciously engineer species or their characteristics, it just happens due to selection pressures in an ever-changing environment.
            Everytime a neural network is modified due to experience (synapses created, re-routed, etc.), that is stored information — period. It is a change of state, due to incoming information/stimuli that cause it to change in some notable way. Learning something over and over, repeats those neural patterns in the brain (neurologists have studies this and demonstrated this to be fact), further reinforcing electrical communication through them, thus storing the experience in a physically evident way.

            “If you are depending on a different meaning of “information” then what does it even mean to say that information is stored? When the auto-mechanic gives my car a tune-up, I would not describe that as storing information. When the piano tuner adjusts the piano, I would not call that storing information. It seems to me that those kinds of tuning operations are closer to what the brain is doing than are information storing operations.”

            The point is that information isn’t restricted to the narrow definition that you’ve given it. When you store information, it means that the information in one way or another (even if coded differently) is able to be retrieved later. Do you not have any memories in your brain/mind right now? If so, then you have stored information. It does not matter if the mechanism of storage is counterintuitive or not EXACTLY like a computer storing a 0 or a 1 in a flash-ram transistor. The brain is complicated and has it’s own MECHANISMS of storage which is the detail that we are uncertain of.
            Now let me analyze your examples regarding the auto-mechanic and piano tuners.

            When the auto-mechanic gives the engine a tune-up (I happen to be a mechanic among other trades myself), what does he do exactly?
            He may replace spark plug wires, swap out spark plugs, change the oil & air filter — and what exactly does this mean regarding information? Well, I will explain. When the car had a specific response upon starting the engine before the tune-up (you communicate to the controls to start the engine) and it starts and runs sluggishly, it could be because there have been interruptions in the communication (due to bad spark plug wires) and the signal to fire the cylinder is retarded or distorted (original repeating information has been lost/changed into new information which is not as useful for the clean operation of the engine). The arrangement of molecules of the spark plug, filter, etc., have stored states in them (memory/information) in order to run properly. After the second law of thermodynamics comes into play, the parts break down and no longer are storing the original molecular arrangements/states (information), for they have been changed due to decomposition, etc. Replacing the parts, is equivalent to restoring original information in the parts. The reason why you are having trouble understanding this is because you are probably limiting your idea of information storage to be like a file in a folder or cabinet, which is an analogy that is used with computers, etc., but is very limited. Think instead of individual atoms arranged/stored in a lattice. If you go back to that lattice, you will see those atoms (“information” in the case of car parts) stored and retained in roughly the same way you left them. If one had the technology to grab a specific atom, it will be stored in the place you last left it (a high probability if it is a high melting point solid such as a metal car part).
            When you tune a piano, then same thing occurs. The original information (arrangement of atoms, tension in the strings, etc.) has been changed due to strain in the piano wires over time. By re-tightening them to a specific tension, you are modifying the strings to return the information back to where it once was (in this case, wire tension / atomic lattice stress/strain is the primary information restored). After restoring the information, the string will become harmonic again and perform it’s function appropriately.

            -Peace and love

            Like

      • Lage
        April 10, 2012 at 21:45 (UTC)

        Neil,

        What ideas have I presented that do not fit into the Shannon/Weaver model of information theory if we want to limit the definition to such a model? What the heck, why don’t we discuss it in terms of Shannon et al. Shannon defines information as a “purely quantitative measure of communicative exchanges”, and that is exactly what the brain stores. In our case, we could quantify information in a number of different ways (including binary code), but we can quantify it in various ways nevertheless (this is arbitrary as long as we stick to rules in how we quantify it. Second, we can look at the term “communicative exchanges” — which simply means that something has to be exchanged (information, potentially coded) in order to communicate something else (an idea or whatever we want to be represented by the information after decoding it). The brain involves several different levels of “communicative exchanges” all the way from sensory input coming from the environment and experience (with the brain on the receiving end) to motor output outwards into the environment (with the brain on the sending end). Our channel of communication within the body is the neurons as well as the nervous system itself (sympathetic and parasympathetic), etc. We not only have these individual interneuronal information exchanges, but we also have clustered packets of decoded information (which we perceive as thoughts, ideas, etc.) that our brain allows us to experience and communicate vocally or otherwise to other individuals who likewise can accept it, have their brain break it down into pertinent components, recode into neuronal pulses and eventually store into dynamic memory or otherwise through the neuronal path once again.

        Like

  5. Neil Rickert
    April 11, 2012 at 20:56 (UTC)

    I’m just adding comments at the end, as we have exceeded the max nesting depth.

    Unfortunately, everything in the universe is ultimately abstract due to our epistemological indeterminism …

    That’s not what people normally mean by “abstract.”

    Also, the value doesn’t have to be a 0 or a 1, that is only the limit of a binary system.

    The point I was making doesn’t depend on the number of possible values.

    The point is that information isn’t restricted to the narrow definition that you’ve given it.

    The point is that what I wrote in the post, and what you disagreed with, was with respect to a narrow conception of “information.” If you want a broader conception, then you aren’t commenting on what I posted and I don’t have any particular interest in debating it.

    Like

    • Lage
      April 11, 2012 at 21:43 (UTC)

      Your initial claim “as best I can tell, the brain doesn’t store information at all.” is what we are mainly discussing here. You seem to disagree with me on how to define information. So I’m asking you how do you define “information”, and how do you define “storage”. Once I know your esoteric definition, then we can see what fits and what doesn’t regarding the brain. Most if not all physicists would agree with my point of view regarding information storage. I’m interested in your precise definitions so we can determine for sure what fits and what doesn’t. What’s the difference between storing a 0 or a 1 on a piece of a hard drive to represent a bit of information vs. storing a Boron atom or a Phosphorous atom in an atomic lattice location to represent a bit of information? If both bits are retreivable, storable, and can be used to represent some information, then what else are you missing here to satisfy your definition (since you seem to accept that information is stored in a computer but not in any ways I’ve described — I admit my engineering mentality may be hard to fully understand at times but my principles still hold).

      Like

      • Neil Rickert
        April 11, 2012 at 23:46 (UTC)

        So I’m asking you how do you define “information”, and how do you define “storage”.

        I am going by Shannon’s definition, which seems the most useful.

        We understand how information is used in computers, and how it is stored. Ordinary human speech and writing seems to be described well as Shannon information. And a case can be made that the content of perception is Shannon information, and includes speech and writing. It is in short term memory while being used. My point was that I don’t see any likelihood that the direct content of perception is ever stored in some kind of long term memory.

        Like

        • Lage
          April 12, 2012 at 13:01 (UTC)

          Again, you say that you are using Shannon’s definition, but what is your definition? I want to know how you interpret Shannon’s definition. What are the basic points that you think are most important when defining “information”?

          Like

          • Neil Rickert
            April 12, 2012 at 15:33 (UTC)

            What are the basic points that you think are most important when defining “information”?

            That information consists of meaningless abstract symbols which are represented in some physical form.

            The point of “meaningless” here is that any meaning comes from the interpretation of the symbols, and is not part of the symbols themselves.

            Like

          • Lage
            April 12, 2012 at 16:58 (UTC)

            “That information consists of meaningless abstract symbols which are represented in some physical form. The point of “meaningless” here is that any meaning comes from the interpretation of the symbols, and is not part of the symbols themselves.”

            So in the case of DNA, for it to contain information, it has to contain “meaningless abstract symbols which are represented in some physical form”. Well it does. It contains “genes”, which is an artificial construct (symbol) that we humans have invented to describe something in a useful way (a set of nucleobases that are a certain length and in a particular order). The symbols need to be represented in some physical form (if you look at the DNA strand, what we symbolize as “genes” are represented physically by the physical nucleobases. The sets of nucleobases don’t really have any genetic property on their own — only when they are in a particular arrangement along with many environmental requirements, hardware, etc., do they demonstrate their familiar properties. We interpret the genetic code a particular way (phenotypes, etc.) which is otherwise meaningless if we just look at a list of nucleobases (CGTAGCTCTCGATCGATCA, etc.) in the DNA strand, just as a bunch of 1’s and 0’s is meaningless (101101010101110001, etc.). If DNA (and it’s information) is sitting on it’s own in a jar, it won’t do anything and it has no meaning other than being a bunch of atoms in a particular arrangement. If those stored 1’s and 0’s are stored in a computer, they don’t do anything. We choose to ascribe meaning to those 1’s and 0’s arbitrarily, just as we choose to say that this genetic sequence is a “gene” for brown eyes (there’s a lot more to it than that with regards to environmental requirements to make this possible, but this is a convenient way to symbolize them). Those 1’s and 0’s only mean what we say they mean because that is our convention. To say that a particular genetic sequence “means” something is also arbitrary in how we choose to label it and the ribosomes used to “translate” it.
            If my 1’s and 0’s (electrical on/off states) are in the correct environment (power available along with other hardware) and in a particular arrangement, a particular action will happen. Likewise, if my “genes” are in the correct environment (sugar and proteins available in aqueous surroundings) and arrangement (genetic sequence), as well as a translating ribosome, a particular action will happen.
            We can analogize the DNA with binary code (one was invented/programmed by humans, and the other through natural selection). The DNA on its own is meaningless with regards to it’s genetic information/potential locked inside. The binary code is likewise the same. The fuel and proteins/enzymes/ribosomes surrounding the DNA allow the information locked away inside code to be realized a certain way (translated and/or give it meaning). They are the translators and the hardware. If we had different types of proteins/ribosomes in existence (if we engineered our own), the DNA could be interpreted/translated differently and it would not be fixed — so it is not abstract.
            So DNA appears to fit your criteria above.

            Like

          • Neil Rickert
            April 12, 2012 at 17:05 (UTC)

            So in the case of DNA, for it to contain information, it has to contain “meaningless abstract symbols which are represented in some physical form”. Well it does. It contains “genes”, which is an artificial construct (symbol) that we humans have invented to describe something in a useful way (a set of nucleobases that are a certain length and in a particular order).

            The way we talk about DNA, including the way we describe genes as sequences of letters, is Shannon information. But it does not follow that the DNA itself is Shannon information.

            Perhaps I seem pedantic. I am trying to make precise distinctions for reasons of clarity. When we fail to make those distinctions, we confuse ourselves and go astray.

            Like

          • Lage
            April 12, 2012 at 17:32 (UTC)

            Neil,

            I believe that we can still consider the DNA to contain Shannon information (it does not need to BE Shannon information, it only needs to contain it), so the main issue is defining what we consider to be the source. It isn’t some body or being necessarily (as in the case of speech, or a letter you mail), rather it is nature herself that is the sender. Natural selection could even be considered the origin of the information. Likewise, it is nature’s ribosomes and the laws of physics that are the current translators that give the seemingly meaningless genes context and meaning and thus a particular translation. We are no more qualified to send or originate a message of Shannon information than nature herself. When you look at the causal chain of events starting from the big bang — every piece of information (even what you consider to be Shannon information) isn’t a result of some single source (humans are not “causa sui” to use the latin term), rather all information results from the previous causal chain (infinite number of events and requirements) that set it into existence and transmission. Once we realize that this makes information (including the Shannon variety) point-sourceless, then previous hurdles that we thought existed preventing us from labeling DNA as a compartment of stored information goes away. One could go so far as to say that Shannon information is untraceable. If I send a morse code message to you that says “The sky is blue”, we could say that I’m the originator of the message, but am I? What if someone told me that the sky was blue, and I was just a propogating medium (channel) to pass that information onto you? What if anything we ever learn and share with anyone else, is just propagation (channeling), then the source could go back ad infinitum. You (as well as many others I’m sure) seem to hold onto a convention that the most recent “being” to transmit a message is the source, but really they are a channel of propagation. If the “original” (previous) message sent to me was “The sky is blue and blue jays are blue”, and I tell you “The sky is blue”, we have lost some information, but for all you know, all of the information has been preserved because you don’t know the totality of it’s past course. My intention may be to only tell you about the sky being blue, but does the Shannon information model specify how to deal with these difficulties in definition or can anyone arbitrarily choose what they want to define as the “sender”, even if they are only propagators (which fits the definition of a channel as well)? It seems overly arbitrary to me and it seems that nothing fits perfectly into this model.

            Like

          • Neil Rickert
            April 12, 2012 at 17:47 (UTC)

            I believe that we can still consider the DNA to contain Shannon information (it does not need to BE Shannon information, it only needs to contain it), so the main issue is defining what we consider to be the source.

            Sigh!

            Shannon information is a human artifact. It is not a natural kind. It is not something that we find in the natural world.

            We can model something as Shannon information. That would be an example of mathematical modeling, and is something that we would do if we found it useful. But to do such modeling, we would need to be precise about the mathematical model we were using, and how we connected it to things that exist in the world.

            Like

          • Lage
            April 12, 2012 at 17:48 (UTC)

            I do agree with you on that point.

            Like

        • Lage
          April 12, 2012 at 13:45 (UTC)

          “My point was that I don’t see any likelihood that the direct content of perception is ever stored in some kind of long term memory.”

          So if I perceive a moment where my mother spanks me for being a “naughty boy”, then 5 years later I remember the incident just like I’m playing a movie in my mind. How is this not long term memory retrieval. 5 years is certainly outside the realm of short term memory, so…?

          Like

          • Neil Rickert
            April 12, 2012 at 15:36 (UTC)

            So if I perceive a moment where my mother spanks me for being a “naughty boy”, then 5 years later I remember the incident just like I’m playing a movie in my mind.

            I already discussed that kind of situation in the 4th paragraph of the post that we are commenting on.

            Like

          • Lage
            April 12, 2012 at 17:07 (UTC)

            You didn’t really discuss the type of example I mentioned. All you did was briefly state that you think that the brain “reconfiguring things” is a mechanism for responding to some tornado alert (even though long term memory of what to do in that situation is why you know to be more alert). So let’s discuss my specific example so you can illustrate it better.

            If I remember a moment where my mother spanks me for being a “naughty boy”, then 5 years later I remember the incident just like I’m playing a movie in my mind (perhaps because it had such a profound impact on me), how is this not considered long term memory retrieval? Where did the memory come from if it wasn’t stored? Did something send me the set of images remotely/telepathically — for me to see that episode?
            Also, 5 years is certainly outside the realm of short term memory, so explain your reasoning with my specific example. There is no need for me to continually think about that incident everyday for 5 years, in fact it may be lost or forgotten (difficulty retreiving it) until something may job my memory or if I search deep enough, I may recover the 5 year old memory. Explain how this is not an example of long term memory retrieval?

            Like

          • Neil Rickert
            April 12, 2012 at 17:38 (UTC)

            If I remember a moment where my mother spanks me for being a “naughty boy”, then 5 years later I remember the incident just like I’m playing a movie in my mind (perhaps because it had such a profound impact on me), how is this not considered long term memory retrieval?

            However, you are not really playing a movie in your mind.

            The concensus from psychology appears to be that memories are reconstructions, and not anything actually recalled from a memory store.

            There was a period when psychology was investigating recovered memories, and psychologists were helping people recover long lost memories. It turned into a scandal. Some of these long lost memories, though quite vivid, turned out to be memories of events that had never occured. People went to jail on false charged based on those recovered memories.

            Like

          • Lage
            April 12, 2012 at 17:44 (UTC)

            Yes, I am a psychology major and I’m aware of the possibility that some memories are inaccurate (just as bits may go bad in a computer and provide inaccurate file data). My point is that there are plenty of memories (even if you want to call them reconstructions) that we all remember that are longer than short term memory time frames (months, years, etc.). Even if they were reconstructions, they’d have to have something to instruct how to reconstruct them. I remember my 5th birthday, and having a pony ride in front of my house and I remember certain details vividly. My parents siblings, and even photographs confirm that my memory is very accurate. So how are those memories reconstructed if there isn’t something stored? If there is nothing stored, then there would be no instructions on how to reconstruct the memory. A computer reconstructs stored data as well. There is never 100% fidelity and people deal with these problems all the time whether it’s lost data, incomplete data, or data that’s been scrambled. So a computer deals with the same issue and yet you are still comfortable with calling it long term storage. Please explain. How are those memories (especially accurate ones) reconstructed if there isn’t something stored?

            Like

          • Lage
            April 12, 2012 at 17:46 (UTC)

            Also, you are incorrect about the Psychology-concensus. There is no concensus that no storage mechanisms are used in memory formation and retrieval. There IS a consensus in Psychology that memories aren’t localized like a bit in a computer, rather they are non-locally stored in neural networks. FYI.

            Like

          • Neil Rickert
            April 12, 2012 at 17:54 (UTC)

            Also, you are incorrect about the Psychology-concensus. There is no concensus that no storage mechanisms are used in memory formation and retrieval.

            You are misreading what I said.

            My comment about consensus was that memories people experience are reconstructions. I am well aware that many psychologists do believe that there is some sort of data storage and retrieval going on.

            Like

          • Lage
            April 12, 2012 at 18:04 (UTC)

            “I am well aware that many psychologists do believe that there is some sort of data storage and retrieval going on.” Then why did you post this? Are you just saying that you’re in disagreement with the psychological concensus that the brain in fact stores information? If so, I still would like an explanation as to how you think I can recall a memory that is long term? How can it be reconfigured if there is no long term storage? You specifically said: “The concensus from psychology appears to be that memories are reconstructions, and not anything actually recalled from a memory store.” Now you’ve said: “I am well aware that many psychologists do believe that there is some sort of data storage and retrieval going on.” You can’t have it both ways. Either memories are retrieved from storage or they are not. That’s really what a memory is by definition. Even if it’s reconstructed, what are the building blocks — you guessed it, memories! Just like if I store 1000 frames of a movie as 1000 individual files. If I watch the movie, you may say, that movie wasn’t stored it was reconstructed, but it was reconstructed from individual components that were stored. The only difference is HOW it is stored, not whether or not it is stored.

            Like

          • Neil Rickert
            April 12, 2012 at 18:20 (UTC)

            “I am well aware that many psychologists do believe that there is some sort of data storage and retrieval going on.” Then why did you post this? Are you just saying that you’re in disagreement with the psychological concensus that the brain in fact stores information?

            Firstly, I am using a narrower meaning of “information.” Secondly, yes I disagree with what the psychologists are saying.

            As for long term configuration – we use tree rings to ascertain the age of trees. But it would surely be a mistake to say that the trees are “storing” age information.

            The computer is a syntactic engine. When I say that information is in the form of abstract meaningless symbols, I am emphasizing that syntactic nature.

            As best I can tell, the brain is primarily a semantic engine. It uses syntactic methods (as in the use of language for communication), but it does that to serve its semantic purposes. AI is around 60 years of age, if we date it from the 1950 “Mind” paper by Turing. It is still going strong, after 60 years of complete failure. The philosophy of mind is still going strong after 2000 years of complete failure. What AI and philosophy of mind attempt, is to give a syntactic account of what the brain is doing. If the brain is primarily semantic, as I suggest, then that is probably doomed to continued failure. Unfortunately, syntactic thinking dominates cognitive science.

            Like

          • Lage
            April 12, 2012 at 20:53 (UTC)

            Neil,

            The way the brain processes (and stores) information is also through the use of abstract symbols that are otherwise meaningless (our perception is completely subjective and has no bearing as to how the world really is).
            It is our perception, brain-wiring, integrative structures, and continued experience that relates all the data together that gives it context and meaning from the time our brains are developing and through adulthood. The brain is taking in information in the form of sensations (due to different forms of energy entering and exiting our body which are presumably non-abstract “things” which we label or represent with abstract symbols and abstract quantities) and produces perceptions which are subjective constructions that are a result of translating the incoming information a particular way (based on many things including the aforementioned brain-wiring, etc.).

            “As for long term configuration – we use tree rings to ascertain the age of trees. But it would surely be a mistake to say that the trees are “storing” age information.”

            Technically I can use that data (rings on the trees) to recall specific information perceived in the past (some form of proof that they experienced one season of “experience” with each ring as vague and non-detailed as that may be). My long term memories allow me to do this as well, and with much greater detail. I can re-experience something I’ve experienced in the past, because of the memories stored (how much detail and how close it replicates that experience can be of varying degrees). So there is a major difference here between the tree analogy and the brain’s storage of experiences/memories, yet I would say that information is still stored in both cases. Technically, the tree is storing age information even if not consciously or purposely doing so (perhaps not “Shannon” information, but information about it’s age in seasonal units nevertheless). There is a stored artifact (which we could call a memory) of what that tree has experienced in that sense. Another difference is that the tree can’t recall or retrieve this stored information, but we can retrieve information stored in our brains.

            Like

          • Lage
            April 13, 2012 at 13:31 (UTC)

            “The computer is a syntactic engine. When I say that information is in the form of abstract meaningless symbols, I am emphasizing that syntactic nature. As best I can tell, the brain is primarily a semantic engine. It uses syntactic methods (as in the use of language for communication), but it does that to serve its semantic purposes.”

            Yes and the key thing to remember here is exactly what you said: “the brain is PRIMARILY a semantic engine”. It is still syntactic in many ways. If memories were stored/retrieved in a different order, once recalled, they may not make sense. The brain will try to make sense of them with it’s semantic potential, but it doesn’t mean it will work every time. If there is enough scrambling of the information (i.e. the syntax is poor), the semantic attributes will not be able to decipher the meaning correctly. I can think of a sentence “Car a drive I”, and my brain will try to decipher the key words through semantics and make sense of it “I drive a car” (perhaps). However, if we consider a sentence such as “Car work a Drive I”, I could interpret that (through Semantic methods) as “I drive a car to work” or “I drive a car for work”, or “I work to drive a car”, or even “I drive a work car”, etc. Syntax still has it’s place in the brain (just as a computer does) and that is what is important to realize here. Of course semantics is primary (and it’s ok if it is), but it does not negate the syntactic property. The same example above could be applied to memories or thoughts. If I am remembering a memory out of order either chronologically and/or sequentially (yet still logically possible), such as: a moment when I was watching TV and had a flashback of hitting my head earlier. I could remember the moment as me hitting my head and watching TV (which would be logically true in terms of absolute chronology), or I could remember the moment as me watching TV and having the flashback of hitting my head earlier (which is the exact memory that I was trying to recall). Syntax matters in these cases, not just semantics (in terms of brain processing) as the order that the potentially meaningless memories are stored/recalled is dependent on syntax and this may be necessary in many cases in order to correctly interpret even if semantic methods are available to try and accomplish the task. BOTH are necessary in many cases.

            Like

          • Lage
            April 13, 2012 at 13:33 (UTC)

            “AI is around 60 years of age, if we date it from the 1950 “Mind” paper by Turing. It is still going strong, after 60 years of complete failure. The philosophy of mind is still going strong after 2000 years of complete failure. What AI and philosophy of mind attempt, is to give a syntactic account of what the brain is doing. If the brain is primarily semantic, as I suggest, then that is probably doomed to continued failure. Unfortunately, syntactic thinking dominates cognitive science.”

            The main reason in my opinion for the A.I. “failure” (which I mentioned previously), is because they are trying to reverse engineer a carbon-based brain that uses non-local parallel processing and make it out of a silicon-based, computer-like template dominated by serial processing methods (even if parallel processing methods are starting to emerge the last couple years). The human brain took millions of years to perfect and realize, and the A.I. “brain” has only taken 60 years of trial and error by people basically trying to push a square peg through a round hole. By comparison, the 60 year time frame is less than the blink of an eye. It’s the fact that their thinking is limited by previously designed computer technology. They think that they can analogize the brain to a computer (in terms of storage and pattern recognition) which is a huge problem. Human brains have incredible patter recognition capabilities, but computers only can recognize what patterns they are programmed to look for. Humans don’t realize how many patterns they actually are identifying everyday, so only the patterns that they know we recognize (if not less) will end up in an A.I. system. I am very troubled by the lack of “thinking outside the box” that most engineers employ. They use old methods, except for a creative few in number who actually revolutionize technology with a relatively new idea (rather than just increasing performance of the old technology with more storage, faster processing, etc.). All of the improvements help to some degree, but when it comes to mimetic technology, there is nothing better than radical new approaches which most A.I. engineers have failed to use.
            So while they may have trouble giving a syntactic account of what the brain is doing, it does not mean that the A.I. technology is impossible, nor does it mean that because of the failure we must assume that the brain isn’t syntactic in nature. It could be primarily syntactic, but perhaps we have failed to see that because we haven’t been thinking outside the box. For years people thought that the Sun rose and set everyday about a fixed earth, as opposed to the slow rotation of the Earth causing the illusion of a moving Sun. People thought this for thousands of years, and finally one day, when able to think outside the box, Copernicus thought of a new possibility — and he was most certainly correct.
            I agree with you that “Unfortunately, syntactic thinking dominates cognitive science”. It is definitely a problem as it narrows the scope of possibilities, but it can also mean that we just aren’t understanding enough of how the brain uses syntax.

            Like

  6. Neil Rickert
    April 11, 2012 at 20:58 (UTC)

    Regardless of what you may think, DNA is definitely stored information.

    You think it is information. I think it isn’t. We can both find plenty of people who agree with us. I don’t see any point in arguing it.

    Like

    • Lage
      April 11, 2012 at 21:36 (UTC)

      Can you tell me why you think DNA isn’t stored information. What part of my explanation (specifically) do you disagree with? Let’s hammer this out, so I can see your point of view more clearly. Why isn’t DNA considered a type of stored information. What prerequisites do you have for something to be information, and for it to be stored? Let’s start there so we can see where we end up.

      Like

      • Neil Rickert
        April 11, 2012 at 23:35 (UTC)

        Can you tell me why you think DNA isn’t stored information.

        It is not abstract. DNA is there for its direct causal role, not as an abstract representation.

        John Wilkins had some good posts about that on his blog, here and here.

        Like

        • Lage
          April 12, 2012 at 13:40 (UTC)

          Unfortunately, one can argue that any other form of information that you can think of is there for it’s direct causal role. It is we human beings that arbitrarily choose to call certain types of information “abstract” representations of something else (just like we give segments of DNA an abstract title such as “genes”, when they are really just something that we’ve conveniently labeled as such to represent them in a certain way that’s useful). That’s all that existence is really — nothing but a large causal chain of events (everything has a causal role — look into the concept of “determinism” and you’ll see what I mean). If you want to dispute this, then tell me: What type of information is not a part of a causal chain (thus has no causal role)? This was one of your criteria for believing that DNA isn’t stored information, so give me an example of information that is not a part of a causal chain/role. Whether or not you want to call it a “direct” or “indirect” role is arbitrary and subjective and thus is irrelevant to the argument. Everything is connected and integral to everything else in the universe so every role is equally important (think of the butterfly effect to illustrate this truth).

          Technically, if one were to name the nucleobases of DNA some abstract labels (such as Adenosine, Guanine, Cytosine, and Thymine — which we have done), and label certain segments that fit our criteria of “genes” (which we have done), then we have plenty of abstract representation of the information stored in DNA already. We’ve chosen to define the information, just as we define it in computers with binary code. Even though the DNA strand works as a whole, we have chosen to break it up into what we call “genes”. In mapping the genome of various organisms, again we have arbitrarily chosen something to represent a “gene” (as if it’s a discrete entity even though it’s connected in a continuous chain) and assume based on what we’ve learned that it will eventually provide instructions for protein formation. Information stored in a computer (bits) also has a causal role. It may eventually be used to execute a program or software. Or it may serve a function for future computation of an equation or other calculation. It is no different than DNA in that sense, except that other mechanisms stored the information in DNA, whereas human beings store the information in a computer (later to be used for some causal role). Information in a computer can serve a function, as well as the information stored in DNA. Information in a computer doesn’t have to serve a function (other than being stored), and DNA also doesn’t have to serve a function. If resources are not provided, as well as the right environment, neither the DNA information, nor the computer information can be used for anything other than potential later use. There is no difference here. You are just failing to see the similarities for some odd reason. Imagine a computer that is programmed to copy it’s main programs (just like DNA does), and all of those main programs have their own functions (just like DNA does) — there are so many similarities, it’s much more of a stretch to say that one is information and one isn’t based on the reasoning I’ve heard from you thus far.

          That link you placed in here shows exactly why DNA is information. Look at the schematic diagram of a general communication system we need an information source (genome or DNA strand), a transmitter (gamete or Diffusion within an aqueous medium), a channel (DNA, aqueous medium such as a cellular environment and/or the inside of a reproductive tract, a gamete could also be considered the channel as it physically moves the DNA from one place to another), noise source (genetic mutation), receiver (zygote or another cell in the case of DNA replication during cellular mitosis), destination (the new cellular nucleus, etc.). So everything needed to fill that model is right there. The only thing that was missing from John Wilkins diagram was the information channel (which is obvious, I don’t know why he didn’t figure it out but perhaps he had a brain fart). If there are any issues with my examples, then we need to disect each term’s definition within that communication model to see if it qualifies (I’m able to do so if necessary).

          -Peace and love

          Like

  7. Neil Rickert
    April 11, 2012 at 21:01 (UTC)

    A thermostat uses logic and makes a decision.

    We use logic in describing the action of the thermostat. But the thermostat is not itself using logic.

    Like

    • Lage
      April 11, 2012 at 21:34 (UTC)

      The thermostat doesn’t need to use logic. The point is some action is performed if a parameter is met beyond a certain threshold. That is all that matters here in the case of making a decision. In our case a decision is made because of several parameters meeting a certain threshold. We choose through similar means, since we are pre-wired just like a thermostat (not my analogy of choice, but we can still use it).

      Like

  8. Neil Rickert
    April 11, 2012 at 21:02 (UTC)

    Also to be clear, “computing” only means to determine an amount or number. The thermostat is determining what the temperature is (albeit through a mechanical method in the case of analog), which satisfies the definition.

    The way the thermostat works doesn’t have anything to do with numbers.

    Like

    • Lage
      April 11, 2012 at 21:30 (UTC)

      It doesn’t have to have anything to do with numbers. The point is that a decision is made (whether mechanically driven, electronically, or some other form) which causes it to perform one of two options (turn on the heat or don’t). Your brain can work the same way, everytime you make a “choice”. It chooses the choice based on your current reference/standard based on previous experiences (stored information), how your brain is wired, etc.

      Like

  9. Neil Rickert
    April 13, 2012 at 16:15 (UTC)

    A final response to Lage:

    It is clear that we have very different ideas about what the brain is doing, and how it works.

    Like

    • Lage
      April 13, 2012 at 16:26 (UTC)

      Final response, huh? Yeah, my wife tells me that I wear her down sometimes. What can I say other than that I am an intellectual debater/philosopher by nature. That and the mixture of my analytical nature and holistic view of the universe can be hard for some people to swallow.
      I would still love to hear your alternative explanation of how I can retreive memories of the long-term past in my mind/brain (that are confirmed to be true with evidence) without information storage used in the process. I don’t think that you have an alternative explanation to explain that example, or I would have heard it already. Anyways, it was a pleasure discussing this with you. While I don’t think you nor I had any free will in choosing to do so, as our actions are a result of randomness and/or determinism, I still am glad we had this experience. It’s always a pleasure to hear new points of view as that is the only way we learn new things or change our positions on issues that interest us. Take care Neil and keep propagating those memes!

      Peace and Love,
      -Lage

      Like

  10. Jeffrey Shallit
    April 29, 2012 at 01:50 (UTC)

    I disagree with “the brain is a computer” idea

    Well, then everything that you are likely to say is probably wrong. So I guess I can stop reading this blog, because when you reason starting from an incorrect model, you are likely to be right only by accident.

    Like

    • Neil Rickert
      April 29, 2012 at 02:13 (UTC)

      Well, then everything that you are likely to say is probably wrong.

      It is no surprise that you are a computationalist. What is a surprise, is to see a mathematician so utterly certain about what is, at best, only a speculative hypothesis.

      Like

    • Lage
      April 29, 2012 at 03:08 (UTC)

      Regardless of whether any of us are “correct” (I don’t believe in true “correctness” as everything is subjective), or whether or not we disagree on certain points, we can still focus on seeing a new point of view and perhaps learn something new. Personally, I find the topic very fascinating and it’s better to discuss a topic on a blog with someone you disagree with (at least on some points) — otherwise it is just an exercise of reinforcing your own beliefs which makes life far too easy and much less interesting. It’s usually the new and seemingly radical ideas that make most people uncomfortable, but getting out of one’s own comfort zone is productive for the potential of significant intellectual discovery.

      Peace and Love

      Like