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

What is thinking?

  1. Mar 19, 2006 #1
    Hello there, I'm by no means a philosophy student but I plan on majoring it in college :rolleyes:. Anyway, I was killing some time, waiting for my introductory books to arrive, by playing with my 8 year old little sister. Well, she asked me this very interesting question. "Phusicist, when a chinese person thinks in his head, is it in chinese?" I replied, "Ummmm, probably" and then she volleyed, "What if he came here and talked in english more, would he still think in chinese?".

    This prompted me to ask a few questions of my own. With all this talk about language, I deduced that thinking, or at least abstract/complex thinking must have something to do with manipulating (So thinking should be intertwined with working memory too) linguistically defined concepts (I separate thinking from abstract thinking because when we drink something we usually don't say, "Hmmm I am thirsty, being thirsty is unpleasant, if I drink x I will not be thirsty, so I will drink x" we usually do this instinctively, maybe we use emotions as shortcuts in reasoning?). It seems possible that simple thoughts can exist without language, but it's very unlikely that complex ideas such as those in philosophy or math could.

    Other groups of people with more primitive languages must have trouble with thinking and this must hinder overall progress! For example, Filipinos don’t have the verb to be in their language!

    Then I thought, what if someone does not learn any language at all, are they incapable of abstract thinking? What would it be like to peak inside their mind? Maybe the answer lies in feral children? http://www.feralchildren.com/en/index.php

    So, what is thinking? I'm also curious because I always see philosophy introductions mention that philosophy can make your thinking clearer, how so? Well, philosophy definitely improves linguistic ability and semantic reasoning, so these alone would be a way I think.
    Last edited: Mar 19, 2006
  2. jcsd
  3. Mar 24, 2006 #2
    I am by no means a neuroscientist, but historically it would seem that there's a correlation between abstract thinking and language. Don't quote me on that, though.
    I have a friend who has English as a first language. He went to Italy for a couple of years, and now he assures me that when he's tired or stressed, he thinks in Italian! I don't know if that's relevant or not, but it might help.
  4. Apr 10, 2006 #3
    It's often said that crows can count pretty high-- IIRC somewhere between 10 and 20. And then there's tool use (just ran across this):

    But you're right, it's unlikely that "higher" or "more complex" thoughts would emerge without language.

    I did my "humanities thesis" in college on the necessity of distinction. At the time, I was interested in Chaos theory, which explained the problem that science has of trying to break down problems smaller and smaller, which essentially can *never* acheive perfection. The thought had crossed my mind, "is there another way of thinking that would allow us to understand the universe without breaking things down into their constituent parts?" I looked a lot at various philosophers at the time, but in the end, I pretty much concluded "no", and got what I think is a pretty good understanding of what thought is and how it evolves.

    I believe thought is essentially the attempt to achieve a solution to a desire without necessarily enacting that solution. The desire starts off as something basic like "eat food" or "have sex", but later (in human thought) evolves into "make money", "understand concepts", etc. Essentially, if you truly had no goals, you wouldn't think.

    Language facilitates thought by giving it a framework that is generic and consistant. It can apply to a general case, and doesn't repeatedly need redefinition. Language helps us break down concepts away from the specific such that concepts can be applied in a much broader scope. Having an increased vocabulary helps, but for the most part, the benefit comes from the ability to break things down into constituent elements. The more we can learn *how* to break things down and classify them generically, the better we can work towards better comprehension.

    Essentially, I believe that the necessary elements for thought are:
    - perception
    - the ability to affect one's environment
    - memory
    - desire
    - the ability to distinguish
    - associativity

    Language per se isn't necessary, but it helps us to distinguish and associate.

    Last edited by a moderator: Apr 22, 2017
  5. Apr 10, 2006 #4


    User Avatar
    Gold Member

    But all those things can be done by some animals pretty low on the complexity scale.

    I am reading Jay Ingram's "Theatre of the Mind" and I am becoming more and more convinced that language and consciousness are very closely linked, if not virtually synonymous.

    Have you ever tried to silence your internal monologue? I mean completely silent? I doubt you can do it for more than about ten seconds or so without practice. But it's a great technique for falling asleep. I tend to fall asleep very shortly after succeeding.

    It's an odd state you find yourself in, once there is no voice in your head. It's very difficult to call it "thinking" at all. I'll bet it's much like the foggy twilight consciousness that lower mammals exist in.
  6. Apr 10, 2006 #5
    Hm. That depends on what you might consider "consciousness" versus "thought". I will constantly maintain that other animals without language are capable of thought.

    Perhaps my favorite example is the case of the octopus. They gave an octopus a jar, with its favorite food inside. The jar was sealed with a cork, which the octopus COULD open, but it wouldn't be immediately obvious how. After all, opening jars isn't something that octopii have an instinct for doing.

    After something like 6-8 hours, the octopus manages to open the jar, and eats the food. Cool! But as if this weren't enough, several days later the team comes back and gives the *same* octopus another identical jar with food inside. And sure enough, it opens it within a matter of minutes.

    I'd call it thought. Not complex thought by any means, but thought nonetheless, without language.

    Consciousness is tricky. I'm not even sure I know how to define it. Some have described it as being "self-aware", but that's a little silly. Does being self-aware necessarily require a distinction between oneself and the outside world? Or a distinction between oneself and another being? Perhaps it requires a memory of *thought*? I'm not sure. But I'm not sure I'd limit my view of consciousness to requiring language, since I view language as something very specific, as opposed to consciousness, which is pretty generic.

    Language, at its root, was NOT devised for "higher thought" it was devised as communication between individuals. The simplest forms of language we wouldn't even *call* language. For example, a beaver thwacks its tail on the water in order to warn others of danger. Communication at a very basic level. It does make me wonder, however-- how to deaf people think? Do they think in terms of hand gestures?

    Anyway, how would you define consciousness?

  7. Apr 10, 2006 #6


    User Avatar
    Science Advisor

    Thought, i believe, is mostly simulation of input. Thoughts may simulate auditory, visual, tactile, olfactory, palatal input. Language is not restricted to any subset of the senses, so thought may simulate any stimulus and still follow a language. One definition of "language" might be "something that can be interpreted". As such i don't think thought requires a language, as you might find yourself thinking about random images, sounds, concepts, etc, not exactly following a language. I'd say that thought has the potential to follow a language, but doesn't require one. Similarly i'd say thought may or may not be logical, depending on whether it follows a language derived from logic.
    One of the advantages of thought is that it is abstract and so doesn't require the physical circumstances. This speeds up learning very much. For example, suppose you are unable to think, and perodically come accross a situation which requires you to come up with a solution which is not trivial. While you are exposed to that situation you are able to learn. If the situation involves opening a jar of food, then actions you perform on the jar will be registered in your brain as either "working", or "not working". Once you open the jar you have solved the problem, have learned the solution and can reproduce it in the future.
    On the other hand, suppose that you're really not very bright and every time you need to open a jar of food you always fail and something always happens that takes you away from the jar, like some other animal eventually comes around and takes it from you. Then, you can only learn how to open a jar every so often, during the time that you are exposed to the jar.
    If you have the ability to think, then you can simulate the situation in your head and attempt to solve it there. This way you don't have to wait until you are exposed to the jar. You can learn faster. But this requires you to be able to reproduce a variety of situations. Might require you to simulate what a jar feels like, how it reacts when hit with a rock, how the rock and the jar interact depending on how you hit the jar with the rock, etc, it requires concepts.
    This is probably why humans are so successful, we learn very fast compared to the other animals. Either because those animals don't think very much, or because they don't have enough concepts that would allow for useful thought.
    I guess we could kind of see humans as "delusional" primates. Perhaps at one point in evolution our species branched from the other primates through a change in the brain that caused previous input to come up again, reproducing a past situation. Depending on whether the specimen could differentiate between the current sensory input and the one the brain was reproducing from past experience, then the being would be more or less fit. Namely, completely delusional primates would live in a fantasy world and be very unfit. On the other hand some level of this behavior might easily have been beneficial.
    Last edited: Apr 10, 2006
  8. Apr 11, 2006 #7
    I can't answer any of those questions you guys gave but yet maybe this will help a little.

    I think ( :blushing: ) or rather it really is, that normal life humans live fully depends on a language. Without it, consciousness of surrounding wouldn't be possible. You think using a language, you express ideas using a language. I agree with this case that more words you know the smarter you get. If you possess enormous vocabulary and you like to think, your thinking will probably be better than of the same person who knows less words than you do.

    I always wrote, write and will write in the poorest language one can ever write, that's the reason why for example thinking about more difficult things is so hard to me. Exception here are physics and mathematics where you don't need to possess Shakespeare-like skills in writing, but again if you know more words it's easier for you to think and you shape your mind for sort of easier thinking.

    As Dave above said, if you stop thinking, you soon will get asleep. Proven in my case. Now one may easily conclude:
    language --> thinking --> Consciousness.
    You may ask now, "wait but what about images? We may also think in images, at least I can" Well, we all can but thinking with pictures much depends on language. Without language our pictures or what we see would be colorless, senseless, just without a meaning. You look outside, you see trees, and say they're beautiful, but wait, how do you know they're beautiful? First of all, it's because you are fully aware of what the words 'beautiful' 'nature' 'life' mean, second of all it's a matter of fancy. You look at the table, and you say it's the table, word you've remembered. It's color for example brown is also remembered. If you want to think about the table, first your mind looks for the word table, and then searches for the picture of table, as far as I know. Yet I know nothing and I'm too lazy to learn :frown:

    Looking for critics, Thanks.
  9. Apr 12, 2006 #8


    User Avatar
    Science Advisor

    I don't think your statement of "the more words you know, the smarter you get" is very accurate. I think that, by far, the most important contributor to an individual's intelligence (my definition of intelligence being for the most part a measure of problem-solving ability) is how much exposure/interaction with the physical world that individual has been subject to, particularly in his/her earlier years. I wouldn't be surprised to find a very strong correlation between levels of playfulness of children and their intelligence at more mature ages. Type of play also contributing, not just quantity.
    Last edited: Apr 12, 2006
  10. Apr 12, 2006 #9
    I would argue that "the more words you know, the more easily you can think". And the more easily you can think, the more prone you are to learning more things. It doesn't necessarily follow that you WILL, but it makes you more prone to it.

    But even then, it's hardly quantifiable. You know 12,644 words, but I know 13,015 words, therefore I have a stronger toolset of words? Nah, not necessarily. Some words are more useful than others. Many words are similar. And even then, we may not have defined words well enough so that we know to apply them when they're applicable. 500 words compared to 5,000 words? Now we're talking.

    The big step is learning how to USE words-- knowing the concept of words. Giving something a name, whether that something be an object, an action, a quality, a relationship, or whatever, allows you to encapsulate a it such that it's not necessary to go back and re-establish it later.

    If, for example, I called "Bob" "the short guy with brown hair who always likes to talk about fish", I'm re-evaluating him every time, which means I'm spending thought-power in re-classifying him, where I could be getting to the point and explaining to you what he did.

    Hence, language is a toolset for thinking. We build better thoughts *with* thinking than without. It doesn't mean we WILL necessarily, but it helps us.

    I would deem "raw" intelligence as a measure of your associatative ability. How good are you at relating one problem to others that you've seen? In what ways can you relate them? How frequently do the associations you make meet with favorable results?

    But, that said, you're right. The greater palette you have of experiences, the greater your capacity is at *making* associations. If you don't have something similar to compare to, you'll make a poorer association than you otherwise could. Hence, experience greatly contributes to better decisions. I don't think it's necessarily *raw* intelligence, but it *does* help us to rate our intelligence.

    I think it's also related to how often you *try* and learn. When you do the same things over and over again, you don't usually get anything new out of it. Routine doesn't prepare you for adaptation. But preparing to adapt means you're trying to solve problems you otherwise would never have known existed. Trying to learn gives you a wider variety of experiences, which again, are tools for your raw intellect to associate with.

  11. Apr 12, 2006 #10
    The scientists that research general intelligence normally define it as reasoning and novel problem solving ability.

    This is true, a person can probably memorize an entire dictionary using memory devices like mnemonics but this doesn't really demonstrate complex thought. Higher thinking requires a person to manipulate linguistic concepts and ideas in ways that transform their meaning and implications; transformation occurs when they combine facts and ideas in order to solve problems, synthesize, generalize, explain, hypothesize or arrive at some conclusion or interpretation, producing new knowledge.

    I posit that verbal analytic ability affects how and what you think. How you think about what you're doing affects how you do it, or whether you do it at all.This may directly affect your mathematical research or medical reasoning.

    Also, here's an interesting link I just found that's semi-related: http://www.lagriffedulion.f2s.com/sft2.htm
  12. Apr 12, 2006 #11


    User Avatar
    Staff Emeritus
    Gold Member
    Dearly Missed

    A, La Griffe du Lion! Be warned, the maintainers of that site accept completely the differences in ethnic national IQs and their genetic bases, and in particular the difference in US white and black IQs which they have many posts about. Their posts tend to be in the form of little stories with a statistical point, as in the one phusicist linked to.
  13. Apr 12, 2006 #12
    Thought is the collective past, carried forward into the collective present. May I suggest the book 'Thought as a System', David Bohm, 1994.
  14. Apr 12, 2006 #13
    What's wrong with that? If it's granted that there are those individuals who are naturally better at certain things, faster, stronger, smarter (I'm sure we'll both agree that neither you nor I have the same mathematical expertise as Albert Einstein or the same critical thinking ability as Bertrand Russell!), etc. Why is it suddenly unreasonable to try and trace the lineage of these people to a common ancestor? The relatively low population of Ashkenazi Jews (2% in the US) compared to their contributions in the fields of highest thought (30ish% of all Nobel Prize Laureates!!@) is particularly striking!

    Thanks for the suggestion, I'll check it out. I'm sure there are a lot of philosophers who have already dealt with this too, maybe in philosophy of mind?
  15. Apr 13, 2006 #14
    Ah, the idea of "thinking"! Now just what is that all about? Well guys, being the "crackpot of the day" on this forum, I take it as my duty to comment. If I exceed your interests, I am sorry! You don't have to read this unless you think you should. :rofl:

    First of all, everyone makes a great issue of "consciousness" so it must be an important concept. On the other hand, it must have little to do with survival as most surviving entities apparently manage to survive without it. I guess one could say that "survival" is an "unconscious" success story! It seems to me that there are two very very different kinds of thought out there: conscious thought and unconscious thought. Perhaps one could think of "conscious thought" as an abstraction layer of thought over and above that of the "unconscious" realm. (There are nuroscientists who hold that such is the actual construction of the brain.) :redface:

    Just as a convenience (plus a little fun) suppose I refer to "unconscious thought" as "squirrel thought" under the presumption that squirrels are not that versed in explicit formal logic which seems to be the great forte of "conscious thought".
    Hmm, is the achievement of language a consequence of "squirrel" thought or "logical" thought. If it is a consequence of logical deduction, you should be able to lay out to me the step by step process which you used to achieve the facility of a language. It seems clear to me that language is just another mental construct created by our subconscious during the first twelve months of your life. No one has ever successfully established a conscious method of solving the problem of "creating a language" from a stream of undefined sounds (something hundreds of millions of children do every year almost without fail). If anyone could solve that problem analytically, don't you think our computers would be talking to us? :confused:

    So language is a "squirrel" construct without which, it seems, the abstraction layer I call "logical" thought (just a name I have given to "conscious" thought because it brings to attention the great facility of conscious thought). Ah, but this implies that any entity which can solve a logical problem is conscious. But is that really true? :wink:
    Yeah, and if you knew some of the people I know you would recognize that language does not bestow the ability to think on an abstract level. Is there perhaps another layer of thought which one might call "abstract" thought which sits atop that abstraction layer I have called "logical" thought. Again it isn't that a facility with "logical" thought bestows the ability to think on an abstract level but rather that without a foundation of "logical" thought, abstract thought is just impossible. Gee, this is beginning to sound like that old joke, "it's elephants all the way down"! :yuck:

    Well, this whole thing is just getting too long (it seems I have a tendency to exceed the attention span of most people) so I will graduate to my central observation. In essence, there are two very different ways of "understanding the universe". There is that emotional feeling that something makes sense; that you understand what is going on and have no doubts as to the validity of your expectations. This is pretty clearly a "squirrel" conclusion (the squirrel understands how to jump to that next branch). Then again there is a very different kind of understanding which allows you to logically defend some set of analytical expectations in intimate detail; even in cases where no emotional feeling exists to defend the validity of those expectations (here I am talking about all those totally counter intuitive deductions so common in analytical work). This is a "logical" conclusion. The next level is understanding on an abstract level which is a whole different story which I will omit for the time being. :biggrin:

    So all thought can be divided into two categories, each with its own strengths and weaknesses. The great strength of logical thought is that the conclusions reached through logical thought are guaranteed to be as valid as the premises upon which they are based. The weakness of logical thought is that it is limited to a very small number of premises: i.e., the specific number of factors which can be included in the analytical statement of the problem. This is a seriously small number when compared to the volumes of information available to us through our senses. :bugeye:

    A further problem with logical thought is that the number of specific steps in the process cannot be excessive as we must be consciously aware of each step. If we are to be truly logical, each and every step must be consciously validated. Anyone who has carefully thought anything out is very well aware of the fact that considerable time is consumed in such an analysis. If a logical process were to involve a hundred steps, I doubt many here would even make an attempt to follow that logic. Now mathematics and formal logic provide us with a certain respite from that last constraint but, even so, logical thought is of very limited applicability. :confused:

    Squirrel thought has its own strengths and weaknesses. Its strength lies in the astonishing number of factors which may be taken into account. Its weakness is the fact that the process can not be validated: i.e., there is no way to prove a squirrel decision is correct. Nevertheless, most of them will be good decisions. Why is that? The answer should be clear. Whatever the mechanisms are, by which those decisions are reached, they have been honed and polished through millions of years of survival; failure to make good "squirrel" decisions has been cleaned from the gene pool by the consequences of the bad decisions. :biggrin:

    Watch a basketball player dribble down the floor, dodging his opponents, sometimes dribbling behind his back, as he jumps suddenly sideways and snicks the ball through the net thirty feet away! Any athlete knows that very little logical thought goes into such a move. In fact, if you try to consciously think about what you are doing, you won't be able to do it. I think it was Buddha who once said all evil comes from conscious thought. :devil:

    What I am getting at is the fact that logical thought is actually a rather worthless endeavor when it comes to life and death decisions. It is often much better to "go with your gut"; let it be a squirrel decision. In fact, in the absence of mathematics, logical decisions are so limited as to be almost entirely inapplicable to any day to day activities. This is why many students can not understand a purpose to learning mathematics. Actually they are quite right, neither math nor logic serve much of a purpose to important problems. I have known very successful people who have never made a logical decision in their entire life. :cool:

    However, when a problem can be approached with math and logic, one can be quite sure of the absolute validity of their conclusions. Well, "absolute" to a certain extent: it is possible that an important factor was omitted or that some axiom thought to be true is, in fact, false. Thus it is important that we understand how those factors came to be established. There is but one answer; squirrel decisions! I have come to the fundamental conclusion that squirrel decisions are the single most important part of thinking; logical thought is not even possible in the absence of squirrel thought. :tongue:

    This is, in fact, the single biggest problem in trying to understand the universe. Most everyone believes the ideas they have arrived at via their personal squirrel decisions are the only possible conclusions which can be reached. The reader should understand that "belief" of anything is a squirrel decision. The ability to communicate (language itself) was acquired through squirrel thought. Accept your squirrel decisions as your best bet when it comes to any serious question, but don't ever think that those squirrel decisions are infallible. You don't have to believe they are infallible before you can follow them; when it comes to life, "you pays your money and you takes your chances". o:)
    I think you oversimplify the issue of thought. "Perception", perceive what? This presumes "the ability to distinguish" and "associativity". All of these presume "memory". That leaves "desire" and "the ability to (affect?? don't you mean effect?) one's environment" which seem to me to be intimately related. Nonetheless, I think the last two are worth considering. It seems to me that the ability to effect one's environment constitutes a division into what can be controlled and what cannot be controlled. This says to me that one should include another abstract layer in that concept of a brain. A layer I would call the expectations layer (desire would be a categorization of one's expectations). However, if the brain could set up a specific array tree of expectations depending upon various future possibilities, then it seems it could set up a decision tree within that tree of expectation which would yield the highest probability to its desires (which is what? survival?). I think I could even write a computer program to do that! Would you say the computer was "thinking"? :confused:

    A lot of science is done in the total absence of logical thought and that has to be so; but scientists should not forget that fact. If they do, science folds over to religion. It may work great, but that does not mean it is valid. Think about that next time you see a "poor squirrel decision". :surprised

    Have fun -- Dick

    Knowledge is Power
    and the most common abuse of that power is to use it to hide stupidity[/QUOTE]
  16. Apr 13, 2006 #15
    I think that's largely correct-- effectively the difference between *established* neural pathways and *new* neural pathways. Established neural pathways are sort of like "instinct" or "subconsious" thought. Decisions you make without really needing to think them through.

    Eh, I wouldn't be so quick to tout "logical" thought as being so great! Check out Blink, a book about "snap" or "subconscious" decisions.

    But perhaps you ought to clear up your definition of "logical" thought as you call it. Let's say you have three invitations for three different events on the same day. A co-worker's wedding, a nephew's bar mitzvah, and a friend's Broadway debut. Which do you go to? You'll have to think about it. And deciding what to do isn't *logical* per se, but it takes a lot more effort than a subconscious decision.

    On the other hand, let's say you have a gloriously abstract problem like "What is the square root of 33124?" or "If all snarfs are round, some snups are snarfs, and all snigs are snarfs, are all snigs round?" Those are strangely different because there's a discrete, logical answer assuming the validity of the givens. But still, you need conscious thought to answer both questions.

    Perceive anything! :) Effectively, you need to be able to have some sort of information to process. And information gets to you in the form of perception. It's theoretically possible that information is self-derived, but I'll hold that that's a practical impossibility.

    How so?

    I think people take those two items for granted SO very much, and yet they're so utterly crucial that they NEED to be specified. After all, THAT'S what we've been lacking with computers all these years. How do we get them to distinguish and associate? We can *tell* them how, but then they're not really doing it themselves, we're just establishing pathways FOR them. They have to be capable of creating NEW pathways.

    Again, I have to ask how? Could you not associate to distinct objects within your perception and forget them an instant later when they leave your sight? I'd say so!

    I believe it is indeed "affect". To "affect" something means to alter that something. "effect" is the change that you create. So, I may affect the soda can by crushing it, and the effect my crushing had was that the can is now compressed.

    Well, yes, that's effectively taking several steps all at once, but that's more or less how thinking is effected (how many times can I use varients of the word?).

    In your example, one needs:
    -perception to build plausible possibilities
    -memory in order to access those possibilities
    -the ability to distinguish in order to tell the difference between one possible outcome and another
    -associativity to relate the current decision with the abstract one, as well as associate probable success rate with past experience
    -desire to evaluate the desireability of the outcomes
    -the ability to affect one's environment in order to realize that the decision will actually HAVE a desireable result, otherwise the excersize is useless

    The problem is that while you could get a computer to carry out a specific thought process, it's not variable and reapplicable. The computer won't know where and when to apply a *similar* process again. You can train it to apply *exact* processes at different times, but it's not *thinking*, because it's following ONLY established instructions.

    But they DO have thinking computers! Just very basic ones. Check out "neural networks". They essentially lay out programs which function the same way that synapses in the brain do. Initially, nobody really understood WHY they worked, but they did (I'm not sure if they've made much progress in that department, either). And they work based on evaluated success and failure-- the "desire" I mentioned above.

  17. Apr 14, 2006 #16
    I think you are slightly off the mark here; but only slightly and I may be misunderstanding you. It is fairly clear that one can create new neural pathways which yield altered "instinctual" or "subconscious" thought. Certainly behavior can be altered by training and, if one is to refer to the neural pathways generating that behavior as "thought", one cannot categorize these neural pathways as established in an absolute sense. However, if you mean the pathways are not essentially altered by a specific event, I would tend to essentially agree with you. My son-in-law did his graduate work in neural network simulations and they do a pretty good job of simulating instinctual thought but they are certainly not what I would call *established*. In most such designs they the circuits continue to be free to adjust their weights even after those weights have pretty well stabilized. :rolleyes:
    I essentially align it with procedures at least trying emulate formal logic and I think you misinterpreted what I was saying if you think I was touting "logical" thought as being "so great". Note my conclusion to that paragraph: o:)
    I see our conscious thought as being composed of two specific elements which could be termed perception and processing. From my perspective, the perceptions of our conscious mind are the conclusions reached by our unconscious mind. :surprised
    What you are talking about here is a complex mix of what I call "squirrel" decisions and "logical" decisions. First, the very fact that you are aware of these three possibilities is a conscious perception created by a squirrel decision. They weren't brought up by any sort of conscious retrieval from your memory. They were presented to you by your unconscious mind in response to your conscious interest in what you were going to do. I am sure your unconscious mind will also present you with a small number of additional factors which might influence you: most presented as emotional factors associated with each of these possibilities. But what it won't do is present your conscious mind with ALL the possibilities on each issue. Your conscious mind can only handle a very small number of things and you depend very much on your unconscious mind to glean the overall circumstance to present you with a relatively small number of relevant issues. Haven't you ever made a decision only to regret it later with a comment something like, "oh gee, I forgot all about x" where x is something logically important to that decision? Was it a conscious decision not to take that issue into account? No, I think not. It was a unconscious failure to retrieve a significant factor. :biggrin:

    As I said, human logical thought is extremely limited; however, it can be examined in detail for errors. Thus the conclusions are as accurate as the elements going into the decision. The squirrel thought seems almost unbelievingly extensive (it can solve problems where you can't even comprehend all the factors on a conscious level, things you can't even enumerate from a logical perspective) but it also has its problems. You cannot examine it for errors and it does occasionally make mistakes. Nevertheless, if my life depends on being right, I think it's safer to go with my gut instincts than it is to go with logic. :wink:
    Not really, it depends very much on what decisions you decide to leave to your unconscious mind and I think you have the power to do that. You can make what seems to you to be a logical decision recognizing that your perceptions might be wrong or you can forget everything and just go with your gut instincts. Idiot savants prove that even square roots can be done by the unconscious mind if it has been trained to do it. :rofl:

    I am firmly convinced that the very choice to be conscious of a problem is a squirrel decision. I think it is very easy to get so in the habit of doing some very complex things that we are not even consciously aware of what we are doing. In most cases, if asked, we can easily explain what we were doing because we have done it so often but is some situations, it is possible to recognize that we are really consciously unaware. Many many years ago, I had a job where there were two essentially equivalent but quite different routes from my house to work. The two routes met at an intersection about eight blocks from home with a traffic light. If the light was green I would go straight ahead, if it was red I would turn right: i.e., in the worst case I had to stop but I never waited for the light (there was almost no traffic when I went to work). That choice set my route to work. A couple of years of that and I discovered that, when I got to work, I often had utterly no idea which route I had taken. The whole thing had turned into a complete subconscious habit. I am firmly convinced that, once the subconscious has discovered a usable solution to a problem, it no longer bothers the conscious mind with the issue. I think that, so long as the unconscious mind is convinced it knows what is supposed to be done, it doesn't even bother the conscious mind. I could comment on this issue for days so I'll drop it here rather than bore you.
    Again, you misperceive my meaning. :rolf: :rolf: My point was that what you call your perceptions are assumed solutions (or "squirrel" conclusions) provided to you conscious mind by your unconscious mind. For example: when you read this, you will perceive the existence of of an image on the screen of your monitor (or perhaps a sheet of paper if you have printed it out). Now, even in your mental model of reality, your actual perceptions should be excited rods and cones on the retina of your eye or at least signals on the optic nerve to your brain. But I will hazard a guess that such a direct phenomena is beyond your conscious mind to perceive directly; it can only examine the phenomena from an analytical perspective. It follows that what you consciously perceive is an illusion created by your unconscious mind. Again, I can talk about this issue forever without beginning to exhaust examples and I have no interest in boring you. :yuck:
    Because, in order to distinguish things, your unconscious mind must have already associated different incoming data with these thing and in order to associate things you need to be able to distinguish them. How can you associate things which cannot be distinguished? Both of these require some underlying mechanism to provide the correct illusion to the conscious mind. If that mechanism vanishes, the illusion must vanish therefore, the mechanism must be remembered and some sort of "memory" is required. What I am saying is that the "reductionist" attack on the problem is the wrong attack. :grumpy:
    Now there I will again disagree with you. The intellectual community is lacking something much more basic than that. They are lacking an analytical solution to a basic problem solved by hundreds of millions of human children every year. These children begin as fertilized eggs with no more conscious awareness of reality than any other single celled being and within about twelve months they have managed to creat a mental model of reality that they will use practically unaltered for the rest of their lives. I think people get this whole issue backwards. The goal of science, as it is done by humanity today, is to justify that mental model that their subconscious handed to their conscious mind before they were two. They totally ignore the question as to how such a mental model is achieved. Consciously or unconsciously acquired, an attitude of achieving results by ignoring relevant issues is called working in ignorance. The important issue here is the fact that the existing solution they think they are searching for is swept under the rug unexamined as the very first move of their analysis. :cry:

    Human intelligence is totally isolated from the outside world in the sense that we cannot possibly know, a-priori, how any information is acquired. The only contact of a human being with reality exists via interactions, the real meaning of which simply cannot be known a-priori. Our mental image of the universe is constructed from data received through mechanisms (which we call our senses) which are also part of that image. It seems to me that any scientist in the world should hold it as utterly obvious that one could not possibly model the universe until after some information about that universe were obtained. It should be clear that the problem with that position is that we cannot possibly model our senses (the fundamental source of that information) until after we have modeled the universe. This is the basic conundrum seemingly ignored by everyone save myself (in fact, being willing to look down that rabbit hole is how I came to be a certified crackpot). :rofl: :rofl: :rofl: :rofl:

    The only problem I have with your attack is that it smacks of exactly the same kind of thinking which has already failed to succeed for many many years. It entirely ignores the very problem I just pointed out. Essentially your solution is organized from the perspective of a very complex scientific explanation; and that explanation was designed to justify a mental model of reality concluded by your subconscious mind at roughly the age of two. It is designed to fulfill the illusion presented to your conscious mind without giving the first thought as to how that illusion could be created. Everyone merely presumes it is the only valid solution to the original problem and that they are directly observing reality itself. As I said above, their approach simply ignores a very fundamental issue which is, in fact, turns out to be the very key to solving their problem. :frown:

    Have fun -- Dick

    "The simplest and most necessary truths are the very last to be believed."
    by Anonymous
  18. Apr 17, 2006 #17
    Sure we can! The epitome of established, fixed neural pathways seems to be the computer. Incapable of altering themselves, except to within the degrees established by the program. I think of computer programs much in the way of the subconscious. Capable of dealing with familiar problems and established patterns. So much so that it's far better at dealing with these problems than your conscious mind.

    Essentially, I would classify thought as the ability to alter your established pathways. When your unconscious (squirrel) mind approaches a problem that it's unsure of to a certain degree, it gives it to the conscious mind. The conscious mind isn't very efficient like the subconscious, but it does unexpected things and is able to arrive at conclusions that the subconscious isn't able to. Once the conscious mind arrives at a conclusion, it's passed off as input into the subconscious, altering the established pathways.

    Hence, without the ability to think, your subconscious would act quite consistantly and would me more akin to a machine.

    I haven't studied the particulars of neural networks, but from what I know, I assume that they will continue to emulate a certain stage of learning, and may be (once more perfected) capable of far better learning than the human mind.

    Well, the point isn't so much that it's impossible, but that it's useless for you or me or your average Joe. Let's take a more complex problem:

    You have an extensive collection of regular octogons called Norps. Each edge of each Norp has two points called Gips. The Gips are evenly distributed along the edge, such that the distance from each Gip to its neighboring vertex and to its neighboring Gip are equal. Further, each Gip is connected to exactly one other Gip by means of a straight line segment. Your extensive collection contains one of every possible such Norp. How many Norps do you have?

    The unconscious is at a loss. Or, at least, *mine* sure is! Maybe you can instinctively guess at a solution without the aid of your conscious mind, but it's at a severe disadvantage, particularly if you've never come across such a problem ever before. If you've dealt a lot with combinatorial mathematics and geometry, maybe your subconscious can make a slightly educated guess, and wind up in the right ballpark, but let's say for the sake of argument, that you ask this of a 4th grader. Their subconscious may pull a number out of thin air, but they'd be fully aware that it was purely a guess, and totally unsure of how to relate the answer with the problem. The CONSCIOUS, however, can begin to deal with the problem, because it isn't based on routine, it's based on innovative ways of applying association and logic.

    That's totally true. For example, I'm told that champion chess players are much the same way. They can play the first dozen moves or so of a chess game VERY well without really needing to think. And when asked "why did you move your bishop there?", it takes them a minute, because suddenly they have to re-evaluate the problem consciously. In fact, as evidence has shown, the conscious re-evaluation might be totally mistaken, for just the reasons you stated above! They're forced to re-evaluate with only what the subconscious mind provides them as input. There may in fact be TONS of other contributing factors, but the conscious mind doesn't necessarily get them all, and may not re-evaluate in the same way that the subconscious evaluates normally!

    Quite right, I think. My personal thought is that the higher the degree of uncertainty that exists in a problem, the more the conscious becomes involved. And when faced with a totally new problem, the subconscious pretty much throws up its hands and passes it off to the conscious.

    Well, that's true, but I'll hold that it isn't always so. Learning words is a great example. Upon hearing the word "soda" my subconscious translates the raw auditory input to my conscious, and my head is filled with the concept of a fizzy, sugary, liquid. I'm barely even aware of the pure auditory sounds, and filter out background noise, and maybe even the accent of the voice which says the word.

    But it wasn't always true. At one time, my subconscious got a strange mixture of what it assumed to be speech, and passed that off to my conscious for evaluation. "Hey, conscious mind-- I heard this word 'soh-dah'. Dunno what it means. Little help?"

    But even earlier in development, I hear a mishmash of raw auditory input, and my subconscious has no idea what to do with it at all. It doesn't even recognize that this thing as speech, or even as a distinct sound. It probably distinguishes it as "a sound that my mother is making", but doesn't even know how to distinguish the word from the surrounding speech. Again, the unconscious passes it off to the conscious, which is pretty helpless at that stage. Especially with a word like "soda".

    And even before that, I may not be able to even distinguish it as a distinct sound. My mother's voice saying "soh-dah" is grouped together with the birds outside, the TV on in the other room, the echo of the room, the rustle of clothing, and anything else that might be going on.

    I think that at each stage, the conscious helps develop the direction of the subconscious-- that's the only way the subconscious grows beyond that which is purely instinctive.

    In such a way, you're right, the conscious mind rarely deals with the rawest form of inputs. But it can. But the point really is that you NEED perception in order to think. Effectively, you need somthing to think about! It doesn't really matter WHERE that perception comes from (IE your subconscious or the raw "outside" world), just so long as you're perceiving *something*.

    No no-- I agree that association presumes distinguishment. And I agree that distinction presumes perception. But I don't see how perception assumes both association and distinguishment, which is I believe what you were asserting way down there... In my list, I was attempting to go from the most basic requirements (IE that which does not presume anything) to the more advanced requirements (that presume much). But I believe each is a useful distinguishable component towards developing thought.

    Well, I think it's sort of right and wrong. I think Hegel was the first I read who started talking about several different aspects of thought evolving *together*, whereas beforehand the philosophical assumption was the concept of the "blank slate" or "compartementalized slate". However, I think that the reductionist explanation may be helpful to understanding. Feel free to disagree, however.

    Actually, I find that a rather unfair critique. Your implication is rather existential in nature, which never seems to result in much. Effectively, reality may not be what we THINK of as reality, but we have no choice but to DEFINE reality as such, because without that assumption, we have nothing. Your desire to assess the development of the mental model of the universe is certainly valuable, but how do you do it? And how is it productive? And how is it verifiable?

    I don't see how-- in fact, I think it addresses the problem pretty precisely. Its failure lies in the definition and enacting of meaningful distinction and association. Those areas need to be honed to quite a degree.

    I'm not sure where you're seeing the "perspective of a very complex scientific explanation" in my explanation. Could you show me where?

    Last edited: Apr 17, 2006
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook