Concept Formation in Terms of Neurological Efficiency

[DarkReaver13]

I've been thinking about how one might describe a strictly neurological basis for proper concept formation.

Assuming that concept formation is a tool used to increase one's cognitive potency, you could argue that proper concepts reduce the amount of energy that needs to be used by one's brain in the long-term (initially extra effort is required in actually integrating and isolating the concept, however once this is done it makes subsequent processing much more efficient).

For example, an improper concept (anti-concept) -- for example subsuming communists, members of the Red Cross and players of Liverpool F.C. into an abstraction based upon their use of the colour red -- would actually be superfluous and rationally unusable. Thus, more energy would be required by the brain in an attempt to make use of such an abstraction than would be simply using the constituent parts by themselves.

Conversely, a properly formed concept -- which eliminates the extra work required to retain all of its constituent parts -- would reduce mental enegry requirements.

Obviously there is no way of proving this at this time, but I thought it was an interesting (and different) way of looking at the issue.

If the above is at all accurate, it provides a way of explaining the objective nature of concepts in that there is a definitive method of forming concepts properly (i.e. that which most efficiently organises knowledge in terms of the amount of energy required to do so).

[brian0918]

Is there any evidence that evolutionary pressure is simply moving towards energy efficiency? Are there no evolutionary changes that, though they may lead to less efficient energy usage, provide a survival advantage?

[aequalsa]

I've been thinking about how one might describe a strictly neurological basis for proper concept formation.

Assuming that concept formation is a tool used to increase one's cognitive potency, you could argue that proper concepts reduce the amount of energy that needs to be used by one's brain in the long-term (initially extra effort is required in actually integrating and isolating the concept, however once this is done it makes subsequent processing much more efficient).

For example, an improper concept (anti-concept) -- for example subsuming communists, members of the Red Cross and players of Liverpool F.C. into an abstraction based upon their use of the colour red -- would actually be superfluous and rationally unusable. Thus, more energy would be required by the brain in an attempt to make use of such an abstraction than would be simply using the constituent parts by themselves.

Conversely, a properly formed concept -- which eliminates the extra work required to retain all of its constituent parts -- would reduce mental enegry requirements.

Obviously there is no way of proving this at this time, but I thought it was an interesting (and different) way of looking at the issue.

If the above is at all accurate, it provides a way of explaining the objective nature of concepts in that there is a definitive method of forming concepts properly (i.e. that which most efficiently organises knowledge in terms of the amount of energy required to do so).

As you acknowledged, we are not quite up to that level of understanding how we store of information, so far as i know. If I were to hazzard a guess though, I think the gain in efficiency might better understood in terms of recall, especially with regard to pertinent, connected information. If it is correct that information is stored in neural networks, conceptualization might allow reuse of similar pathways. If you imagine an over simplified version where a string of compressed neurons named, x->y->z is how we store the concept mammal, then it would be easy to see how x->y->x->c meant cat and x->y->z->d meant dog. The benefits of accurate concept formation would be that all of those cells that store information about mammals would not need to be redundant and that because they are in the same network, utilizing the same trunk lines, it would be easy to recall conceptually similar concepts from long term memory. Activating one concept in the category brings all the items in that category a step closer to consciousness.

Energy usage is more relevant to short-term and working memory so I don't think proper concept formation would affect it. Having proper categories would probably make storage of new info easier since the category it belonged to could be quickly pulled up. That could affect how much time you spent learning a new thing which would probably mean less energy usage.

[DarkReaver13]

Is there any evidence that evolutionary pressure is simply moving towards energy efficiency? Are there no evolutionary changes that, though they may lead to less efficient energy usage, provide a survival advantage?

Good point, however we're talking quite specifically about the benefits of conceptual knowledge. How does conceptual knowledge make one more likely to survive? Well, I'd argue that one of the main reasons (if not, the reason) is that it reduces a vast amount of perceptual data into a small amount of conceptual data, thus making cognitive processes much easier and more efficient. Surely a reduction of energy consumption would lead to an increase in processing speed?

Of course, there may be numerous other factors at work that I haven't considered yet.

[DarkReaver13]

As you acknowledged, we are not quite up to that level of understanding how we store of information, so far as i know. If I were to hazzard a guess though, I think the gain in efficiency might better understood in terms of recall, especially with regard to pertinent, connected information. If it is correct that information is stored in neural networks, conceptualization might allow reuse of similar pathways. If you imagine an over simplified version where a string of compressed neurons named, x->y->z is how we store the concept mammal, then it would be easy to see how x->y->x->c meant cat and x->y->z->d meant dog. The benefits of accurate concept formation would be that all of those cells that store information about mammals would not need to be redundant and that because they are in the same network, utilizing the same trunk lines, it would be easy to recall conceptually similar concepts from long term memory. Activating one concept in the category brings all the items in that category a step closer to consciousness.

Energy usage is more relevant to short-term and working memory so I don't think proper concept formation would affect it. Having proper categories would probably make storage of new info easier since the category it belonged to could be quickly pulled up. That could affect how much time you spent learning a new thing which would probably mean less energy usage.

Aha! I've actually just written an essay which touched upon some of these ideas - the "associative network model" of cognition, which talks about "priming" neural pathways and the strengthening and/or weakening of links between particular ideas.

Perhaps there is an optimal arrangement or strength of any given connection, however would this not still ultimately lead to an increase in processing speed / power / efficiency? Is there an end result beyond that?

[DavidOdden]

Assuming that concept formation is a tool used to increase one's cognitive potency, you could argue that proper concepts reduce the amount of energy that needs to be used by one's brain in the long-term (initially extra effort is required in actually integrating and isolating the concept, however once this is done it makes subsequent processing much more efficient).

I think this is the wrong way to approach the question. If you have a device that will actually measure energy used by the brain, then you might attempt to prove this scientifically, but the suggestion that "you could argue" belies the implied advantage of reducing the question to low-level brain physiology. The best hope for this approach would be not that it requires more energy to operate in terms of lousy concepts, but rather that lousy concepts limit your existence. Your assumption seems to be that people will expend whatever energy necessary to reach some goal, so that people with lousy concepts will have to work much harder to live the good life. I bet, if you were to study the question scientifically, that you'd find that people who have lousy concepts simply can't live the good life -- they have to settle for less.

The question is, what is the basis in fact for assuming that expending more mental effort will make up for a bad epistemology?

How does conceptual knowledge make one more likely to survive?

This, btw, is a rather different question from the one about the effect of invalid concepts. People who rely on bad concepts do nevertheless have plenty of conceptual knowledge. On the other hand, animals have no conceptual knowledge at all. Conceptual knowledge makes it possible to create medicine, plant crops, build houses, predict weather, generate electricity, sanitize water, compose and record opera, shop for clothing online, and so on, all of which are essential to living.

[aequalsa]

Aha! I've actually just written an essay which touched upon some of these ideas - the "associative network model" of cognition, which talks about "priming" neural pathways and the strengthening and/or weakening of links between particular ideas.

Perhaps there is an optimal arrangement or strength of any given connection, however would this not still ultimately lead to an increase in processing speed / power / efficiency? Is there an end result beyond that?

I would think that it would, but only indirectly. If the correct associations are made, when you try to recall something from long term memory, I would guess that you would spend less time "looking" for it because one concept would lead naturally to a related concept rather then into a bunch of dead ends....This would be as compared to a disconnected list of words which lacked any context.

Not sure what you mean about an end result.

[Thomas M. Miovas Jr.]

We don't know the relationship between cognition and neurology, so that question has to be left to future scientists, but what we can say is that conceptualization is more efficient than dealing with everything on a case by case perceptual level. Man is much more efficient at having enough food to eat because he has a conceptual consciousness and can know when to plant crops and when to reap the benefits. Other animals just have to wait until they fall off the branches or eat them if they can reach them. So I would say the argument for greater efficiency is there, but not yet on the neurological level, since we don't know that relationship. Unfortunately, most scientists who study neurology discount consciousness and so they never ask that question of neurology and consciousness. Trying to improperly integrate sensory material into pseudo-concepts or anti-concepts is certainly not efficient in that it will lead to confusions, but we don't have to get all the way to the neurological level to understand that. So, until we know the connection between consciousness and neurology, that is an unanswerable question.

Added on edit: Also, we already know that conceptualizing takes more energy than not conceptualizing, because it takes a concerted effort to think versus just going with the flow. So, it may not be proper to think of it in terms of efficiency in that sense, rather it might be better to say that thinking, even though it takes more effort,is more efficacious than not thinking, even though one uses more energy in the process of thinking.

Edited October 15, 2009 by Thomas M. Miovas Jr.

[aequalsa]

We don't know the relationship between cognition and neurology, so that question has to be left to future scientists,

Maybe left to current scientists....

http://www.skewsme.com/implants.html

The field of cognitive neuroscience is progressing very rapidly.

[DavidOdden]

The field of cognitive neuroscience is progressing very rapidly.

There have been many remarkable advances in understanding the low-level neural underpinnings of our sensory apparatus and of motor control. What examples do you have in mind that pertain to cognition?

[aequalsa]

There have been many remarkable advances in understanding the low-level neural underpinnings of our sensory apparatus and of motor control. What examples do you have in mind that pertain to cognition?

I think concept formation on the neural level is primarily about memory, in that to form it must first be transferred into memory and be transferred in such a way so as to allow recall later. Primarily this would, I think, be achieved, by utilizing the same pathways which where developed earlier. So visually, the basic parts might be stored as Geons, for example, and then all that would be required for a new memory about a shape, would be a list of the parts and how they fit together.

One relevant example in particular, which I am having trouble finding again, where they attached an electrode to a single cell in the visual cortex of a cat, and the one cell would respond with varying levels of intensity to a rotating line in its field of vision. Also another where an individual cell would activate in response to seeing a picture of michael jordan in a human subject. The website I posted provides quite a few interesting experiments that relate in varying degrees as well.

[Thomas M. Miovas Jr.]

Modern mechanical or electrical implants do not help cognition, but they do help perception, at current levels. Even the electro-mechanical aids in paraplegics don't help cognition directly. The only area I know that we can help cognition is for chemical drugs for things like schizophrenia and other such ailments. In these cases, the drugs do help control the neurology of the patient, though I'm not sure it can be pinned down to better efficiencies of the neurological process. They do help, so something is going right, but I'm just not sure about the efficiency angle. I do think it is clear that if the neurology isn't working correctly, cognition is more difficult; and finding the right drugs for improving neurological functioning is a great fix for these ailments.