how brain works
Jan. 25th, 2006 08:32 am![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
sab123I wrote a post on some message board about how I think the brain works. And it looked good, so I've felt that I should not let it go to waste but store it somewhere more permanent. It's funny how these biologists are arguing about things which seem pretty obvious to anyone who worked on computer and software development. So here we go:
I'd say that all the actual thinking is done with the "abstract images" (or "abstract entities", if you prefer, to avoid any terminological confusion with visual images). The language is merely a way of input-output of the information, and the translation between the internal representation and the external one (language) happens on the way.
I think the best illustration of this independence is that there are obvious cases where the translation unit breaks or can't keep up in speed with the thinking unit but the thinking unit works fine. Personally I often forget words for what I want to say (and end up saying "that thingy" or "this thingy").
In math the formula language is really a way to structure information and split the big problem into a number of small problems, and then solve each of them separately one by one without the need to remember the intermediate results (since they are written on paper, and looked up from the paper when they are needed for further calculation).
I feel that the information is stored inside the brain by associative connections. That is, each abstract image has a number of connections with other abstract images. Some of these abstract images are actually the visual images. For example, the abstract image which we would label "table" is connected to the abstract image "table in my dining room" which is connected to the visual images of the said table. Naturally, that abstract image "table" would be also connected to the abstract image of the word "table" in the language (or many such images if this particular person speaks multiple languages - or, specifically in English, probably separate abstract images for the spoken word and the written word).
And the thinking itself seems to work by turning indirect associative connections into direct ones. As a simple example, "here are two abstract images, is there any path connecting them together?".
If you think about the IQ tests, that's what they measure: the ability to find such connections between the presented units of information.
An interesting consequence is the model why it's so difficult to explain or to learn the complicated things. Inside the head all these connections exist in parallel and are used in parallel. But when explaining something, it must be translated to the language, which is sequential by its nature. So everything has to be serialized when it's being translated to the language. Which is a pain to do well, as anybody in the computer programming business knows. The same problem comes in when reading (or hearing) the explanation, to translate from the language to the internal abstract images and form the correct connections between these images.
And things get complicated further since the same words of the language are connected to somewhat different abstract images in the heads of different people. That's why you can often see that you take some explanation and give it to two people, and the first one would understand it well, and the second one won't. Then you take some other explanation and give it to the same two people, and the first one doesn't understands it, and the second does. It's because the abstract images involved in the first explanation when translated to and from the language have matched well to the abstract images in the head of the first man but not the second man, and the opposite has happened with the second explanation.
I'd say that all the actual thinking is done with the "abstract images" (or "abstract entities", if you prefer, to avoid any terminological confusion with visual images). The language is merely a way of input-output of the information, and the translation between the internal representation and the external one (language) happens on the way.
I think the best illustration of this independence is that there are obvious cases where the translation unit breaks or can't keep up in speed with the thinking unit but the thinking unit works fine. Personally I often forget words for what I want to say (and end up saying "that thingy" or "this thingy").
In math the formula language is really a way to structure information and split the big problem into a number of small problems, and then solve each of them separately one by one without the need to remember the intermediate results (since they are written on paper, and looked up from the paper when they are needed for further calculation).
I feel that the information is stored inside the brain by associative connections. That is, each abstract image has a number of connections with other abstract images. Some of these abstract images are actually the visual images. For example, the abstract image which we would label "table" is connected to the abstract image "table in my dining room" which is connected to the visual images of the said table. Naturally, that abstract image "table" would be also connected to the abstract image of the word "table" in the language (or many such images if this particular person speaks multiple languages - or, specifically in English, probably separate abstract images for the spoken word and the written word).
And the thinking itself seems to work by turning indirect associative connections into direct ones. As a simple example, "here are two abstract images, is there any path connecting them together?".
If you think about the IQ tests, that's what they measure: the ability to find such connections between the presented units of information.
An interesting consequence is the model why it's so difficult to explain or to learn the complicated things. Inside the head all these connections exist in parallel and are used in parallel. But when explaining something, it must be translated to the language, which is sequential by its nature. So everything has to be serialized when it's being translated to the language. Which is a pain to do well, as anybody in the computer programming business knows. The same problem comes in when reading (or hearing) the explanation, to translate from the language to the internal abstract images and form the correct connections between these images.
And things get complicated further since the same words of the language are connected to somewhat different abstract images in the heads of different people. That's why you can often see that you take some explanation and give it to two people, and the first one would understand it well, and the second one won't. Then you take some other explanation and give it to the same two people, and the first one doesn't understands it, and the second does. It's because the abstract images involved in the first explanation when translated to and from the language have matched well to the abstract images in the head of the first man but not the second man, and the opposite has happened with the second explanation.
