2007-04-22
Thinking soup
My daughter fell on the ground and had hurt her hand. One full drop of blood came out of the palm of her hand. A little catastrophe. After many tears and some tender care there was a serious problem. ‘Daddy, is it my left-hand or my right-hand?’ ‘Don’t you know anymore? Which hand do you take your pencil when you make a drawing?’ ‘My right-hand. But is this hand my right-hand or my left-hand?’ ‘Do you use that hand for writing?’ ‘I don’t know.’ I handed her a pencil. Of course she took it immediately with her right hand. ‘So, this is my right-hand?’
This is one of the smallest experiences of what I think might be called embedded cognition. Jelle can tell you much more about this. See his weblog ant on the beach.
In brief my understanding is that our knowledge is not completely in our brain. Our brain contains little pieces of knowledge, associations between events. We don’t have a complete program. The responses of our brain depend on our previous response or thought, the signals it gets from the body and the situation we are in. In fact it depends on the complete environment.
Another example is the performance of a dance, for example salsa. A dance is the result of the actions of 2 people, the music and small events in the environment, such as a slippery floor. It’s very difficult for me to tell exactly what all the moves are in certain figures. But still I can dance them. I know the move when it’s time to make it. I happen to be in a certain position, my lady happens to be in a certain pose and then I know what to do. Sometimes it happens that my lady is in a slightly different pose and then (when I’m lucky) I also ‘know’ what to do and make another move that I thought was the right one. The figure suddenly evolves into another figure. After such a figure my dance partner sometimes asks to do that again. Quite embarrassing, because I can’t even remember what I did. The figure just emerged.
So how do we think? Do we think? Are our actions just reflexes? (Ah! finally I’m back on reflections/reflexions). Last week I was browsing through one of those piles of papers that are scattered through my home and I found a reference to an interesting idea that has not yet been taken seriously enough. It’s very simple mathematical model of the associative brain by NG de Bruijn: A model for information processing in human memory and consciousness (1994).
In brief, the model contains two mechanisms: a huge number of simple cells that are able to store a simple relation A->B and an activation mechanism that enables those cells randomly for a certain time window. Enfin, the result is a ‘thinking soup’ that is able to store associations and might be able to (re)produce knowledge. With the computing power we have today it must be possible to make a simulation of this ‘thinking soup’. And I think this thinking soup might be fit into the embedded cognition theory.
This is one of the smallest experiences of what I think might be called embedded cognition. Jelle can tell you much more about this. See his weblog ant on the beach.
In brief my understanding is that our knowledge is not completely in our brain. Our brain contains little pieces of knowledge, associations between events. We don’t have a complete program. The responses of our brain depend on our previous response or thought, the signals it gets from the body and the situation we are in. In fact it depends on the complete environment.
Another example is the performance of a dance, for example salsa. A dance is the result of the actions of 2 people, the music and small events in the environment, such as a slippery floor. It’s very difficult for me to tell exactly what all the moves are in certain figures. But still I can dance them. I know the move when it’s time to make it. I happen to be in a certain position, my lady happens to be in a certain pose and then I know what to do. Sometimes it happens that my lady is in a slightly different pose and then (when I’m lucky) I also ‘know’ what to do and make another move that I thought was the right one. The figure suddenly evolves into another figure. After such a figure my dance partner sometimes asks to do that again. Quite embarrassing, because I can’t even remember what I did. The figure just emerged.
So how do we think? Do we think? Are our actions just reflexes? (Ah! finally I’m back on reflections/reflexions). Last week I was browsing through one of those piles of papers that are scattered through my home and I found a reference to an interesting idea that has not yet been taken seriously enough. It’s very simple mathematical model of the associative brain by NG de Bruijn: A model for information processing in human memory and consciousness (1994).
In brief, the model contains two mechanisms: a huge number of simple cells that are able to store a simple relation A->B and an activation mechanism that enables those cells randomly for a certain time window. Enfin, the result is a ‘thinking soup’ that is able to store associations and might be able to (re)produce knowledge. With the computing power we have today it must be possible to make a simulation of this ‘thinking soup’. And I think this thinking soup might be fit into the embedded cognition theory.
