Owen Nieuwenhuyse wrote:

> I don't think that is particularly relevant to what I am trying to do.

> 

> I am building a more comprehensive model of human mental

> processes in symbolic terms.



I've got some stuff you could be interested in.



You need to seperate what you're dealing with to:

 - Emotions

 - Emotion-type learning

 - Processing (includes info keeping process)

 - Social interactions.



I'd say it is clear what emotions are and how they effect our behaviour.

They're chemicals that effect our behaviour analog-wise (usualy by a

gauss curve with an unknown offset; nothing, little, more, lots, less,

little, nothing).



I've once simulated emotion-type learning on a computer: A pixel on a 2D

plate full of coloured squares (representing situations). The pixel

would wonder around the plate, some colours would mean a crash, some a

nothing; the pixel would disslike crashing. It had a memory bank of

about 20 slots, each slot could hold one colour with a good or bad

association. If the pixel encountered a colour that resulted in a crash,

it would associate it to bad. If a long time no crash was sensed on a

specific colour, it was associated to good. The more enteries in the

memorybank, the higher the priority. When the pixel was heading towards

something associated to bad, it would panic (quickly alternate

direction), if it was heading towards something associated to good, it

would keep the direction and speed up. It was quite a perfect simulation

of this behaviour, when I later compared it to actual cases, I found out

the simulation was right.



The processing in preformed by corexes in the brain. Everything is

hardwired, except the contens of the memory banks of course. Much like

as in a computer.



Social interactions are one of the tougher parts in this. I've lately

focused on the master-slave type interactions, since they are quite

common and have the toughest effects. It's terrible and yet great. It's

easiest described with this question: "Would you allow me to manipulate

with you, if I assured you a perfect life?"



> The objective is to build a "belief structure" that is acceptable to

> me, so that I can cast commands within that structure.



You're complicating things without a need for such. There are some

truths around that what you belive is completely meaningless and that a

mind can reprogram itself in the process of being programmed by

something else. The results can be catastrophic, so I suggest you stay

out of these terms.



I got myself screwed-up in the process of programming my understanding

that has been programed at the time by my environment. All my actions,

including the programming I was preforming were controled by the one

that has been programming me (including myself of course). This spirale

of reprogramming deleted most of the meaning in one of my personalities. 



But who am I to judge what is "screwed-up" and what "normal"? One of my

personalities has it's own backup and is running fine and stupid, the

other personality however now understands the entire system but knows

less meaning. It feels depressive to the personality that has a backup,

but the other one knows it is all just nonsense. Is it?



What am I saying? Never think about the "belief structure" or anything

allike!



> I think that the brain will then adapt it's methods to suit the belief

> structure.



Hm. The brain has one and only method of functioning and does not adapt

it to anything. It's like a PC: One can adapt the signal routing method

(for example from DMA to Programmed I/O), but it can't possibly change

the fact that the cards are sitting in slots on the mother-board.



I gather from here that your brain does not have to do any adaptations.



> It is a little hazardous, because I run the risk of locking myself into

> a mind-set that may make some normal behavior more difficult to

> access, rather than less difficult.

> 

> To counter this effect, I intend to continue with standard re-training

> routines as well.



This is true, however do keep in mind that that risk appears everywhere

and is most dangerous because you cannot detect it.



--



Don't feel bad about asking/telling me anything, I will always gladly

reply.



Digging for info? Try AI Meta Search:

Http://WWW.AIMetaSearch.Com



GTSC4 -- If nobody else wants to do it, why shouldn't we?(TM)

Meson & GTSC4 are now merged:

Http://WWW.MesonAI.Com