Re: Economy of ideas

Lars wrote:
Economy-Problem
---------------------------

1. Ideas are representations of perceptions. An idea is represented by
multiple neurons.
2. Human thinking is based on assoziativity of ideas.

Most people will agree with these two sentences. But they lead to the
following problem:
Imagine A is an idea that is associated with the ideas B and C.

A ------>B
|
|-------->C

Imagine A is conscious. How does the human brain manage that only B or
C is conscious at a time? How does the human brain manage that not B
or C are both conscious at a time?

There is almost the same problem if you deny that consciousness is
something special. The question is then: How does the human brain
manage that only one action is done at a time?

There must be a biological mechanism that adjusts the number of active
ideas at a time. Has anyone heard of such a biological mechanism? Has
anyone been confronted with this problem in AI-programming? (Note that
A,B,C are not single neurons but representations of ideas.)

Same as in any other evolutionary system. The number of individuals is trimmed to fit the resources available by killing off the 'weaker' ideas.

Inhibitory links in neural networks have been used to model this process, but in an actual brain there are probably multiple complex pruning mechanisms, some of which use reinforcement learning to determine what is a good idea.

The best heuristic I can think of at the moment would be something like
'An idea that can derive the same conclusions as a more complicated idea/group of ideas gets to replace the more complex idea/group.'
.

Relevant Pages

  • Economy of ideas
    ... Ideas are representations of perceptions. ... multiple neurons. ... How does the human brain manage that not B ... There is almost the same problem if you deny that consciousness is ...
    (comp.ai.philosophy)
  • Re: Curts AI epiphany
    ... how consciousness works? ... a cognitive agent (such ... as a person) begins with representations. ... Harnad wrote a paper on "The Symbol Grounding Problem" (Physica D, ...
    (comp.ai.philosophy)
  • Re: Curts AI epiphany
    ... how consciousness works? ... a cognitive agent (such ... as a person) begins with representations. ... There is no symbol grounding problem. ...
    (comp.ai.philosophy)
  • Re: The Evolutionary Challenge of Self Referencing Representations
    ... Why couldn't the midbrain house representations? ... If other things are needed, too, to have this that or the other kind of consciousness, that would be no problem for us. ... My own inclination would be to think that a midbrain alone could be conscious of the world around it and some states of its own body, but not of being conscious of these things or or most of its representations. ... I think it most unlikely that any small group of neurons could be any conscious state. ...
    (sci.psychology.consciousness)