From newshub.ccs.yorku.ca!torn!cs.utexas.edu!usc!sol.ctr.columbia.edu!destroyer!uunet!trwacs!erwin Mon Aug 24 15:41:24 EDT 1992
Article 6661 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!torn!cs.utexas.edu!usc!sol.ctr.columbia.edu!destroyer!uunet!trwacs!erwin
>From: erwin@trwacs.fp.trw.com (Harry Erwin)
Newsgroups: comp.ai.philosophy
Subject: Re: Freewill, chaos and digital systems
Message-ID: <704@trwacs.fp.trw.com>
Date: 20 Aug 92 14:47:09 GMT
References: <Bt4xt1.MA0.1@cs.cmu.edu> <702@trwacs.fp.trw.com> <1992Aug19.144240.8058@syscon.rn.com>
Organization: TRW Systems Division, Fairfax VA
Lines: 76

don@syscon.rn.com (Don McLane) writes:

>erwin@trwacs.fp.trw.com (Harry Erwin) writes:

>>rudis+@cs.cmu.edu (Rujith S DeSilva) writes:

>>>(1) Does freewill arise solely through the mechanism of chaos?

>>All the answers you'll get on this one will be speculative. I suspect
>>chaos is part of it, but not all.

>I was just thinking about this, so here's my speculation: I think the
>feeling of freewill may arise from our disconnectedness with the
>past.  The past doesn't seem to determine us; we feel like
>autonomous agents.  But, if the world is chaotic, our present
>mental state is the result of uncountable insignificant details.
>We can't trace the causal chain backwards, so we feel free.

>So, I say yes, the feeling (illusion?) of freewill arises solely
>through the mechanism of chaos.

>		   ---Don

The purest case of free will occurs when you're playing a social game that
involves outlasting your opponent where the winner is determined by some
parameter that can be learned during the game. In this case, the goal is
not to outstay the value of the game to you. If no information is
collected during the play of the game, you should stick with it until the
cost so far equals your expected gain. If information collection is added,
then bluffing strategies become viable, where you bluff by staying in. The
population strategy for this game is unstable and actually evolves
chaotically.

Imagine yourself playing this game and approaching the point where
bluffing strategies kick in. It is unwise to base your strategy on the
occurance of some event that can be predicted, so you just wait until the
time is "right," where you quit. But meanwhile you have to keep a poker
face. When do you quit? When your free will says to quit.

Has anyone studied the mental processes underlying bluffing strategies in
poker? I think that's the closest we've probably come to understanding
free will. I do know about the experiments that demonstrated that the
unconscious prepares for actions some time before the will freely decides
to implement them or not, but I'm more interested in the decision
processes than in the preparation for decision. Are there signals
detectable that always occur before a poker player drops out and never
otherwise? It would be interesting to find out.

BTW, I expect that chaotic dynamics are used in the brain for rapid
pattern matching and for novelty generation. They may be used as well for
controlling free-will by having the abandonment of a quasistable state
trigger the free decision.

I'm looking at a model of the free will as a quasiperiodic chaotic
process with a dominant periodic component of about 7 independent
substates. Our perception of "guiding" this process is hard to explain.
We appear to have control over a collection of goals with payoffs
that this process generates plans to achieve, but when I observe the
presentation of substates, I find I have less control over their
generation that I would expect. Rather, I seem to be able to emphasize
some that seem of value and deemphasize others. Eventually the cycle
stabilizes and I can then store it and generate another cycle to observe.
Creativity seems bound up in the generation of candidate cycles, but I
strongly suspect that generation is a chaotic process that I have little
or no control over. Free will is selection, and may reflect the common
experience of having many nearly equivalent substates to emphasize or
deemphasize. Or it may reflect the common experience of waiting for the
"right" time.

Cheers,
for achieving.
cognition
-- 
Harry Erwin
Internet: erwin@trwacs.fp.trw.com



