From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!news-server.csri.toronto.edu!rpi!usc!cs.utexas.edu!uunet!tdatirv!sarima Thu Dec 26 23:57:04 EST 1991
Article 2268 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!news-server.csri.toronto.edu!rpi!usc!cs.utexas.edu!uunet!tdatirv!sarima
>From: sarima@tdatirv.UUCP (Stanley Friesen)
Newsgroups: comp.ai.philosophy
Subject: Re: How does one model anesthesia in AI?
Message-ID: <336@tdatirv.UUCP>
Date: 19 Dec 91 01:02:48 GMT
References: <59809@netnews.upenn.edu> <12665@pitt.UUCP> <60022@netnews.upenn.edu> <12686@pitt.UUCP> <60318@netnews.upenn.edu> <322@tdatirv.UUCP> <60554@netnews.upenn.edu>
Reply-To: sarima@tdatirv.UUCP (Stanley Friesen)
Organization: Teradata Corp., Irvine
Lines: 43

In article <60554@netnews.upenn.edu> weemba@libra.wistar.upenn.edu (Matthew P Wiener) writes:
|In article <322@tdatirv.UUCP>, sarima@tdatirv (Stanley Friesen) writes:
|>Most neural blockers only block *one* neurotransmitter, or perhaps a family
|>of related ones.  Perhaps the neural transmitters that transmit the 'learn'
|>signal are not effected by the currently used general anesthetics.  Or they
|>do not effect the modulation effects of the glia.
|
|So you are proposing multiple networks using the same wiring but different
|signals and weights?  Admit it, you don't believe in Occam anymore!

Goodness no!  I am just observing that the biological NN's are not composed
of a single, uniform node type.  Rather they have different types of nodes,
each of which has different functional properties.  They are still connected
up as a single network (unless someone has performed a lobotomy).

This is *not* a proposal, it is an observational fact.  That *is* how brains
are structured.  It is a standard concept of neurology that each neuron has
a single primary neurotransmitter, and that each neurotransmiter has a
different function in the brain.  [This has been modified recently to allow
the existance of secondary transmitters in a single neuron - such as
endorphins].

It is also the single most ignored aspect of brain anatomy in digital NN
research.

|>In short, the 'odd' features of anesthetics are due to the particular physical
|>mechanism used in living brains, and is not relevant to cognition per se.
|
|On general principles, I am dubious of biological intelligence existing
|without consciousness and vice versa.  Why not?  Evolution seems to have
|given us both together.  I conclude that if an animal has enough intelli-
|gence, it will categorize itself in a unique manner, leading to conscious-
|ness, and that if an animal is conscious of itself, it will accelerate the
|selection process for intelligence.

Yes, I quite agree.  At least if you take 'intelligence' and 'consciousness'
as properties with varying degrees, rather than simple binary properties.

I also suspect that we will find the same holds for AI entities.
-- 
---------------
uunet!tdatirv!sarima				(Stanley Friesen)



