From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!utgpu!cs.utexas.edu!sun-barr!olivea!spool.mu.edu!wupost!darwin.sura.net!gatech!ncar!noao!amethyst!organpipe.uug.arizona.edu!NSMA.AriZonA.EdU!bill Thu Feb 20 15:22:18 EST 1992
Article 3885 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!utgpu!cs.utexas.edu!sun-barr!olivea!spool.mu.edu!wupost!darwin.sura.net!gatech!ncar!noao!amethyst!organpipe.uug.arizona.edu!NSMA.AriZonA.EdU!bill
>From: bill@NSMA.AriZonA.EdU (Bill Skaggs)
Newsgroups: comp.ai.philosophy
Subject: Re: Brain waves
Message-ID: <1992Feb19.233912.15326@organpipe.uug.arizona.edu>
Date: 19 Feb 92 23:39:12 GMT
References: <17872@castle.ed.ac.uk> <66025@netnews.upenn.edu> <3413@novavax.UUCP> <13493@pitt.UUCP>
Sender: news@organpipe.uug.arizona.edu
Reply-To: bill@NSMA.AriZonA.EdU (Bill Skaggs)
Organization: Center for Neural Systems, Memory, and Aging
Lines: 25

gowj@novavax.UUCP (James Gow) writes:
>Does a neural net produce something akin to a brain wave? [ . . . ]

geb@dsl.pitt.edu (gordon e. banks) writes:
>I'm not sure how well the neural nets reproduce EEGs.
>I think they have shown some promise.

EEG's are produced by electric current flowing between different
parts of neurons, for example from dendrites into the cell body.
Most of the neural nets being studied today use "connectionist"
units, which have no spatial structure, so they don't generate
anything like an EEG.

With almost no exceptions, units with spatial structure are used
only by people interested in biological realism (like me).  There
are several computational models of parts of the brain that
reproduce reasonably accurately the EEG's recordable in those
parts.  Probably the best known is Roger Traub's model of the
CA3 region of the hippocampus, which is described in a book
he recently wrote ("Neuronal networks in the hippocampus", by
Traub and Miles).

By and large, the functions of EEG rhythms remain a mystery.

	-- Bill


