From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!news-server.csri.toronto.edu!bonnie.concordia.ca!uunet!boulder!boulder!eddy Mon Dec 16 11:01:55 EST 1991
Article 2122 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!news-server.csri.toronto.edu!bonnie.concordia.ca!uunet!boulder!boulder!eddy
>From: eddy@boulder.Colorado.EDU (Sean Eddy)
Newsgroups: comp.ai.philosophy
Subject: Re: From neurons to computation; an example worm
Message-ID: <eddy.692726825@beagle>
Date: 14 Dec 91 16:07:05 GMT
References: <1991Dec14.110633.28844@oracorp.com>
Sender: news@colorado.edu (The Daily Planet)
Organization: University of Colorado, Boulder
Lines: 58
Nntp-Posting-Host: beagle.colorado.edu

I've been reading this thread with a great deal of interest.  As a
molecular biologist, it has been striking and eye-opening to me that
there seems to be a great deal of resistance to the idea that we can
learn a lot about the human brain (and, possibly, eventually, the
human "mind") from simpler biological systems.

I thought I might try to clarify some of Gordon Banks' points
by better describing the model slug and worm nervous systems
he's been bringing up.

The slug in question is almost certainly _Aplysia californica_, a
large and ugly marine snail. Aplysia has about 100,000 total neurons.
Eric Kandel and coworkers at Columbia University have studied one
particular neural circuit in fine detail, a circuit responsible for a
gill withdrawal reflex. This circuit contains about 24 mechanosensory
neurons, 6 motor neurons, and some other interneurons. The cells are
large enough to impale and obtain intracellular electrical recordings.
The circuit shows characteristics of habituation and sensitization,
which Kandel has been able to study.

The worm in question is probably _Caenorhabditis elegans_, a small (1
mm long) common soil nematode which has proven to be an immensely
useful model system for studying processes of animal development. The
advantage to working in C. elegans is that the worm only has 959 cells
-- and every individual seems to have very nearly the same number,
identity, and location of cells. Other advantages are that C.  elegans
is suitable for the sorts of sophisticated genetic analysis you're
used to hearing from the fruit fly (Drosophila) people, and that
molecular biology in the worm is probably the most accessible of any
model animal system.

C. elegans has 302 total neurons. We have rough guesses for the
functions of nearly all of them, based on their morphology and what
happens to the worm's behavior when you shoot the neuron out with a
laser microbeam. The placement of the circuitry and 10,000 or so
synaptic connections are also mostly invariant between individuals,
and have been mapped for all 302 neurons by John White and coworkers
at the MRC.  In the past few years, several people -- including myself
:) -- have become interested in exploiting this invaluable groundwork
and using C. elegans for a large-scale genetic attack on the molecules
that make a small nervous system function.

Now, I'm not going to argue that we will discover much about human
consciousness from how a worm writhes. But I do think that it's fair
to say that we biologists don't yet fully understand some basic first
principles of how a neuron functions or a set of neurons interconnect
(*particularly* how they interconnect), and that we may as well learn
those basic principles in the simpler and more accessible systems
first. It may, in time, be necessary to posit new emergent properties
to explain "mind" --- but I, and probably most other molecular
biologists, am not going to worry terribly about that 'til I get
there. :) 


- Sean Eddy
- Department of Molecular, Cellular, and Developmental Biology
- University of Colorado at Boulder
- eddy@boulder.colorado.edu


