FleshFactor FleshFactor
logo; AEC FORUM

[Prev][Next][Index][Thread]

FleshFactor: -----the brain is a wet machine



---------------------------------------------------------
A E C  F O R U M - "F L E S H F A C T O R"
(http://www.aec.at/fleshfactor/arch/)
---------------------------------------------------------


I have enjoyed reading the ideas presented by Laurie McRobert, Rick Nance,
Brian Molyneaux, Robert Adrian, Dean Pignon and others on chaos, wet brains
and artificial intelligence. I note with appreciation Laurie McRobert's
reference to our work on protein-based computing, but wish to clarify one
thing first.  One of the reasons my group (and many others in Europe, Japan
and Israel) are exploring the use of bacteriorhodopsin is that this protein
is not limited to binary logic, but in fact, defaults (because of its
multi-state photocycle) to three-state or higher logic.  Thus, most of the
devices that we are working on that simulate neural processes rely on
either three-state behavior (yes,no,maybe) or holographic association
(Fourier logic).

I agree with Rick that "protein based circuits (wetware) aren't enough".
We are not even close to a level of artificial intelligence even
approaching animal behavior, let alone human behavior.  We can do a fairly
good job of simulating human memory recall, in that large-scale holographic
associative memories not only simulate the response characteristics of
humans in pattern recognition, but show apparent creativity in responding
to new stimuli, when no directly relevant data are present.  Some of us
believe that human creativity is best described as "faulty" association,
the coupling of two or more disparate concepts in an apparently unjustified
fashion from different realms of experience to create a new concept.
Digital computers and standard (If:then:else) software fail at this,
because they are binary in nature, and "faulty association" is quite
impossible (the program either fails to find anything, or loops
indefinitely).  Attempts to force neural responsiveness via software which
is constrained to digital machinery is extremely difficult, and has yet to
succeed (many think it never will).

I end with a personal observation.   It returns to the issue of whether we
are machines or not, and ties into our discussion of the brain.   We are
biological machines based on complex symbiosis.  That is, our large-scale
behavior is the complex sum of the behavior and interactions of all the
small organisms that make up our body.  Each of these organisms can be
analyzed as simple machines, and our complexity is simply due to the
combination of many simple (and elegant) architectures.  Evolution has
combined millions of different simple organisms into a complex system that
has been further refined by trial and error to produce what we are today.

We think because evolution had enough time to create a thinking system by
trial and error.

We have a soul because we think we have a soul.

But the bottom line is we are machines, and there is nothing wrong with
being a machine.  After all, it took nature 3.5 billion years to evolve
from simple one-celled organisms to what we are today.  Given enough time,
humans can make thinking machines with personalities and what, for all
intents and purposes, are souls.

The question that remains to be answered is how long this will take, and
will the human race survive long enough to create intelligence and machines
with souls.  A second question is will these artificial souls have a
quality worth preserving.


Bob Birge


------------------------------------------------------------------
        Robert R. Birge, Distinguished Professor of Chemistry,
             Director, CASE and W. M. Keck Centers,
             Syracuse University,111 College Place,
                  Syracuse NY 13244-4100, USA
                        email: rbirge@syr.edu
        voice: 315-443-1900; fax: 443-4070; lab: 443-5928
------------------------------------------------------------------


--------------------------------------------------------------------
to (un)subscribe  the Forum just mail to
fleshfactor-request@aec.at (message text 'subscribe'/'unsubscribe')
send messages to fleshfactor@aec.at
--------------------------------------------------------------------


[FleshFactor] [subscribe]