From newshub.ccs.yorku.ca!torn!utcsri!rpi!zaphod.mps.ohio-state.edu!moe.ksu.ksu.edu!ux1.cso.uiuc.edu!mp.cs.niu.edu!rickert Wed Aug 12 16:51:53 EDT 1992
Article 6525 of comp.ai.philosophy:
Newsgroups: comp.ai.philosophy
Path: newshub.ccs.yorku.ca!torn!utcsri!rpi!zaphod.mps.ohio-state.edu!moe.ksu.ksu.edu!ux1.cso.uiuc.edu!mp.cs.niu.edu!rickert
>From: rickert@mp.cs.niu.edu (Neil Rickert)
Subject: Re: Memory and store/retrieve.
Message-ID: <1992Jul30.154557.24844@mp.cs.niu.edu>
Organization: Northern Illinois University
References: <1992Jul28.194953.7337@puma.ATL.GE.COM> <1992Jul29.165648.1525@mp.cs.niu.edu> <1992Jul29.225334.750@dirac.physics.sunysb.edu>
Date: Thu, 30 Jul 1992 15:45:57 GMT
Lines: 80

In article <1992Jul29.225334.750@dirac.physics.sunysb.edu> charles@dirac.physics.sunysb.edu (Charles Ofria) writes:
>I think that the main problem we are suffering from here with determining if
>memory and store/retrieve is required for intelligence is our definitions of
>these terms.

  Definitions seem to be a continual problem in the discussion of
intelligence.

>              In his last post Neil mmade it clear that he was talking about
>stores that will last and can be retrieved at any point in time.

  I think this is what most people mean when they use the term "memory".
The term has "memory" has been muddied recently.  Twenty years ago most
computer books used the term "memory" for disk, tapes, punch cards, etc,
and used the term "storage" or "core" for the short term storage in the
machine.  Unfortunately usage has changed, and we no longer have a good
choice of words to make a simple but clear distinction.

>                                                                  This is a
>very strict definition and indeed is not needed for intelligence.  Some form
>of memory certainly is though.  You must at least be able to carry some idea
>from one thought to the next - or even from the begining of a thought to the
>end - and this IS memory!

  I actually question this too.  But only because my personal definition
of intelligence is very liberal, and would include bacteria as organisms
exhibiting some (very small) degree of intelligence, although I would
not credit them with having thoughts.

>When I suggested that commmunication was not needed, Neil immediately pointed
>out all the minor communications of neurons and such in the human mind.

  Our difference here is that I do not agree this is minor communication.
I see it as major.  As I see it, all substantial intelligence is
constructed with communication.  But then I see a computer as a network
of communicating AND and OR gates.

  From my perspective, we do not know anything about how to create
intelligence.  But certain natural physical phenomena (semi-conductors,
protein chemistry) already exhibit small degrees of intelligence.  We
construct more advanced intelligence by harnessing these physical
phenomena.  This requires complex communication networks.

>                                                                         This
>form of communication is certainly needed, but I think we suffered from the
>same problem there with differing ideas of what we meant by communication.

  I admit I probably shortchanged your comments on communication.  As
I have indicated, I cannot consider communication between neurons as
'minor'.  But I think we can reasonably distinguish between internal
communications and external communications.

  In your earlier comment, I believe you were discussing external
communications.  Since you did not make this precise I answered with a
comment about internal communications, mostly to point out the lack of
precision.  But let me now comment on external communications.

  My instinctive reaction is the same as yours - namely that if we sever
the communications (external communications, that is), the intelligence
remains.  But instincts can be wrong, and there is much evidence to the
contrary.  If a child suffers severe sensory deprivation, it is known
that his mental development (and probably his physical development) will
be highly abnormal.  It is my understanding that if a deaf child is not
recognized as deaf by his parents, so that no compensatory communication
(sign language, reading, etc) is substituted, then the physical growth
of the brain is abnormal.  So, at least with humans, communication seems
to be quite important.

  If we look at a computer, we can ask if external communication is
important there too.  But if you cut off all external communication,
you remove the ability to change the programming, or modify the data.
The computer is relegated to running the same computations over and
over on the same data.  I think most of us would say it was a dumb
automaton, and was not exhibiting intelligence.

  I guess that what this boils down to, is that we judge intelligence
by the interaction of the intelligent unit with its external world.  In
this sense, perhaps a behavioral definition of intelligence is the only
valid definition.



