Newsgroups: comp.ai,comp.ai.alife,comp.ai.philosophy
Path: cantaloupe.srv.cs.cmu.edu!nntp.club.cc.cmu.edu!hudson.lm.com!netline-fddi.jpl.nasa.gov!elroy.jpl.nasa.gov!swrinde!pipex!uunet!timbuk.cray.com!walter.cray.com!mwd
From: mwd@cray.com (Mark Dalton)
Subject: Re: Thought Question
Message-ID: <1995Jan2.203350.4362@walter.cray.com>
Followup-To: comp.ai.alife,comp.ai.philosophy,comp.ai,alt.consciousness
Lines: 81
Nntp-Posting-Host: pajarito.cray.com
X-Newsreader: TIN [version 1.2 021193BETA PL3-CRIb]
References: <3dfhkq$gov@news.worldlink.com> <Shanks.179.2EFFA0C1@us.net><3dprpk$6a@bbs.pnl.gov> <34@reservoir.win-uk.net>
Date: 2 Jan 95 20:33:50 CST
Xref: glinda.oz.cs.cmu.edu comp.ai:26183 comp.ai.alife:1612 comp.ai.philosophy:24236

Shane McKee (shane@reservoir.win-uk.net) wrote:
:  
: Marty Grogan (ml_grogan@pnl.gov) writes:
: >I doubt that machines will ever independently discuss "humelligence" using
: >"machelligence"-developed languages.  If they did, who or what would know
: >or care?
: >
: I still haven't had anybody (including Roger Penrose) give me a
: good reason why the human brain can't be regarded as a computer.
: OK, so the mind sometimes doesn't work algorithmically, but its
: constituent molecules do obey known (or at least simulable)
: physical laws. Why -can't- intelligence be an emergent property of
: an underlying algorithmic process involving these constituents?
:  

I am crazy, but I will take a stab at a little part of this. (I should not
touch, but oh well.  (^8 ).

	Using the analogy of a computer is a 'valid' analogy, but that is
all it is.  There are weaknesses in any anology.  Yes, they both process
information, but they are very different, and the parellelism makes them
a little more similar.  However, brains (from my LIMITED understanding)
are very different in that they change 'on the fly' - rewire, change
conformations, etc..

  1. When a path in the brain is used for task 'A' repeatedly, then (I think)
     task 'A' will then need less of a threshold to cause action 'A'.
  2. As for memory, (I saw something on PBS about this but may simplify or be
     wrong on, feel free to clear this up anyone), when a memory occurs there
     are structural changes in the brain that store the memory, and I would
     guess the more this is accessed, the 'easier' the path, and the more
     structural integrety will occur.
  3. Associations are very strong (many cross-references).  Orange to me
     will cause immediate associations of the smell, image and taste of the
     fruit causing my mouth to water; and also bring to mind a orange VW bug
     with a little rust.
  4. There is 'intensity' and 'repeatedness' that cause the memory to 'stick'
     more.  I think it was Aristotle or Plato that used to slap thier students
     when they were to remeber something.  Strong/sharp/strange associations 
     are what they use in many of the memory techniques to help people memorize.
     On Old Olympus Towering Tops a Fin and German Viewed Some Hops (this worked
     until they changed the names of one of the Cranial nerves).
     It is probably the oddity/intenseness that causes a faster structural
     change.  If you saw a dinosour walk down your street you would probably
     remember that, but if you saw a dog, you may not even notice, because
     of the dogs commoness and the dinosour would be VERY rare.
	- Computers don't distinguish between the two very well.
  5. Certain parts have certain functions and can modify themself to take
     over tasks 'on the fly'.
  6. Susumption architecture robots by Rodney Brooks was probably one of
     the closer ideas.  The idea is that different areas have different
     functions and different priorities.  Planning is less important than
     avoiding the truck comming straight at you.
	I would have like to seen that increased to little cpu/transitor
	clusters, that would work toghether (and were repeatitive) so that
	they could take the place if some failed, or a lot needed to be
	done at once.  Or the next function over was 'zapped' and needed
	to be replaced.
   7. Image analysis is a real hard problem for computers, but little babies
      can do this very fast.  I think this is done by taking SMALL features
      and placing them together to form a 'thing'.  Sort of like a hash tree.
      Has 4 legs, is black and white, and about 4 feet tall - a cow.  Because
      some kids will if you are driving down the road will see cows but because
      of the distance look smaller so they say 'doggies'.  But they maybe
      for other reasons.  I don't know!

Well these are just a few ideas differences.

Anyone else??  I will think more about this, and see what I cannot formulate.

Mark  	(Disclaimer: I am sure my company would not take anything I say
		     seriously, so don't hold them accountable for me (^8  ).
------------
Mark Dalton       CH3-S-CH2 H H                    H      O       H
Cray Research,Inc.      |   | |                    |       \      |
Los Alamos,NM 87544     CH2-C-COO    //\ ---C--CH2-C-COO    C-CH2-C-COO
mwd@cray.com                | |     |  ||   ||     |       //     |
                            H NH2    \\/ \ / CH    NH3    O       NH3
                                          NH
URL = http://lenti.med.umn.edu/~mwd/mwd.html

