Newsgroups: comp.ai.alife,comp.ai.philosophy,comp.ai,alt.consciousness
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!howland.reston.ans.net!ix.netcom.com!netcom.com!jqb
From: jqb@netcom.com (Jim Balter)
Subject: Re: Thought Question
Message-ID: <jqbD2o9po.7L3@netcom.com>
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
References: <3eh97p$8oh@agate.berkeley.edu> <sa209.104@utb.shv.hb.se> <1995Jan12.022935.26572@Princeton.EDU> <3fmba3$nmf@vixen.cso.uiuc.edu>
Date: Thu, 19 Jan 1995 21:44:12 GMT
Lines: 54
Xref: glinda.oz.cs.cmu.edu comp.ai.alife:1859 comp.ai.philosophy:24805 comp.ai:26632

In article <3fmba3$nmf@vixen.cso.uiuc.edu>, Jeff Smith <smithjj@cat.com> wrote:
>In article <1995Jan12.022935.26572@Princeton.EDU>, schechtr@flagstaff.princeton.edu (Joshua B. Schechter) writes:
>|> 
>|> I believe the issue is not whether or not a computer can simulate a
>|> brain. It seems that a majority of people here seem to agree (whether
>|> or not they are correct) that a computer can simulate a brain. The
>--maybe -- we haven't yet simulated a brain.
>|> hardware of a brain seems to be accepted to be a type of universal
>|> turing machine and as such, can be simulated (as soon as it is
>|> understood) by any other turing machine.
>
>Why do some people accept the hardware of the brain to be a turing machine?
>Please define what you mean by a turing machine and express how the human
>brain fits this definition.  It's hard for me to see how anyone can have 
>definitive answers on this question when we don't fully understand how the 
>brain works, at the sub-neuron, neuron, or higher levels.

It is not a matter of whether the human brain is a Turing machine.  Obviously
there is no infinite tape in a human brain.  But neither is there one in
the PC on my desk.  The question is whether the human brain is computationally
equivalent to some Turing machine.  Since, as Neil Rickert, Darryl McCollough,
and others have argued, the activities of the human brain are finite, then,
if we view such activities as computations, the human brain is indeed
computationally equivalent to some Turing machine.  We can throw in such
issues as that there are many human brains and that any one human brain varies
over time, but we can still view the entirety of activities of all human brains
that have existed to date as a finite computation.  The whole discussion about
Turing machines and all the pronouncements by Penrose, Searle, etc. are really
rather boring from the point of view of the theory of computation (something
with which they all seem inadequately familiar), since the only activities
that are not computable by some Turing machine involve infinite abstractions
that occur nowhere in the real world.

>|> 
>|> The issue seems more to be "Can a simulation of a brain think?"
>|> And, of course, this brings up the question of what we mean by thinking...
>|> 
>|> 
>Turing test roughly defines thinking as conversing in a manner indistinguishable
>from a human being.  Thinking is what humans do.

Indeed.  The discussions here tend to forget Turing's purpose for the Turing
test.  His intent was to respond to various arguments against the proposition
"Machines can think."  He argued that "think" was too poorly defined to be
worth discussing, so he proposed substituting a more tractable proposition,
[this is my formulation, not an actual quote from Turing]
"Machines can pass for human beings within the confines of the Turing test."
He then set forth to show how the arguments that had been proposed at that
time against "Machines can think." could be rebutted when launched against
the second proposition.  The TT was a rhetorical tool (a "thought experiment")
devised in order to rebut certain abstract arguments.  It was never intended as
a methodology.
-- 
<J Q B>
