Re: Aaron Sloman's "The Irrelevance of Turing Machines to AI" article

From: Neil W Rickert (rickert+nn_at_cs.niu.edu)
Date: 08/04/04 

Date: Wed, 4 Aug 2004 14:17:32 +0000 (UTC)

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Wolf Kirchmeir <wwolfkir@sympatico.ca> writes:
>Neil W Rickert wrote:
>> Wolf Kirchmeir <wwolfkir@sympatico.ca> writes:

>> If the way [an ability] changes is fixed, then it is innate. If the way it
>> changes depends on the environment, then it is only partially
>> innate.

>This seems to me to be the old notion of "instincts" in new clothing. I
>don't see any point to it - it's a disttraction from the real question:
>how do abilities develop?

Up to now, you seem to have been using "develop" as a way of
dismissing "how" questions.

>>>Now you might want to claim that the child always had the ability to
>>>make these fine distinctions,

>> No, I don't make that claim. My view is that learning to
>> discriminate is a large part of what we consider to be learning.

>If by that you mean that discrimination develops, that's what I've beens
>saying. If you mean something else, then kindly explain what you mean by
>"learning."

There you go again. You say it develops as a way of dismissing it
as anything that requires explanation.

>>>>This does not make a lot of sense either. Operant conditioning
>>>>already presupposes some notion of similarity, so cannot be used
>>>>to explain how that notion is acquired.

>No, operant conditoning does not presuppose a notion of similarity, it
>only presupposes the ability to discriminate.

Six of one, half a dozen of the other -- as you have been pointing
out to Lester.

> Abilities aren't notions,
>nor do they require notions.

"Operant conditioning" isn't an ability. Its an explanatory term (or
it claims to be explanatory). Notions are used in explanations.

>What, exactly, do you mean by "notion", anyway? Unregenerate behaviorist
>that I am, I think of "notion" inn terms of your behaviour. The ability

As an unregenerate behaviorist, you should take me as behaving by
exercising my fingers. That just happens, by some strange
coincidence, to make marks on your screen.

The notion of "notion" comes up when we decide that these marks
are actually conveying intelligible information. But I guess
unregenerate behaviorists are incapable of comprehending that.

>>>I don't think operant conditioning pre-supposes some notion of
>>>similarity _in the conditioned organism_. The notion is part of a theory
>>>about what we observe, which means it's an abstraction from behaviour.

>> No two circumstances are identical. What would be the benefit of
>> conditioning, if the circumstances where that conditioning applies
>> would never occur again. It seems that there is a dependence on
>> similarity of circumstances if conditioning of any kind is to be
>> useful.

>I didn't talk about identity, I talked about similarity. Your notion of
>"operant conditioning" is silly.

You are using "notion", which you questioned above.

No, the problem is not with my notion of operant conditioning -- it
is with your evasion.

Your notion of operant condition clearly depends on similarity, as
when you say "I didn't talk about identity, I talked about
similarity". Except, that is, when you say "I don't think operant
conditioning pre-supposes some notion of similarity _in the
conditioned organism_."

Or maybe you were again evading the issue. It doesn't matter whether
the conditioned organism has any conscious notion of similarity. It
does matter that the conditioned organism has behavioral abilities
which we would consider as dependent on similarity.

>[snip plaintive call for a definition fo "similarity"]

>I'm getting tried of this. "Similarity/difference" derive from

I'm getting tired of it too.

So let's drop it. I will just conclude that you were completely
unable to support the arguments you were making. We can leave it at
that.

>non/identity. I gave you a definition in terms of a numbetr computed
>from comparison of the elements of two sets.

Defining similarity in terms of comparison would be circular.

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.3.6 (SunOS)

iD8DBQFBEO/5vmGe70vHPUMRAiEeAKDIMRsZxwf2m/XxpH3qt/3oq1nXrgCcCupQ
5xesjKmNkYRwqnW6THQh1UU=
=BDRK
-----END PGP SIGNATURE-----

Relevant Pages

  • Re: Aaron Slomans "The Irrelevance of Turing Machines to AI" article
    ... > discriminate is a large part of what we consider to be learning. ... No, operant conditoning does not presuppose a notion of similarity, it ... only presupposes the ability to discriminate. ... > conditioning, if the circumstances where that conditioning applies ...
    (sci.cognitive)
  • Re: Patterns of evolution in intellegince
    ... > which, according to Hume, is simply comparing, is a development ... > sequential aspects of both conditioning and reasoning, ...
    (sci.bio.evolution)
Loading