From newshub.ccs.yorku.ca!torn!cs.utexas.edu!swrinde!sdd.hp.com!zaphod.mps.ohio-state.edu!pacific.mps.ohio-state.edu!linac!mp.cs.niu.edu!rickert Fri Sep  4 09:41:00 EDT 1992
Article 6697 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!torn!cs.utexas.edu!swrinde!sdd.hp.com!zaphod.mps.ohio-state.edu!pacific.mps.ohio-state.edu!linac!mp.cs.niu.edu!rickert
>From: rickert@mp.cs.niu.edu (Neil Rickert)
Newsgroups: comp.ai.philosophy
Subject: Re: what is consciousness for?
Message-ID: <1992Aug25.141057.19246@mp.cs.niu.edu>
Date: 25 Aug 92 14:10:57 GMT
References: <1992Aug23.231434.28446@mp.cs.niu.edu> <BtJJ7x.1AL@watdragon.uwaterloo.ca>
Organization: Northern Illinois University
Lines: 60

In article <BtJJ7x.1AL@watdragon.uwaterloo.ca> cpshelle@logos.uwaterloo.ca (cameron shelley) writes:
>rickert@mp.cs.niu.edu (Neil Rickert) writes:

>> But the formal structures of conscious knowledge are usually encoded
>> linguistically, and the unconscious can learn to recognize the patterns
>> of associated words without regard to whether they represent a formal
>> structure or not.
>
>By "encoded linguistically" I take you to mean for the purpose of
>communication.

No, that was not my meaning.  Language is far more important to us than
mere communication.  Doubtless our language capability evolved because
of the advantages of communication.  But we use language far more for
thought than for communication.  Language, particularly language based
thought, acts as a memory refresh mechanism.  That is, we can repeatedly
return to the same thoughts, and every time we do so we strengthen our
memory of them.  I believe we would have almost no memory recall ability
if it were not for this phenomenon.

>> I would say that conscious learning is less accurate than unconscious
>> learning.  Perhaps this is because it is rapid, so based on less
>> experience.  As for rapidity of conscious learning, this is due to
>> consciousness providing a feedback system which reinforces learning.
>> That is, since we are aware of our conscious thoughts, they are in
>> effect recycled as input, and this constitutes a reinforcing
>> feedback.

>I find the idea of consciousness as feedback quite interesting.  But
>it does seem to render the term "conscious learning" something of a
>misnomer.

The point here is that we use consciousness to reinforce our learning.
Unconscious learning is based on extensive experience.  But consciousness
provides us the ability for thought, and that allows us, through a
thought based simulation, to repeat a single experience many times, and
get the same learning effect as if there had been multiple real repetitions
of the identical experience.

>           And if the consciousness does represent feedback for the
>unconscious mind, how does that imply it should have a more formal
>representation?

There is nothing in the conscious which requires a more formal
representation.  But, it happens that most of our conscious learning is
language based, so what we learn tends to be formalized into the type
of structures most easily represented by language.  If you like, it is
because language is discrete rather than continuous, that we tend to
structure it.

>                 If they are so closely related, how does
>consciousness abstract the representation of unconscious knowledge?

If I understand what you are asking, this is the big question which has
baffled many people.  Namely, where does semantics come from?  I believe
that unconscious learning is very broadly oriented, and can perhaps best
be described as a contextually oriented learning to recognize complex
patterns of associations.  As we thus learn to associate a word with
sensory input, or with other words, all in some contextual setting, we
build up those semantics.


