Date: Sun, 26 Mar 1995 20:56:22 -0600 (CST)
From: Allucquere Rosanne Stone 
Subject: Re: visualization
To: Jerry Prothero 
Cc: jsp@biostr.washington.edu

On Sun, 26 Mar 1995, Jerry Prothero wrote:

> I would be interested in references on the theories of consciousness
> you mentioned.

Hi, I only have a second, because I've got to finish editing these page 
proofs tonight.  So here's a very brief answer, to be elaborated tomorrow 
sometime...

I'd start with Pat Churchland's stuff.  I don't know if she has a later 
edition of "Neurophilosophy", but that's a good place to start.  Also 
look up recent work by Llinas, and for something completely different you 
should taste Humberto Maturana and Francesco Varela's work on structural 
coupling, which is about nonrepresentational approaches.  Mmm, I've left 
out some important folks here...more later.

> If the current expertize on the subject of controlled visualization is
> tied up in Eastern mysticism, I think it's time to translate it.

Yeah, right *giggle*.  Just try.  Start with A. C. Bhaktivedanta Swami 
Prabhupada's "Easy Journey to Other Planets".  Then try some of the 
esoteric Buddhist treatises on the Taras.  That's as close as you're 
going to get to controlled visualization, and it's very far away.  These 
folks are not into accurately reproducing visual information.  I don't 
think they'll be of much help, because in the normal human (and even in 
most visual and menmonic savants) visual recall is notoriously tricky, 
heavily influenced by subjective stuff.  Neurologically, vision may be 
describable in some sense, but _seeing_ is a partial, social act; only 
somebody like Commander Data is going to have much success at detailed, 
repeatable visualizations.

> If Jeff's latest rant is correct, the complexity problem will not
> evaporate

I agree, in spades...and go so far as to add that any system of n>1
modules which is complex enough to be useful already contains at least n>0
fatal combinatory states.  And we can't prove they're there until they
bite us. 

, in which case being able to visualize in great detail may
> be an advantage as complexity becomes more and more what limits
> engineering.

Here I disagree.  I think the complexity of real systems already exceeds 
any possibility of forming a mental image of them.  You could probably 
hold an image of the generative algorithm, if the system in question had 
one, but that's precisely the can of worms, isn't it?:)

> I had a particularly annoying (in that it was uncontrollable) dream
> this "morning" (as Jeff once said, morning is when I get up) in which
> I had time to leisurely count the threads on a screw.  The image
> vanished instantly as soon as I became too conscious.

Hmm, I tend to be extremely suspicious of memories of dreamstate events 
of that nature.  Occasionally they do seem to be accurate, but there's a 
big difference between training yourself to remember complex things by 
mnemonic devices, and stunning your cortex sufficiently to get it to quit 
partying with your thalamus for a while.  I think you're talking about 
the latter, really, and I don't think it's doable, because the 
uncertainties of visual recall are part of what makes consciousness work.

Gotta run.  Nice to meet you!  More tomorrow...

--Zoooom!--

--Sandy

______________________________________________________________________

Allucquere Rosanne Stone
Advanced Communication Technologies Laboratory (ACTLab)
Department of Radio, TV and Film
The University of Texas at Austin
CMA 6.118, Austin, TX 78712-1091

Vox: 512.471.6499  Fax: 512.471.4077

"Damn the tortillas, full beer ahead!" -- Anon.
______________________________________________________________________
Backups?  We doan *NEED* no steenking baX%^~,VbKx  NO CARRIER
______________________________________________________________________