Mike, I have a problem with saying vision is a matter of graphics.  if you look 
how non-artists draw things, it isn't like they are using imagery,  People 
think in terms of schematic parts.  When a regular person draws a face, for 
example, they put a head, nose, mouth, not a pixel accurate rendition.  It 
takes a very different skill to twist things into a pure image.  I think even 
your visual emphasis is not quite right, though I do recognize the importance 
of sensory information.  In essence, I think you want to get _too_ close to 
sensation, when there are categorization and feedback steps in front of it 
which are needed for intelligent perception.  We recreate, we don't recall.  
And I think you are missing important subtleties in that.
andi


On May 14, 2013, at 1:04 PM, "Mike Tintner" <[email protected]> wrote:

> To repeat : no question that multiple sense, multilevel processing is going 
> on, and is all vital. I'm just arguing that graphics are central (not the 
> whole tree) - and above all central for conceptual processing - what is 
> normally called "language" . And AI isn't yet capable of handling real 
> language or animal-level concepts. so we have to have an explanation of why.
> 
> -----Original Message----- From: Andrew G. Babian
> Sent: Tuesday, May 14, 2013 6:25 PM
> To: AGI
> Subject: Re: [agi] A General O.D. (Operational Definition) for all AGI 
> projects
> 
> While I think it's valuable for Mike to be here emphasizing sensory input, my 
> position as well, Ben is clearly right that it isn't a matter of graphics 
> being missing.  An intelligence must learn to handle all sensory inputs, and 
> possibly just touch can feed into the maps, with some work.  In addition, 
> it's a learned sensory-motor system-- there is also feedback between senses 
> and motor output.
> andi
> 
> Can I help?
> 
> On May 14, 2013, at 11:35 AM, Ben Goertzel <[email protected]> wrote:
> 
>> What is your evidence that people unconsciously reason graphically,
>> even when they feel like they're not?
>> 
>> Of course, it's obvious the brain maintains multiple spatial maps
>> (e.g. the allocentric map in hippocampus, and the egocentric maps in
>> parietal cortex), and links this with visual cortex which is good at
>> visual pattern recognition -- but what's your evidence that this sort
>> of graphical/visual representation is universally widely used by
>> people as the main tool for concept representation?
>> 
>> 
>> 
>> On Wed, May 15, 2013 at 12:23 AM, Mike Tintner <[email protected]> 
>> wrote:
>>> !. "You seem to be taking your own personal experience of thinking and


-------------------------------------------
AGI
Archives: https://www.listbox.com/member/archive/303/=now
RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-f452e424
Modify Your Subscription: 
https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-58d57657
Powered by Listbox: http://www.listbox.com

Reply via email to