On 30/06/2011 10:55 PM, Darren New wrote:
> http://www.evl.uic.edu/swami/crabcanon
>
> My favorite prelude in Godel Escher Bach.

My a startling coincidence, last week I opened GEB on a random page, and 
this just happened to be the page I opened on.

This text is a pristine example of why I stopped reading GEB. I got the 
book hoping to learn something new and interesting. What I got instead 
was tedious riddles. Lots of them.



Opening the book a few more times at random, I did eventually come 
across something actually mildly interesting. A discussion of the human 
brain. Specifically, visual processing. Apparently there are cells which 
detect light, cells which process differences in light, right the way up 
to cells which fire in response to particular visual patterns. Simple 
basic shapes and textures.

But what about higher levels of complexity? Taking this to its logical 
conclusion, is there a "grandmother cell", which fires if and only if 
you happen to be looking at the image of your grandmother?

It's an interesting question. And the discussion mentions some 
interesting experimental results (e.g., cells which fire when you're 
standing at a particular point in space). But, to me, asking whether 
there's one cell which fires when you see your grandmother is a bit like 
asking whether there's one transistor somewhere in a computer which 
switches on only if you happen to be running MS Word.

What you put it that way, the absurdity of the question becomes obvious. 
"MS Word" is an abstraction created not by one transistor, but by the 
concerted activity of millions of them. I would expect that complex 
high-level phenomena like recognising a person's face is a similar 
matter for the human brain.



So far, that's about the only moderately interesting thing I've managed 
to get from GEB.

Post a reply to this message