IUBio Biosequences .. Software .. Molbio soft .. Network News .. FTP

machine brains

Michael Edelman mje at mich.com
Mon Feb 8 10:57:27 EST 1999



William Thomas wrote:

> I often wonder how an eagle, with very little brain power left over after
> accounting for optics, can accurately make the complex physics calculations
> required for a 100mph+, 1000' dive to snag a fish that is swimming under
> water while the defecated light creates an optical illusion as to the
> position of the fish that varies depending on the fish's depth, the water
> clarity, and the intensity and angle of the sun.

Simple. Well, almost simple ;-)

A lot of the processing is hard wired. By the time the signals from the eye
reach the visual cortex and limbic system you've already done a lot of
processing, beginning in the retina, extracting time and frequency doman
informations, etc. The intensive processing is already done by the time the
signals arrive at the brain. And eagles don't get it right the first time. They
have to learn, too. Overe time all the environmental variables get included in
tthe training. And even eagles fly into windows from time to time.

Brains are of course very plastic machines- even eagle brains. They can train to
perform complex tasks requiring a lot of feedback and correction, and eventually
pass the tasks over to a background process that doesn't require active
monitoring. Humans can drive a car, smoke a cigarette, drink coffee and carry on
a phone conversation simultaneously.






More information about the Neur-sci mailing list

Send comments to us at biosci-help [At] net.bio.net