(cont).
then you've got something related to memory, no matter how you describe it.
Hubel and Wiesel (sp?) found some pretty interesting (although incomplete)
stufff about the organization of the occipital cortex and probably never even
heard of PCT. Do you want to claim that they are wronb because they have the
wrong paradigm? We know alot of visual processing occurs in occipital. We
know that attention mechanisms are involved in at least prefrontal areas. We
know that the hippocampus is involved in at least one form of memory. WE
didn't need PCT to find that out, because we are working at a levelof analysis
below PCT. Sure PCT could help inform the processs but it isn't as necessary
as it would be for levels above PCT--the rest of psychology.
Making good models occurs from going both bottom-up and top-down. I think its a
mistake to think we can come up with a hierarchy without looking at what the
brain is doing. The sorts of processes that can be disassociated via lesions I
think is "weirder" than anything we would imagine without such research. Ya
gotta have both.
Mark
[Martin Taylor 921215 19:30]
(Mark Olsen (cont) 921214)
Hubel and Wiesel (sp?) found some pretty interesting (although incomplete)
stufff about the organization of the occipital cortex and probably never even
heard of PCT. Do you want to claim that they are wronb because they have the
wrong paradigm? We know alot of visual processing occurs in occipital.
You've put your finger on one of the reasons why I decided in graduate school
and later never to take neurophysiologists at their word. We knew (as
experimental psychologists) in 1959 about the need for spot detectors and
line detectors, and the like, because we got it from the requirements
imposed by psychophysical results. We discussed it in our group discussions.
At that time neurophysiologists said there was no such thing in the brain,
so we said "OK, but it's a functional requirement." A few years later
Hubel and Wiesel said "there are these funny specialized detectors in the
visual system--wonder what they are for" and got a Nobel prize for it.
This happens over and over in neurophysiology. The psychologists needs a
function to explain something, the NP (complete?) says "No way," and a few
years later another NP says "Look at this neat thing--wonder what it does."
Think of linear correlators, for example (1970s. I think). I can't think
of other specific instances at the moment, but I remember many occasions of
"Oh, so now they agree that the brain can do what we said it must but they
said it couldn't" over the last three decades.
I think the situation is the same as regards PCT and psychology or
neurophysiology. If a PCT analysis suggests that some function is necessary
(such as the passive parallel monitoring and active quasi-serial control
structures), then I will believe it, and I will tend to accept it when
people say by other methods that they have found how such structures are
made at some other level of description. I will not believe them when
they say that they know that these things can't be, or are not, done in
the brain.
It's nice to know what neurons CAN do, singly or in groups. But I would
never trust a neurophysiologist to tell me what they DO do, and I would
not believe one who told me what they CANNOT do.
Martin