perceptual organization

CSGnet Archive CSG1998
1

[from Mary Powers 980504]

Bruce Gregory (980501.0605 EDT)

You said (to Rick) "The different perceptual worlds expression simply
includes different perceptual organization in addition to same perceptual
world, different reference levels. To understand what is happening, we
still need to model (and Test) further."

I guess I'm wondering what it is about the control theory model of
perceptual organization that you don't like.

Maybe input signals should go through the comparator first? Reference
output to the perceptual function?
Perhaps you'd rather have intensity signals processed higher in the system,
or relationships lower down?

Or do you mean something else by "different perceptual organization"

We may indeed have to model further, but I think that can wait until the
model we have is _thoroughly understood_ and tested. It accomodates more
than most people think it can. I think that there is enough difference in
perceptual worlds taking into account different settings of reference
signals; different loop gain; possibly differences in the type of feedback -
leading, lagging or proportional (to bring up a totally unexplored aspect of
the model); plus differences in experiences and memories - to keep people
busy with _the organization as proposed_ for some time.

I suppose you meant something else, but I'm responding to what you seemed to
be saying. And I guess I get ticked off at people who want to start
tinkering with the model because they think it doesn't cover their special
questions, rather than learning enough about the model to find that it does.
This is one major reason psychologists who have looked into control theory
have made such a hash of it.

Is the _model_ deficient in the detail and accuracy "that true understanding
requires", or do its limitations and deficiencies reside in the eye of the
beholder?

Mary P.

2

[From Bruce Gregory (980504.1203 EDT)]

Mary Powers 980504

Bruce Gregory (980501.0605 EDT)

You said (to Rick) "The different perceptual worlds expression simply
includes different perceptual organization in addition to same perceptual
world, different reference levels. To understand what is happening, we
still need to model (and Test) further."

I guess I'm wondering what it is about the control theory model of
perceptual organization that you don't like.

I'm afraid I was unclear. There is nothing about the control theory model of
perceptual organization that I am unhappy with. I was unhappy only with the
idea that different people need to be controlling "the same" perceptions in
order for conflict to occur. It seemed to me that this requirement was
overly restrictive. Furthermore it seemed unlikely to me that at the neural
level the inputs making up a perception are ever identical in different
people. This issue was resolved for me when Bill posted:

"There won't be any conflict unless the environmental variable in my control
loop affects the environmental variable in your control loop. They don't
have to be the same environmental variable, but the more closely they're
related, the more likely is a conflict."

We may indeed have to model further, but I think that can wait until the
model we have is _thoroughly understood_ and tested. It accommodates more
than most people think it can. I think that there is enough difference in
perceptual worlds taking into account different settings of reference
signals; different loop gain; possibly differences in the type of
feedback -
leading, lagging or proportional (to bring up a totally
unexplored aspect of
the model); plus differences in experiences and memories - to keep people
busy with _the organization as proposed_ for some time.

Yes, I completely agree. I hope that is clear now.

Best Offer