<Bob Clark (940624.2125 EDT)>
Tom Bourbon [940616.1640] Subject: Re: RKC-PROGRAM CONTROL
Maybe you feel that you are "butting in on a conversation," but I am
glad to hear another voice! It seems to me that anything seen on the
Net is open for anyone's comments!
Apparently I failed to make my point clear to Rick because he
responded:
*I have controlled the perception of a _program_ of numbers by
*generating outputs that are opposed to a disturbance that changes
*the program. etc.
I intended to direct his attention to the arbitrary decisions that he
made (whatever they might have been) in the process of creating that
program. Instead, he focussed on the operation of the program after
it was completed. Of course the program works the way he said it
would -- he designed it that way!
In my later post, [Bob Clark (940621.1012) Subject: RKC-PROGRAM
CONTROL], I tried to direct his attention to the arbitrary decisions
he was making as he composed his comments. As he did this, his
attention seemed to be on the post itself, rather than the process of
composing it.
From your comments and discussion, Tom, I conclude that your
attention was attracted by Rick's program, reminding you of the
program you had distributed as: Tom Bourbon [940603.1108: Program,
sequence, relationship.] Your summary of your demonstration suggests
that your program was rather more complex than Rick's, and included
more than one level.
The variable(s) you were controlling differed from Rick's. You
focussed on ("paid attention to") the "relationship between the
cursor and the target." You arranged for the required relationship to
change from time to time.
You were not paying attention to your own arbitrary decisions as you
designed the program.
After summarizing a variation on your original model, you noted:
That model duplicated the results from both of my runs; both times,
it created the same sequence of relationships between cursor and
target, but it did not create the same sequence of simulated
movements for the control handle-- it also eliminated the effects of
the disturbance during the second run.
It performed just as you intended! You are a skilled modeler!
You conclude:
there is no need to postulate programmed actions (movements,
outputs) when a person (or as in my demonstration, a rudimentary
PCT model) controls program-level perceptions.
Of course there "is no need" -- unless you happen to prefer to use
those concepts in your descriptions. Such control systems can be
described in many ways, depending on their complexity, the nature of
the variables to be included, etc. The treatment of temporal
variables particularly affects the form of the description.
Description in terms of "programmed actions etc" is only needed if
those higher levels being included are based on perceptions of that
kind. Otherwise, most of the actions can be treated as "simply
happening," thus:
The movements of the person, and those of the PCT model of the
person, simply "happen" as outputs from a system that is part of a
high-gain negative feedback loop that has its higher-level
references set to perceive programmed sequences.
Of course, you need not include those outputs that "simply 'happen'"
as long as you are confident that they are "in working order."
After all, if your model works as you designed it to work, you are
satisfied with your skill as a modeler. But if the model should
somehow fail one or more of your tests, what "corrective action" do
you take? Do you re-examine the tests, or the model, or the theory
on which it is based, or your "modelling skills?"
The purpose that you seem to have arbitrarily selected is to design a
model of some portion of a Hierarchical Network of Negative Feedback
Control Systems.
I have no problem with that purpose, nor with your modelling skill.
Rather, my interest is in describing the ARBITRARY DECISIONS YOU MAKE
AS YOU DESIGN YOUR MODEL!
Regards, Bob Clark