[From Rick Marken (2003.04.02.0800)]
Bruce Nevin 2003.04.01 22:44 EST)
> Rick Marken (2003.03.31.1510)]
> > Bruce Nevin (2003.03.31 13:18 EST)--
> > Suppose perception of the pattern in the behavior of others (perhaps in
> > terms of themes or features) sets references for the way many things are
> > done (music, warfare, etc.)
>Sorry, sir, but as the PCT policeman it is my duty to cite you for a
>minor PCT violation. Perceptions don't set references in PCT.
If systems set references based on memory but not on observation, it is
considerably more difficult to understand how to model imitation.
I think that there is no question that imitation is based on observation. That's
what imitation is, by definition. You were cited for saying that
"perception...sets the reference". Judge Powers gave you a pass under the
assumption that you misspoke (I'm incline to go along with the Judge in this
case). I'm sure you meant that references are set by the actor _on the basis of_
observation of the to be imitated perception.
How would you model two autonomous control systems such that one can
imitate the other? It seems to me that this can only be done by the
imitator perceiving the behavioral outputs of the other and setting
internal references accordingly
Yes. That shows that my paraphrase is exactly what you meant to say.
- either (a) references for perceptions of
one's own outputs, or (b) references for control of other variables by
means of outputs that mimic those of the other. What do you propose?
Exactly that. Though implementing such a model would be an interesting challenge.
Maybe I'll try it. I agree that imitation is an important aspect of learning. I
think there are two distinct kinds of imitation: one in which the imitated
behavior can be seen in nearly the same way by both the imitatee and the imitator
(this would be the kind of imitation where I move a cursor on the screen in some
pattern and you imitate making the same pattern with the cursor) and the far more
interesting care in which the imitated behavior is seen quite differently by
imitatee ad imitator (as when I try to dance like Fred Astaire).
Richard S. Marken, Ph.D.
Senior Behavioral Scientist
The RAND Corporation
PO Box 2138
1700 Main Street
Santa Monica, CA 90407-2138
Tel: 310-393-0411 x7971