[From Bruce Abbott (941213.1820 EST)]
Our internet link went down again, so I'm just getting mail from Sunday and
Monday.
Bill Powers (941212.0730 MST)
Bruce Abbott (941211.1330 EST) --
Behavior varies (reorganization in progress), affecting the
environment, changing the state of a perceptual variable, changing the
error between that variable and its reference level. If the error is
reduced, that behavior, which a higher-level system had "selected,"
remains selected and thus becomes part of the output function of the
lower-level system controlling that perceptual variable.
This is not the relationship between levels that is proposed in HPCT.
Higher level systems do not select or specify or provide reference
signals for _behaviors_. Reference signals specify perceptions (inputs)
not actions (outputs). This is just as true of the reorganizing system's
effects as of any other control systems. All control systems operate to
control their inputs, not their outputs.
Well, I sure got myself into trouble with that one! I'm afraid I'm just too
used to thinking in terms of operants--behaviors defined by their common
(perceptual) consequences. While it is true that every lever-press or string-
pull is unique, the fact that we can perceive and describe them as lever-
presses and string-pulls means that there is some common denominator by which
we recognize the essential "sameness" of these acts. I took it as implicit
that these perceptions are repeatable because the behavioral output will vary
to compensate for disturbances.
So what I envisioned (but communicated badly) was that a series of different
"acts" (organized sequences under feedback control) would be initiated by the
cat during reorganization. Successful acts would remain "selected" in the
behavioral output function, because they would halt reorganization. Does this
capture your conception of reorganization, or do I still have it wrong?
I'm going to have to be more careful to use that term "behavior" in its more
restrictive PCT sense as mere output. On the other hand, having to repeat the
whole litany of controlled perceptions and reference levels every time one
wants to refer to a stable perceptual consequence of behavior like "pulling
the string" is terribly awkward. Is there a good term available that would
serve to connote all this? How about "operant"?
Mary Powers 941212
But I don't think you have described it correctly. The giveaway
is that parenthetic (reorganization is in process). Behavior
always varies, whether or not reorganization is going on.
Mary, the parenthesis was meant to clarify that I was talking about what goes
on during reorganization, as opposed to the variation in behavior that occurs
in response to disturbances during ordinary control. I was not trying to
suggest that behavior varies ONLY during reorganization, but rather to state
that WHEN behavior varies during reorganization, certain things follow.
Regards,
Bruce