[From Bill Powers (920925.2000)]
Martin Taylor (920925.1100) --
Whoa, there. One of the consequences of the PCT approach is the
idea that the same end can be achieved by many different means. If
one possibility is blocked, another can be used. And often the
various means are mutually exclusive in any case.
This is a "levels" problem, which I was avoiding during the argument
to keep from getting more complexities into the act.
A control system that acts by choosing one action from a selection of
others can't also choose the amount and direction of that action that
ensues from disturbances. That takes two levels (more than one,
anyway). Achieving the same end by variable means (which is required
by disturbances of various kinds) is the result of a control system's
reacting to error by altering its output. If the system is an action-
choosing system, then the disturbance would amount to a change in the
environment of a rather complex sort, that requires switching from one
KIND of action to another KIND (through sending reference signals to
THIS system instead of THAT at a lower level). In order for this level
of control to work well enough to achieve real control, this switching
must be quick and smooth: automatic. If it isn't -- if there are false
starts and wrong choices -- then we would probably guess that
reorganization is going on.
My point is that even when we speak of switching KINDS of control, the
situation is quite analogous to what happens at a lower level where
the dimension of change of output is strictly HOW MUCH of the
perceptual variable is to be maintained. I'm using this HOW MUCH idea
in a pretty metaphorical way, because I just want to mean a shift
along whatever dimension of change is appropriate at the level under
discussion. Obviously, at higher levels the dimension along which
change of output takes place isn't the simple sort of continuous thing
we see at the level of intensities. But it can still be ordered to
create a dimension of stepwise change.
Taking away one of the possible choices is directly analogous to
disturbing a lower-level system by tying a rope to one arm. It alters
the feedback function connecting the output (which selects choices) to
the input. The analogy would be to altering the schedule of
reinforcements, not to directly adding or subtracting reinforcments.
This kind of parametric disturbance shows up only when the control
system tries to act; then it discovers that the effect on the
controlled variable is different or missing. In a choosing-type
system, the result of the increased error would be a further change in
the choice of action -- picking the next one in order of ease,
handiness, effectiveness, or whatever the ordering scheme is. If the
bicycle is missing take the car; if the car is missing take the bus;
if the bus doesn't come, take a taxi.
If there IS no natural "next" choice, then the error will not be
corrected. The perception in question won't be controlled, and the
next higher level will be disturbed. Instead of an automatic next
choice of a category of action, we might see the person reasoning out
some alternative, or dropping the goal that called for making the
choice (if both the bicycle and the car are missing, call in sick).
Failing that, the system might be thrown into a state of
reorganization.
If that isn't affecting the operation of B's control system, I >don't
know what is. And I think this is the sense in which Greg >was using
the phrase, not in the sense that A changes the linkages >or functions
in B's control systems.
I have never argued that control systems can't be affected, even on
purpose. But so far I don't see what the term "linkage" describes that
isn't already described in any basic analysis of relationships that
uses all the resources of HPCT.
ยทยทยท
-------------------------------------------------------------------
Jeff Hunter (920923 - 1) --
This gives us two common meanings for the word "control"
a) I can make something do what I want it to
b) I can make something do what it doesn't want to
(or I can make it choose the lesser of two evils)
I agree with Rick Marken on this. This is really just control. The
word that's changing its meaning between these two definitions is
"make." If you ask me for the salt, I can "make" you reach out a foot
(that is, 12 inches) to receive it if I hold it one foot from where
your hand is. If you don't want to let go of the salt, I can still
"make" you give it to me, by prying your hand open using a force too
great for you to resist. In both cases, I'm controlling; it's just
that in the second case I have to use a little more effort to overcome
environmental resistances.
Of course in the second case, a conflict is created, while in the
first case there is none (as Rick said).
-------------------------------------------------------------------
Martin Taylor (920925.1100) --
But for a lot of parents with unhealthy children, changes in diet
often are among the changes that they try, until they find what the
kid was (say) allergic to. The kid becomes healthier, and they no
longer try to change the diet.
Sure. In this case there's a symptom that can be perceived, so to
control that perception one varies the diet to try to make it go away.
If you have a stock of remedies at hand, it's an ordinary control
process. If you don't know what to do, you just try different things
until you hit on one that makes the symptom go away -- reorganization.
Feeding healthy food as a matter of principle isn't quite the same
thing. If the kids develop a symptom, you can't feed them still
healthier food -- if you could you'd be doing that all along. When you
don't actually have any control, all you can do is imagine that you
have control, or give up.
I know people using herbal remedies and megavitamins. I asked recently
how one of them was doing. "Wonderful," the answer came back. "He's
vomiting and coughing up a lot of black stuff, really getting rid of
the poisons." I said "That's nice." He's a sensible fellow; I'm pretty
sure he will soon tell his helpers that he feels a whole lot better
now and they can stop giving him that stuff.
-----------------------------------------------------------------
We use "reference" and "percept" and "error" with technical
meaning. Why not "linking" which seems to describe a phenomenon
that is important in interpersonal interaction.
Because I don't think it describes anything we can't already handle,
and because I like to use "link" in other ways, as in "the feedback
function links the output to the controlled variable" and "properties
of the environment link the disturbance to the controlled variable." I
don't want to have to keep explaining that I mean one kind of linkage,
not some other restricted and made-up sense of the term. I have to do
enough of that already. I think that creating a lot of new terms just
makes for jargon, and conceals simple relationships by lumping special
cases into categories of their own. Maybe later.
------------------------------------------------------------------
Dennis Delprato (920925) --
RE: Analysis of Skinner's World.
That's a really grown-up job. It's the sort of analysis we need done
on PCT. Why not give it a whack yourself? Maybe you'd be a good person
to edit the forthcoming issue of Closed Loop on influence and control,
with editorial and historical comments!
Mary and I both think that we aren't done with this thread yet, and
that the Closed Loop on this subject shouldn't be put out prematurely.
I guess Greg reached the same conclusion. We seem to be headed slowly
in some direction; let's wait until an obvious pausing place appears.
Congratulations on the "in press." Somebody around here knows how to
get published.
--------------------------------------------------------------------
Best to all,
Bill P.