[From Bruce Abbott (941218.1650 EST)]
Rick Marken (941217.1415)
target. When I made this change (which I called "getting rid of an
artifact" from the experiment) Bruce's selection by consequences
model no longer worked because consequences (rewards and punishments) no
longer selected probabilities of pressing that acheived the goal. The
selection by consequences model did what Bill Powers said it did.
The control model (a selection OF consequences model) and human subjects
did NOT fail (to control) under this change of circumstances.
The failure of Bruce's selection by consequences model under this
"fairly simple change of conditions" is very important, not only
because the control model did not fail under these conditions but also
becuase people did not fail under these conditions either.
I provided a fairly detailed reply as to WHY this happened, which was ignored.
Basically it came down to two points: (a) people use other variables to assess
the consequences of their actions than the limited one provided to my little
learning e. coli and (b) Rick's model performs well in this situation because
it is not a learning model. It does not describe how human participants are
able to adapt to changing circumstances. A correct model of human performance
on Rick's e. coli simulation would require implementing several layers in the
hierarchy--for which my simple two-level system would be no match.
Furthermore, I never offered this model as a serious alternative to PCT. You
may recall that what you were asserting at the time was that a "reinforcement"
based model COULD NOT control in this situation. I demonstrated that one
could, and that is all. You then went on to demonstrate that this model is
not as resistant to changes in contingency as the simple negative-feedback
model is, which I agree is true.
The result of this set of demos should have been the realization that
selection by consequences is incompatible with the control exhibited
by people in the E. coli demo. The subject cannot vary actions (the interval
between bar presses or the probability of a bar press) as necessary in order
to produce a consistent result (the spot remaining near the target) if those
actions are selected by their consequences. In order to control, actions
must be free to vary, as necessary (that is, as determined by the error
signal in the control loop).
I fully understand--and agree with--this. (Why you don't think I do is an
endless source of mystery to me.) But let's talk about learning. How do your
human participants learn to react in such a way as to bring e. coli under
control? How about producing an actual MODEL of this, rather than a verbal
description of how such a model might work? Your current e. coli model does
not learn, and thus does not embody the full capability of your human
participants to adapt.
Also, from your own description, it took you quite a while to discover a set
of conditions under which my learning e. coli would fail. How about listing
all those variant contingencies you tried in which the model stubbornly
refused to fail?
If one is dedicated to (controlling for) this point of view (system
concept) then demonstrations that are not consistent with it are
disturbances to be dealt with in any way possible. That's why Bruce has
not yet said: "Geez. You're right.
I was afraid you still felt this way. Rick, I accepted feedback-regulated
control as a general principle of behavior (and rejected the reinforcement
model) before I ever heard of Bill Powers and his book. Five years in an
industrial setting working with proportional furnace controllers, plus reading
W. Ross Ashby's _Introduction to Cybernetics_, Herbert Simon's _Sciences of
the Artificial_, and several other related sources on feedback-regulated
control had prepared me to recognize the value of Bill's contributions as soon
as I encountered them. Various commitments, constraints and choices kept me
from actively pursuing PCT research, but none of these reasons have anything
to do with a dedication to the TRT point of view.
So, you must be asking yourself, if all this is true, why do I seem to be
pushing for a traditional viewpoint? Probably for a number of reasons. One
is that, unlike you, I am not yet convinced that all knowledge accumulated
during the past 100 years under the reinforcement tradition becomes
immediately irrelevant once one takes a PCT perspective. I keep looking at
that Necker Cube, first from the TRT viewpoint, then from PCT. They are both
descriptions of the same cube, although you have to undergo a kind of
perceptual reorganization (as the Gestaltists put it) to see it. [The analogy
cannot be carried too far; the two Necker cube constructions are both valid.]
This examination leads me to ask how certain phenomena observed and explained
in traditional research can be interpreted from a PCT viewpoint.
This prompts me to pose certain questions, and to provide, to the best of my
understanding, what I think would be the traditional explanation. In some
cases I have also attempted what I would view to be a PCT explanation. The
explanations and corrections you, Bill, Tom, and others have provided have
been generally helpful and illuminating.
In my view it's a win-win situation. If PCT can be shown to handle a
phenomenon (e.g., the empirical law of effect), great, I've learned something.
If not, then I've exposed a weakness in need of attention, and we've all
learned something.
Recognize, too, that I am still learning, and that I am to some degree
handicaped by my training and experience (as are we all). I will get it
wrong--old habits are hard to break. In addition, I've never been one to
accept anything uncritically. Don't mistake my being critical for rejection.
I have to prove each step to my own satisfaction before going on to the next.
I also have in mind the ultimate acceptance of PCT within my field of
research. The failure of PCT to take the field by storm is due, in my
opinion, to more than simple reluctance to adopt a new way of thinking about
behavior. To gain acceptance, PCT must demonstrate an unequivocal ability to
account for the known phenomena of the field, and to do so in a more natural
and parsimonious way. I fully recognize that research conducted from the PCT
perspective must evaluate different variables under different conditions than
those customarily employed in this area. Yet TRT and PCT both claim to
explain behavior. Because TRT was there first, PCT must be able to show how
well-known phenomena of instrumental and operant conditioning can be accounted
for, so that traditional accounts become unnecessary and inelegant by
comparison. The operant conditioning project Bill and I are beginning is a
step in that direction. So is our current debate about the law of effect.
Rick, you said that you wish you could find just ONE more research
psychologist who, like Tom Bourbon, would take up research from a PCT
perspective. You've got that guy on the end of your fishing line, and you
don't even know it.
Regards,
Bruce