[From Rick Marken (960916.1050)]
Bruce Abbott (960916.0905 EST) --
I have what I believe are excellent reasons for disagreeing with your
premise, but I have no desire to argue with you about it for a second time.
I hear ya!
Indeed, I've already spent far too much of my time trying fruitlessly to
present a couple of Simon's ideas I still think are relevant to modeling and
support the PCT approach to analysis
It seemed like time well spent to me. I think it's good mental exercise. Kind
of like discussing evolution with creationists.
Suffice it to say that we continue to disagree; I'm going to leave it at
that.
That's up to you. But it seems like it would be very useful to have a
discussion of this issue. I think it is extremely important (that is, if one
thinks that PCT is extremely important). I think PCT's failure to have any
impact on conventional psychology is a direct result of the fact that PCT has
been seen (erroneously) as an alternative explanation of what psychologists
already know. In fact, PCT is an explanation of something that psychologists
never knew anything about at all -- purposive behavior.
Hank Folson (960916) --
my perception that in several threads parties on both sides do not seem to
be getting much satisfaction out of the discourse
I've been extremely satisfied myself. The only thing that has been
unsatisfying is that Bill and Mary Powers keep saying what I want to
say -- only better;-)
Bruce Gregory (960916.1015 EDT) --
While EAB may have limited to non-existent value, I have learned a lot from
less experimentally oriented approaches. For example, observations of the
different communication styles of men and women have a practical value, but
they also offer a fertile field for interpretation using PCT.
I think observations like these (and the observations of EAB and other fields
of conventional psychology) provide a good _staring point_ for _testing_ from
a PCT perspective. Such testing might eventually lead to a determination of
the variables that are actually controlled in a conversation and to tests of
whether a particular man or woman is controlling for these variables. But I
have very strong reservations about giving PCT interpretations of _any_
conventionally obtained data. It's OK to do this if the PCT interpretation is
just the _start_ of a research effort to test the interpretation (which
means testing for controlled variables, because a PCT interpretation of
behavior means guessing at what variables an organism might be controlling
that would give rise to what we see as their behavior). But it's really a
dead end if it's just a game of coming up with a PCT "just so" story that
seems plausible.
The problem with plausible PCT interpretations of existing data is nicely
illustrated by Bill Powers' efforts, over the last couple years, to come
up with a plausible PCT model of EAB ratio schedule data. The data Bill used
(from Staddon and Motherall) seems very clear; on non-demanding ratios, rats
press _less_ as the ratio becomes _less demanding. It looks like rats are
controlling their food input and that the ratio schedule is just part of the
feedback function that determines the rate of response needed to maintain a
reference level of food input. On more demanding ratios there is an apparent
positive feedback effect which can be accounted for by assuming that the
food input control loop is nested inside an outer energy expediture control
loop.
Anyway, the PCT explanation fit the data extremely well. But then Bruce
Abbott discovered (as Bill Powers (960916.0630 MDT) noted) "that 'rate of
pressing' is determined, in a very common type of experiment, primarily by
how long the animal takes to collect and consume food, with the actual rate
of pressing being essentially constant at a high rate otherwise". So now we
don't even know if what we are seeing in the EAB data (reinforcement rate as
a function of pressing rate) is a control phenomenon because pressing rate
might not be a variable that could be making up for the ratio requirement;
pressing rate is possibly just an artifact of how often the animal eats.
So here we have data that "makes sense" from a PCT perspective. It is data
that seems to include a measure of a controlled variable (reinforcement rate)
under disturbance (ratio requirement). The desire to include "conventional
data" among the things that "PCT could handle" led to a plausible (but
almost certainly incorrect) control theory explanation of a phenomenon that
might not have involved control at all.
The problem of explaining behavior is not that it's so hard to do but that
it's so EASY. That's why competing theories of psychology have no problem
explaining _any_ behavior. Unfortunately, plausible seeming explanations of
behavior are usually wrong -- and they are often wrong in a way that creates
problems for the person whose behavior is being explained (like explaining a
behavior as a result of "hyperactivity"). PCTers (well, this PCTer) want this
crap to STOP. We want people to understand that behavior is the control of
perception; and that a correct explanation of behavior cannot be achieved by
developing plausible stories about what is seen (the actions and results
produced by the behaving system); a correct explanation of behavior can only
be acheived by figuring out (using The Test) what perceptual variables an
organism is controlling. This requires work (that conventional psychologists
have not yet put in) rather than story telling skills (at which conventional
psychologists excel).
Bruce Gregory (960916.1120 EDT) --
What Newton discovered is that his invention described the way the world
behaves. PCT is Bill Powers invention. What Bill and many others have
discovered is that this invention describes the way living systems work
That's how I would say it, too. Very nice!
Best
Rick