[From Rick Marken (931104.1000)]
Norm Holland (931103.2150) --
Welcome to CSGNet, Norm!
I am puzzled why you say (in
the introductory material Gary sent me) that PCT reearch "meets
resistance from all sides."
I'm puzzled too -- by the resistance, that is. But it is there,
usually in the passive form; PCT is resisted by being ignored.
PCT is rarely criticized or celebrated. In my field (experi-
mental psychology) there are VERY FEW references to Powers' model
in the refereed literature. Where PCT IS discussed, it is nearly
always described or used incorrectly (some names involved in the
"misuse" of PCT in experimental psychology: Fowler, Turvey,
Carver, Scheier, Locke, Karoly, Hyland).
Or "The phenomenon of control is largely unrecognized in science
today."
Again, the sad fact is that control IS an almost completely
unrecognized phenomenon; in my own field, anyway. If control
WERE recognized, then psychological researchers would be busy
trying to figure out what variables people (and other organisms)
control. They are not. If anything, researchers are looking for
the variables that control behavior -- variables which PCT says
do not exist (at least not in the world of innanimate objects).
My impression is that
the "constructivist" movement in psychology is now the dominant
paradigm. It has, startng in the 1970s, gradually replaced
behaviorism, and I would say it is exactly in accord with
Powers' view.
The "constructivist" or "cognitivist" movement IS now dominant in
psychology. But this makes it even tougher for PCT because everyone
thinks that "the revolution" in the behavioral sciences has already
occurred; it hasn't. Cognitivism was good inasmuch as it made it
respectable to include "mental processes" in theories of behavior.
Unfortunately, however, cognitivism is built on the same model of
behavior as that used in behaviorism. Simplifying shamelessly,
the behaviorst model was:
s -->r
External stimuli (s) lead, via the nervous system (-->) to
behavioral responses (r). The cognitivist model is:
m -->r
Internal mental processes (m) such as memories, plans,
intentions, frames, productions, etc, lead to behavioral
responses.
So the basic model of behaviorism and cognitivism is the same;
it is what we PCTers call a "lineal causal model". The essence
of the model is that behavior (r) is the LAST STEP in a causal
chain; behavior is output. Cognitivism simply moved the start
of this chain from outside to inside the organism and, at the
same time, added all kinds of complex processes that might make
up the "m" that leads to behavior.
The proof that behaviorism and cognitivism use the same model is
in their methodology. The "correct" way to do a psychological
experiment is to manipulate an independent variable (IV) and measure
the consequences of this manipulation on behavior (the dependent
variable, DV). Look through any psych journal and you will see that
all experiments, whether they are done by behaviorists (and there
are still plenty around) or cognitivists, use this IV-DV approach.
PCT shows that behavior occurs in a circle of causation and that
what occurs in this circle is "control". What is controlled when
organisms behave is not behavioral output but perceptual INPUT.
In order to understand behavior (according to PCT) we must try
to figure out what perceptual inputs are being controlled, not what
behavioral outputs are being caused. PCT is based on a new method-
ology (called "the test for the controlled variable" -- which is
similar to IV-DV but also quite different) that is aimed at
determining what an outganism is "doing" (ie. what perceptual
variables it is controlling).
Is there a relation between this group and the cybernetics people
whom I associate with Sir Stafford Beers?
I'd give that a "no". We (some of us PCTers) have attended
cybernetics meetings but found that very few of the cybernetics
people had any interest in what we had to say.
Now Neisser suggests that we have two systems.
We use constructivist processing for most things, particularly
complicated things. We use "information processing" only for
spatial maneuvers (catching that baseball that Gibson made so
much of).
Neisser never got quantitative about the "perceptual cycle" that
he discussed in his 1976 book (I forgot the title) so he never
understood that such an arrangement acts to control its
perceptual input relative to internally specified reference
standards, resisting variable disturbances in the process.
Everything was spelled out for Neisser three years earlier by
Powers (in BCP) but Neisser chose to ignore it. C'est la vie.
Again, welcome to the net, Norm. I sure hope you get a chance to
participate a bit; you have some very fascinating interests.
Best
Rick