Behavior and Control

[From Rick Marken (950714.1220)]

Bruce Abbott (950714.0920 EST) --

Me:

I think your approach to PCT is based on the notion that PCT is an
alternative model of "behavior".

Bruce:

Well, perhaps Bill needs to retitle his book, as it SEEMS to offer an
alternative model of behavior. It says "Behavior: the Control of
Perception."

If you read the title carefully you will realize that it is completely
appropriate. The important part is the ":" which stands for "is not what it
seems; it is actually".

Conventional psychology takes it for granted that behavior is what we see an
organism "doing"; behavior is any observable result produced by an organism:
bar press, arm movement, rating response, drink of water, spoken word, etc.
Measures of these results are the dependent variables in conventional
psychological reserach.

The main point of B:CP is that this view of behavior is wrong: the results we
call behavior are not produced by the organism alone but by the organism in
concert with independent environmental influences -- disturbances. The fact
that organisms produce certain results consistently shows that these results
are under control; the organism must be varying its contribution to the
result to compensate for disturbances: behavior IS control.

Control theory shows that, when a result is controlled, what is actually
controlled is a perceptual representation of the result. So behavior is NOT
an observable result produced by organisms; behavior is a perceptual result
produce by the organism in concert with independent environmental
disturbances.

The PCT view of behavior knocks the foundation out from under all
conventional psychologies. All these psychologies -- behaviorism,
cognitivism, psychoanalysis, etc -- are based on the idea that behavior is an
objective phenomenon; you can't see mind but but you can see behavior. PCT
shows that you can't see behavior either: behavior is NOT on objective
phenomenon; it is a subjective phenomenon. Behavior is, ultimately, the
behavior of perceptual variables that are under control.

Although behavior is a subjective phenomenon, there are objective methods for
determining what an organism is doing (what perceptions it is controlling).
These methods are collectively known as The Test for Controlled Variables.
This appraoch to studying behavior has never been used in conventional
psychology so the data of conventional psychology are, for the most part,
useless to those of us who are interested in the study of control (the
operant data is an exception because organisms are placed into situations
where they have at least some control over a variable; but the
organisms ability ot control is usually constrained to be so poor -- so that
it doesn't look like it's controlling -- that the data is of only minimal
value).

Most of the data of conventional psychology are measures of the visible side
effects of control; they tell us almost nothing about what organisms are
actually doing (controlling). This is what my "mind reading" demo is
about. Have you ever seen that demo, Bruce? If so, I would like to hear what
you think of it. I think that is shows what happens when you deal with
behavior as an objective phenomenon. The behavior of the five numbers on the
screen is an objective phenomenon; the behavior of each number is produced
by the subject. The behavior of only one of the five numbers is actually
under control but there is no way to tell, by looking at the behavior of the
five numbers, which number that is under control. In other words, you can't
tell what a person is doing (controlling) by looking at their behavior (the
objective results they produce). This little demo shows that the basic
assumption of scientific psychology -- the assumption of objective behavior:
that you can see what what an organisms is doing - - is wrong.

I thought you were going to describe your view of my approach to behavior.

I described what I percieve as your approach to behavior. You seem to accept
the conventional view of behavior -- that behavior is an objective
phenomenon; that it is any measureable result of an organism's actions. I
may be wrong; perhaps this is not your view of behavior. If it is not your
view of behavior, then I presume that you are noting this fact in the latest
edition of your research methods text and explaining why conventional
aproaches to studying behavior tell nothing about what an organism is
actually doing (controlling).

I think the hardest thing to accept about PCT is that it requires a complete
re-conceptualization of that one thinks of as "behavior". Behavior seems
like such an obvious thing; so obvious that few psychology textbooks
even bother defining it (those that do say little more than "behavior is what
organisms do"). It is taken for granted that people know what behavior is.

Behavior is so obvious that it is difficult to believe that every behavioral
scientist who has ever lived has been wrong about what it is; that behavioral
science has been built on a mistake -- the mistaken idea that behavior is
an objective phenomenon. It is hard to believe that everyone from B. F.
Skinner to Herb Simon, from Sigmumd Freud to Noam Chomsky, everyone who has
taken it for granted that "behavior" is what you see people doing, is wrong.
It is hard to believe that all research that is based on "objective" measures
of behavior tells us little or nothing about what organisms are doing
(controlling perceptions). It is hard to believe it -- but it's true. That's
what all the PCT demos and experiments are about. We don't ask people to take
it on faith that behavioral science is based on an illusion; we expose the
illusion facts.

PCT is about the fact that behavior IS the control of perceptual input. PCT
is not a theory of the objective behavior studied in conventional behavioral
science; PCT is a theory of the subjective behavior of controlling
perceptual variables. Once one understands this fact about behavior one
simply stops doing psychology the old way and starts studying control.

Tom Bourbon (950713.1108) --

Skinner is source enough for us to see the inconcistencies and
contradictions in his ideas.

Wonderful post Tom!

I look forward to reading Bruce Abbott's judgement of Skinner's understanding
of Skinner's own theory.

Best

Rick