Having PCT and eating SR too/Methodology

[From Oded Maler (930407.1650 -ET)

···

*
* [Rick Marken (930407.0900)]
*
* I think what is needed now is a nice, concrete discussion of how to
* DO PCT research -- both in the field and in the lab. A good field
* research example of "The Test" was given by Bill Powers in his analysis
* of Greg William's net behavior. I admit that I have spent a lot more
* time kvetching about the uselessness of IV-DV research than describing
* ways to go about doing PCT research.

There is no information in Rick's posting about anything except
his mental disturbances (and bad consciences as an IV-DV ex-merchant).

:slight_smile:

--Oded

--

Oded Maler, LGI-IMAG, Bat D, B.P. 53x, 38041 Grenoble, France
Phone: 76635846 Fax: 76446675 e-mail: maler@imag.fr

[From Rick Marken (930407.0900)]

I said:

The living being I see
portrayed by the IT group is the old computer image of the
"information processing" school of psychology: perceptual
input, p, is transformed by a computer program into the output,o.

Martin Taylor (930406 18:30) replies --

All of what I have been writing has been based on the idea that the
above description is just as false as Rick says it is.

You can't have your cake and eat it too. I know that you SAY that
you don't like the SR or information processing concept of behavioral
organization. But your entire argument in this "information about
the disturbance in perception" debate has been based completely on
an SR view of the operation of a control system. You want to have
your PCT and eat SR theory (dressed up as IT) too. This
is a VERY COMMON affliction. We've seen it MANY times -- witness
the Karolyans, etc. Our experience with this "have PCT and eat
theory X too" is, I imagine, the reason why Bill Powers
posted the following, painfully true complaint:

Under the best of circumstances, people who believe in
X learn about control theory, and say "Wow, sure, PCT is the most
revolutionary thing since sliced bread, but you know that we Xers
have been saying the same things all along, and when you get
right down to the science of it, X is fundamental. We're glad to
see that PCT fully supports X."

PCT and IT (as you've described it) are NOT compatible. If PCT is
right about how living systems work, then IT (or, at least, the
aspect of it that you have described as being related to the functioning
of a control system) is wrong. The fact that you think I am invoking
"magic" as the basis for the operation of a control system when I say
that there is no information about the disturbance in perception (now
proved quantitatively with the data posted by Bill and myself) shows
that you don't understand how a control system works. There
is no magic -- unless you consider a computer program to be magic.

Your inability to shuffle off the IT coil prevents you from seeing
the revolutionary implications of PCT for understanding and STUDYING
living systems. If you could get past IT you would see that a statement
like:

I assert that investigation of the
informational characteristics of the things to be controlled is an
essential part of achieving [the goals of PCT research]

makes no sense in terms of the PCT model. There are no "informational
characteristics" of "things to be controlled". Control systems
tell the "things to be controlled" what to do -- not vice versa.
More importantly, if you got all the way out of IT and all the way
into PCT you would see what I mean when I say:

IV-DV studies may provide hints about the existence of controlled
variables -- but they cannot be a basis for understanding the
behavior of control systems.

In behavioral science, IV-DV studies are studies of the effects of
environmental variables (potential disturbances, in PCT terms) on
measures of behavior (potential outputs, in PCT terms). So, under
the BEST of circumstances in behavioral science, the IV-DV researcher
is manipulating a variable (or variables) that is actually a disturbance,
d, to some controlled variable and measuring an output, o, that also has
an influence on that controlled variable. The BEST you can hope to find
in such studies (according to PCT) is a relationship of the form:
      -1
o = g (d)

The dependent variable, o, is related to the independent variable, d,
according to the inverse of the feedback function, g(), relating
o to p. Thus, under the BEST of circumstances, the results of IV-DV
research tells you ZILCH about the organism and plenty about the
connection between the organism and a controlled variable. After you have
completed this research you end with no idea what variables the subjects
is controlling for. That is, you don't know anything about the purpose
of their behavior. For a nice, tangible explanation of what this means in
the context of real research, see Powers (1971) Behavioral Science article
on a rat experiment (I think's in LCS I, I hope).

In actual practice, d and o have only moderate effects on a common
controlled variable. Moreover, each of the typically numerous subjects
in these experiment probably has a slightly different reference for the
state of that variable. The result of this is that the relationship
between o and d in IV-DV studies appears to be VERY noisy -- ergo,
statistics are (mis)used to find the "true" relationship between
o and d (IV and DV). More likely, statistics are used to see if it
is reasonable to conclude that there is ANY relationship between
o and d). Sound familiar?

I think what is needed now is a nice, concrete discussion of how to
DO PCT research -- both in the field and in the lab. A good field
research example of "The Test" was given by Bill Powers in his analysis
of Greg William's net behavior. I admit that I have spent a lot more
time kvetching about the uselessness of IV-DV research than describing
ways to go about doing PCT research.

So I'm going to defer sociology theory for a bit in favor of some
tangible suggestions about how to test for controlled variables.

It might be best to start with an actual example of an IV-DV
experiment, show what might be considered "hints" about controlled
variables (CVs) and suggest ways to nail down these CVs using the
test. I'll try to find a reasonable target study today. I sure
hope someone helps me out in this discussion.

I'm prepared to stop bashing IT; what I think we (PCTers) should
be doing now is explaining how to DO PCT.

OK?

Best

Rick