Doing The Test

[Stefan_Balke]

From Stefan Balke (960806.1100 CET)

Rick Marken (960805.0740) --

After all, the not very subtle subtext of the paper is that the experimental
method, as currently used in the behavioral sciences, tells us nearly nothing
about the behavior of organisms and that the only way to find out about
behavior is by Testing to determine the variables organisms control.

Hi Rick,

in your post to Bruce Gregory you pointed out again the importance of doing
The Test as a 'conditio sine qua non' for psychological research and,
needless to say, I agree that this is right. So, we have to do The Test, ok.
But, as it has been mentioned before, their are some difficulties in the
practice.

To make things easier I started to ask/observe myself what perceptions I was
controlling for within the last 15 minutes, instead of doing The Test with
another person where I don't have such an easy access. I could tell you all
of my controlled perceptions except those, which are unknown even for
myself, maybe because they do not exist at all, maybe because they exist
only in form of non aware rules, ideal, standards, norms, biological
constrains, etc, but I fear I don't have the literary qualities, which are
needed to make it readable (I highly recommend T. C. Boyle's very close
descriptions of his actors perceptions). Altogether I can say, that their
had been at least 5 perceptions where I did some counteraction in order to
control a perception in the last 15 min.

1. to perceive a pleasant room temperatur -> closing the window - easy to
observe and easy to Test.
2. to perceive a pleasant position at the chair -> finding a good position -
easy to observe and easy to Test.
3. to be helpful and to be competent -> answering the question of a
colleague in a nice and smart way - easy to observe, not as easy to Test.
4. to write this note -> concentrating and selecting the important ideas for
my intention - hard to observe, impossible to Test.
5. to write in proper English -> searching for the right words in my mind
(') and in the dictionary ('') - (') not observable, ('') easy to observe,
easy to Test.

And there are lots of other perceptions I do control for which remain
totally invisible for me and the observer, because there was no disturbance
and therefore no visible action or a thought.

But we could imagine in a mind experiment, that we would be able to perceive
the controlled perceptions of another person (and this is not so absurd, if
you think of a very close person) what would be the use? Direct
communication without misunderstandings? Ok, that would be useful enough,
but what else? Would it bring to light anything new about the human nature?
What is to expect?

Best, Stefan

[Richard_Marken]

[From Rick Marken (960806.0900)]

Stefan Balke (960806.1100 CET) --

in your post to Bruce Gregory you pointed out again the importance of doing
The Test as a 'conditio sine qua non' for psychological research...So, we
have to do The Test

We only _have to_ do The Test if 1) we are convinced that organisms are
living control systems and 2) we want to understand what these organisms are
doing (controlling) and how they are doing it. The Test is the only approach
to the study of living control systems that makes sense when one understands
how control systems work.

(I highly recommend T. C. Boyle's very close descriptions of his actors
perceptions).

I'd like to see this; what book are you referring to? I've never heard of
T. C. Boyle.

But we could imagine in a mind experiment, that we would be able to perceive
the controlled perceptions of another person (and this is not so absurd, if
you think of a very close person) what would be the use? Direct
communication without misunderstandings? Ok, that would be useful enough,
but what else? Would it bring to light anything new about the human nature?
What is to expect?

I think these are _excellent_ questions. I believe that psychologists have
problems with The Test, not only because it is an unfamiliar procedure but
also because it is used to achieve unfamiliar goals. The goal of conventional
methodology is the "prediction and control of behavior"; the "usefulness" of
conventional methodology is built into this goal. There is no question about
the usefulness of knowing how IQ scores correlate with GPA or how the number
of bystanders affects the chances that any one will help in an emergency or
how delay between reponse and reward affects the probability of learning the
response, etc. Clearly, these kinds of results -- the results of conventional
methodology -- help us predict (things like GPA and "helping") and control
(things like "learning").

The goal of PCT research is not to predict and control behavior; the goal is
to _understand_ behavior. This goal must seem rather "fluffy" to people who
want to use their knowledge to improve human behavior (ie. control it).
Psychologists (like all people) are controllers so they tend to evaluate
behavioral knowledge in terms of control: behavioral knowledge is "useful" if
it can help people control other people. "Understanding" makes sense to such
psychologists only in terms of how well it helps them control behavior.

PCT psychologists are controllers too. But their goal is not to control
behavior; their goal is to control the relationship between a model (PCT) and
observation (the controlling observed via The Test). I call this relationship
"understanding". To the extent that the model mimics observation, the PCT
psychologist is in control of his understanding of behavior.

Understanding is under control when the relationship between model (invented
reality) and observation is maintained at a reference value such as "nearly
identical" (model and observation behave in just about the same way).
Understanding is controlled by continuously Testing (observing) and comparing
the results of these Tests to the behavior of the model. If the model stops
behaving like the new observations, we have lost control of our
understanding. Understanding is brought back under control by improving the
model or finding a new model whose behavior matches all observations.

I believe that the goal of understanding behavior (in the sense described
above) is worthwhile. I can use my understanding to predict and control
behavior; but I can also use my understanding to see the limitations of
prediction and control when dealing with the behavior of a hierarchical
organization of control systems. I can apply this understanding to myself and
see when I am creating problems for myself unnecessarily. I can apply this
understanding to others and be aware that what bothers me about their
behavior is often just an irrelevant side effect of their controlling; I can
also be aware that irrelevant side effects of my controlling are likely to
disturb others. I can see the value in trying to cooperate with others in a
world where everyone is trying to control their own perceptions as best as
they can; I can see the problems created for _everyone_ when some people
cannot control what they need to control or when they cannot find
means of controlling that don't interfere with the ability of others to
control; I can see the terrible waste created by conflict and I can see that
conflict is a natural (and unncessary) consequence of control.

I feel like it has been VERY worthwhile (for me) to try to control for
understanding behavior (instead of trying to control for aspects of other
people's behavior); but that's just me. If PCT says _anything_ it's that
people control for perceptions that work for them (as means of controlling
for higher order perceptions). Understanding works for me; I can only hope
that it works for many others as well.

Best

Rick