Controlled Variables (was Quantitative & Qualitative...)

[R_Marken]

[From Rick Marken (990603.0730)]

Me:

You use The Test to see which variables are controlled.

Marc Abrams (990602.1908)--

The Test with another individual is not a practical way to
do this

See Bill Powers (990503.0232 MDT) and Marken, R. (1997) The
Dancer and the Dance: Methods in the Study of Living Control
Systems, _Psychological Methods_, 2, 436-446

Me:

The first step in the study of purposeful behavior is the
identification of CVs.

Marc

And you can't do that until you understand the _nature_ of a CV.

See Bill Powers (990503.0232 MDT) and Marken, R. (1997) The
Dancer and the Dance: Methods in the Study of Living Control
Systems, _Psychological Methods_, 2, 436-446

Me

We can observe controlled perceptions...just as well as we can
see actions or disturbances.

Marc:

Fine, But how do you _know_ it's a CV.

See Bill Powers (990503.0232 MDT) and Marken, R. (1997) The
Dancer and the Dance: Methods in the Study of Living Control
Systems, _Psychological Methods_, 2, 436-446

Marc:

When I say we cannot observe them (CVs), I mean we cannot tell
them apart from uncontrolled actions, unless we do the test,
and doing the test is no guarentee that we will uncover the CV.

See Bill Powers (990503.0232 MDT)

How many times have you done the Test? Can you relate one or
two experiences and tell me if those CV's were components of
other CV's. Did these CV's have to "cooperate with other CV's
to get things done? What else did you learn about CV's from
doing the test?

See http://home.earthlink.net/~rmarken/ControlDemo/HP.html
and paper starting on p. 159 of _Mind Readings_.

I wonder what answers the test would provide with regard to
what you were controlling for with regard to Aquiring PCT
knowledge.

Why not try it? I think you will find that you are able to
see CVs too. In fact, I think you are seeing some (of mine),
though the image may be a bit fuzzy still. Keep asking
questions or proposing answers (both being potential disturbances)
and watch to see what variables I protect from these disturbances.

Best

Rick

···

---
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken

[Marc_Abrams2]

from [ Marc Abrams (990603.1131) ]

[From Rick Marken (990603.0730)]

Me:

> You use The Test to see which variables are controlled.

Marc Abrams (990602.1908)--

> The Test with another individual is not a practical way to
> do this

See Bill Powers (990503.0232 MDT) and Marken, R. (1997) The
Dancer and the Dance: Methods in the Study of Living Control
Systems, _Psychological Methods_, 2, 436-446

I may be wrong, but In my opinion, the Test, as presently configured, is
difficult to do and will _never_ produce the kinds and amounts of data you
need, It is too time consuming to do, and useful research on the hierarchy
and CV's. need to be well documented. As long as the experiments remain on
the computer and tracking tasks are the main research vehicle, we are ok. If
we move to non-computer, non-tracking tasks, things get a bit more
complicated. If that were not the case. It would have already been done, or
at least some efforts would have been made. As you suggested to me, read
Bill's post (990603.0232) _CAREFULLY_. There is _much_ that needs to be done
for the Test to be a _practical_, _research_ tool. The Test as presently
configured is an excellent way to show the _phenomenon_ of a CV. It's an
entirely different matter when you are trying to understand the nature of
the hierarchy and CV's.

Me:

> The first step in the study of purposeful behavior is the
> identification of CVs.

Marc

> And you can't do that until you understand the _nature_ of a CV.

See Bill Powers (990503.0232 MDT) and Marken, R. (1997) The
Dancer and the Dance: Methods in the Study of Living Control
Systems, _Psychological Methods_, 2, 436-446

You keep on throwing up Bill's post. I suggest _you_ read it.

Me

> We can observe controlled perceptions...just as well as we can
> see actions or disturbances.

Marc:

> Fine, But how do you _know_ it's a CV.

See Bill Powers (990503.0232 MDT) and Marken, R. (1997) The
Dancer and the Dance: Methods in the Study of Living Control
Systems, _Psychological Methods_, 2, 436-446

Bill doesn't think you can see controlled perceptions I quote directly from
his post;
"...
It's the other way around. ALL we can observe of another person's
controlling is the CV; what we can't observe is the "CP" -- the controlled
perception in the other person."

Marc:

> When I say we cannot observe them (CVs), I mean we cannot tell
> them apart from uncontrolled actions, unless we do the test,
> and doing the test is no guarentee that we will uncover the CV.

Rick

See Bill Powers (990503.0232 MDT)

Me:
No, _you_ see that post. Another quote from Bill:

"We can be mistaken about the CV, in that the other person might be
controlling a perception derived from the environment in ways different
from our way of perceiving the CV, or might be controlling something
closely related to our concept of the CV but not identical to it, and so
on. We must, of course, be aware of this and not settle for the first
definition that we stumble across."

> How many times have you done the Test? Can you relate one or
> two experiences and tell me if those CV's were components of
> other CV's. Did these CV's have to "cooperate with other CV's
> to get things done? What else did you learn about CV's from
> doing the test?

See http://home.earthlink.net/~rmarken/ControlDemo/HP.html
and paper starting on p. 159 of _Mind Readings_.

Thanks for the reference. I'll read the paper and check out the demo. Btw,
can you send me the Dancer paper?

> I wonder what answers the test would provide with regard to
> what you were controlling for with regard to Aquiring PCT
> knowledge.

Why not try it? I think you will find that you are able to
see CVs too. In fact, I think you are seeing some (of mine),
though the image may be a bit fuzzy still. Keep asking
questions or proposing answers (both being potential disturbances)
and watch to see what variables I protect from these disturbances.

Not right now. I have a few things on my plate. Maybe you will do a SDMOL
with regard to them?

Marc

[Isaac_Kurtzer1]

i.kurtzer (990603.1430)

>From [ Marc Abrams (990603.1131) ]

> [From Rick Marken (990603.0730)]
>
I may be wrong, but In my opinion, the Test, as presently configured, is
difficult to do and will _never_ produce the kinds and amounts of data you
need, It is too time consuming to do, and useful research on the hierarchy
and CV's. need to be well documented. As long as the experiments remain on
the computer and tracking tasks are the main research vehicle, we are ok. If
we move to non-computer, non-tracking tasks, things get a bit more
complicated. If that were not the case. It would have already been done, or
at least some efforts would have been made.

A. Method is the logic of how you design an experiment. It is independent of
whether you are studying "being a good communist", or "catching a baseball".
The complications comes with the experimental procedures which are peculiar to
each subject, questioning the viability of a surplus economy versus adding a
tradewind, but the logic is the same: push on something W you know has a
connection to X and see if X stays the same relative to Y which also is pushed
by W.
B.The alleged limits of the Test are not what stops people from using the Test.
It the limits of their own imaginations or an unwillingness to try something
different.

As you suggested to me, read
Bill's post (990603.0232) _CAREFULLY_. There is _much_ that needs to be done
for the Test to be a _practical_, _research_ tool. The Test as presently
configured is an excellent way to show the _phenomenon_ of a CV. It's an
entirely different matter when you are trying to understand the nature of
the hierarchy and CV's.

You have made a blanket statement about the insufficiency of the the "TEST as
currently configured", but this is an empirical question. I cannot see any
reason why the Test is insufficient. Do have a particular example in mind?

i.

[Marc_Abrams2]

from [ Marc Abrams (990603.1846)]

i.kurtzer (990603.1430)

A. Method is the logic of how you design an experiment. It is independent

of

whether you are studying "being a good communist", or "catching a

baseball".

The complications comes with the experimental procedures which are

peculiar to

each subject, questioning the viability of a surplus economy versus adding

a

tradewind, but the logic is the same: push on something W you know has a
connection to X and see if X stays the same relative to Y which also is

pushed

by W.

OK. Did I say something that countered this argument?

B.The alleged limits of the Test are not what stops people from using the

Test.

It the limits of their own imaginations or an unwillingness to try

something

different.

I, I don't know if or why people do or don't use the Test. I was making a
statement about the practicality ( not the validity ) of getting the right
kinds of data, in the amounts we need. To better understand the nature of
CV's and levels,

> As you suggested to me, read
> Bill's post (990603.0232) _CAREFULLY_. There is _much_ that needs to be

done

> for the Test to be a _practical_, _research_ tool. The Test as presently
> configured is an excellent way to show the _phenomenon_ of a CV. It's an
> entirely different matter when you are trying to understand the nature

of

> the hierarchy and CV's.
>

You have made a blanket statement about the insufficiency of the the "TEST

as

currently configured", but this is an empirical question. I cannot see

any

reason why the Test is insufficient. Do have a particular example in

mind?

OK, What have we learned about the nature of CV's from doing the test so
far? What do I mean by "nature"? How are they changed how are they formed,
reformed, etc. The purpose of the test has _ONE_ goal. To _identify_ a CV.
It does that well, _with_ a well trained individual facilitating.

Marc

[Isaac_Kurtzer1]

i.kurtzer (990604.0030)

>From [ Marc Abrams (990603.1846)]

> i.kurtzer (990603.1430)
>
> A. Method is the logic of how you design an experiment. It is independent
of
> whether you are studying "being a good communist", or "catching a
baseball".
> The complications comes with the experimental procedures which are
peculiar to
> each subject, questioning the viability of a surplus economy versus adding
a
> tradewind, but the logic is the same: push on something W you know has a
> connection to X and see if X stays the same relative to Y which also is
pushed
> by W.

OK. Did I say something that countered this argument?

When you say"The Test as presently
configured is an excellent way to show the _phenomenon_ of a CV. It's an
entirely different matter when you are trying to understand the nature of
the hierarchy and CV's."

and
" the Test, as presently configured, is
difficult to do and will _never_ produce the kinds and amounts of data you
need, It is too time consuming to do, and useful research on the hierarchy
and CV's. need to be well documented. As long as the experiments remain on
the computer and tracking tasks are the main research vehicle, we are ok. If
we move to non-computer, non-tracking tasks, things get a bit more
complicated. "

The difficulty of the Test is its implementation, not the logic, i.e. how its
"presently configured" . There are a bevy of peculiarities that still need to
be worked out methodologically but the fundamentals seem pretty airtight. Just
as one example, lets say that we want to test the perception "staying between
the lanes". There are many enviromental variables that this might correspond to
and we start out by being little demons and pushing on the car. A conventional
psychologist might pick up the car momentarily and put it at different positions
on the road. Then they would ask "is this "staying between the lanes" ?" and
you could say "yes" or "no". With three lanes you would find three peaks of
yes's fading to no's in between. With the Test using a continuous disturbance
you would find the car staying put, and you might infer that the regions that
the variable would be in had the distubance not been countered are regions that
do not match the reference. This is the problem of local minima. If you gave a
step pulse followed by a continuous disturbance you would see the system setting
into and maintaining a new position on the road since the CV was not any
particular lane but between a lane. These are experimental difficulties, but
the logic is the same.

OK, What have we learned about the nature of CV's from doing the test so
far? What do I mean by "nature"? How are they changed how are they formed,
reformed, etc. The purpose of the test has _ONE_ goal. To _identify_ a CV.
It does that well, _with_ a well trained individual facilitating.

Those are some very lofty goals. I think that we are not in the position to
worry about that just yet since we do not even have enough CV's to define a
level. But tell me if you make any progress.

i.

[Marc_Abrams2]

from [ Marc Abrams (990604.2307) ]

i.kurtzer (990604.0030)

The difficulty of the Test is its implementation, not the logic, i.e. how

its

"presently configured" .

Thats what I said, or at least meant to say.

There are a bevy of peculiarities that still need to
be worked out methodologically but the fundamentals seem pretty airtight.

I thought you said the "method" was the logic. You just contradicted your
last statement. Can you please clarify.

Just as one example, lets say that we want to test the perception "staying

between

the lanes". There are many enviromental variables that this might

correspond to

and we start out by being little demons and pushing on the car. A

conventional

psychologist might pick up the car momentarily and put it at different

positions

on the road. Then they would ask "is this "staying between the lanes" ?"

and

you could say "yes" or "no". With three lanes you would find three peaks

of

yes's fading to no's in between. With the Test using a continuous

disturbance

you would find the car staying put, and you might infer that the regions

that

the variable would be in had the distubance not been countered are regions

that

do not match the reference. This is the problem of local minima. If you

gave a

step pulse followed by a continuous disturbance you would see the system

setting

into and maintaining a new position on the road since the CV was not any
particular lane but between a lane. These are experimental difficulties,

but

the logic is the same.

Isaac, Can you give me a an actual Test that was performed between _two_
people, where ther were no computers involved. Out side of _possibly_
identifying one or more CV's what else was accomplished? You seem to be
taking this as an attack on the test. It's not. I am _trying_ to get some
discussion going on how we might be able to possibly refine the test, and
MOL and possibly start being able to look at the nature of CV's and levels.
I haven't been feeling all that well over this past week so I haven't been
able to finish my work on the SDMOL material. I am shooting for Sunday. I
will probably be going into the hospital for a series of tests and
monitoring next week, that could last anywhere from 3 to 7 days. So I will
not be able to respond to any posts until after I get out. I will spend the
rest of the time I can this weekend finishing the work on the SDMOL. ( In
between the Mets-Yanks and Pacers-Knicks series )

I think the SDMOL _could_ ( as in; maybe, it's got a chance, etc. ) be
important. You asked me to explain how it could possibly be both a "casting
the net" and a "testing the speciman" method. I explained it and received no
feedback from you or anyone else on it. Isaac, I ask questions because I
sincerly want to know the answers to them. If you think what I am attempting
to do is BS say so and tell me why.

> OK, What have we learned about the nature of CV's from doing the test so
> far? What do I mean by "nature"? How are they changed how are they

formed,

> reformed, etc. The purpose of the test has _ONE_ goal. To _identify_ a

CV.

> It does that well, _with_ a well trained individual facilitating.
>

Those are some very lofty goals. I think that we are not in the position

to

worry about that just yet since we do not even have enough CV's to define

a

level. But tell me if you make any progress.

What _are_ we in a position to worry about? You haven't seen the SDMOL, yet
you seem to dismiss it out of hand.

You also seem to think that there are a finite number or types/kinds of
CV's. How come?
How many CV's comprise a level?. What is a level?

What have _you_ done to put us in a better position to answer some of these
questions. Are you actually going to get involved in some PCT research in
your Phd program or does that present to many "political" problems? What are
some of your interests vis a vie PCT, and the work you will ultimately be
doing when you finish school.

Lofty goals? Maybe, maybe not. But they are some goals. But I bet that even
if I can't reach them now, I'll end up knowing why, which is never to far
away from being able to reach them.

and finally,

...But tell me if you make any progress

What are you attempting to control for with that statement? Are you really
that smug?

Marc

[Isaac_Kurtzer1]

i.kurtzer (990607.1300)

From [ Marc Abrams (990604.2307) ]

> i.kurtzer (990604.0030)

> The difficulty of the Test is its implementation, not the logic, i.e. how
its
> "presently configured" .

Thats what I said, or at least meant to say.

Good.

>There are a bevy of peculiarities that still need to
> be worked out methodologically but the fundamentals seem pretty airtight.

I thought you said the "method" was the logic. You just contradicted your
last statement. Can you please clarify.

Method is the logic, but five statements are not a fully-fleshed out
methodology. Our premises seem pretty tight but further developments of
the method are needed. Airtight on the premises and fully-formed are not
synonyms.

I think the SDMOL _could_ ( as in; maybe, it's got a chance, etc. ) be
important. You asked me to explain how it could possibly be both a "casting
the net" and a "testing the speciman" method. I explained it and received no
feedback from you or anyone else on it.

I am reserving judgement. Introspection on controlled variables seems to
be very sketchy, and circuitous.

What _are_ we in a position to worry about? You haven't seen the SDMOL, yet
you seem to dismiss it out of hand.

See above.

You also seem to think that there are a finite number or types/kinds of
CV's. How come?
How many CV's comprise a level?. What is a level?

How is it that you ask these questions when you know noone has done the
research? Are you asking my personal opinion?

What have _you_ done to put us in a better position to answer some of these
questions.

I don't have to do anything for you or any "us". I have no obligation
and noone can tell me i havn't paid my dues.
If you want experiemntal examples of my contribution to PCT via
experiment check out the CSG conference tapes '92 and '96. For theory
check out '98. If that isn't sufficient then tough shit.

i.

[Marc_Abrams2]

from [ Marc Abrams (990617.1201) ]

i.kurtzer (990607.1300)

> >There are a bevy of peculiarities that still need to
> > be worked out methodologically but the fundamentals seem pretty

airtight.

How do we move from the "fundamentals" to a tangible methodology? The
question is rhetorical. If we ever hope to reseaerch levels and CV's we have
got to come up with someway of looking at these things.

Method is the logic, but five statements are not a fully-fleshed out
methodology. Our premises seem pretty tight but further developments of
the method are needed. Airtight on the premises and fully-formed are not
synonyms.

You just reiterated my point. Thank you.

I am reserving judgement. Introspection on controlled variables seems to
be very sketchy, and circuitous.

As well you should. But i don't need or want your "judgement". I'd love to
see you contributing to the effort ( i.e. of understanding levels and CV's )
If introspection is tough how can you possibly begin to think you could make
heads or tails from what you observe from others?

> What _are_ we in a position to worry about? You haven't seen the SDMOL,

yet

> you seem to dismiss it out of hand.

See above.

I have. You don't present an altermative means, nor do you go beyond the
simple statement that introspection of CV's seems to be sketchy and
circuitous.

Why do you believe this?

> You also seem to think that there are a finite number or types/kinds of
> CV's. How come?
> How many CV's comprise a level?. What is a level?

How is it that you ask these questions when you know noone has done the
research? Are you asking my personal opinion?

I asked the first because your statement suggested to me that you felt there
were a finite number of CV's, and that you could "classify" them into types.
If I was mistaken. so be it. Your personal opinion ( when stated as
such ) is always valuable. We would have a tough time building theories if
that wasn't so.

> What have _you_ done to put us in a better position to answer some of

these

> questions.

I don't have to do anything for you or any "us". I have no obligation
and noone can tell me i havn't paid my dues.
If you want experiemntal examples of my contribution to PCT via
experiment check out the CSG conference tapes '92 and '96. For theory
check out '98. If that isn't sufficient then tough shit.

My friend, you have a problem. You wrote a post basically dismissing
something you haven't even seen yet. You did not come back with _any_
constructive statements. Such as:
1) We have already tried that and it didn't work or
2) I think you might want to ....

Get my drift? My question above was aimed _specifically_ at what I was
talking about ( i.e. CV's and levels ) I wasn't questioning your manhood ar
your committmernt to PCT.

As for your last statement. Chew harder.

Marc