Another Call for Controlled Variables

[From Rick Marken (981120.0845)]

Although I was trained as a conventional psychologist (and even
received a PhD degree in experimental psychology signed by soon
to be President Reagan), I am (as other conventional psychologists,
sociologists, etc on this list can attest) not a very good one. At
least, I am not very good at being able to find all the controlled
variables that have been discovered by conventional psychologists.

As I have noted before, I believe the concept of a _controlled
variable_ -- Powers' recognition that there are variable aspects of
one's own experience that correspond to variables being controlled
by other controllers -- is the central contribution of PCT to
understanding human (and animal) nature. It is, I believe, a concept
with which conventional psychologists are completely unfamiliar.

And their unfamiliarity with the concept of the controlled variable,
I believe, is why there is no evidence of systematic attempts by
conventional psychologists to discover what variables organisms are
controlling when they are engaged in various observable behaviors.
This, I believe, is why it's hard (impossible?) to find any
conventional psychological research studies aimed at the testing
for controlled variables. It is also, I believe, why PCT has been
(and continues to be) ignored by the behavioral science community;
it's not surprising that there would be very little interest in
a theory (PCT) that explains an unknown phenomenon (controlled
variables).

Nevertheless, as I am currently in the process of writing a paper
on the concept of controlled variables for an audience of
conventional psycholgists (I plan to submit the completed paper
to Psych Bulletin or Psych Review) and I would like to be able to
present some examples of research that could at least be interpreted
as revealing the possible existence of controlled variables. So I
would appreciate any references to behavioral research articles that
describe research that seems to indicate (or, better,tests for) the
existence of controlled variables.

I did put out a previous call for references to conventional
psychology research papers that seem to reveal controlled variables.
The response was not overwhelming; I would like to get some more
references to studies of _human_ behavior that seem to reveal
controlled variables. So if anyone can think of such references
(I'd like author, title, journal, date on the article) I'd really
appreciate it.

Thanks

Rick

···

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

[From Chris Cherpas (981120.1010 PT)]

Rick Marken (981120.0845)--

As I have noted before, I believe the concept of a _controlled
variable_ -- Powers' recognition that there are variable aspects of
one's own experience that correspond to variables being controlled
by other controllers -- is the central contribution of PCT to
understanding human (and animal) nature.

So, controlled variables is a subset of controlled perceptions,
ones that "correspond to variables being controlled by other
controllers?"

... I am currently in the process of writing a paper
on the concept of controlled variables...

I very interested in the relations between intrinsic variables,
controlled perceptions in the behavioral hierarchy, and controlled
variables (as above). There's been a lot of great work already,
any new statements, revisions, etc. could help me sort these out
better.

In general:
Control puts such an interesting perspective on "nature-nuture"
questions. For example, since reading Powers' work on evolution,
I have a new concept of myself as carrying around parts of the
evolutionary soup of original chemical control systems, as opposed
to merely some (plan-and-execute or reactive style) genes that
are just "programmed" to the do the right thing to build bodies
12 ways, as the old Wonder Bread commercials used to say. Whether
my cartoon version here is really what the author intended is
another question, of course.

Anyway, it's a reason why I'm asking about relations with intrinsic
variables. I don't know how metaphorical I'm really being about
"carrying around" the beginnings of life, obviously, but Bill's
work has profoundly altered my view of the potential for
conceptually synthesizing evolutionary biology and psychology, in
particular, and, in an integrative Theory of Life, which has the
scope of the big theories of physics.

Best regards,
cc

[From Rick Marken (981120.1440)]

Me:

I believe the concept of a _controlled variable_ -- Powers'
recognition that there are variable aspects of one's own
experience that correspond to variables being controlled by other
controllers -- is the central contribution of PCT to understanding
human (and animal) nature.

Chris Cherpas (981120.1010 PT)]

So, controlled variables is a subset of controlled perceptions,
ones that "correspond to variables being controlled by other
controllers?"

Sorry. I guess I was unclear. Controlled variables _are_ controlled
perceptions as seen from the point of view of the observer; they
are perceptions in the observer that correspond to perceptions
_controlled_ by the controller. For example, temperature is a
perception that I have. My perception of temperature _corresponds_
(closely but not exactly) to the perception controlled by
another controller (the thermostat) when I am in my office; my
perception of temperature corresponds to the theremostat's
controlled variable. My perception of temperature does not
correspond to a perception controlled by another controller when
I am outside; the temperature I experience is just caused by all
those weather variables (pressure, solar radiation, etc).

I very interested in the relations between intrinsic variables,
controlled perceptions in the behavioral hierarchy, and controlled
variables (as above). There's been a lot of great work already,
any new statements, revisions, etc. could help me sort these out
better.

Do you mean there is published research on the relationship between
intrinsic and controlled variables? If so, I would like to get the
reference to it.

Best

Rick

···

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

[From Chris Cherpas (981120.1530 PT)]

Chris Cherpas (981120.1010 PT)--

So, controlled variables is a subset of controlled perceptions,
ones that "correspond to variables being controlled by other
controllers?"

Rick Marken (981120.1440)--

Sorry. I guess I was unclear.

You were clear. Maybe my use of the term "subset" isn't what
you expected; the set of controlled perceptions having this
correspondence would be a "subset" of all controlled perceptions:

Rick Marken (981120.1440)--

Controlled variables _are_ controlled perceptions as seen from
the point of view of the observer; they are perceptions in the
observer that correspond to perceptions _controlled_ by the
controller.

Which makes sense for the Test.

Chris Cherpas (981120.1010 PT)--

I very interested in the relations between intrinsic variables,
controlled perceptions in the behavioral hierarchy, and controlled
variables (as above). There's been a lot of great work already,
any new statements, revisions, etc. could help me sort these out
better.

Rick Marken (981120.1440)--

Do you mean there is published research on the relationship between
intrinsic and controlled variables? If so, I would like to get the
reference to it.

No, I just meant what Bill, for the most part, has already written
in describing the overall theory, as in B:CP. I wondered if there
were any implications that had come up, independent of any specific
experimental work. For example, how would one approach the problem
of "Testing for an intrinsic variable?" In other words, what special
considerations are there for even _approaching_ experimental work with
intrinsics?

Best regards,
cc

[From Erling Jorgensen (981121.1315)]

[Rick Marken (981120.0845)]

I would like to be able to
present some examples of research that could at least be interpreted
as revealing the possible existence of controlled variables. So I
would appreciate any references to behavioral research articles that
describe research that seems to indicate (or, better,tests for) the
existence of controlled variables.

Rick, I don't know if the following references are really what you're
after, or whether they will meet your [rigorous & generally loud ;)]
standards for what counts as acceptable research. For one thing,
they deal with perceptions high up in the proposed hierarchy,
admittedly a difficult region to test systematically. For another,
most of them do not come from mainstream psychology -- whether that
is conceived as "conventional," "behavioral," or "experimental" --
so they may not carry any weight with your intended audience. A
few seem more from the "social constructivist" camp, and that would
be a strike against them right there (perhaps in your own eyes as
well, I'm not sure.)

But ever since coming across these articles, I have been struck by
their ingenuity and clarity, and the sense that they somehow attend
to control processes at work in interactive settings; so, see what
you think. (In what follows, I'll capitalize words if I'm making
an explicit reference to Bill's hierarchy, [although he may be
getting tired of seeing those labels bandied about!]).

Fivaz-Depeursinge, Elisabeth (1991). Documenting a time-bound,
circular view of hierarchies: A microanalysis of parent-infant
dyadic interaction. _Family Process, 30_, 101-120.

Deals with how certain parent-infant episodes are constructed and
modified (which sounds like a control process to me.) I see
episodes here as consisting of behavioral Events which are brought
into Relationships of temporal correlation. If the parent is
varying perceptions of Events and Relationships in order to
stabilize a perception of an episode (e.g., reciprocal gaze),
presumably episodes can be thought of as either Sequence or Program
perceptions.

This article suggests that some episodes are being nested inside
other episodes. A compelling strength of the article is that it
uses time frames to determine those relative hierarchical positions.
In other words, longer (and slower) episodes are higher up in the
hierarchy, framing briefer (and faster) episodes. This is exactly
the relationship PCT would predict -- slower control systems set
the terms, while faster control systems become the means for
carrying out the higher level control. I believe the Plooij's
used a similar methodology/insight in their observations and
classification of the developmental control systems of chimpanzees.

Tannen, Deborah & Wallat, Cynthia (1987). Interactive frames
and knowledge schemas in interaction: Examples from a medical
examination / interview. _Social Psychology Quarterly, 50_ (2),
205-216.

This article seems to show a doctor controlling for, and changing
the Sequence of, communicative episodes (Programs?) consisting of
examining a child, consulting with her mother, and teaching
residents via a simultaneous videotaping of the session. She
frames and signals a change between these different aspects of the
interview, primarily through shifts in her vocal register -- e.g.,
"motherese" vs. normal conversation vs. flat intonation.

Cools, A. R. (1985). Brain and behavior: Hierarchy of feedback
systems and control of input. In P.P.G. Bateson & P.H. Klopfer
(Eds.), _Perspectives in Ethology, Volume 6: Mechanisms_ (109-168).
New York: Plenum Press.

You may know of this article already, since it is explicitly based
on Powers' work (B:CP, & the _Science_ article in 1973). It
identifies brain areas (viz., neostriatum, substantia nigra, and
colliculus superior) involved in controlling Programs via "the
ordering and sequencing of behavioral states" (p.112). It is not
a study of humans, as you requested, but one using invasive
techniques with rats and cats. However, its methodology and
reasoning are quite impressive.

Baumeister, Roy F. & Newman, Leonard S. (1994). Self-regulation
of cognitive inference and decision processes. _Personality and
Social Psychology Bulletin, 20_ (1), 3-19.

I haven't had a chance to look at this closely, but it seems to
involve what may be control of Principles (inference processes).
It suggests that a scientist vs. a lawyer would use different
Programs as they gather evidence, assess & reassess implications,
and integrate the data, in order to self-regulate their inferences.

Gergen, Kenneth J. & Gergen, Mary M. (1988). Narrative and the
self as relationship. _Advances in Experimental Social Psychology,
Vol. 21_, 17-56.

Deals with how narratives of the self are constructed and maintained,
which I see as one example of control of high-level System Concepts.
Different Principles are offered as what may get varied to construct
coherent narratives, as well as how narratives are negotiated in the
social arena (protected against disturbances) via the accounts people
give for their actions. It also presents the structure of several
scenarios for expressing emotions, which are written as interactive
choice points -- (I was reminded of the decision nodes of Bill's
conception of Programs.)

Rick, while only one of the above authors seems to know of Bill's
work, I don't think we should insist that everything of value has
to use PCT terminology and concepts. I think it is perfectly
acceptable if a given author is moving in the right direction (in
terms of implicitly realizing the importance of control as a
phenomenon), _as long as_ they are not making huge errors due to
what they don't yet know about control processes. At any rate,
good luck with your publication efforts.

One other thought occurs to me. There seems to be a lot more
openness to cybernetics / control / systems concepts in certain
fields other than those mainstream psychology journals you
mentioned. I wonder if we should build more linkages to them,
rather than keep hammering at a fortress that doesn't want to
let down the gates. You might consider gearing your paper in
a different way for a more receptive audience.

All the best,

        Erling

[From Rick Marken (981121.2300)]

Erling Jorgensen (981121.1315)--

Rick, I don't know if the following references are really what
you're after, or whether they will meet your [rigorous & generally
loud ;)] standards for what counts as acceptable research.

They are exactly what I'm after. Thanks! I'll take a look and see
if there is anything in them that I can use. All I need is
exactly what you've given me -- suggestions.

Rick, while only one of the above authors seems to know of Bill's
work, I don't think we should insist that everything of value has
to use PCT terminology and concepts.

Nor do I. I'm actually more interested in work that was done
without knowledge of PCT. The McBeath et al study that I refer
to in my "Dancer..." paper is an example of what I'm looking
for: research that represents an attempt to determine a controlled
variable and is done without the researcher explicitly knowing or
saying that this is what is being done.

One other thought occurs to me. There seems to be a lot more
openness to cybernetics / control / systems concepts in certain
fields other than those mainstream psychology journals you
mentioned. I wonder if we should build more linkages to them,
rather than keep hammering at a fortress that doesn't want to
let down the gates. You might consider gearing your paper in
a different way for a more receptive audience.

I'm just interested in the scientific study of the behavior of
living control systems. Whoever is open to doing such studies is
OK with me. I've geared my papers to many different audiences
(cyberneticians, sociologists, biologists, addiction researchers,
etc) via many different forums, in an effort to reach people
who might be interested in doing this kind of research. I think
I have managed to be rejected by all audiences; I guess I'm an
equal opportunity rejectee;-)

By the way, I am not "hammering at the fortress" of mainstream
psychology because I have some particular interest in conquering
that fortress; nor do I think that that fortress is of particular
strategic importance. In fact, I don't even feel like I am hammering
anymore at all. I am just trying to publish papers in an archive
(mainstream psychology journals) that seems appropriate. I think
future psychologists who _do_ want to find out about scientific
studies of living control systems are more likely to look through
past issues of Psych Bulletin or Psych Review than through the
Journal of Deconstructed Sociological Cybernetics or whatever;-)

Mainstream psychology is supposed to be about the scientific
study of how living systems work; I think I am doing scientific
studies of how living systems work. So I try to publish my papers
in the places where I think such papers are supposed to go. I know
that the scientific psychologists who read these journals now cannot
do much more than ignore my work; they've got careers and reputations
to maintain. But I still think its better for me to put my stuff
before an audience (like mainstream psychologists) that is at least
ostensibly interested in the same things as me. And maybe some
some graduate student in 2010 will read, say, my 1997 Psych Methods
paper and become the next rigorous and loud PCT researcher;-)

Best

Rick

···

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

Rick,

I've been impressed by some of the stuff surveyed in

Dunbar, Robin. 1996.
  _Grooming, gossip, and the evolution of language_.
  Harvard U. Press.

For example (pp. 85-86, about as far as I've got in the book):

"Psychologists have developed a crucial test for theory of mind, known as
the 'false belief test'. It asks the key question: is the child aware that
someone else can hold a false belief (or at least6 a belief that the child
supposes to be false)? The now-classic example of this is the so-called
'Sally and Ann' test. Sally and Ann are two dolls who are presented to the
child and formally introduced. the child is shown that Sally has some
sweets, which she then places under a cushion on a chair. Having done this
(perhaps with the child's help), Sally leaves the room. Then, while Sally
is out of the room, Ann takes the sweets from under the pillow and puts
them in the pocket of her dress. When Sally comes back into the room, the
child is asked, 'Where does Sally think the sweets are?' Up to the age of
four, children invariably answer, 'In Ann's pocket.' But after about
four-and-a-half, they invariably say, 'Under the pillow,' adding with
conspiratorial glee, 'but they aren't there!'"

This looks like a new level or kind of control coming in around age 4.
Dunbar claims that a deficit in this respect is what we see in autism and
(less severely) in Asperger's syndrome, and infers from that and other
things that this capacity for social awareness and social function may be
more important than the (later-arriving) physical-environmental capacities
for conservation and so on that so exercised Piaget. There are per-chapter
lists of references that look like they would provide what you need.

It looks to me like there are numerous other suggestions in this book about
controlled variables. Aside from being thought provoking and a good read.

I'm sending this message direct because I decided to limit my posts to
CSG-L to models, and I have not yet had time to get going with Vensim.
However, the dissertation has been accepted, degree next month, so I hope
to get there in the next few months.

  Bruce Nevin

Hi Bruce --

I've been impressed by some of the stuff surveyed in

Dunbar, Robin. 1996.
  _Grooming, gossip, and the evolution of language_.
  Harvard U. Press.

Thank you! Thank you!

I'll take a look at it ASAP.

Congrats on the thesis.

Hasta luego

Rick

···

--

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