How to teach psychology?

[From Rick Marken (970310.1750)]

Dan Miller (970310)--

This sounds like a great course, Rick.

Thank you. Thank you.

Would you make the same decision [to leave academia] today?

I think so.

I did teach a couple of special courses on PCT when I was at Augsburg
College. But I still had to teach the regular Psych courses as well:
Intro, Research Methods, Statistics, Developmental, Learning,
Sensation/Perception/Cognition, Social, etc etc. I certainly would not
have been willing to teach all those courses from a PCT perspective,
even if it could be done. It would not have been fair to the students,
many of whom were hoping to go on to graduate school. You sure can't
pass the GRE if you've only taken PCT psychology.

And I certainly couldn't have gone on dishing out the conventional
wisdom in the courses I had to teach. When I did the research methods
course, for example, I felt like an astrophysicist teaching a course in
Sun signs;-)

So, yes. Given what I perceived (and still perceive) as the academic
resonsibiliites of a professor of psychology, I would make the same
decision now as I made then; I'd stop being a professor of psychology.

Best

Rick

from Tracy Harms (970310.1350 PST)

Rick Marken (970310.1100)

I heartily agree with your recommendations for teaching psychology as
informed by control-theory. My only disagreement arises here:

And, finally, I would discuss what one might expect to see when independent
control systems interact in the same environment; this would be the "social
psychology" section of the course.

In contrast, I would from the beginning nurture an awareness of ecological
interaction. By the time one gets around to the "social psychology"
section the principle of feedback among control systems should be familiar.
Not so familiar as the feedback within control systems, I suppose, but
enough so that the distinctly *social* aspects of this area are seen as a
special case of the more general pattern of ecological resolution.

Tracy Bruce Harms
harms@hackvan.com

[From Rick Marken (970310.1100)]

Lars Christian Smith (031097 12:30 CET) --

You have often stated your skepticism towards conventional psychology.

Well, it's not really "skepticism". It's just that, given what I now
understand about how control systems work I know that you can't find out
much about them using conventional methodology.

how would you actually teach an introductory psychology course? How
would you organize the course? What would you include, other than PCT?

I would start by pointing out the ubiquity of the controlling done by living
systems. I would present tons of evidence that organisms shape (control) the
environment, rather than vice versa. I would point to achievemments as
simple as walking and as skillful as writing fugues as evidence of
control. I would have all kinds of simple demos to illustrate the nature of
control (the ability to consistently produce preselected results in the face
of inconsistency). I would spend a LOT of time showing the students what
control IS, that it's all around them (and what they do) ALL the time, and
that it is what distinguishes the behavior of living systems from that of
non-living systems.

Once the phenomenon of control (purposeful behavior) is understood I would
describe the nature of systems that control. This would be done in the
context of simple control tasks (tracking tasks), but the appropriateness of
these control tasks as a model of ALL controlling would be clear (or should
be made clear) to the student at this point. In this section of the course
the student should come to understand what "control of perception" means. I
would also show that the perceptions that organisms control can be as
"simple" as the position of a line or as complex as one's position on
abortion. I would probably point out examples of the students' control of
complex perceptions at this point.

Once the student understands the phenomenon and the model of control I would
desribe how to study control; how to determine what living systems are
controlling. I would use demos like the "coin game" to illustrate this
methodology. I would also use some "real life" Tests of control;
conversational approaches to determine what is controlled, etc. To the extent
possible I would try to find studies in the "conventional" literature which
are either suggestive of POSSIBLE controlled variables (such as the
ethological studies Gary mentioned) or that actually are tests for controlled
variables (like the baseball catching article in Science). The idea would
be to help the student learn how to GUESS (based on observation of overt
behavior) what variables an organisms MIGHT be controlling and to suggest
studies (based on their new understanding of the nature and theory of
control) that would TEST these hypotheses.

I would then explain a bit about possible hierarchical relationships
between control systems in individuals (using the "portable demonstrator",
my spreadsheet model and some of the demos in my unpublished "Hierarchical
behavior..." paper). I would then do a little section on learning and
development of new control systems. I would probably rely heavily, here, on
the work done by the Plooij's on chimp and infant develpment of control.

And, finally, I would discuss what one might expect to see when independent
control systems interact in the same environment; this would be the "social
psychology" section of the course.

As you can see, there would not be much in this class about the "classical"
findings in psychology. I would use some ethological data but this would be
included in order to illustrate what we DON'T yet know and what we WANT to
know - not what we do know (since, as I noted in a previous post, the
research to determine what we want to know -- controlled variables --
hasn't been done yet).

I might also include some "classical" perceptual demos, but only to help
students understand what perception IS; and why we need models of the
perceptual process that converts what physics tells us is on our sensory
surfaces into what we experience as the outside (and inside) world.

I left academia because I knew that, even if the College would allow me to
teach an Intro to Psychology course like this (they would have), it would
not have been fair to the students. So I've been teaching this course on the
net for the last seven years instead;-)

Best

Rick

[Dan Miller (970310)]

Rick Marken (970310):

This sounds like a great course, Rick. If I had taken it I might
have become a psychology major, instead of sociology. This might
have been true of others, as well.

I left academia because I knew that, even if the College would allow me to
teach an Intro to Psychology course like this (they would have), it would
not have been fair to the students. So I've been teaching this course on the
net for the last seven years instead;-)

This decision was made a number of years ago. Would you make the
same decision today? I ask this, because I took an introduction to
sociology course from a professor who had discarded conventional
sociology (if he ever embraced it) for a brand amicable with
perceptual control theory. There remains a strong elective affinity.
Without young professors teaching undergraduate and graduate classes
it is difficult for a scientific movement to be successful.

I'm making no accusations, but just wondering about your present
thinking on this important topic.

Dan Miller

Dan Miller
miller@riker.stjoe.udayton.edu

from Tracy Harms (970310.1800 PST)

[From Rick Marken (970310.1750)]
[...] You sure can't
pass the GRE if you've only taken PCT psychology.
[...]
So, yes. Given what I perceived (and still perceive) as the academic
resonsibiliites of a professor of psychology, I would make the same
decision now as I made then; I'd stop being a professor of psychology.

This reminds me of Alan B. Scrivener's humorous (or bitter?) comment that
''today we call schools where more practical skills are taught "Junior
Colleges," while all the really useless ideas are confined to
"Universities."''

(The section where that quote occurs is reproduced below, taken from
A Curriculum for Cybernetics and Systems Theory,
URL A Curriculum for Cybernetics and Systems Theory)

I don't personally agree with the entirety of these assertions by
Scrivener, but I found them well worth reading.

Tracy

ยทยทยท

____________________________________________________________________________
The blind spot in our civilization when it comes to wholeness and loops
goes back a long way. Plato said, in Georgics, that "Helmsmanship is
unassuming and modest and does not boast and does not behave as if it had
done something wonderful." This contempt for piloting fits in with the
general ancient Greek aversion to any linking truth-seeking with useful
activity. This meant the literate philosophers and the skilled workers
were kept apart. The Romans were by contrast ruthlessly practical, had
literate engineers (who could write down their skills and plans). They
conquered the known world with their "practicality," which had toxic
social, religious and ecological side effects. Christinanity swept the
Roman empire partially as an antidote to the Roman toxic pragmatism. The
dark ages were, in Europe, a period of avoidance of practical applications
of technology in favor of the search for religious grace. Our civilization
didn't deviate from this pattern until the 13th Century, when literate
monks were forced to garden, and dragged out a bunch of Roman engineering
books, rediscovering hydraulics and inventing the windmill. Similar
integrations of learning and labor helped bring about the Renaissance and
the Industrial Revolution. And yet today we call schools where more
practical skills are taught "Junior Colleges," while all the really useless
ideas are confined to "Universities." Greek logic was made symbolic by
nineteenth century academic George Boole, who -- like the Greeks --
prohibited loops in logic chains. But in this century electronic engineers
found that you can build logic gates physically, and you can hook them up
in loops, resulting in both digital memories and digital oscillators.
These humble engineers were "off the map" of Western civilization's
mathematics and philosophy. One of the simplest of these circuits was a
one-bit memory, which was named a "flip flop" -- indicating the lack of
academic tradition for the whole idea.

Our traditions cling to the idea that explanations can be built of short
causal chains: event A causes event B, which causes event C. Loops are
prohibited because they are hard to analyze, introducing non-linear terms
into the equations. Therefore our current scientific method has become a
form of pretend madness in which we deny that anything is connected to
anything else unless we can prove that it is. We use this method because
we can so much more easily start with assumed isolation and then prove the
connectedness of the components of a system than do the opposite. But our
simplifying assumption has become an article of faith, and this false faith
is aggravating the "externalities" of our society: pollution, crime,
alienation, illiteracy, the decay of our infrastructure, the decline of
our industry. All of these problems have been amplified by the success of
our technology in achieving narrowly defined goals. Cybernetics and
systems theory are part of the antidote to the toxic byproducts of our
short-sighted reductionism.

                                                Alan B. Scrivener

[From Bruce Gregory (970310.2115)]

Rick Marken (970310.1750)]

And I certainly couldn't have gone on dishing out the conventional
wisdom in the courses I had to teach. When I did the research methods
course, for example, I felt like an astrophysicist teaching a course in
Sun signs;-)

You mean I _shouldn't_ be teaching my course in Sun signs?

Bruce Gregory

[Martin Taylor 970311 12:20]

Tracy Harms (970310.1350 PST)

... I would from the beginning nurture an awareness of ecological
interaction. By the time one gets around to the "social psychology"
section the principle of feedback among control systems should be familiar.
Not so familiar as the feedback within control systems, I suppose, but
enough so that the distinctly *social* aspects of this area are seen as a
special case of the more general pattern of ecological resolution.

This makes sense to me, and it could be used both as a lead-in to the idea
of "reorganization" and, having got that idea straight, it could be used
to show how individual reorganization can lead to social conventions,
language, economic activity, and so forth. Could be done in early high-
school, I would think.

The ecological feedback web is a good observable example of the effect that
the activity of one control system can have on the abilities of another to
control. The same happens inside a body, but can rarely be seen (it can,
perhaps, in pathologies). By pointing it out in the external ecological
context, the interactions might lead to experiments involving The Test
on individuals in social contexts, and thence to the (sorry, Rick) dryer
and more boring, if more scientific, applications of The Test in conditions
where measurement is more precise (tracking studies).

If you really understand PCT, the whole is a seamless web, both within and
among control systems. In teaching, I would hazard a guess that the easiest
place to start picking apart the web is a place that students can observe.
One such place is inside themselves, but since each student is different,
some may not "get it" as easily as others. In social-ecological action,
everybody can look, and the teacher may perhaps be more able to determine
why a student doesn't "get it" than when the problematic perception is
of the relationships among the student's own control activities.

There's a nice side-effect as well, of having students who are sensitive
to possible problems with apparently obvious environmental interventions.

Maybe nonsense, but it seems reasonable to me.

At this moment.

Martin

[From Rick Marken (970311.0930)]

For Bruce Abbott (970307.1940 EST), when he returns from vacation:

Me:

Bad choice. Gary NEVER annoys me:-)

Bruce:

Well, that's about to change...In fact, Gary has already gone to
work. (See Gary Cziko 970308.2011 GMT.)

As you shall see, Gary has, indeed, successfully annoyed me. Its' nice
to know, however, that he was put up to it. I'd hate to think that he
actually believed that stuff about conventional methods giving us useful
information about control;-)

I'm actually glad that you put Gary up to this because it gave me a nice
idea for an improved version of the "Behavioral Illusion" java demo
which I hope to have up on the net soon. Thanks.

Me:

Norman Anderson was able to obtain measures of the open loop
perceptual function using his method (anaysis of variance)?
You crack me up, Bruce!

Bruce:

Naw, it wasn't his use of ANOVA that impressed me. His techique
works perfectly well without it -- it's essentially graphical.

Anderson's technique is still done with people who are in a closed loop
relationship with respect to the evoked meanings of their judgements, is
it not? In Anderson's case, the subjects are in a closed loop with
respect to their judgements of a match between perceptions of
combinations of factorially varied "stimuli" and the real or imagined
perceptions evoked by their overt (or imagined) judgement of these
combinations. Therefore, a graph of the average overt judgement made to
each stimulus does not represent the
open-loop perceptual function relating stimulus to judgement --
any more than it does in the case of magnitude estimates.

Hans Blom (970311e) --

My main gripe with "the simplest form of PCT" is that it is based on
models as well...

So PCT is just a version of MCT? But now I am even more puzzled. As far
as I can tell, MCT is the brand of control theory used by most
psychologists today (if they use any brand of control theory at all).
If MCT = PCT then why haven't the experts in MCT (like yourself)
explained to these psychologists that control systems control
perceptions, not outputs; that the only way to determine what
perceptions are under control is by applying disturbances to
hypothetical controlled variables and looking for lack of effect;
that controlled perceptions can be as simple as the sensed tension
in a muscle or as complex as a theory of economic behavior, etc etc?
Why, in other words, haven't the experts in MCT worked to change the way
scientific psychology is done?

Best

Rick