[From Bryan Thalhammer (2006.02.25.0930 CST)]
Editorial:
First of all, when you teach a course in Research Methods in Psychology, I presume that the institution has some standards of what constitutes academic work for course credit. A course designer would work from the basic course description and the articulation needs, namely, its prerequisites, and followup courses that build on the content presented in the prerequisites and present course. That determines the basic structure of the course. In addition to that basic structure an individual professor can introduce new interpretations of existing methods and new research methods entirely.
Secondly, and I think most important, is to present PCT Research methods in the context of existing methodology, with the intention of showing how it may produce better and more usable results than traditional methods. Without a hook into traditional methods, the PCT presentation may not make sense. Since most psychological concepts are still very much grounded in linear causal chains, students may not be able to perceive much less control for the concept of cycles of causation that is incipient in a PCT explanation.
For that reason, Rick's plan of 3 weeks traditional / 1 week PCT has a much better chance of showing the contrasts, the advantages and the possibility of fruitful outcomes of the Test for the Controlled Variable, MOL, etc. It is a major paradigm shift, looking within the actor rather than from the point of view of the observer.
In that context, I see Bjorn's contributions as helpful.
Also in that context, I offer this blend:
* Present the basic stuff, no different than Henry Perkinson does in Learning through Mistakes and what Phil Runkel does in Casting Nets. However, and this is a big however, at the end of each chunk, present questions or challenges of how the traditional method may not really answer the question of a person (not a group) behaves and on the basis of what factors (reference settings?). Set up the last week with a presentation of the holes or failings of traditional methods. And it goes without saying that this is not a bait-and-switch, but a true intellectual and scientific challenge to the students.
* There is a sales aspect to this, of course, where the challenges or questions open the students to the possibility that the pyschoanalytic, behavioral, cognitive science, neuroscience, constructivist, and other cause/effect models may not adequately explain the behavior of living organisms. At the acceptance that there are some failures in these previous explanations, a proposal of the PCT explanation can be introduced.
* Then, of course, we know the rest: Excerpts from Casting Nets, MSOB, from Things We Do by Cziko, and Robertson, Goldstein, Mermel & Musgrave (1987/1999), "Testing the self as a control system: Theoretical and methodological issues, in Int. Journal Human-Computer Studies (1999) 50, 571-580. This last could be an interesting social psychology demonstration that could be shown in class?
It is, I guess, possible to begin the course with a direct proposal of the PCT explanation, but I believe this is risky, because it puts the presentation of a new paradigm during the initial honeymoon period of students with prof, and may start off the relationship on a shaky basis. So I think that the foregoing could be better.
Really, I think that Phil has the instructional model and his firm basis in stats shows the way. Bill's standard Rubber band demo, Dick and David's experiment and David and Tim Carey's MOL work might be the vehicles that help students to the basic PCT understanding?
--Bry
See:
Wiki Psychology:
<http://en.wikipedia.org/wiki/Psychology>
Wiki Cognitive Science:
<http://en.wikipedia.org/wiki/Cognitive_science>
Wiki Perceptual Control Theory:
<http://en.wikipedia.org/wiki/Perceptual_Control_Theory>
···
[Bjorn Simonsen (2006.02.25,14:50 EST)]
[Rick Marken (2006.02.24.1050)]
I'd particularly like hearing from those of you who have
had experience teaching PCT methodology.
Now you are hearing from one who doesn�t have experience teaching PCT methodology. But, nevertheless. Read it or not, it depends on your references.
I guess someone tell you to spend three weekends going over "conventional" research methods and one weekend on studying individuals. The world asks to be deluded.
May I introduce this mail proposing you to start your first lesson saying: �I will now spend three weekends telling you the way to obtain statistical knowledge about the behavior exercised by a fictitious concept called a group of human beings. This knowledge will be appreciated among other scientists you meet. And I will weight something I call a misused concept, �significance�. But I am sure you will get the last word using this concept. Open your ears and listen when I advice you to not explain individual perceptions of behavior by referring to such tests. If you are going to explain individual behavior, I advice you to listen to my methods lectured the last weekend�.
But I want to keep it simple, clear and relevant.
I'm writing to ask for suggestions regarding how to
teach PCT research methods. What are good
(ie. interesting) examples of PCT-based research
that have been done or that COULD BE done?
Are there studies in the "conventional" literature
that could be used to illustrate PCT research in
an interesting way? I though I might discuss the
MOL as an example of individual research based
on PCT. What do you think?
I think you know better what I am going to write. I propose you organize your lectures around the Rubber Band Experiment. Hire or buy a web camera and a program where you can see stills on your computer together with an indication of time in 100 parts of a second. Organize the Rubber Band Game over a millimetre scaled paper and the web camera above.
I don�t feel very well continuing, because I know you know better, Rick. But I do continue.
1. Ask the student C to look at the student E�s hand and counteract what he is doing. Make a spreadsheet where you register the coordinates for E�s hand and C�s hand and make a table below where you have functions describing the motions and the motion direction of E�s and C�s hand (a2 + b2 = c2 and tangent angle b = b/a) and correlate the actions of E and the actions of C. Do the experiment when E uses 5 different Rubber Band Strengths and C uses the same rubber band.
There are many statistical methods you can use.
Make the experiment as a nil hypothesis experiment and ask for the conclusion.
2. Do the same experiment and ask C to keep the knot over the dot.
3. Do the experiment as 2) but transform the dot to a circle with a diameter of about 5 cm. Tell them about tolerance. You and Martin taught me about tolerance before Christmas.
4. Use a Pen and put a dot representing a knot on E�s Rubber Band and on C�s Rubber Band 2 an d 3 cm from the original knot. Now there are three �knots� on the Rubber Band. Do the same experiment as 1 (C counteracts E�s movements). Use the web camera and after about 15 seconds they read the coordinates from the three �knots� and from E�s hand. Use the spreadsheet and get the observations of the movement to the three �knots� and correlate each of them with the movement of E�s hand.
5. Do the same experiment as 4, but ask C to choose one of the �knots� over the dot as in 2). Study the statistics and ask why there is less correlation between the movement of E�s hand and one of the movements to the �knots�. Introduce the test.
6. Do the same experiment as 5, but ask E to do the test relative to the three �knots�. Tell the students about a better way to find out which perceptions people control. A better way than statistics.
7. I will not tell you how it is possible for C to control his knot relative to a configuration or a relationship or a program. I will neither not mention about the Rubber Band Game and an Extern Conflict, because you know it better. But I will tell you an idea about Extern Conflict and MOL. Let us say C is trying to put the knot over dot M and E is trying to put the knot over dot N. After a short conflict, you may ask how they agree about something that includes the dots M and N. And one of the students will answer: �Maybe we can agree about moving the dot along a line between M and N both together�. Now you have nice opportunity to tell them about External conflicts and MOL.
Once again, I shouldn�t send this mail because Bill, Tom and you (and other) know this much better than me.
Have a nice day.
bjorn