Request for Tracking Demo

[Erling Jorgensen (950930.2150CDT)]

To Bill Powers or Tom Bourbon

I have been raising PCT (sort of like a mosquito) with some of my
profs and colleagues here at UND. I've been able to explain the
logic of things like the tracking tasks fairly credibly, but I
don't have any demos of them.

Would it be possible for one of you to post the code for a full
tracking demo (e.g. following cursor position), complete with
disturbance table(s), means of plotting, etc.? What kind of
compiler would be needed? Does it work with a mouse (or only a

My advisor, whose specialty is research methods and testing, is
already suspicious of the low correlations that typically show up
in psychological studies, but he's not sure what to make of the
.99 correlation claims of some of the PCT studies. I think he
buys into the logic of a generative model. We've discussed
the kinds of perceptions suggested by the Hierarchy, and brain-
stormed a little about how to spot some of these things in the
ongoing flux of counseling. There's precedent for this way of
thinking to the extent that counseling recognizes multiple layers
of context, and steps back into metacommunication all the time.

From a practical standpoint, introducing disturbances (the Test)

is not seen as the best way of building a "therapeutic working
alliance." The hard part for my advisor, I think, is in seeing
how goal-directed _everything_ is, that they are perceptual
goals, and that specific behaviors do not matter (in terms of
the functioning of the system.) The other hard part is in
conveying that "interventions," or any other type of
jostling/disturbances, lead to corrective actions of a
_system_, and not just lineal changes in a variable.

I think the demo would be helpful in showing that the "behaviors"
(of the model and the person) are _different_, but the results are
the same. The model changes pixel numbers (or some such thing, I
presume) to vary the cursor, while the person changes muscle
lengths to move a mouse which corresponds with different cursor
positions. But both of them are able to stabilize the _same_
relationship perceptions, to a very high degree of correlation.

Any help you could give me would be appreciated.

(Windmill-tilting in Academia...)