Visualizing behavior through HPCT

[From Shannon Williams (960226.15:30)]

Hans Blom, 960222--

... HPCT ... has absolutely no predictive power that I can see.
Since it cannot be used to predict, I do not think it is a valid

Maybe not. Most of HPCT is concerned with model building, in this
case the determination of which function with which parameters
provides a best fit with the actual data. Prediction, however, is
something else. You will have to freeze the model that you have
obtained in the model building stage and compare what it does with
the results of an _independent_ group of test subjects and show that
the fit in your "test population" is (almost) as good as the fit that
you obtained in your "learning population". If that is true, you have
a predictive model and you can predict.

In the current PCT model, if I think that John Doe has some perception
that he wishes to maintain, then I can predict his behavior when I
disturb or threaten to disturb his perception (assuming that I am
familiar with his behavior repertoire).

In the current PCT or HPCT model, I cannot predict what perceptions John
Doe's brain will "decide" that John needs to maintain.

This is the prediction that I am interested in.



Rick Marken (960222.1200)--

could you present the kind of learning data that you feel HPCT should
be able to predict?

If a person is presented with a problem (he is having difficulty making
some perception match its reference), and I present some partial solution
to the problem, I want to be able to predict if my partial solution gives
him enough information to solve his problem.

For example, say you are trying to extract a ring from a puzzle. I can
question you to determine your current strategies. When I give you a hint,
I want to be able to point to different constructs in my HPCT model and
predict whether you can now solve the puzzle.

(BTW- if you do not believe that such predictions are possible, then
of course, you will not see any weaknesses in the HPCT model.)