[FROM CHUCK TUCKER 941129]
I have enjoyed reading about the modelling that has been
done on the net during the last month not because I can
do it (I can't) and not because I understand it (I don't
completely) but that it is the first clear demonstration
of what it means to "model" and it has shown how at least
four people DO modelling. I would hope that before y'all
get on to other modelling projects (such as what do human
beings do in the various models) that y'all would summarize
what you have done and provide us with the "final" models
and give us some instructions on how we might be able to
use these modelling programs to demonstrate the differences
between the two models. PLEASE
While reading the posts on modelling on the net (as I am
typing this post my screen shows Bruce's post "The Party's
Over" - I wonder what that says!) I have thought that all of
this would be quite useful to test the many "theories" that
we have in sociological social psychology which use the
ideas of S-R (e.g., Expectation States; Exchange Theory).
Each of those "theories" imply a model of the human being
which can be programed and tested to see if the results of
their studies can be replicated. it seems to me that it
can be written so that two similarily programmed human
beings can interact with each other and see what the results
might be - do they match up with the studies? (even the
silly aggregate results that are published!). If modelling
is going to be useful and if PCT wants to show its worth
and utility I can't see a better way than showing how the
model can produce the same results as the studies. Then
PCT research would be interesting and useful and I would
be surprised if there would not be just a few people who
would take notice of it.
Regards, Chuck
I was interested in Chuck Tucker's plea for PCT models that replicate
experimental results done in the context of other theories of behavior
than PCT. In fact, I am working on such a model that I would be willing
to share in the near future (as some additional touches need to be made).
This model attempts to replicate the network exchange experiments, but
from a PCT point of view. So far, the initial results are very promising.
The model is a three level hierarchical model that attempts to control
the perceived level of participation in the exchanges (the paradigm is
one of excusion. If some subset of persons are involved in the exchanges,
some other subset is excluded, and must make some increase in their
offerings to get to be included in future rounds of exchanges). The
program involves setting up 3 to 7 such pct systems where the outputs of
each are disturbances to the inputs of others.
I will need to put more documentation into the program, and add some
addition documentation about the program and what it is trying to do
before it can be posted. My usual programming is in C++, but for you guys
who use Pascal, I have written the program in that language.
For those who have done a lot of programming of PCT models, I have a
general question: how can you model parallel processes (the interaction
of several PC systems) using sequential programming methods. I have tried
to randomize the order of processing, and take small steps to minimize
the effects of order, but those effects are still present. Any Ideas?
Peter J. Burke
<[Bill Leach 941201.18:55 EST(EDT)]
[CHUCK TUCKER 941129]
Chuck, I am also glad to "hear" you comment. I believe that most of the
participants in this modeling activity do not realize just how much many of
us "by-standers" are gaining from the effort.
All the joviality aside, it seems to me that a very honest effort is being
made "on both sides" to be objective and thorough. No one appears to be
willing to let assumptions "slide by". This is probably an excellent
example of the effort required to come to an understanding in a modeling
effort.
While Bruce may well be in the position of a "defender" here, I am very
impressed by his considerable efforts in making sure that he understands
others and that they understand him.
I believe that regardless of whatever "ultimate conclusions" may be draw by
what parties, all will have to conclude that significant progress in
understanding the process and especially subtle complexities of modeling
has been gained by all.
-bill