Retro/Sponte/Perfaction

[From Bruce Abbott (960209.1630 EST)]

Rick Marken (960208.1000) --

Bruce Abbott (960208.0935 EST)

Some studies provide strong support for the spontefaction of certain
variables; consider escape and avoidance experiments, for example.

See my post [Rick Marken (960205.1000)] and, in particular, my description
of the problem of determining, from observation of the IV-DV relationship,
what is controlled in the knee jerk reflex. I agree that conventional studies
strongly suggest that organisms control. But moving from that qualitative
observation to the quantitative meat of the matter -- determining _what_,
exactly, is being controlled -- requires careful testing for controlled
variables.

I agree; I'm not suggesting that these studies can _replace_ careful testing
for controlled variables.

I continue to believe (no doubt contrary to your view) that available
[EAB] data will provide considerable guidance in formulating spontefaction
models of the behaving organism

You bet this is contrary to my view. Available data does provide guidance
for further _research_; but it provides little if any guidance for
formulating control models because controlled variables have only been
_suggested_ -- not _identified_. Moreover, there is no data on control
against which any model based on such data can be tested.

When you begin testing for controlled variables, it helps to have some
particular set of observations in mind that one is attempting to explain by
means of a model. If those observations also provide strong hints about
what variables are being controlled and how, so much the better. Just to
suggest that a particular variable is being controlled it tantamount to
proposing a model. What I'm saying is that there is no need to begin from a
position of complete ignorance when there is a wealth of good observations
to provide guidance.

Your attempts to build control models of existing data are not a complete
waste of time -- but very nearly so. I suppose I get upset about this
because I see you wasting your time on this obsession with useless data when
when you could be making a significant copntribution to the development of
PCT science by doing the research needed develop and test models of the
spontefacting that is so obviously occurring in the operant situation. I
suppose you'll have to find out on your own that the attempt to "account
for" conventional data with control models is a waste of time. I just wish
you were spending this time in the lab, where I think you belong.

I certainly want to move to the lab work as quickly as possible, but have
been facing certain constraints that have made this difficult to carry out.
Meanwhile, the model-building exercises have been a way for me to learn how
to apply control models to the sort of data I am interested in and also
develop some working hypotheses to pursue in the lab work. I don't see this
effort as a substitute for the lab work, but as preparatory to it and as
something that can be done until the lab gets back on line. It sure beats
just sitting on my hands.

Regards,

Bruce