Martin's Criteria?

CSGnet Archive CSG1994
1

Tom Bourbon [940809.0850]

[From Bill Powers (940808.1845 MDT)]

Martin Taylor (940808.1200 MDT) --
RE: one-way control

Martin:

If the reference signal is excitatory and the perceptual signal
inhibitory, you can't get that function. In that arrangement, the
error signal is limited to a maximum value of R.

Bill:

Right. Glad I wasn't totally confused. Better you than me.

Martin:

Then of course if R=0, there will never be any error. A most
uninteresting and unuseful control system. ... There might be reasons
for such systems in nature, but I can't think of one.

Bill:

Well, I think that the spinal reflexes (which are organized that way,
although not entirely obviously) are both interesting and useful.

. . .

This is the actual
arrangement in the spinal cord. The same arrangement will work at any
level.

Martin, I have a brief question, inspired by your occasional assertion that
a particular example of control, or of control systems, is "uninteresting."
Most often that assertion indicates the writer has unstated criteria for
determining that which is interesting. In each case where you have made the
assertion, I have been left wondering why the example did not meet *your*
unidentified criteria. The present instance of all the spinal reflexes is a
good example. Can you tell us something about the criterion, or criteria,
you use when you decide an empirical example is "uninteresting?" I really
haven't the foggiest idea what it, or they, might be.

Later,

Tom

2

[Martin Taylor 940809 11:50]

Tom Bourbon [940809.0850]

A really thought-provoking question, Tom. And insightful (for me at least).

Martin, I have a brief question, inspired by your occasional assertion that
a particular example of control, or of control systems, is "uninteresting."
Most often that assertion indicates the writer has unstated criteria for
determining that which is interesting. In each case where you have made the
assertion, I have been left wondering why the example did not meet *your*
unidentified criteria.

Well, you set me to wondering, as well. I might ask you, in turn, how
you manage to come up from time to time with these wonderful, creative,
questions?

I think I use "interesting" rather as in the Chinese curse: "May you live in
interesting times." (We do). An "uninteresting" system to me is, in some
cases, one that presents no problem of interpretation and is therefore
uninteresting intellectually. In other cases, a system may be uninteresting
because it seems to be less useful as a part of the control hierarchy
than a related system would be, as in the case of inhibitory reference
signals in one-way systems. In this specific case, if Bill is right that
the spinal cord does involve inhibitory reference signals, that kind of
system may turn out to be more interesting than I had thought.

In general, what I think of as "interesting" seems to have two qualities:
it seems to be an effective and important element of a hierarchic control
system, and it has some apparent intellectual challenge. I may change
this description as I think more about your question.

Come to think of it, there's a third aspect: the "interesting" thing may
account in a natural way for some observation that otherwise stood on
its own, explained only in its own little context. That's why the PCT
degrees-of-freedom explanation of alerting seems interesting to me.

Which reminds me: I constructed, but have yet to post, a further discussion
on the meaning and counting of "degrees of freedom." I must polish and send
that before I go on Thursday. (Among many other things to be done before
then :slight_smile:

···

======================

Tom and Bill--Your data from the sleep-deprivation study are (almost) all
on floppies waiting to be sent to you. I hope to do that, too, before
leaving on Thursday.

Martin