[From Rick Marken (920813)]
Bill Powers (920813.1400) says;
Just saying "CONTROL WON'T WORK (in the long run) precisely because people
ARE controllers" is enough to convince anyone who already understands the
principles of control. To others it just sounds like a slogan.
That's why I keep harping on
disturbances -- they're the main reason that control is necessary, and the
main reason why other non-feedback models can't work in the real world.
OK. I was waxing slogany.
The point of my post, however, was to emphasize the distinction between models
of appearances (which I called imitations) and models of processes (which
I called models). The fact of the matter is that the main reason for this
distinction is the existence of disturbances -- in particular, disturbances
that are not detectable by the behaving system OR by the modeller of the
behaving system. If one does not know that these disturbances are (or
could be) affecting the observed results, then s/he will tend to build models
that simply produce the results -- ie. that imitate what the modeller sees.
But we know (thanks to you) that disturbances (in many different forms) are
always having effects on behavioral results. Thus we know that consistently
produced results must be controlled -- and what is controlled is a perceptual
representation of those results. So we need a control model to produce the
results we see. An imitation model won't work because any change in the
disturbing conditions would lead to inconsistent results. So imitation
models only work in environments where there are no disturbances. Only a
control model will produce consistent results when there are disturbances.
So, indeed, understanding that disturbances can influence every kind of
result that organisms are known to produce consistantly is crucial to
acceptance of the control system as the basic organizing principle of the
nervous system. And it is crucial to seeing what imitation models are and
why they don't really work.
Regards
Rick
···
**************************************************************
Richard S. Marken USMail: 10459 Holman Ave
The Aerospace Corporation Los Angeles, CA 90024
E-mail: marken@aero.org
(310) 336-6214 (day)
(310) 474-0313 (evening)