[From Bill Powers (960412.0600 MDT)]
Stefan Balke (960412.1115 CET)) --
To control the state of a perceptual variable I permanently compare
the state of that variable with my reference state for that
variable. If this comparison leads to the result that there is a
deviation from the reference level than an error signal will be the
result. This error signal changes the reference level for one or
more lower order variables.=20
...
I think in order to reduce the state of the error variable the
error signal doesn=B4t change randomly any reference level for any
lower order variable.= It changes the reference level of
that/those lower order variable/s that will bring the controlled
perceptual variable into its reference state....
Wrong beliefs will lead to poor control.=
=20
Your understanding of the control process is fine, for the most part.
The only problem with your cognitive symbolic representation of your own
control processes is that they ARE symbolic representations, and the
processes you are talking about are not caused to work by symbolic
descriptions of them -- not unless the process of which you are speaking
is specifically the process of producing symbolic representations.
Wrong beliefs will lead to wrong control at the level of perception
where the beliefs are relevant. But belief exists only at the highest
levels of experience; you can control the positions of your limbs even
if the way you believe those control systems to work is very different
from the way they actually work. Of course beliefs do enter into all
behaviors -- this is a hierarchical system. But the way they enter, for
the lower processes, is in deciding what to control, not in making the
lower control systems operate properly. If you believe that a door opens
by pushing on it, you will set reference signals telling the lower
system to achieve the perception of a door swinging open away from you.
The lower system will attempt to produce this perception, but of course
it will fail if the door actually opens by pulling on it. This is not a
failure of the lower systems, however; it is a failure of the higher
belief system, which has to be changed so it believes that that
particular door opens by pushing. The lower system did not fail because
it was a wrongly-designed control system; it failed because it was given
a reference signal that couldn't be satisfied in the current
environment.
The way we talk and think about our own control systems is only a
behavior called talking or thinking. You can probably improve the way
you talk and think through making cognitive-verbal models of these
processes. But this will not improve your handwriting, or your skill at
riding a bicycle, or your ability to shoot free-throws in basketball, or
other nonverbal skills. The only way to improve these control systems is
through practice, failure, and continuing practice -- all in inner
silence. Cognitive-verbal skills are of no help with this.
When a golf pro tells you to "keep your left arm straight" it will do
you little good if you just repeat those words to yourself. What you
have to do is translate the words into a particular way the left arm
_feels_, and set up that _feeling_ as a reference signal. Neither your
golf pro nor you can tell the lower systems HOW to keep your left arm
straight. That is accomplished by the lower control systems, in ways of
which you have almost no direct experience. The best you can do is to
keep imagining a straight left arm, and keep swinging while the lower
systems teach themselves how to vary the muscle tensions to keep that
perception matching the reference condition all during the critical part
of the swing.
Of course as you do this, the lower systems will also teach the higher
systems what the words mean, experientially. You will discover that the
golf pro doesn't really mean that you should keep your left arm
straight. As you try to do this, you will find that it's impossible to
keep your left arm straight and complete a normal-looking and -feeling
golf swing. Watch the professional golfers on television. They don't
keep their left arms straight -- except in parts of the golf swing where
the arm should remain fully extended. You have to convert the coach's
words into achievable nonverbal reference conditions.
Just remember, you don't control the names of perceptions or verbal
descriptions of perceptions: you control the perceptions themselves, and
the perceptions are nonverbal.
···
-----------------------------------------------------------------------
P.S. Note strange characters in your posts, like
the error signal doesn=B4t change
Is this a function of how your word processor works?
----------------------------------------------------------------------
Best,
Bill P.