[From Rick Marken (960411.0820)]
Stefan Balke (960410.1630 CET) --
You're right, of course it's necessary to try to notice the reason for the
error, otherwise it's impossible to counteract effectively.
I think you missed my point. Noticing the reason for the error (which
requires noticing the variable controlled and the reference state for that
variable) allows you to notice _that_ you are controlling and _what_ you are
controlling (or trying to control). But it won't necessarily help you control
(counteract the effects of disturbance) more effectively.
I thought your "error scale" was a good way for people to become aware of
their own controlling but I don't think it's a particularly good way to
improve control. If you are not controlling as well as you think you should
be (if error is chronic) then you are probably in conflict and the onlu the
way to improve control is to solve the conflict; this is done by becomming
aware of you own controlling (which is what your "error scale" helps with)
from the perspective of the systems that are setting the references
(goals) that produce the conflict.
The advantage of the self monitoring is that I become aware and much more
sensible to the many control operations that I (and other people) carry out
continuously.
Yes. And this can be very valuable. But it can also be disruptive; while you
are aware of your own controlling the quality of control tends to break down.
This is what Zen is about, I think: learning to keep consciousness _out_ of
the systems that are working well.
If I do the self monitoring I'm usually in a situation doing something...At
this moment I look at the error monitor (this means I generate a figure
that represents the actual stress level) and a figure comes up to my mind. I
do nothing, but something like saying to myself: figure? and the figure
representing the error level appears.
My question is, should this figure be called an intrinsic signal?
No. I think it's an imagined (self-produced) perception; like thinking of an
apple when there are no apples around.
Is it an intrinsic signal with a reference level
I would say it's an imagined perceptual signal (based, presumably, on error
signals resulting from the comparison of real perceptions with real reference
signals); and if you want that figure to be something in particular (like
zero) then you have a reference for that (imagined) perception.
I simply can't understand that it is obviously possible to produce a
figure that represents actual error level.
Why not? I can easily imagine numbers (figures) that represent the level of
many perceptual variables -- the beauty of different pieces of music, the
quality of prose selections. What I am probably doing is what you are doing;
assigning an imagined number to the degree to which the perception matches my
reference for that perception; that is, I am assigning a number to error.
What mechanism could judge which figure would be the best?
The figure is just an (imagined) perceptual variable (call it p). The "best"
perception is the one that matches your reference specification for that
perception (r). So the mechanism that judges which figure is the best is the
comparator mechanism characteristic of a control organization: e = r-p. When
e is 0, the figure (p) is best.
David Wolsk (960410) --
I suspect that the tickle response bears some resemblance to these
other response systems.
Try to think of them as control systems. You seem to be thinking of tickling
in S-R terms:
tickle-->laughter
The tickle is a stimulus that causes a laughter response.
PCT suggests that this must be an illusion; what we see as stimuli are
disturbances to controlled variables; what we see as responses are related to
the actions that compensate for what would be the effects of disturbances
to the controlled variable. What must really be going on in tickling is:
tickle -->CV--|person|-->laughter
^ |
>----------------|
What perception (CV) might be affected by both tickle and laughter? Actually,
there are probably several levels of perception controlled in this situation,
but I suspect that one controlled perception might be something like
"approval"; when the person doing the tickling is someone you like, you laugh
in order to perceive approval of the result of the tickle. The fact that the
laughter is not a "response" to the tickle is shown be the fact that the
response when the person doing the tickling is someone you hate or fear is no
longer laughter. At least, I don't think it is. Somhow, I can't imagine
laughing too hard if I were being tickled by Rush Limbaugh.
Best
Rick