gap between error and action

CSGnet Archive CSG1996
1

From Stefan Balke (960219.1630 MEZ)

Rick Marken encouraged me in a privat post to ask my questions on the net.
Here is one of them.

I posted to Rick:

I know from myself, that I want to do the "right decisions", I control for
"right decisions". In order to do so, I guess based on my experiences and on
the available situational informations, which of the possible alternatives
would be the best for me to reach my goal. I do not calculate anything, but
I have to _decide_, which (counter-)action I will carry out. Steering the
wheel is a very direct behavior, but deciding which route will be the
fastest at a special daytime is a more complicated process.
My question is, how the thought to be best of the possible counteractions is
detected.

Rick answered to me:

I know from myself, that I want to do the "right decisions", I control for
"right decisions".

This is a very interesting situation. I do it too, of course. But I don't
think I am controlling when I do it; when we consciously have to decide
something, we are _hoping_ to get the desired results, perhaps even
_expecting_ to get the desired results but, unless we are in control, we
are not necessarily _going_ to get the desired results. I'm sure you will
find this difficult to accept; but I think if you look carefully at what
you accomplish with your conscious decisions you will see that conscious
decision making occurs only when ypu are not firmly in control.

This has been discussed before on the net; I think it's worth it to bring it
up again.

I´ll add the following question to make my request a little clearer.
Since everything we do (and think) is control (as Rick told to me in an
earlier direct post - and I agree with this view), it should be possible to
describe the process of conscious decision making as a (H)PCT negative
feedback model. I´m interested to know, how a PCT-model for making concious
decisions would look like.

Kent McClelland [960215.1445 CST] has pointed out

We control our perceptions, not our actions.
In order to control a perception . . .
      We must have a memory or mental model of what we want (OUR REFERENCE

STANDARDS).

      Our PHYSICAL ACTIONS must affect the thing to be controlled.
      We need PERCEPTUAL FEEDBACK about to the thing to be controlled.

We act to correct the errors when our current perceptions don't match our
reference standards.

I agree with this statements, but I want to know, how we close the gap
between the mismatch and the selected action.

Best

Stefan

2

[From Kent McClelland 960220.1045 CST]

Stefan Balke (960219.1630 MEZ)

Kent McClelland [960215.1445 CST] has pointed out

We control our perceptions, not our actions.
In order to control a perception . . .
      We must have a memory or mental model of what we want (OUR REFERENCE

STANDARDS).

      Our PHYSICAL ACTIONS must affect the thing to be controlled.
      We need PERCEPTUAL FEEDBACK about to the thing to be controlled.

We act to correct the errors when our current perceptions don't match our
reference standards.

I agree with this statements, but I want to know, how we close the gap
between the mismatch and the selected action.

Stephan, as I see it, the gap isn't between the mismatch and the selected
action, but between our current perception and the reference standard we're
using. Our control loop operates to select an action that will close the
perceptual gap. If we're successfully controlling our perception, we're
continually closing that gap (most of the way, at least). The course of
action we select opposes the moment-to-moment disturbances in the
environment and keeps the controlled environmental variable (CEV) stably in
line with our reference standard. Our action isn't the thing being
controlled (made stable), because it has to be constantly changing to cope
with the constant changes in the environmental disturbances we encounter.
If we always did exactly the same thing in response to every disturbance,
our perceptions would go out of control and end up varying wildly because
of the changing disturbances (unless we happen to be in a highly
"controlled" environment, like those the engineers try to provide for
industrial robots).

This is a pretty basic PCT idea. Does it make sense to you?

Kent

3

From Stefan Balke (960222.1230 MEZ)

Kent McClelland (960220.1045 CST)

Stephan, as I see it, the gap isn't between the mismatch and the selected
action, but between our current perception and the reference standard we're
using.

I agree with the idea that (mostly) there is a gap between the current
perception and the reference standard and that this gap is called the error
or perceptual gap. I meant another gap, the gap between an error value
unequal zero and the selection of an action, which might be called the
selection gap.

Our control loop operates to select an action that will close the
perceptual gap.

I was asking for this: How exactly does a control loop operate to select an
action that will close the perceptual gap. Do you or do any other
PCT-theorists have an idea?

This is a pretty basic PCT idea. Does it make sense to you?

Yes, I confess, it makes great sense for me ;-). You are right, I´m a
newcomer to PCT, but I already learned the basic lessons and am now trying
to close my knowledge-gaps.

Best regards

Stefan