A Thorny Question

[From Fred Nickols (2006.11.11.1213 EST)] --

As I reviewed some of the material I've been writing about the GAP-ACT
model, I posed myself a question I can't answer. So, I'll try to restate it
in PCT terms.

What I'm calling a goal (G) refers to what is called the "reference
condition" in PCT. In either case, they serve to specify some desired or
required state for what I call the "target variable" (T) and which is known
in PCT as the "controlled quantity" or "controlled variable."

In my GAP-ACT model, T is "out there" or part of the environment. Ditto for
the controlled quantity in PCT. PCT, of course, identifies those controlled
quantities as "remote physical phenomena" (see p. 61).

What if the controlled quantity or targeted variable isn't some "remote
physical phenomena" but instead refers to some aspect of me? What if the
controlled quantity isn't "out there" but is instead "in here." If I set a
goal of "being a better person" I obviously need to sharpen that goal or I
won't have a snowball's chance of achieving it. So perhaps I express it in
terms of my own, overt, observable behaviors (something like "snap at people
less often" or "make fewer snide comments" or "listen instead of criticize"
and so on). My overt behaviors are as "out there" as anything else.

In any case, if someone says, "Fred, your GAP-ACT (and that PCT stuff) is
all well and good but I hope you've noticed that it's focused on 'out there'
and doesn't deal with what goes on 'in here'," then I'm struggling with what
to say. I know PCT deals with internal systems, especially at the lower
levels so "remote physical phenomena" obviously doesn't apply to everything
- or does it.

I'm looking for help again - as usual ...

Regards,

Fred Nickols
"Assistance at a Distance"
nickols@att.net
www.nickols.us

[From Rick Marken (2006.11.11.11.11)]

Fred Nickols (2006.11.11.1213 EST) --

In my GAP-ACT model, T is "out there" or part of the environment. Ditto for
the controlled quantity in PCT. PCT, of course, identifies those controlled
quantities as "remote physical phenomena" (see p. 61).
...
In any case, if someone says, "Fred, your GAP-ACT (and that PCT stuff) is
all well and good but I hope you've noticed that it's focused on 'out there'
and doesn't deal with what goes on 'in here'," then I'm struggling with what
to say.

Just say: What is actually controlled by a control systems is an internal representation of aspects of what's out there. The remote physical phenomena that are controlled are controlled because they correspond to the perceptual variable that is actually controlled. Explain it in terms of a thermostat. The thermostat doesn't control room temperature; it controls the the size of a coil. The size of the coil is a perceptual representation of the temperature of the air near the coil. So as the thermostat controls the size of the coil, it controls the temperature of the air around the coil (the remote physical phenomenon) as a side effect. And since the temperature of the air near the coil depends on the temperature of the rest of the air in the room, the thermostat can be said to control room temperature (an even more remote physical phenomenon). To avoid going through all this rigamarole it's easier to just say that the thermostat controls room temperature. But sometimes it is very important to understand the epistemology of control because their is a tendency for people to assume that there are things out there in the real world to control. In fact, the real world is just a set of possibilities; what we control depends on what aspect of the world we perceive. If the size of the thermostat coil depended on humidity (a variable that exists only as a perception, temperature * moisture) rather than temperature then humidity -- a variable that doesn't even exist out there, only in here are a perception -- would be controlled.

A listener who is still around after _that_ explanation is definitely an up and coming PCT fanatic.

Best

Rick

Richard S. Marken Consulting
marken@mindreadings.com
Home 310 474-0313
Cell 310 729-1400

[Martin Taylor 2006.11.11.14.14] (actually it was 11.11.14.14.14 when I started to write the date and time!)

[From Fred Nickols (2006.11.11.1213 EST)] --

What if the controlled quantity or targeted variable isn't some "remote
physical phenomena" but instead refers to some aspect of me? What if the
controlled quantity isn't "out there" but is instead "in here."

In our earlier discussion, I said you were confusing the inside and outside of a control loop with the inside and outside of the person in whom the control loop exists. You are doing it again.

A simple (canaonical in PCT) control system consists of a loop that has two places affected by signals from outside the loop. The loop itself has two segments, which we can call an "inside part" and an "outside part". The "inside part" is connected to one signal from outside the loop (the "reference signal"), while the "outside part" is connected to the other (the "disturbance signal").

The "inside part" of a canonical control loop consists of three things and only those three with their connecting links: (1) an interface to the "outside part" called the "perceptual input function", which transforms data from the outside part of the control loop into a variable PCT calls the perceptual variable; (2) a comparator that compares the value of the perceptual variable with another variable coming from outside the loop (PCT calls it the "reference variable"); and (3) an output function that produces a variable that influences the "outer world".

The "outside part" of the control loop consists of the effects on the whole universe generated by the signal emitted by the output function, insofar as those effects influence the perceptual input function inputs. The perceptual input function is also influenced by a second signal from the outer universe, which PCT calls "the disturbance variable".

Usually, the control loop is drawn without a clean separation between inside and outside parts of the control loop. But when puzzles like yours arise, it is important that the separation be marked. And it even more important to note that this separation is NOT at the person's skin, except possibly for the most peripheral control systems. For almost all control loops, BOTH the reference signal and the disturbance signal come from within the person's skin, though outside the control loop. The control loop has no way to know whether their ultimate source lies inside the skin or outside it. The person's skin is simply irrelevant to the operation of the control loop.

Now, an external observer is, by definition, outside the person's skin, though if that person is a doctor or physiologist they may have probes that look beneath the skin. For most of us, and most of the time even for a doctor, we can perceive only what happens outside a person's skin. When we think of a person controlling anything, even their (imputed) internal perceptions, it's very easy to be seduced into thinking that the "outside parts" of all the control loops fall outside the person's skin.

That seductively easy, and almost always erroneous, equation of the inside and outside of a control loop with the inside and outside of the skin seems to be at the root of most of the puzzles you have brought up in the last couple of weeks.

In any case, if someone says, "Fred, your GAP-ACT (and that PCT stuff) is
all well and good but I hope you've noticed that it's focused on 'out there'
and doesn't deal with what goes on 'in here'," then I'm struggling with what
to say.

If you forget entirely about what Bill P. calls the "imagination loop", you will have a problem. If you forget that one can perceive some of one's bodily states, you will have a problem. If you take only the strict hierarchy of inputs from the outer world generating one layer after another of perceptual signals starting only from effects at the skin periphery, you will have a problem. But even the most rigorous versin of HPCT accepts that internal signals and cross-feeds from references to perceptions do exist. So, too, should GAP-ACT. The "out there" to a control loop can very well be "in here" to the person.

I'm looking for help again - as usual ...

I hope the foregoing is more help than confusion!

Martin

[From Fred Nickols (2006.11.11.1443 EST)] --

From Rick Marken (2006.11.11.11.11)]

Just say: What is actually controlled by a control systems is an
internal representation of aspects of what's out there.

<snip the rest>

I understand that, Rick. I was asking about a situation in which what they're trying to control is "in here," not "out there."

The remote
physical phenomena that are controlled are controlled because they
correspond to the perceptual variable that is actually controlled.

I understand that, too, which is why in my follow-up post I said ignored my first post because I am assuming the same principle applies; namely, what is being controlled is a perception of whatever the controlled quantity might be - whether it's "in here" or "out there."

I am also assuming that you must be able to form a perception of whatever it is - wherever it is - or there's no possibility of controlling it.

Thanks for taking the time to reply...

Regards,

Fred Nickols
nickols@att.net

[From Fred Nickols (2006.11.11.1454 EST)] --

[Martin Taylor 2006.11.11.14.14] (actually it was 11.11.14.14.14 when
I started to write the date and time!)

>[From Fred Nickols (2006.11.11.1213 EST)] --
>
>What if the controlled quantity or targeted variable isn't some "remote
>physical phenomena" but instead refers to some aspect of me? What if the
>controlled quantity isn't "out there" but is instead "in here."

In our earlier discussion, I said you were confusing the inside and
outside of a control loop with the inside and outside of the person
in whom the control loop exists. You are doing it again.

Thanks, Martin, that's a useful distinction and one I'd not made. I do seem
to be confusing the "skin" with the loop.

<snip>

That seductively easy, and almost always erroneous, equation of the
inside and outside of a control loop with the inside and outside of
the skin seems to be at the root of most of the puzzles you have
brought up in the last couple of weeks.

>In any case, if someone says, "Fred, your GAP-ACT (and that PCT stuff) is
>all well and good but I hope you've noticed that it's focused on 'out
there'
>and doesn't deal with what goes on 'in here'," then I'm struggling with
what
>to say.

If you forget entirely about what Bill P. calls the "imagination
loop", you will have a problem. If you forget that one can perceive
some of one's bodily states, you will have a problem. If you take
only the strict hierarchy of inputs from the outer world generating
one layer after another of perceptual signals starting only from
effects at the skin periphery, you will have a problem. But even the
most rigorous versin of HPCT accepts that internal signals and
cross-feeds from references to perceptions do exist. So, too, should
GAP-ACT. The "out there" to a control loop can very well be "in here"
to the person.

>I'm looking for help again - as usual ...

I hope the foregoing is more help than confusion!

It's helpful, Martin. I guess I'd better go back to PCT school.

Regards,

Fred Nickols
nickols@att.net