Conflict in HPCT

[Bruce_Gregory4]

[From Bruce Gregory 92003.11.27.1255)]

[David Goldstein (2003.11.27.1209)]

Both Richard Robertson and I, both of whom work with real conflicts in
real people, gave our answers. Richard Robertson actually gave a better
answer in terms of the current PCT model. It is not a conflct.

Bill Powers gave his theoretical answer and referred to some
demonstrations he has created to illustrate his points.

The problem situation was about the brake/no brake situation. Is this a
conflict? The answer from a clinical and theoretical perspective was
no.

I read Bill's post quite differently. There is conflict but the gain on
the collision avoidance system is much higher and overwhelms the other
system.

Bruce Gregory

[Bruce_Gregory4]

[From Bruce Gregory (2003.11.27.1305)]

David Goldstein (2003.11.27.954 EST)

An internal conflict exists when a person wants to control a perception
in mutually incompatible ways. Who would give this situation even a
second thought? Would you feel a pull in the direction of continuing at
your current speed?

It is "giving that a second thought" that I was concerned with. The
reaction occurs too quickly for higher-level systems to be involved. As
Bill pointed out, all that is required is that the collision-avoidance
system have a sufficiently high gain that it can overwhelm the "press
on" system. There is conflict. but it disappears when the crisis is
past.

Also, you said something that I don't think is true, at least according
to my understanding of PCT. I don't think the the Reoganization System,
which contains innate(genetic), biological control systems that
regulate
chemical and physiological goals, sets any particular goals at the
systems level.

I agree. I meant to say that reorganization system effects the
reference levels of the highest-level perceptions in a way that
minimizes intrinsic error.

My understanding is that it controls something like total
error signals in the perceptual control system hierarchy, and has no
knowledge of what the error signals mean. I think that Bill Powers is
working on a model of this situation and will make it available on
CSGnet in the near future.

That's my understanding, too.

Bruce Gregory

[Richard_Marken2]

[From Rick Marken (2003.11.27.1000)]

Bruce Gregory (2003.11.27.0812)--

One more try. No "model." A few simple questions.

All your questions seem to be about a model. I don't understand why
you say "No 'model'"?

The reference level for the location
and position of my foot is established by a higher level perceptual
control system for "getting to work."

"Reference level" and " higher level perceptual control system" are
components of the PCT model.

This reference in turn is
established by a still higher-level control system "keep my job."

Again, "reference" and "higher-level control system " are components of
a model.

According to HPCT, does a conflict exist?
If so, how does this conflict get resolved
If not, why not?

Since Bill has explained how to go from the PCT model to the phenomenon
of conflict, I'll try to answer this question from the other direction:
how to go from the the phenomenon of conflict to an explanation in
terms of PCT.

I think the Powers, Clark & McFarland (1961) paper describes how to
observe conflict using the "portable demonstrator". Just ask a person
to track, using the index finger of one hand, the movements of your
_two_ index fingers, left and right, as they are held together next to
each other. Slowly move your paired fingers in front of you so that it
is fairly easy for you partner to track the movements.

After a few seconds, when your partner is tracking pretty smoothly,
abruptly move your two fingers apart. I just did this with my son and
the result was that his finger briefly froze between my two fingers.
Then his finger went off to my left (his right) finger, though not all
the way since he was laughing too hard to continue. The freezing of the
finger is the phenomenon of interest. PCT views this as the visible
symptom of internal conflict.

My son was presumably controlling for tracking _both_ fingers during
the initial part of the demonstration. When each of my target fingers
went to a different position, my son presumably had two references for
the position of his own tracking finger. One reference was for his
tracking finger to track my left target finger, the other was for his
tracking finger to track my right target finger. Since his tracking
finger couldn't be in two places simultaneously, there was a conflict.
According to PCT, this conflict existed because two control systems in
my son, of about equal gain, were trying to get the same perception --
the perception of the position of his tracking finger -- into two
different states. One system wanted this perception near my left target
finger, the other wanted this perception near my right target finger.
As long as both of these systems tried, with equal gain, to get the
tracking finger into these two different states, the tracking finger
remained immobile.

This conflict, where the tracking finger remains virtually frozen,
usually lasts for only a couple seconds (my son stopped for about 1
second). Very quickly the subject does something else like move the
tracking finger to one of the target fingers or just lowers the
tracking finger and stops the "game". From a PCT perspective, the
partner seems to "go up a level" and dismantles goals so as to
eliminate the conflict. My son just laughed when he saw his finger
frozen between my two fingers and that ended the "game"; laughter is, I
think, a sign of his having gone up a level.

I think this conflict demo is a good way to get a grasp of the
_phenomenon_ of conflict that PCT explains. I don't know how one could
explain this phenomenon other than using at least a two level hierarchy
of control systems with at least two higher level systems (the two
systems controlling for keeping the tracking finger aligned with the
left and right target fingers, respectively) that set the reference
for one lower level system (the one controlling the visual position of
the tracking finger).

Best regards

Rick

···

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

[Bruce_Gregory4]

[From Bruce Gregory (2003.1127.03)]

Bill Powers (2003.11.27.0659 MST)

There are several problems here, the first being the idea that intrinsic reference levels are at the highest level of the hierarchy. They are not. Intrinsic reference levels are not in the hierarchy at all. Intrinsic error signals result in reorganization, which is a series of alterations of parameters such as gain, or of the connections by which higher systems in the hierarchy contribute to which lower reference signals. or of other aspects of the physical circuitry.

I was speaking too elliptically.

There could even be some selection of reference signals at higher levels where no higher system exists, but these would be randomly selected, perhaps from memories at that level, not directed toward "survival" (unlikely to be at the highest level) or any other specific goal. The only goals of the reorganizing system are the intrinsic reference signals, none of which is "survival."

I was under the impression that the reorganizing system's goal was to minimize error in intrinsic control systems. A subset of these intrinsic control systems are concerned with variables that are closely linked to survival.

Understand that I am describing what the theory proposes, not claiming that these are established facts. But you're asking what the theory says.

The second problem is that conflicts can exist only between systems at the same level. If there were a higher learned goal of survival, whatever that means, it could not conflict with pushing on the brakes. Neither could the goal of "getting to work on time." Only another system directly concerned with operating the brakes could issue a contradictory reference setting for the position of the braking foot and thus create a direct conflict. This follows from the formal definition of conflict in HPCT. Conflict exists if and only if one variable must be in two different states at the same time in order for two control systems to correct their own errors. There are also degrees of conflict, from harmless to devastating.

Exactly, the conflict I was referring to is the perceived location of my foot, either on the accelerator or the brake pedal.

We might say that the system that wants to get to work on time tries to increase the forward speed of the car (if its ETA predicts being late), and the system that wants to avoid collisions tries to decrease the forward speed of the car. Now we have two different reference levels for the same variable, the forward speed of the car, and this constitutes a conflict for the systems trying to adjust the speed of the car. The car can't go at two different speeds at once. Below the level of controlling the forward speed of the car there is no conflict.

O.K. That's another way to look at the problem.

Unless some higher-level system intervenes by altering the reference signals for one or both of the conflicted systems, those systems will produce opposing adjustments of the reference signal for forward speed, and the result will depend on the design details of the control systems involved (see Kent McClelland's simulations of conflicted systems). If the two systems have identical static and dynamic characteristics, the net reference speed, and the actual speed, will be the sum or perhaps the average of the two reference settings (depending on exactly how the two reference signals combine).

O.K.

If a higher order system or set of systems exists that can perceive the existence of the conflict and has learned how to resolve such conflicts (say, by momentarily turning off the goal of getting to the meeting on time), the conflict will immediately be resolved.

I'm not sure that i understand how a higher-order system can "perceive the existence of the conflict." It is also unclear to me that such a higher-order system could perceive the need to act and then act in a short enough time to prevent conflict during the braking process.

Most conflicts are resolved in this way and cause only momentary inconveniences. But if the situation prevents any easy solution or there is no already-known solution, the conflict will simply persist. If the result of its persistence is to create significant large errors in other control systems, reorganization will probably start, and there is then no way to predict what the solution will be, if a solution is found.

Yes, I doubt that the emergency braking scenario leads to reorganization.

Again, these are predictions from the theory that are _roughly_ borne out in my experience, but that have never been formally tested. One has to know what the theory predicts, of course, before there can be any test of it.

In the Crowd program, there is a conflict between collision avoidance and the two other possible goals, following a person or getting to a destination. The conflict is never resolved. What happens is that the collision avoidance system experiences very large changes in error for small adjustments of direction of travel, while the other systems' errors change much less, because the other person or the destination is much farther away than the nearby obstable. In effect, the collision avoidance system has much higher loop gain than the other two systems have, and it overpowers the other two systems when a collision is imminent. When its error returns close to zero, the other two systems simply go on operating and progress continues.

A similar mechanism seems to be at work in the braking example. The collision avoidance system has a much higher gain than the get-to-work system. In that case, higher level systems need not be considered at all.

These conflicts can persist when there is a large number of obstacles present. However, the Crowd agents do have a very minor kind of "reorganization" working, a small random signal added to the perceptual signals. This signal assures that no action ever simply repeats exactly; there is always a slight variation. So when an agent gets stuck in a closed path, the path is randomly perturbed by a small amount, and most of the time there is an eventual escape. It's possible that there wouldalways be an escape, but I have never waited more than five or ten minutes to see if that would happen.

I am curious, now. Was the above question a request for information, or for something you could then sneer at or attack?

I'll leave that for you to decide. As Sigmund Freud said, "Sometimes a cigar is only a cigar."

You seem to be in conflict about letting the true believer lie there, or joining him.

I'm not the true believer type.

Bruce Gregory

[Marc_Abrams1]

from [Marc Abrams (2003.11.27.1420)]

[David Goldstein (2003.11.27.1209)]

Look at Bruce's last post. It starts off with something like "Models
aside..." And then it goes on to ask some questions about internal
conflict.

[From Rick Marken (2003.11.27.1000)]

Bruce Gregory (2003.11.27.0812)--

One more try. No "model." A few simple questions.

All your questions seem to be about a model. I don't understand why
you say "No 'model'"?

Both Richard Robertson and I, both of whom work with real conflicts in
real people, gave our answers. Richard Robertson actually gave a better
answer in terms of the current PCT model. It is not a conflct.

[From Bruce Gregory 92003.11.27.1255)]

I read Bill's post quite differently. There is conflict but the gain on
the collision avoidance system is much higher and overwhelms the other
system.

The problem situation was about the brake/no brake situation. Is this a
conflict? The answer from a clinical and theoretical perspective was no.

David, the questiion was what does the _HPCT model_ say. Bill was only able
to provide a theoretical answer. The reason for this is that this particular
problem would need to be handled by reorganization and currently that is not
part of the model but is certainly part of the theory. _ALL_ of my questions
have to do with the _MODEL_, not the theory (well, maybe a little bit of the
theory :-))

If I understand what you are saying, you are saying that PCT is all
post-hoc.

ABSOLUTELY NOT. The PCT illusion is.

Tom Bourbon did a demonstration one time in which he predicted
a person's behavior on a tracking task one year later.

That is not the kind of prediction I am talking about. The model (not only
the theory) should predict (explain) any controlled behavior. That is, given
an example or problem like the one Bruce gave, you should be able to use the
hierarchy and explain what might happen. The HPCT _model_ does not currently
have that capabilty, although the theory _CAN_.

The behavioral illusion refers to something different than what you seem
to mean.

let me try an example here of what I mean.

from Rick's post;

[From Rick Marken (2003.11.27.1000)]

"...I think this conflict demo is a good way to get a grasp of the
_phenomenon_ of conflict that PCT explains. I don't know how one could
explain this phenomenon other than using at least a two level hierarchy
of control systems with at least two higher level systems (the two
systems controlling for keeping the tracking finger aligned with the
left and right target fingers, respectively) that set the reference
for one lower level system (the one controlling the visual position of
the tracking finger)...."

_This_ from Rick is what I call the PCT illusion.

What kind of hierarchy is Rick speaking of here. He says 'higher' and
'lower' levels , but what kind of levels are these? Can we plug this into
the HPCT hierachy model and see results? I don't think so. So what is
normally done is a PCT _theoretical_ explanation is given and the HPCT
'model' is bypassed. Rick and Bill do this all the time. The theory and
model are two distinct animals here. You would hope that someday, both would
be equivalent, but that won't happen if you don't discuss it.

So the questions about HPCT take on two forms. One are the ones addressed to
the theoretical model, the other to what the model is capable of predicting.
These are _SEPERATE_ issues but are often confounding on CSGnet.

Why don't you try to model that conflict?

For the same reason no one in the past 30 years has been able to. The HPCT
model will not allow it.

See if you can come up with a
predictive model which predicts when you will stay on and when you will
go off. Consider this to be a challenge. If you need any help with this,
I am sure that Bill Powers and Rick Marken would be glad to help.

If only they could they would. But I bet they can give me one great
theoretical explanation for it.

How about you? You seem to have some answers.

marc

[Bill_Powers]

[From Bill Powers (2003.11.27.1258 MST)]

Bruce Gregory 92003.11.27.1255)--

I read Bill's post quite differently. There is conflict but the gain on
the collision avoidance system is much higher and overwhelms the other
system.

That is from my description of how the Crowd demo works. It is not a
description of the example you gave. Your example might have worked that
way, depending on how you set it up, if you gave enough details to judge.
Your assumption of a conflict between two different levels of goals was
incorrect. If any conflict resulted, it would have to show up at a single
level, and your example didn't show that.

You are correct in saying that if a conflict did somehow arise between
getting to work and avoiding collisions, it might be resolved if the
effective loop gain for the collision avoidance system were much higher
than that for the getting-to-work system. Avoiding the collision might
cause a momentary increase of error in the getting-to-work system, but then
the immediate danger would be over and the getting-to-work system would
continue controlling. That is how the Crowd program works.

Bill P.

[Bill_Powers]

[From Bill Powers (2003.11.27.1309 MST)]

Bruce Gregory (2003.1127.03)--

I was under the impression that the reorganizing system's goal was to
minimize error in intrinsic control systems. A subset of these intrinsic
control systems are concerned with variables that are closely linked to
survival.

In the opinion of an observer, yes. In actual fact, yes. But I don't think
that any of these intrinsic control systems controls for survival. Survival
is the result of controlling all the intrinsic variables successfully. Some
people get at idea of what survival means and control for that, but in fact
nobody actually perceives "survival" per se. There can be no correction of
errors in such an imaginary control system. You can't control against
failure to survive, because you're dead.

When people talk about controlling for survival, I think they really mean
controlling for specific things on which they believe survival logically
depends. But that's a learned sort of control, not the intrinsic kind.

Exactly, the conflict I was referring to is the perceived location of my
foot, either on the accelerator or the brake pedal.

If this was the intended conflict in your example, you needed one system
which was trying to brake at the same time another system was trying to
accelerate, thus requiring the foot to be in two different locations at the
same time. Each system would have its own reasons for choosing a direction
of acceleration, and those reasons might trace back to the higher systems
you mentioned: getting to work, and avoiding collisions.

I'm not sure that i understand how a higher-order system can "perceive the
existence of the conflict." It is also unclear to me that such a
higher-order system could perceive the need to act and then act in a short
enough time to prevent conflict during the braking process.

I agree that the most likely outcome is for the collision-avoidance system
to act before the get-to-work system even realizes there is a problem --
after all, up to the point of actually stepping on the brake, progress
toward work is normal. I think people have been known to drive a little
faster after a momentary stop to avoid a problem -- the other system's
error is a little bigger for a while.

A higher-order system can perceive a conflict by perceiving desires to do
two contractory things at once: be a nice guy and win at all costs, for
example. Often people do not realize they are in a conflict; they see the
"good" side and forget that there are reasons for supporting the "bad"
side, too (not all conflicts are between good and bad things, of course).
During MOL demos, people often discover conflicts, as in the tape from the
first Vancouver workshop on MOL. The conflict most often resolves very
shortly after the person manages to observe the conflicting desires or
intentions simultaneously. I am mentioning these details in case some of
our listmates who are concerned with conflicts haven't realized this yet.

A similar mechanism seems to be at work in the braking example. The
collision avoidance system has a much higher gain than the get-to-work
system. In that case, higher level systems need not be considered at all.

I think you're probably right about that, and about the momentary conflict
being insufficient to call for reorganization.

I'll leave that for you to decide. As Sigmund Freud said, "Sometimes a
cigar is only a cigar."

I salute your cigar, sir, and wish you enjoyment of it. I found nothing
offensive in your post.

Best,

Bill P.

[Bruce_Gregory4]

[From Bruce Gregory (2003.11.27.1602)]

Bill Powers (2003.11.27.1258 MST)

Bruce Gregory 92003.11.27.1255)--

I read Bill's post quite differently. There is conflict but the gain
on
the collision avoidance system is much higher and overwhelms the other
system.

That is from my description of how the Crowd demo works. It is not a
description of the example you gave. Your example might have worked
that
way, depending on how you set it up, if you gave enough details to
judge.
Your assumption of a conflict between two different levels of goals was
incorrect. If any conflict resulted, it would have to show up at a
single
level, and your example didn't show that.

I didn't mean to leave the impression that the conflict arose between
different levels. It arose between systems having different references
for the position of my right foot (on the accelerator and on the
brake).

You are correct in saying that if a conflict did somehow arise between
getting to work and avoiding collisions, it might be resolved if the
effective loop gain for the collision avoidance system were much higher
than that for the getting-to-work system. Avoiding the collision might
cause a momentary increase of error in the getting-to-work system, but
then
the immediate danger would be over and the getting-to-work system would
continue controlling. That is how the Crowd program works.

Yes, I think the Crowd approach is the one to take in my example. It is
simple and it accounts for my behavior!

Bruce Gregory

[Martin_Taylor2]

[Martin Taylor 2003.11.27.1710 EST]

[From Bruce Gregory (2003.11.27.1305)]
...I meant to say that reorganization system effects the
reference levels of the highest-level perceptions in a way that
minimizes intrinsic error.

That's not MY understanding of the reorganizing system--which
suggests that it is not too well understood.

My understanding is that the reorganizing system affects the
long-term characteristics of the structure of the perceptual control
hierarchy, such as the inteconnections among levels, possibly the
gains of individual loops, and perhaps the transport lags in
individual loops. In no way does the reorganizing system affect (let
alone "effect") reference levels anywhere in the hierarchy.

Over evolutionary time-scales, reference levels WITHIN the
reorganizing system may themselves be varied, but that's a far cry
from the moment-by-moment changes of reference levels that happen
throughout the perceptual control hierarchy that exists within a
behaving individual.

As for the "accelerate-or-brake" conflict that started this thread, I
think Bill has it right. The ONLY conflict is in a control system
that perceives the forward speed of the car, and its reference level
is affected (1) by a slowly changing (or invariant) reference for the
time to arrive at work in a control system that perceives the
currently projected time of arrival, and (2) by a rapidly changing
PERCEPTION of the distance to the leading car where the function
relating error to output is highly non-linear (it doesn't matter if
you are a long way behind the car in front of you, but if it's
looming up closely and quickly, it matters a lot:-).

So, as Bill said, as soon as the distance to the car in front gets
shorter than the current (speed-related) reference distance, the
output of the distance-to-trouble elementary control unit (ECU)
rapidly overwhelms the slowly changing output of the
"getting-to-work-on-time" ECU in setting the reference for the ECU
that controls perceived forward speed.

···

-----------------

We used to have this kind of discussion when I was a more active
contributor to CSGnet years ago. I had thought at that time that the
conventional PCT answers had been more or less agreed. Do you
remember my distinctions among "core PCT", "agreed, conventional
PCT", and "speculative PCT"? To challenge "core PCT" requires really
heavy evidence, which is unlikely to be found. "Core PCT" includes
things like "behaviour is the control of perception, not the
reverse". "Conventional PCT" is where one is most likely to run into
emotional challenges when one suggests alternatives. "Conventional
PCT" includes the hierarchy and the reorganizing system. "Speculative
PCT" includes the exact description of the levels in the hierarchy
(at least I think so, though they may belong in the category of
"conventional PCT"), and such things as mechanistically how the
category level is produced from analogue variations in lower-level
perceptions.

Challenging "core PCT" without evidence is likely to be treated as
ignorant comment. Challenging "conventional PCT" gets strong
emotional reaction from those who think the convention is part of the
core. Challengin "speculative PCT" is likely to get interested and
interesting comment from all sides.

All of which is separate from demonstrations or discussions of the
usefulness or uselessness of PCT in the complexities of real-world
behaviour.

Happy Thanksgiving, US Americans!

Martin

[Bruce_Gregory4]

[From Bruce Gregory (2003.11.27.1920)]

I want to thank everyone who contributed to this thread. I gained some
valuable insights from your comments and responses, and I hope i was
not the only one to benefit.

Bruce Gregory

[Bill_Powers]

[From Bill Powers (2003.11.27.1729 MST)]

Bruce Gregory (2003.11.27.1920)--

I want to thank everyone who contributed to this thread. I gained some
valuable insights from your comments and responses, and I hope i was
not the only one to benefit.

Spoken like a gentleman and a friend. Hard to keep it that way, but worth
the effort.

Best,

Bill P.

[Bill_Powers]

[From Bill Powers (2003.11.27.1731 MST)]

Martin Taylor 2003.11.27.1710 EST --

Thanks, Martin. Good distinctions and wise words. Just what we need.

Best,

Bill P.

[Philip_Runkel]

This is Phil Runkel attempting a reply to Marc Abrams's of 27 Nov at
11:55:34 AM PST.

Marc, you wrote that PCT "should predict ... any controlled behavior.
[In] an example .. like the one Bruce gave, you should be able to use
[HPCT] and explain what might happen."

If I am understanding what is in your mind behind those words, it may be
that there is a crucial point that is allowing you to feel frustrated
with what some of us tell you.

Namely: NOBODY CLAIMS that PCT (or HPCT) enables you to predict
behavior, controlled or uncontrolled. Even in the Skinner box, where
the rat can do nothing to relieve its hunger other than press the lever,
we cannot predict when the rat will do it or in what rhythm. NOBODY
CLAIMS that living creatures control their behavior. Instead, PCT
claims that we control perception, and that we can usually control a
perception by means of any of a variety of unpredictable actions.

If by "what might happen," you mean particular acts that might occur,
PCT does not undertake to predict any. PCT does, however, predict
certain manners of functioning. One example is the pause of the finger
of Rick's son when Rick suddenly pulled his own fingers wider apart.
That illustrates the well documented delay when the person moves from
one level of control to a higher one. What HPCT predicts is internal
FUNCTIONING, not external action.

Maybe you have all along been expecting HPCT to give you a capability it
is incapable of giving you?

--Phil R.

[Marc_Abrams1]

from [Marc Abrams (2003.11.27.0027)]

Hi Phil I hope your feeling well and thank you for jumping into this.

This is Phil Runkel attempting a reply to Marc Abrams's of 27 Nov at
11:55:34 AM PST.

Marc, you wrote that PCT "should predict ... any controlled behavior.
[In] an example .. like the one Bruce gave, you should be able to use
[HPCT] and explain what might happen."

You misinterpreted what I was saying and I did not clearly state my
thoughts. Thanks for allowing me to clarify.

Yes, PCT does NOT predict behavior. I was talking about predicting the
behavior of the _MODEL_, not an individual.

It all came together for me today and I believe I can explain what my
concerns and hopes are. But first I need to provide a disclaimer. There are
people on this list who think I have malice in my heart and evil intentions.
They hold Bill up as a near deity and _ANY_ questions about the validity of
his work are not acceptable. I have no desire to hurt Bill, nor do I have a
desire to prove him wrong. I wish that what I am about to say was _not_
true, but that would mean I would need to lie to myself and make believe
these problems did exist and that I will not do. I say these things
_BECAUSE_ I want things to work. I asked David Goldstein today how you fix
something when you are unwilling to talk about it? I got _no_ answer and I'm
not holding my breath waiting for one.

I will pose the same question to you and everyone else on this list. Unless
of course you feel nothing is broken.

Sorry, I'm not in that camp.

This is what I have come up with.

The HPCT model is _not_ wrong. It just doesn't exist. There is NO HPCT
model. There is an HPCT _theory_ and a _PCT_ model. They are _NOT_ the same
nor do they represent the same processes.

That is the major reason why 95% of the discussions on CSGnet degenerate
into shouting matches. It is not because of a few cuss words. As long as the
discussion adheres to the PCT _MODEL_, everything is fine. When the
discussion starts moving into untested ground (HPCT theory), and it usually
does, all hell breaks loose. You can't validate HPCT by the PCT model. No
matter how much you holler and scream, wave your hands and question
someone's intentions, it cannot be done. We _NEED_ a HPCT model. _NO_ I
don't expect Bill to build it. I do expect Bill to acknowledge the need and
help where and when he can. We need discussion on CSGnet on both the theory
and model of HPCT. An HPCT model will not be built by osmosis or by some
great intervention. I feel I am getting stonewalled on CSGnet in my honest
attempt to help build an HPCT model. That Phil, is what is frustrating

If I am understanding what is in your mind behind those words, it may be
that there is a crucial point that is allowing you to feel frustrated
with what some of us tell you.

Have I clarified my thoughts?

Maybe you have all along been expecting HPCT to give you a capability it
is incapable of giving you?

No,if we had a functioning HPCT model we would not be having this
discussion.

I hope you and your family had a wonderful Thanksgiving. I expect to be
getting your book shortly and after hearing the raves about it I'm looking
forward to reading it. Again, thanks for asking for clarification. I hope I
provided it.

Marc

[David_Goldstein1]

[David Goldstein (2003.11.28.751)]
[Marc Abrams (2003.11.27.0027)]

Marc: I asked David Goldstein today how you fix something when you are
unwilling to talk about it? I got _no_ answer and I'm not holding my
breath waiting for one.

Answer: Marc, the problem is that you don't seem to really know how to
talk to people without being aggressive, gradiose, and/or disrespectful.

In spite of this, and hoping that you prove me wrong, I would be
interested in hearing what you have to say about HPCT. Bill has reminded
us in his post "For General Interest" what he means by the perceptual
hierarchy.

You say that we need a "working model of HPCT." What do you have in
mind? How will you recognize that we have a working model of HPCT? Can
you name a "working model" in any field of study, so that we have a
better idea of what you have in mind? The only one I can think of are
weather models. Would this satisfy your idea of a working model? Weather
models predict what the weather will be in the future. What will a
working HPCT model predict? What a person will be controlling in a
particular situation? Whether a person will be successful in controlling
well in a particular situation?

David
David M. Goldstein, Ph.D.

···

-----Original Message-----
From: Control Systems Group Network (CSGnet)
[mailto:CSGNET@listserv.uiuc.edu] On Behalf Of Marc Abrams
Sent: Friday, November 28, 2003 12:45 AM
To: CSGNET@listserv.uiuc.edu
Subject: Re: Conflict in HPCT

From [Marc Abrams (2003.11.27.0027)]

Hi Phil I hope your feeling well and thank you for jumping into this.

This is Phil Runkel attempting a reply to Marc Abrams's of 27 Nov at
11:55:34 AM PST.

Marc, you wrote that PCT "should predict ... any controlled behavior.
[In] an example .. like the one Bruce gave, you should be able to use
[HPCT] and explain what might happen."

You misinterpreted what I was saying and I did not clearly state my
thoughts. Thanks for allowing me to clarify.

Yes, PCT does NOT predict behavior. I was talking about predicting the
behavior of the _MODEL_, not an individual.

It all came together for me today and I believe I can explain what my
concerns and hopes are. But first I need to provide a disclaimer. There
are people on this list who think I have malice in my heart and evil
intentions. They hold Bill up as a near deity and _ANY_ questions about
the validity of his work are not acceptable. I have no desire to hurt
Bill, nor do I have a desire to prove him wrong. I wish that what I am
about to say was _not_ true, but that would mean I would need to lie to
myself and make believe these problems did exist and that I will not do.
I say these things _BECAUSE_ I want things to work. I asked David
Goldstein today how you fix something when you are unwilling to talk
about it? I got _no_ answer and I'm not holding my breath waiting for
one.

I will pose the same question to you and everyone else on this list.
Unless of course you feel nothing is broken.

Sorry, I'm not in that camp.

This is what I have come up with.

The HPCT model is _not_ wrong. It just doesn't exist. There is NO HPCT
model. There is an HPCT _theory_ and a _PCT_ model. They are _NOT_ the
same nor do they represent the same processes.

That is the major reason why 95% of the discussions on CSGnet degenerate
into shouting matches. It is not because of a few cuss words. As long as
the discussion adheres to the PCT _MODEL_, everything is fine. When the
discussion starts moving into untested ground (HPCT theory), and it
usually does, all hell breaks loose. You can't validate HPCT by the PCT
model. No matter how much you holler and scream, wave your hands and
question someone's intentions, it cannot be done. We _NEED_ a HPCT
model. _NO_ I don't expect Bill to build it. I do expect Bill to
acknowledge the need and help where and when he can. We need discussion
on CSGnet on both the theory and model of HPCT. An HPCT model will not
be built by osmosis or by some great intervention. I feel I am getting
stonewalled on CSGnet in my honest attempt to help build an HPCT model.
That Phil, is what is frustrating

If I am understanding what is in your mind behind those words, it may
be that there is a crucial point that is allowing you to feel
frustrated with what some of us tell you.

Have I clarified my thoughts?

Maybe you have all along been expecting HPCT to give you a capability
it is incapable of giving you?

No,if we had a functioning HPCT model we would not be having this
discussion.

I hope you and your family had a wonderful Thanksgiving. I expect to be
getting your book shortly and after hearing the raves about it I'm
looking forward to reading it. Again, thanks for asking for
clarification. I hope I provided it.

Marc

[Fred_Nickols3]

[From Fred Nickols (2003.11.28.0805)] --

Bruce Gregory (2003.11.27.0812)]

One more try. No "model." A few simple questions. My foot is on the
accelerator maintaining 75 mph. The reference level for the location
and position of my foot is established by a higher level perceptual
control system for "getting to work." This reference in turn is
established by a still higher-level control system "keep my job." This
reference level is established in terms of intrinsic survival goals. A
car unexpectedly cuts in front of me. I slam on the brakes.

According to HPCT, does a conflict exist?
If so, how does this conflict get resolved
If not, why not?

Interesting question. It will force me to go back to B:CP and the other
PCT books I have to see what, if anything, is said about
priorities. Here's some speculation.

It seems to me that the lower levels of the PCT hierarchy are quite finite,
consisting of the body and its systems. As you move up the hierarchy, the
range of possibilities expands. Toward the top, at the levels of
principles and concepts, it seems to me that there is an almost unlimited
set of possibilities, all of which rely on a much smaller set of lower
level possibilities for the attainment and maintenance of reference
conditions. A couple of thoughts (still thinking out loud):

It would seem to me that conflict could most certainly exist but "up a
level" so to speak. Your foot on the accelerator, maintaining the car's
speed at 75 mph, seems to me to be a configuration and, owing to
adjustments, also involves transitions, the level above
configuration. Above that are levels that address getting to work and
keeping your job. If, as you say, they are established in terms of
intrinsic survival goals, my bet is that "staying alive" is on such
intrinsic survival goal as well, one that is disturbed by a car cutting in
front of you.

I'm not sure where time (and thus urgency and priorities) fit into the PCT
model but it seems to me that the car cutting in front of you represents a
clear and present danger (Thanks, Tom Clancy) whereas getting to work late
or even losing your job poses a lesser threat. Moreover, the threat to job
is off a ways in time and the car is right there right now. The foot on
the accelerator moves off and onto the brake pedal in a manner you describe
as "slamming on the brakes" (which I think I understand perfectly
well). I'll happily accept the experts' view that there is or isn't
conflict in the PCT hierarchy as well as their explanation of how it is or
isn't resolved but, right now, I don't see any inherent requirement for
conflict. It seems to me that there's some kind of switching function
going on that, plainly put, says, "Hey, putting on the brakes right now is
much more important to my really high-level goal of staying alive than the
goal of maintaining 75 mph so as to get to work on time, so let's move my
foot off the accelerator and put it on the brake pedal; like fast and hard,
man!"

Said a little more succinctly, reference conditions are probably infinite
but the resources for achieving them are finite. There has to be some kind
of prioritizing and switching arrangement lurking in there somewhere else
we'd all have walked off cliffs a long time ago.

Regards,

Fred Nickols, CPT
"Assistance at A Distance"
Distance Consulting
nickols@att.net
www.nickols.us

[Bruce_Gregory4]

[From Bruce Gregory (2003.1128.0906)]

Fred Nickols (2003.11.28.0805)

Said a little more succinctly, reference conditions are probably
infinite
but the resources for achieving them are finite. There has to be some
kind
of prioritizing and switching arrangement lurking in there somewhere
else
we'd all have walked off cliffs a long time ago.

Yes, I think you are right. Gain is one important discriminator. The
gain on the collision-avoidance system is doubtless very high. I
suspect that attention plays a role in determining gain. I like nuts,
but the gain on that system seems to be very low until I spot a bowl of
nuts on the kitchen counter.

Bruce Gregory

[Fred_Nickols3]

[From Fred Nickols (2003.11.28.0905)] --

A postscript to my earlier response to Bruce G....

After reviewing Chapter 17 in B:CP, it seems to me that there are two kinds
of conflict lurking in Bruce G's example below:

First, there is the conflict in which the car cutting in front constitutes
a disturbance to Bruce G's goals of driving at 75 mph, getting to work and
keeping his job. The locus of this conflict is BETWEEN the two
drivers/autos. Second, there is the conflict BETWEEN reference conditions
WITHIN Bruce (i.e., between staying alive and driving at 75 mph etc).

In reviewing Chapter 17, I note that Bill comments on both kinds of
conflict but it's not clear to me if they're resolved in different ways.

Finally, I should have re-read Chapter 17 before posting my earlier
response. Duh!

···

At 09:00 AM 11/27/2003, you wrote:

[From Bruce Gregory (2003.11.27.0812)]

One more try. No "model." A few simple questions. My foot is on the
accelerator maintaining 75 mph. The reference level for the location
and position of my foot is established by a higher level perceptual
control system for "getting to work." This reference in turn is
established by a still higher-level control system "keep my job." This
reference level is established in terms of intrinsic survival goals. A
car unexpectedly cuts in front of me. I slam on the brakes.

According to HPCT, does a conflict exist?
If so, how does this conflict get resolved
If not, why not?

Happy Thanksgiving.

Regards,

Fred Nickols, CPT
"Assistance at A Distance"
Distance Consulting
nickols@att.net
www.nickols.us

[Marc_Abrams1]

from [Marc Abrams (2003.11.28.0904)]

[David Goldstein (2003.11.28.751)]

[Marc Abrams (2003.11.27.0027)]

Marc: I asked David Goldstein today how you fix something when you are
unwilling to talk about it? I got _no_ answer and I'm not holding my
breath waiting for one.

Answer: Marc, the problem is that you don't seem to really know how to
talk to people without being aggressive, gradiose, and/or disrespectful.

Is this your answer to my question?

I'm sorry if you don't like my 'style'. You keep on making these
unsubstantiated claims about me. What is 'gradiose'? if you meant grandiose,
please show me where I have ,made exaggerated or wild claims? You call it
'aggressive'. I call it up front, honest, and no nonsense, and as I said to
you yesterday I show as much respect as I get. Have I shown you
'disrespect'? if you feel I have please point it out to me.

In spite of this, and hoping that you prove me wrong, I would be
interested in hearing what you have to say about HPCT. Bill has reminded
us in his post "For General Interest" what he means by the perceptual
hierarchy.

Yes, his "For general interest post was outstanding. Not for the review of
his current ideas about the hierarchy but his seeming willingness to discuss
it.

You say that we need a "working model of HPCT.

_YES_ (Not shouting, just really emphasizing)

What do you have in mind?

Nothing specific and concrete at this point, just _a lot_ of ideas. For
instance, I believe we need to bring emotion and memory into the model. I
think Bill's take on emotions is great but incomplete. Meaning, I agree with
Bill that emotions are the _result_ of control. I also believe that they are
a _major_ contributing factor to _what_ perceptions we control. I also
believe they are important to what kind of reference conditions we have. I
think memory and imagination are _extremely_ important _throughout_ (that
means with regard to every function in the model) the model. I think we need
to discuss HPCT theory with an eye and focus on modeling it.

How will you recognize that we have a working model of HPCT?

When the model can produce the predicted results of the theory

Can you name a "working model" in any field of study, so that we have a
better idea of what you have in mind?

Sure, The PCT model is one. Physics is loaded with them

The only one I can think of are weather models. Would this satisfy your

idea of a working model?

Sure.

Weather models predict what the weather will be in the future.

They _attempt_ to predict the weather. Sometimes they do, and sometimes they
don't.

What will a working HPCT model predict?

That whatever the HPCT theory says has a plausable chance of happening

What a person will be controlling in a particular situation?

No. You would need a _specific_ model of _that_ individual to try that.

Whether a person will be successful in controlling well in a particular

situation?

You could but it would not be practical. You would need to model the
specific person and environment in any attempt. But you are right in
thinking that a good model would, and could be used proactively, rather than
reactively by looking at plausable outcomes given given certain parameters
(properties and conditions of) of the situation.

But the main purpose of a model is to validate a theory. It allows you to
collect data (like the tracking experiment) and see how rhe theory
'matches-up' with the data. Any HPCT model will _never_ be truly finished.
It will be an ongoing project until the model can in fact predict what will
happen to a specific person in a specific situation.

Any HPCT model will be highly speculative and in need of _constant_ revisons
and edits as new facts are learned and new data is acquired. An HPCT model
might not have been practical (given the level of understanding) 30 years
ago. I think that has changed. Some may not agree with me.

Marc

The model would be used by saying; Given the current state of the person
(system) a certain set of inputs, reference condition and disturbances what
are the likely outcomes

[Marc_Abrams1]

from [Marc Abrams (2003.11.28.1057)]

[From Bruce Gregory (2003.1128.0906)]

Yes, I think you are right. Gain is one important discriminator.

When you say 'gain'. I think emotion. Where would the 'gain' originate in
your model?

Marc