Arcvhive Post and Question

from [ Marc Abrams (990715.1614) ]

Came across this exchange. Rick _What_ would you _actually_ tell people to
do to take advantage of your advice?

Marc

[From Rick Marken (940303.2300)] >Dag Forssell (940303 1630)

How about:
When you have accepted that we are all living control systems, you realize

that actions >depend on wants as they relate to present
perception--influenced by the environment, >disturbances and your own
actions--in the circular chain of control.

I still do not like it. How about:

When you have accepted that we are all living control systems, you realize
that people act only to produce intended perceptions; how people act in
order to do this is determined mainly by circumstance. People control the
results of their actions (their perceptions) not the means used to produce
those results (their observable actions).

[From Rick Marken (990715.1440)]

Marc Abrams (990715.1614) --

Came across this exchange. Rick _What_ would you _actually_
tell people to do to take advantage of your advice?

I guess you are taking the following as advice:

When you have accepted that we are all living control systems,
you realize that people act only to produce intended perceptions;
how people act in order to do this is determined mainly by
circumstance. People control the results of their actions
(their perceptions) not the means used to produce those
results (their observable actions).

Based on this advice, I would tell people not to tell people
what to do;-)

But if they _ask_ me to tell them what to do, I would ask them
why they want me (of all people) to tell them what to do?

Best

Rick

···

--
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates mailto: rmarken@earthlink.net
http://home.earthlink.net/~rmarken

[From Tim Carey (990716.0805)]

Thanks Marc, it was a great post.

I think though that Rick has revise his stance on people controlling others
behaviour. My understanding is that he no longer believes:

The problem is that PCT shows that we CAN'T really control the behavior of
another human controller. We might luck out (like your manager) and have
people act, as we want --but PCT shows that it's just that -- LUCK. The

more

common result of controlling other people is CONFLICT -- because the
controllee is always controlling the controller right BACK.

There was a post earlier this year or late last year where Rick presented
his revise ideas.

Thanks again for the post.

Cheers,

Tim

from [ Marc Abrams (990715.1753)]

Ah ! Patience is a virtue. ( one I sorely lack sometimes :slight_smile: ) In my
continued work on the archives I came across this reply from Rick.

_All_ of you _applied_ folks out there pay some heed to this _excellent_
post. Rick sometimes you amaze me. :slight_smile:

[From Rick Marken (990715.1440)]

Thanks Rick, but what I was looking for you did over 5 years ago :slight_smile:

Marc

[From Rick Marken (940303.2300)] >Dag Forssell (940303 1630)

How about:
When you have accepted that we are all living control systems, you realize

that actions >depend on wants as they relate to present
perception--influenced by the environment, >disturbances and your own
actions--in the circular chain of control.

I still do not like it. How about:

When you have accepted that we are all living control systems, you realize
that people act only to produce intended perceptions; how people act in
order to do this is determined mainly by circumstance. People control the
results of their actions (their perceptions) not the means used to produce
those results (their observable actions).

Me:

If "you are concerned about how he acts" don't you think PCT might suggest

that YOU are >the one who should learn to "think more effectively" -- rather
than getting the other person to >"think more effectively".

Dag:

Dr Spock, as one who manages nobody, and cares not one whit what your kids

do that will >get them expelled from school, in jail, sick, or whatever
because of their foolish ways, you >should know. "Let the little darlings
operate with infinite degrees of freedom" is utopia, not the >real world. I
doubt very much that you walk like your foolish talk. I cannot wait for your

constructive suggestions, based on your real world experiences rather than

black and white >theorizing (dreaming).

I see. The "real world", that place where "the rubber meets the road", is a
place where we must control people or be considered "foolish". You described
a situation where you were "concerned about how [someone] acts". Your
concern indicates an error -- there is a discrepancy between an image of how
the person should act and how they are acting. You want the person to learn
to "think more effectively" so that you will no longer be concerned about
his behavior (you will no longer have an error signal). You are framing this
in terms of helping the other person by getting them to "think effectively".
Would you really stop being concerned if the person's way of "thinking
effectively" involved taking a fake spill down the stairway, suing for
workman's comp and then staying at home with the in-laws. Probably not,
because you have references for what you consider a reasonable solution to
the other's person's "problem"; your concern is REALLY about what YOU
perceive because YOU are a controller -- we are all controllers. That is
what PCT is about.

What your little "role play" described was one possible way to get a person
to behave in a way that would stop causing you concern; it was a
demonstration of control; and one that worked unrealistically well.

There is nothing wrong with control; the point of PCT is that EVERYONE
controls-- we cannot help it. We can no more stop controlling than we can
stop breathing. That's why the manager in your "role play" is "concerned".
It is because the manager is a control system -- only control systems can be
"concerned". In addition, one of the things that human control systems are
usually concerned about is the behavior of other people. So people try to
control the behavior of other people; you do it, I do it, and we ALL do it
because we are all controllers. This IS the real world. If you think I am
advocating that people stop trying to control then I can see why you think I
am being unrealistic. That is not only unrealistic -- it is IMPOSSIBLE
according to PCT. People "know" the way things "should" be because they are
a mass of reference signals -- and nothing can stop them from trying to make
their perceptions match those reference signals. And many of those reference
signals specify the "right" level of perceptions of other people's behavior:
the right amount of time to do homework, the right amount of drinking and
smoking, the right this and that.

The problem is that PCT shows that we CAN'T really control the behavior of
another human controller. We might luck out (like your manager) and have
people act, as we want --but PCT shows that it's just that -- LUCK. The more
common result of controlling other people is CONFLICT -- because the
controllee is always controlling the controller right BACK.

···

Date: Thu Mar 03, 1994 11:03 PM PST
Subject: PCT and Management

----
Side note: I think those role-plays are lousy ways of illustrating human
interaction. They are completely ridiculous from a PCT perspective --
because the actions in the role-play (unlike those in the REAL WORLD) always
produce the expected result (compliance from the other person, for example).
I can demonstrate the virtue of any management strategy using this
technique. For example, here is an S-R management technique that I just
"role-played" with my daughter:

Me: Lise, please clean up your room right now.
Lise: But Daddy, I have homework to do.
Me: I said clean up your room right NOW.
Lise: OK Daddy. I'll do it right now. My homework can wait.
Me: Thank you Lise
Lise: You're welcome, Daddy.

Pretty good, eh. In addition, I used a real situation with a REAL teenager.
We may not be talking about where the rubber meets the road but we're sure
right there where the lipstick meets the lips.

Leave "role plays" to the third graders, Dag.
-----
Dag continues:

In the real world, not all control systems manage to satisfy themselves

Some of them have not >grown up to be as capable as you are. Some of them
have not grown up yet. Ed Ford >shows how to help them become satisfied in
a real world of agreements, work, rules, family >groups, teams, and whatever
where there is a need to get along with others.

The kind of "helping" I saw described in your post looked a lot like the
kind of helping we saw described by Tom's student at the last CSG meeting.
The kind of "helping" you describe comes very close to being controlling. It
is very hard to help another person without interfering with that other
person's efforts to control; but a good example of the way NOT to help a
person was shown in the "role play" you described. I don't have any great
ideas about how you can help another person -- but I would imagine that it
starts with the "helpee" expressing a need for help; the person in your role
play did not obviously want help -- at least you didn't say that he did. The
person who had the problem in the role-play was the manager -- the person
who was "concerned" about the worker. That's why I said that the manager
should seek help; and it's why I called it "godfather" PCT. You obviously
think of it as a situation where you are informing a person of a possibly
undesirable consequence of their action -- being chronically late leads to
demotion or firing. However, if that's all it is, then just give the
information and let the person deal with it as they wish. The way you
presented it, it was more like an ultimatum: find a way to work things out
at home or be demoted.

I think that people who want to be helped can be helped -- using techniques
based on PCT. However, I don't think that managers need to learn to help
people in this way. I think the problems of managers usually have to do with
coordinating the efforts of many WILLING workers; managers who spend a great
deal of time "helping" problem workers are probably rare. So the application
of PCT to management should deal mainly with what PCT says about how to
successfully coordinate the efforts of many people in order to achieve
common goals. Here are some ways that I think PCT can be applied to
management:

1. Managers must recognize that people control what they perceive. Thus, an
important task of management is to try to insure that everyone is able to
perceive the common goal in terms of all the perceptual variables that make
it up. Description of the dimensions of the common goal is important; what
are the variables you want the workers to control? These variables can be
given names -- but education is needed to insure common perception: what is
"quality", "efficiency", etc.

2. Managers must try to establish common reference levels across workers for
the intended states of the controlled perceptions. Once workers know what
perception(s) to control they must know the appropriate levels at which to
control them. Once we know that we are to control the "closeness of joins",
we have to know just how close we want them.

These first two points are about the importance of "training" to achieve
cooperation; an important component of this training is the education of
perception; teaching worker's what variables to perceive.

3. Managers must be aware of the fact that all workers are controlling a
whole constellation of perceptions; the perceptions to be controlled at work
are just a subset of the perceptions people control. This means that you
can't expect to be able to arbitrarily tell a person to "do this" (meaning
"control this variable at this reference level"). Manager's must be
sensitive to the fact that control of certain variables will conflict with
control of other variables. A manager must always be willing to be flexible
about who does what, when and with whom. The manager in your story might
have been a bit more flexible about when the worker shows up -- especially
if the worker is otherwise getting things done --unless the main goal of the
company is to produce people who "show up" at a particular time.

4. Managers must understand that worker's control perceptions, NOT actions.
Manager's are there to help people understand what results are to be
produced -- NOT HOW they are to be produced. PCT shows why "micro-
management" is so manifestly unsuccessful.

These second two points are about the importance of "flexibility" in
management. A manager who is committed to "doing it the GM way", or "doing
it the Japanese way", or even "doing it the Deming way", (if the later is
conceived of as a particular set of activities that must be carried out in
order to have a successful company) is committed to eventual failure. What
works now will not necessarily work in 5 years. We might not always know WHY
something no longer works -- but we can tell THAT it is no longer working
when we are no longer getting the results we want. PCT suggests that
managers (leaders) must always be willing to try something new when things
are no longer working; leaders committed to doing it "the right way" will
eventually be "history".

So training and flexibility are two important practical implications of PCT.
Managers who provide workers with training and flexibility will succeed;
those who don't won't (in the end).

One last implication of PCT is that you DON'T need to provide managers or
workers with control -- they already have it. If a manager wants a worker to
show up at a particular time, he can be counted on to try what he can to get
it to happen (like the manager in your role-play). Managers don't have to
learn how to control workers -- nor do they have to learn to help workers
control (they will just be interfering). What managers have learned is to be
aware of their OWN controlling; they must be aware of the fact that they are
controllers. Such an awareness would have helped the manager in your role
play understand his own controlling -- and might have kept him from acting
like such a jerk. He would have known that his "concern" comes out of his
own nature as a controller and his goals about what the worker "should do"
might not help the worker at all. A manager aware of him/herself as a
controller can spend more time worrying about coordinating efforts for the
common good and less time trying to "help" (euphemism for control) people
for their own "good"

Best Rick

PS. Lise did NOT clean her room; but she did do her homework (as usual) and
gave me a big kiss for being such a silly daddy for doing the stupid role
play.

[From Rick Marken (990715.1600)]

Tim Carey (990716.0805)--

Thanks Marc, it was a great post.

I think though that Rick has revise his stance on people
controlling others behaviour. My understanding is that he
no longer believes:

> The problem is that PCT shows that we CAN'T really control
> the behavior of another human controller. We might luck out
> (like your manager) and have people act, as we want --but
> PCT shows that it's just that -- LUCK. The more common result
> of controlling other people is CONFLICT -- because the
> controllee is always controlling the controller right BACK.

I have revised my ideas slightly but I still agree with the
basic thrust of my earlier statement. It just needs some
qualification. PCT shows that we _can_ control behavior
under some very special circumstances. First, we can control
behavior by brute force (or the threat thereof); but this
control only works when we are more powerful than the person
being controlled. So it generally won't work on our peers.
Second, we can control actions by disturbing controlled
variables that are influenced by these actions (this is the
kind of control demonstrated in my "Control of Behavior" demo at:

http://home.earthlink.net/~rmarken/demos.html

But this kind of control only works as long as the actions
you are controlling don't interfere with the controllee's
ability to control other variables (for example, as Bill
notes in B:CP, you can control the finger position of a
person controlling the knot in the rubber band demo, but
you lose control when you try to place the person's finger
where the person doesn't want it -- like on the tip of a hot
soldering iron).

So I would augment my statement above to say "PCT shows that
we can't _arbitrarily_ control the behavior of another
controller (we can't make a person do _whatever_ we want them
to do; we must take the other person's hierarchy of references
into account) except by physically overpowering them".

The fact is that in most real life situations, our attempts to
control another person's behavior will simply put us in
conflict with that person. If we are strong enough, the
conflict won't matter -- we win. If we notice the conflict
and try to avoid it by getting the person to do something
that doesn't produce conflict then we are not _arbitrarily_
controlling the person; we are making the person do either
what they want to do anyway or what they don't care about;
we are taking the other person's hierarchy of references into
account (that is, we are doing what I call "respecting the
other person").

In order to avoid saying all this, I'd be happy to stick
with my original statement above. Just try to avoid
controlling other people; PCT shows that trying to control
others will _most likely_ lead to conflict (that is, to a
big fight).

Best

Rick

···

---
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates mailto: rmarken@earthlink.net
http://home.earthlink.net/~rmarken

from [ Marc Abrams (990715.1839) ]

Glad you enjoyed it Tim. I uncovered another one ( I promise it'll be the
last for awhile :slight_smile: )

from Bill, addressing Rick about the same thread. In response to Rick's
post.

Btw, have a good safe flight in and an enjoyable conference. I really wish I
could have made it

[From Tim Carey (990716.0805)]

I think though that Rick has revise his stance on people controlling

others

behaviour. My understanding is that he no longer believes:

Yes, but I think the jist of the post was an important one. I don't think it
was on how you could not manipulate people, as much as it was on a PCT
perspective of the kinds of considerations a manager, teacher, etc. should
have from a PCT perspective. I don't think Rick's stance has changed on
_how_ PCT might be helpful. I think his stance has changed on whether you
could in fact control others. Rick, what are your thoughts on that post?

Bill's post ( you'll find on the bottom ) adds a few interesting pieces

>The problem is that PCT shows that we CAN'T really control the behavior

of

>another human controller. We might luck out (like your manager) and have
>people act, as we want --but PCT shows that it's just that -- LUCK. The

more

>common result of controlling other people is CONFLICT -- because the
>controllee is always controlling the controller right BACK.

There was a post earlier this year or late last year where Rick presented
his revise ideas.

Yes, I agree, but I think Rick is presenting two ideas here. One is the
ability to control another. That we know and Rick now concedes to be true.
The other is counter-control which _also_ takes place :slight_smile:
( the last thought in the above paragraph )

Marc

Part of a much longer post from WTP addressing a few folks.

Rick Marken (940303.2300) --

I liked your restatements of what managers must learn. Control is a fact and
we all do it. Managers are controlling for getting the people, who work for
them to behave, as the manager wants them to behave. The basic fact is that
if they don't comply, the manager will find someone who will comply. A
worker who fails to satisfy the manager doesn't have the option of staying
on the job.

As you say, this is the basic framework within which all interactions
between managers and workers take place. Dag acknowledges this.

So, what can "management philosophy" do to smooth the path of progress? The
first step has to be taken when a worker is hired. The manager has to
explain that his/her role in the company is to see that certain things are
achieved, by assigning objectives to those who contribute to the
achievement. The worker being hired has to agree to satisfy the goals set by
the manager, to meet the standards set by the manager and by the time
specified by the manager. The manager might indicate an openness to
negotiate details and consider suggestions, but the responsibility for
achieving the manager's goals is the manager's, and the worker is hired only
to achieve those goals. It is up to the worker whether taking the job under
those conditions is agreeable. If this agreement is explicitly reached, it
forms the basis for all future interactions.

Unfortunately, this simple and rational approach runs into all sorts of
snags. Many workers would like to be paid without doing any work, and many
managers would like to get the services of workers without paying any wages.
The goal of maximizing profit means minimizing costs; the goal of living the
good life means obtaining the goodies without onerous labor. Under normal
conditions, there is a built-in labor-management conflict. This conflict is
made worse by pressure on the manager from above and pressure on the worker
from the needs of daily living.

There is no formula that can resolve all the conflicts inherent in the
labor-management structure. What PCT can do is to teach people about human
nature, so they will recognize conflict when it happens and not be surprised
or outraged. The only ultimate solution is for all the people in a company,
from CEO to sidewalk-sweeper, to recognize that they're all in the
enterprise together and depend equally on its success (this also has to be
TRUE). They all have to understand the structure of the company and what
their role is in keeping it viable. This means understanding how people
work, and especially how they don't work -- they don't respond well to
forcing. It means understanding that each person in a company has needs and
interests other than the job, and that to enable either workers or managers
to function well, all these needs have to be taken into account.

All of that's easy to say. What's hard is to generate the good will that's
required to make any of this understanding useful. My own feeling is that
simple understanding of PCT is about the best help we can offer; it's better
than the superstitions, prejudices, habits, and falsehoods under which most
business and other human affairs are conducted today. I go on the principle
that it's better to understand what's going on than not to understand it or
to understand it incorrectly. What you do differently with that
understanding is pretty much up to each person to work out.

I don't really much care how people like Dag Forssell and Ed Ford go about
teaching PCT, as long as they're teaching it. Those who pay attention seem
to get a lot out of it, even without being told what they should get out of
it. In contrast to what exists now, that becomes pretty obvious to people.
People who don't pay attention won't get it, of course, but I think we have
to focus on a quality few rather than trying to mass-produce PCTers. If each
one teaches two, PCT will spread all by itself. All we have to do is keep
the message from degrading.

Best to all, Bill P.

[From Tim Carey (990716.16150]

From [ Marc Abrams (990715.1839) ]

Glad you enjoyed it Tim. I uncovered another one ( I promise it'll be the
last for awhile :slight_smile: )

Thanks again. It was great reading too.

Btw, have a good safe flight in and an enjoyable conference. I really wish

I

could have made it

Thanks, I'm really excited about it. It should be great.

Cheers,

Tim

[From Tim Carey (990716.1615)]

[From Rick Marken (990715.1600)]

Thanks for your reply Rick.

under some very special circumstances. First, we can control
behavior by brute force (or the threat thereof); but this
control only works when we are more powerful than the person
being controlled. So it generally won't work on our peers.
Second, we can control actions by disturbing controlled
variables that are influenced by these actions (this is the
kind of control demonstrated in my "Control of Behavior" demo at:

I noticed here that you changed from using the word "behavior" to using the
word "action". Did you mean to do this? Are you referring to two different
things in the two points?

So I would augment my statement above to say "PCT shows that
we can't _arbitrarily_ control the behavior of another
controller (we can't make a person do _whatever_ we want them
to do; we must take the other person's hierarchy of references
into account) except by physically overpowering them".

Yep, good point.

In order to avoid saying all this, I'd be happy to stick
with my original statement above. Just try to avoid
controlling other people; PCT shows that trying to control
others will _most likely_ lead to conflict (that is, to a
big fight).

Yep, again.

Cheers,

Tim

[From Rick Marken (990716.0740)]

Me:

First, we can control behavior by brute force (or the threat
thereof)... Second, we can control actions by disturbing
controlled variables that are influenced by these action

Tim Carey (990716.1615) --

I noticed here that you changed from using the word "behavior"
to using the word "action". Did you mean to do this? Are you
referring to two different things in the two points?

Yes. You can control _any_ aspect of behavior -- actions or
results -- by brute force. You can only control actions by
disturbing controlled variables.

Best

Rick

···

--
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates mailto: rmarken@earthlink.net
http://home.earthlink.net/~rmarken

[From Tim Carey (990717.0545)]

[From Rick Marken (990716.0740)]

Yes. You can control _any_ aspect of behavior -- actions or
results -- by brute force. You can only control actions by
disturbing controlled variables.

OK, thanks. That's a useful distinction. When using brute force to control
another's behaviour, the control of their actions is pretty easy to
understand. When you use the word "results", however, are you referring to
results from the perspective of the brute force person or the person being
brute forced?

Cheers,

Tim

[From Bruce Nevin (990716.1640 EDT)]

Tim Carey (990717.0545)--

[From Rick Marken (990716.0740)]

Yes. You can control _any_ aspect of behavior -- actions or
results -- by brute force. You can only control actions by
disturbing controlled variables.

OK, thanks. That's a useful distinction. When using brute force to control
another's behaviour, the control of their actions is pretty easy to
understand. When you use the word "results", however, are you referring to
results from the perspective of the brute force person or the person being
brute forced?

This leaves out intimidation by threat of force. I suppose that could be
described as controlling either actions qo' or results qi'=qi by disturbing
some other variables QI'=QI that are controlled by the victim.

qo' = qi -- actions of victim are results for coercer
qi' = qi -- results from both perspectives
            (victim can't control qi' successfully)
QI' = QI -- other variables disturbed (or imagined to be disturbed)
            as a threat (victim controls QI' against disturbance,
            or while imagining losing control to threatener)
    
Where variable x is in the threatener's control loop and x' and x" are in
the victim's control loop.

[From Rick Marken (990716.2150)]

Me:

Yes. You can control _any_ aspect of behavior -- actions or
>results -- by brute force. You can only control actions by
>disturbing controlled variables.

Tim Carey (990717.0545)]

OK, thanks. That's a useful distinction. When using brute force
to control another's behaviour, the control of their actions
is pretty easy to understand. When you use the word "results",
however, are you referring to results from the perspective of
the brute force person or the person being brute forced?

from the perspective of the PCT person; the brute force person
and the person being brute forced don't make this distinction.

Think of it this way. I can control the position of your finger in
the rubber band game by disturbing a controlled variable (knot
position). In this case, I am treating your finger position as
an action and controlling it as such. But your finger position
is also a result (of muscle forces). If I grabbed your finger and
tried to control its position by force then I would be treating
finger position as a "result" and trying to control it as such;
if I am stronger than you, I will succeed; if not, we just have
a fight (conflict).

Note that finger position is the same variable in both cases; it's
an action that controls the position of the knot and, at the
same time, it's a result of muscle forces. In one case you control
it as an action by disturbing a variable affected by that action
(knot positoin); in the other you control it (or try to) by acting
directly on the variable (dealing with it, unknowlingly, as any
other physical variable; a result of physical forces -- muscle
forces in this case).

Best

Rick

···

--

Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken/

[From Tim Carey (990717.1630)]

[From Bruce Nevin (990716.1640 EDT)]

This leaves out intimidation by threat of force. I suppose that could be
described as controlling either actions qo' or results qi'=qi by disturbing
some other variables QI'=QI that are controlled by the victim.

I'm still lagging behind Bruce. When you talk about controlling results by
either threat of force as you mention or brute force as Rick mentioned,
whose results are you talking about? Is the forcer controlling their results
by brute force or threat or are you saying that the forcer is controlling
the forcee's results by brute force or threat?

Cheers,

Tim

from [ Marc Abrams (990716.0736) ]

[From Tim Carey (990717.1630)]

I'm still lagging behind Bruce. When you talk about controlling results by
either threat of force as you mention or brute force as Rick mentioned,
whose results are you talking about? Is the forcer controlling their

results

by brute force or threat or are you saying that the forcer is controlling
the forcee's results by brute force or threat?

Tim, I think we are dealing with memory ( imagination ) here. "Threats" in
and of themselves are meaningless unless the person doing the controlling
either "perceives" them as being "real" or sets a "reference level"
according to an imagination/memory.

Marc

[From Bruce Nevin (990717.1506 EDT)]

Tim Carey (990717.1630) --

When you talk about controlling results by
either threat of force as you mention or brute force as Rick mentioned,
whose results are you talking about? Is the forcer controlling their results
by brute force or threat or are you saying that the forcer is controlling
the forcee's results by brute force or threat?

The two parties are in conflict over control of the state of a variable x.
In one case, the forcer is overwhelming the other by brute force. In the
other cases (threat), there is an if-then contingency controlled by both
the forcer and the victim: if the victim does not control x acceptably then
the forcer will control y by brute force. The two parties are in conflict
over control of the state of this variable y, where y = x or y is an
independent variable. If y is not x, then it has to be something that the
victim cares about more than he cares about losing control of x. It could
be more than one variable, including for instance perceptions of pain,
humiliation, etc.

Tim Carey (990717.1630)--

Tim, I think we are dealing with memory ( imagination ) here. "Threats" in
and of themselves are meaningless unless the person doing the controlling
either "perceives" them as being "real" or sets a "reference level"
according to an imagination/memory.

The if-then contingency (I suppose a program-level perception) must involve
memory and/or imagination since both parts of it are hypothetical, they are
not currently happening. (If the victim is currently controlling
acceptably, then the contingency has the form "if you stop controlling as I
wish, then ...".)

How do the two parties come to be controlling the same if-then contingency
perception? It seems to follow pretty directly from the conflict over
control of environment variable x. In general if my control of an
environment variable is disturbed and my present means of control cannot
overcome the disturbance I seem to start controlling if-then perceptions,
"problem-solving" processes. Can't open the door. Is it stuck? Here, the
disturbance is due to another person. She's holding the door. Why?

  Bruce Nevin

from [ Marc Abrams (990717.1939)]

[From Bruce Nevin (990717.1506 EDT)]

Tim Carey (990717.1630) -- >>> Marc Abrams (990716.0736)

When you talk about controlling results by
either threat of force as you mention or brute force as Rick mentioned,
whose results are you talking about? Is the forcer controlling their

results

by brute force or threat or are you saying that the forcer is controlling
the forcee's results by brute force or threat?

The two parties are in conflict over control of the state of a variable x.
In one case, the forcer is overwhelming the other by brute force. In the
other cases (threat), there is an if-then contingency controlled by both
the forcer and the victim: if the victim does not control x acceptably

then

the forcer will control y by brute force. The two parties are in conflict
over control of the state of this variable y, where y = x or y is an
independent variable. If y is not x, then it has to be something that the
victim cares about more than he cares about losing control of x. It could
be more than one variable, including for instance perceptions of pain,
humiliation, etc.

Tim Carey (990717.1630)--

Actually Bruce this was said by me not Tim in my post:

[ Marc Abrams (990716.0736) ] to Tim

Tim, I think we are dealing with memory ( imagination ) here. "Threats"

in

and of themselves are meaningless unless the person doing the controlling
either "perceives" them as being "real" or sets a "reference level"
according to an imagination/memory.

The if-then contingency (I suppose a program-level perception) must

involve

memory and/or imagination since both parts of it are hypothetical, they

are

not currently happening. (If the victim is currently controlling
acceptably, then the contingency has the form "if you stop controlling as

I

wish, then ...".)

This is one of the reasons why modeling is so important. There are a number
of concepts you speak of here that are context sensitive, and a model would
hopefully help clairify. For instance. I may be "acting" one way and
"controlling" any number of other things. Your statement "if you stop
controlling as I wish, then ..." is vague. Are you speaking of someone's
"actions"? If not. how would you know if in fact I _did_ stop controlling
for what you wished.
When you speak of "hypothetical" here you seem to be implying that
"hypothetical" is an imagined theory, rather then a former experience held
in memory. Both can happen, which are you referring to?
And finally, you speak of " ... stop controlling ... " as if we control
_one_ thing at a time. This is where the context comes into play. we are
_always_ controlling many things. What happens to be important to you ( vs.
the person who is actually doing the controlling ) may or may not be of any
concern to the other. Your statement would be more accurate if you said, "if
you stop acting as I wish, then ...". There are more things I can do to
influence your actions, rather then what you are controlling for.

How do the two parties come to be controlling the same if-then contingency
perception?

I'm not sure there is an "if-then" contingency in perceptions. I think the
if-then contingency comes into play in our output function when an error has
been detected and an action choosen to eliminate or reduce the error. The
"if-then contingencies" sounds like plans of action we have learned and have
stored in memory.

It seems to follow pretty directly from the conflict over
control of environment variable x. In general if my control of an
environment variable is disturbed and my present means of control cannot
overcome the disturbance I seem to start controlling if-then perceptions,
"problem-solving" processes. Can't open the door. Is it stuck? Here, the
disturbance is due to another person. She's holding the door. Why?

First, we do not only control "environmental variables" sometimes we control
perceptions that arise out of memory. But for the moment lets stick with
your example. If your current set of "learned" actions ( which include,
plans, sequences, events, etc. ) are not sufficient to correct the error,
you go into reorganization. Reorganization is the process of developing new
control process to deal with the current "problem"

Marc

[From Rick Marken (990717.1930)]

Marc Abrams (990717.1939)

I'm not sure there is an "if-then" contingency in perceptions.

They exist in my perceptions. This is why I am able to do things
like play bridge, which requires that I be able to control
perceptions like "_if_ trup is led, _then_ you must throw trump
(if you have one)".

I think the if-then contingency comes into play in our output
function when an error has been detected and an action choosen
to eliminate or reduce the error.

This is the "output generation" model of behavior that is popular
today in "cognitive psychology" and "modern control theory".
Unfortunately, it won't work in a disturbance prone world. If a
particular "if-then" contingency must be produced, then there
must be reference for the state of that contingency, a way of
sensing the current state of the contingency and a means of acting
to keep the sensed state of the contingency in the reference state.

The "if-then contingencies" sounds like plans of action we have
learned and have stored in memory.

"If-then contingencies" must be stored in memory as _references_
for "if-then contingency" perceptions. The "if-then contingency"
I described above (the "follow suit" rule) is stored as a reference
for what I should perceive when I play bridge (in fact, it is one
of the rules of bridge). When I play bridge, this _reference_ (and
many others) is replyed from memory. When I see (perceive) that
someone has failed to follow suit (a "renig") it is a disturbance
to my perception of this "if-then contingency" and I act to correct
the perception.

First, we do not only control "environmental variables" sometimes
we control perceptions that arise out of memory.

I wouldn't call this control because there are no disturbances
to the perception; you are simply playing a reference signal
back into the perceptual function. No actions are needed to
produce the desired perception or protect it from disturbance.
I would call perceptions that arise from memory "imagined"
rather than "controlled". When I recall (from memory) my finger
pointing to different locations on the screen I am _imagining_
this perception; when I actually point my finger toward different
locations on the screen I am _controlling_ this perception because
I have to actually act on the world (produce forces and muscle
tensions) to produce the desired perception.

Best

Rick

···

---

Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken/

from [ Marc Abrams (990717.0105) ]

[From Rick Marken (990717.1930)]

Marc Abrams (990717.1939)

> I'm not sure there is an "if-then" contingency in perceptions.

They exist in my perceptions. This is why I am able to do things
like play bridge, which requires that I be able to control
perceptions like "_if_ trup is led, _then_ you must throw trump
(if you have one)".

The entire control process can be viewed as an "if-then" contingency plan.
Not just one but _all_ processes that are going on at any moment of time. As
Bill conjectured in Chap 15. Not all of the loops are "controlling" although
they are all involved in the "controlling process".

If your view is that a "perception" is a signal that enters from the
environment through our senses and continues around through the entire
hierarchy and all related loops, then I would say that you are correct. But
i don't think it works like that. We need to distinguish between the
"controlling process", which includes things like imagination, and passive
observation. And the act of control which is _part_ of the "controlling
process" but not the only part. So I will restate my position. "If-Then"
contingencies are part of the "controlling process."

> I think the if-then contingency comes into play in our output
> function when an error has been detected and an action choosen
> to eliminate or reduce the error.

This is the "output generation" model of behavior that is popular
today in "cognitive psychology" and "modern control theory".
Unfortunately, it won't work in a disturbance prone world. If a
particular "if-then" contingency must be produced, then there
must be reference for the state of that contingency, a way of
sensing the current state of the contingency and a means of acting
to keep the sensed state of the contingency in the reference state.

I'm not sure what the latest and greatest is in the world of convential
Psych, But your take on my logic is not mine. As I stated above. If you
view the "if-then" contingency as the _entire_ control process I have no
problem. I thought Bruce was making the contingency a function of the
perceptual signal alone.

> The "if-then contingencies" sounds like plans of action we have
> learned and have stored in memory.

"If-then contingencies" must be stored in memory as _references_
for "if-then contingency" perceptions. The "if-then contingency"
I described above (the "follow suit" rule) is stored as a reference
for what I should perceive when I play bridge (in fact, it is one
of the rules of bridge). When I play bridge, this _reference_ (and
many others) is replyed from memory. When I see (perceive) that
someone has failed to follow suit (a "renig") it is a disturbance
to my perception of this "if-then contingency" and I act to correct
the perception.

> First, we do not only control "environmental variables" sometimes
> we control perceptions that arise out of memory.

I wouldn't call this control because there are no disturbances
to the perception;

I am talking of the "controlling process". Which can and ( I believe does )
contain non-controlling loops ( such as imagination mode, automatic mode,
etc. ) that have a tremendous ( and as of yet an unvalidated ) effect on
both what we are controlling and how ( i.e. is it real or is it memorex ) we
control. We know what a control loop does. But controlling involves ( i.e.
has been postulated to have ) other processes beside the control loop. Very
little of which we currently understand. I think Bruce was looking at the
"controlling process" rather then a series of loops in "control mode".

you are simply playing a reference signal
back into the perceptual function. No actions are needed to
produce the desired perception or protect it from disturbance.
I would call perceptions that arise from memory "imagined"
rather than "controlled".

Be careful here Rick. Perceptions that arise from memory can very easily
become "controlled" at another level or "switch" back into "control mode".
All of this is conjecture of course. Just like your statement above.

When I recall (from memory) my finger
pointing to different locations on the screen I am _imagining_
this perception; when I actually point my finger toward different
locations on the screen I am _controlling_ this perception because
I have to actually act on the world (produce forces and muscle
tensions) to produce the desired perception.

I don't agree with this if what you are saying is that something must be
"physical" in order for "control" to be involved. I can be controlling for
having certain thoughts just as well as I can for being in a certain
physical position(s).

Marc

[From Rick Marken (990718.1000)]

Marc Abrams (990717.0105)

The entire control process can be viewed as an "if-then"
contingency plan.

Yes, it can. And the fact that you can see (perceive) the control
process this way shows that you can perceive if-then contingencies.
If-then contingencies are a perceptual variable; they are a variable
that can be in a least two possible states: true and false. True
means that an if-then contingency is present; false means it's not.

Although you _can_ view the entire control process as an if-then
contingency plan, I don't think this is a good way to view it.
Looking at control as an if-then contingency (as, for example, Miller,
Galanter and Pribrum did in their classic book on "Plans and the
Structure of Behavior" in which they described the control process
as a TOTE: Test, Operate, Test Exit) leads to the notion that
control is a _sequential_ state process; it ifnores the fact that
the control process occurs over time. The TOTE, for example, is
a sequential state control process; the Test determines _if_
there is an error; if there is, _then_ the system acts in the
Operate phase; then there is another Test for error; _if_ there
is still error _then_ the system Operates again; _if_ not, then
the process stops (Exits).

The problem with this sequential view of control is 1) it is
only stable when system gain is very low and 2) it ignores the
fact that the process results in the input tracking the reference
(control of perception).

In a real control system, variables in the loop change simulataneously,
not sequentially. This fact is captured better by viewing the
control process as the simultaneous solution of the following pair
of equations (it would be more appropriate if these were differential
equations but that's getting me out of my league):

     o = ko (r-p)
     p = ki (o+d)

rather than as an if-then process such as

    if (r-p) <> 0 then o = ko (r-p)
        else o = 0

So I will restate my position. "If-Then" contingencies are part
of the "controlling process."

Well, you know more about this than I do. But I would still suggest
that it's better to view control as the simultaneous solution
of a pair of equations rather than as an if-then process.

The point of my previous post, however, was to show that the
occurance of a contingency _is_ a perception -- and is, thus,
controllable. PCT would say that the state of this contingency
is represented as a perceptual signal; if the contingency is
being followed then the perceptual signal is, say, ~100 pulses/second;
if the contingency is not being followed, the perceptual signal is,
say, ~10 pulses/second. The point is: a contingency is a controllable
perception. I think this is important to understand because it
shows how PCT explains some of the more interesting human behavior
we see, such as following rules. A rule is the state of an if-
then contingency variable. According to PCT the state of degree
to which this rule is being followed can be represented as a
perceptual signal. So an individual can set a reference for the
degree to which they want to see the rule followed, compare their
perception of the contingency to the reference and act, as
necessary to keep their perception of the contingency (rule) at the
reference. Complex behavior, like following the rules of bridge,
can then be seen as another example of people controlling perceptual
variables -- in this case, perceptions of if-then contingencies.

Best

Rick

···

--

Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken/

from [ Marc Abrams (990718.0139) ]

[From Rick Marken (990718.1000)]

Marc Abrams (990717.0105)

> The entire control process can be viewed as an "if-then"
> contingency plan.

Yes, it can. And the fact that you can see (perceive) the control
process this way shows that you can perceive if-then contingencies.
If-then contingencies are a perceptual variable; they are a variable
that can be in a least two possible states: true and false. True
means that an if-then contingency is present; false means it's not.

Nice way of twisting my words. :slight_smile: _You_ defined the control process in this
way. In lieu of this could you please explain the following:

[From Rick Marken (931001.0900)]

Perception IS controlled (some of them -- OK Martin?); perception does NOT
control, guide, cause or tell people what to do). We use working models to
show that this phrase (BEHAVIOR IS THE CONTROL _OF_ PERCEPTION) is, indeed,
a correct description of what is going on in purposeful behavior.

Although you _can_ view the entire control process as an if-then
contingency plan, I don't think this is a good way to view it.

I never said it was.

Looking at control as an if-then contingency (as, for example, Miller,
Galanter and Pribrum did in their classic book on "Plans and the
Structure of Behavior" in which they described the control process
as a TOTE: Test, Operate, Test Exit) leads to the notion that
control is a _sequential_ state process;

Do you believe I think it's a sequential process. If not, who are you
addressing?

it ifnores the fact that
the control process occurs over time. The TOTE, for example, is
a sequential state control process; the Test determines _if_
there is an error; if there is, _then_ the system acts in the
Operate phase; then there is another Test for error; _if_ there
is still error _then_ the system Operates again; _if_ not, then
the process stops (Exits).

Again, I'm not sure who you are addressing here. It certainly isn't my post.

The problem with this sequential view of control is 1) it is
only stable when system gain is very low and 2) it ignores the
fact that the process results in the input tracking the reference
(control of perception).

Ok.

In a real control system, variables in the loop change simulataneously,
not sequentially. This fact is captured better by viewing the
control process as the simultaneous solution of the following pair
of equations (it would be more appropriate if these were differential
equations but that's getting me out of my league):

     o = ko (r-p)
     p = ki (o+d)

rather than as an if-then process such as

    if (r-p) <> 0 then o = ko (r-p)
        else o = 0

Do these equations take into account the 3 other modes proposed by Bill? Or
are these equations of the "control mode" only?

> So I will restate my position. "If-Then" contingencies are part
> of the "controlling process."

Well, you know more about this than I do. But I would still suggest
that it's better to view control as the simultaneous solution
of a pair of equations rather than as an if-then process.

Evidently I do. You are talking about and continue to focus on "control
mode". That is, the process of control. But Bill has postulated in Chap 15.
that there are 3 other modes a "control" loop may be in. These other modes
are _not_ control modes. But they are nevertheless involved in information
processing for controlling. If you choose to make believe that there is no
possibility of there existence without testing, that is a problem you need
to deal with. I view control as a simultaneous process. But the "controlling
process" involves ( at least I believe so ) other non-control processes
(loops) like memory. I do not view the "controlling process" as an if-then
process. _You_ postulated that yesterday.

The point of my previous post, however, was to show that the
occurance of a contingency _is_ a perception -- and is, thus,
controllable.

Just in case you missed it the first time. Please clarify the statement made
in your post. In lieu of the statement below. How do you know that a
contingency is occuring?

[From Rick Marken (931001.0900)]

Perception IS controlled (some of them -- OK Martin?); perception does NOT
control, guide, cause or tell people what to do). We use working models to
show that this phrase (BEHAVIOR IS THE CONTROL _OF_ PERCEPTION) is, indeed,
a correct description of what is going on in purposeful behavior

Marc