Perceptions and Conflicts

[R_Marken]

[From Rick Marken (980427.0930)]

It does seem that many conflicts result from the fact that the
parties to the conflict perceive the same situation differently.
We argue, for example, about whether or not some rat shock
experiments done many years ago involved the Test for the
Controlled Variable. The parties to this conflict seem to see
the same situation (the experiments) quite differently. Let's
assume that they do. Is this the reason for the conflict?

Not according to PCT. If there is a conflict it must be because
the parties want equivalent perceptual representations of the same
variable in different states. In the case of the argument over
the Test, PCT should predict that there would be no argument because
the parties are controlling _different_ perceptual representations
of the same variable (the rat shock experiments). Yet there is an
argument. It seems that the PCT prediction fails.

I think the problem with this analysis is that it provides an
incorrect description of what constitutes the "same variable" in
this situation -- the variable controlled by both parties. The
variable controlled by both parties is not really their perception
of _the rat shock experiments_; that variable cannot be controlled
because it cannot vary; the experiments are a fait accompli; nothing
about them can change. The variable that is actually in contention
in this situation is, I think, something more like "how these rat
shock experiments are described". One party has a reference for
hearing these experiments described as being "like The Test"; the
other party has a reference for hearing these experiments described
as being "unlike The Test". The description of the experiments _is_
a variable; it can change and people can try to control it -- bring
it to the preferred state.

So the conflict is over the same variable: description of the rat
shock experiments. Each party is acting (by yelling louder and
louder on the net) to get this variable in the preferred state.
Of course, it doesn't work becuase the harder each party pushes to
get the description he wants, the harder each party pushes back to
keep the description he wants.

The reason _why_ the parties want this variable (description of the
rat experiments) in different states may have to do with the fact
that they are controlling different perceptual aspects of these
experiments. For example, one party may not want to hear the
experiments described as examples of The Test because this person
perceives these experiments as examples of The Test; the other
person may _not_ want to hear them described as examples of The
Test because this person perceives these experiments as not being
examples of The Test.

But this difference in how a situation is perecived creates no
problem per se. The parties could go on happily perceiving the
rat shock experiments quite differently, with no conflict at all.
Conflict occurs only when the parties want to control equivalent
perceptions of the _same_ variable (description of the rat shock
experiments); in particular, it occurs when they control this
variable relative to different reference levels.

Best

Rick

···

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

[Bruce_Gregory]

[From Bruce Gregory (980427.1406 EDT)]

Rick Marken (980427.0930)

I'm going to be sorry.......

So the conflict is over the same variable: description of the rat
shock experiments. Each party is acting (by yelling louder and
louder on the net) to get this variable in the preferred state.
Of course, it doesn't work because the harder each party pushes to
get the description he wants, the harder each party pushes back to
keep the description he wants.

Isn't it stretching things to call "description of the rat shock
experiments" a controlled environmental variable? I don't say this
sarcastically, but simply to point that we are _way_ beyond the rubber band
and JAVA demos.

But this difference in how a situation is perceived creates no
problem per se. The parties could go on happily perceiving the
rat shock experiments quite differently, with no conflict at all.
Conflict occurs only when the parties want to control equivalent
perceptions of the _same_ variable (description of the rat shock
experiments); in particular, it occurs when they control this
variable relative to different reference levels.

My view is that at the heart of most "conflicts" is the sense that the other
is not perceiving you as a human being worthy of respect. I don't present
this as an alternative to HPCT, I think it is perfectly consistent with
HPCT. It would help my understanding considerably if you would take a crack
at casting this intuition into an HPCT framework.

By the way Martin's recent post on conflicts and interference sent me back
to chapter 14 in B:CP. I read it for the third time yesterday and am
re-reading it again. What I find interesting is that the chapter says very
different things now then it did when I first red it. (I can tell by my
earlier highlighting as well as recollection.) I recommend the experience.
The chapter is very worth re-reading and it has been illuminating for me to
see the extent to which what I read is conditioned by what I believe.

Best Offer

[Bruce_Gregory]

[From Bruce Gregory (980427.1657 EDT)]

Rick Marken (980427.1320)

My guess is that some people control with very high gain for a
perception that they call "respect"; a side effect of some
conflicts must be a disturbance to this perception (just as a side
effect of arm wrestling is a disturbance to perceptions of potassium
concentration in muscle cells).

How can we tell if the Catholics in Northern Ireland are controlling for a
perceived lack of respect from the Protestants and vice versa, in which case
there is no conflict, although there is "trouble". Or if both groups are
controlling for a common perception of the status of northern Ireland (I
have a little trouble imagining a variable with "part of the UK" on one end
of a continuum and "part of the Republic" on the other end) in which case
there is conflict? Assuming of course that we can't carry out the Test on
millions of people.

Best Offer

[R_Marken]

[From Rick Marken (980427.1320)]

Bruce Gregory (980427.1406 EDT)--

Isn't it stretching things to call "description of the rat shock
experiments" a controlled environmental variable? I don't say this
sarcastically, but simply to point that we are _way_ beyond the
rubber band and JAVA demos.

I think this is an excellent question!

The rubber band and Java demos show how control works. In these
demos, the controlled variable is easy to perceive and quantify.
We select these easily observed and quantified variables to
illustrate the _principles_ of control. One principle is: control
systems act to bring perceptual _variables_ to predetermined
states and protect those _variables_ from the influence of
disturbance variables.

Once you get this principle, you can see that the variable that is
controlled doesn't have to be the position of a cursor, the size
of a square or the angle of a line (as they are in the rubber band
and Java demos); all they have to be is _something variable_ that can
be perceived and _influenced_ by an organism. The "description of
the rat shock experiments" qualifies as a potential controlled
variable on all three counts; I can perceive it; it is variable
(I can perceive different "descriptions of the experiment") and
I can influence the description (by adding my own comments).

Of course, just because "description of the rat shock experiments"
is a _possible_ controlled variable doesn't mean that it actually
_is_ a controlled variable. Someone would have to Test that. But
I think it's perfectly reasonable to _speculate_ about the variables
that people _might be_ controlling; this is how we determine (prior
to actually doing the experiments) how PCT might explain particular
behavioral phenomena (in this case, an instance of the phenonomen
of conflict). If we confined ourselves to thinking about behavior
only in terms of the variables that we _know_ people control (the
variables controlled in the rubber band and Java demos) then we
would have very little to say about much "real life" behavior.
Skinner would have had the same problem if he had stuck to talking
only about variables he had determined (by experiment) to be
reinforcers. He would have had to explain all of human behavior
(as he tried to do in "Behaviorism", I think) in terms of giving
or withholding Noyes food pellets;-)

My view is that at the heart of most "conflicts" is the sense that
the other is not perceiving you as a human being worthy of respect...
It would help my understanding considerably if you would take a crack
at casting this intuition into an HPCT framework.

Obviously some people in conflict feel like they are not being
perceived as "worthy of respect". Jeff Vancouver reported having
this experience. I've heard others report this same experience.
I have not had this experience myself but I believe people do
have it. I don't believe that this experience is "at the heart
of conflicts"; it's certainly never been at the heart of any of the
conflicts I've been in; and it's not part of the PCT model of
conflict.

My guess is that some people control with very high gain for a
perception that they call "respect"; a side effect of some
conflicts must be a disturbance to this perception (just as a side
effect of arm wrestling is a disturbance to perceptions of potassium
concentration in muscle cells). Thus, a good way to keep "respect"
under control for some people is to avoid the kinds of conflicts that
disturb this variable. I suppose that's why we don't hear from a lot
of people who used to post on the net; disagreement is just too much
of a disturbance to some variables they are controlling.

From my own point of view, conflicts are unpleasant, not because

they are a disturbance to my feeling of being respected but because
they create chronic error (chronic as long as the conflict lasts)
which I experience as stress. Conflicts get me nowhere. That's why
I like science.

When I get into a conflict with someone (like Bill Powers) who
(like me) is controlling for a scientific system concept, we can
work out a mutually acceptable way to resolve the conflict; we
achieve conflict resolution by coming to an agreement about what
we would observe, in a model and/ or an experiment, if either of
us were right. Then we see what happens.

Bill and I have gone through this exercise several times; sometimes
I've been right; sometimes Bill has been right. So the few conflicts
Bill and I have had were not only resolved amicably (what happens is
what happens) they were also learning experiences for _both_ of us.
I think this kind of conflict -- the kind that is solved by modeling
and observation -- is the basis of science; it's "good" conflict;
but it's only good if all parties to the conflict are willing to
subject their preferences regarding what _should be perceived_ to
a test to determine _what is perceived_.

Best

Rick

···

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

[R_Marken]

[From Rick Marken (980427.1440)]

Bruce Gregory (980427.1657 EDT)

How can we tell if the Catholics in Northern Ireland are controlling
for a perceived lack of respect from the Protestants and vice versa,

Ask them. I imagine that some people will say they just can't get no
respect and some (probably a minority) could care less.

in which case there is no conflict, although there is "trouble".

Just because people are controlling for being respected doesn't mean
that there is no conflict. My point was that a _side effect_ of the
actions used to control a conflicted controlled variable may be a
disturbance to a perception of "being respected". When an Israeli
tractor mows down a Palestinian home (because the Israeli is
controlling for living on the exact same piece of land the
Palestinean is controlling for living on -- there is a _conflict_
over the perception of living on _this_ land) a side-effect of
this action may be a disturbance to the Palestinean's feeling of
being respected by the Israeli. Similarly, when an Israeli child
is blown to bits by a Palestinian terrorist, a side-effect of this
action may be a disturbance to the Israeli's feeling of being
respected by the Palestinean. The disturbance to "being respected"
isn't the problem here; the problem is that two groups of people
want the same variable (living on _this_ piece of land) in two
mutually exclusive states.

(I have a little trouble imagining a variable with "part of the UK"
on one end of a continuum and "part of the Republic" on the other
end).

Maybe it's not a continuum. Maybe it's a variable with only two
states; most Catholics want it in one state ("part of Republic");
most Protestants want it in the other ("part of UK"). I think the
peace agreement (as Bill said) is an attempt to show people that
there is a possible third state of this variable ("part of _both_")
in the hopes that people can settle for keeping this variable in
_that_ state.

Best

Rick

···

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

[Bill_Powers1]

[From Bill Power (980428.0300 MDT)]

Bruce Gregory (980427.1657 EDT)--

How can we tell if the Catholics in Northern Ireland are controlling for a
perceived lack of respect from the Protestants and vice versa, in which case
there is no conflict, although there is "trouble". Or if both groups are
controlling for a common perception of the status of northern Ireland (I
have a little trouble imagining a variable with "part of the UK" on one end
of a continuum and "part of the Republic" on the other end) in which case
there is conflict? Assuming of course that we can't carry out the Test on
millions of people.

The way we can tell is by doing the Test on the people involved. If you
don't consider the Test worth carryinbg out on millions of people (due,
perhaps, to the time and cost involved), then you'll just have to say you
don't know why the Catholics and Protestants are fighting. The means is
there; if you don't use it, that doesn't change the fact that the Test is
the way to answer the question.

Best,

Bill P.

[Bruce_Gregory]

[From Bruce Gregory (980428.0925 EDT)]

Rick Marken (980427.1440)]

Just because people are controlling for being respected doesn't mean
that there is no conflict. My point was that a _side effect_ of the
actions used to control a conflicted controlled variable may be a
disturbance to a perception of "being respected". When an Israeli
tractor mows down a Palestinian home (because the Israeli is
controlling for living on the exact same piece of land the
Palestinian is controlling for living on -- there is a _conflict_
over the perception of living on _this_ land) a side-effect of
this action may be a disturbance to the Palestinian's feeling of
being respected by the Israeli. Similarly, when an Israeli child
is blown to bits by a Palestinian terrorist, a side-effect of this
action may be a disturbance to the Israeli's feeling of being
respected by the Palestinian. The disturbance to "being respected"
isn't the problem here; the problem is that two groups of people
want the same variable (living on _this_ piece of land) in two
mutually exclusive states.

What sort of evidence might lead you to question the adequacy of this story?

Best Offer

[Bruce_Gregory]

[From Bruce Gregory (980428.0950 EDT)]

Bill Power (980428.0300 MDT)]

The way we can tell is by doing the Test on the people involved. If you
don't consider the Test worth carrying out on millions of people (due,
perhaps, to the time and cost involved), then you'll just have to say you
don't know why the Catholics and Protestants are fighting. The means is
there; if you don't use it, that doesn't change the fact that the Test is
the way to answer the question.

No problem. How do we deal with people on CSGnet who are certain that they
_know_ the answer to the question without employing the test. So certain
that they know everyone else must be wrong, in fact.

Best Offer

[Bruce_Gregory]

[From Bruce Gregory (980428.1140 EDT)]

Rick Marken (980428.0800)

So, after all that, the answer to your question is the one Bill
[Bill Powers (980428.0300 MDT)] gave to your earlier question about
you could tell whether the Catholics in Northern Ireland are
controlling for a perceived lack of respect:

>The way we can tell is by doing the Test on the people involved.

Is there a name for the aggregate of an individual's controlled perceptions?
When the individual stops controlling a perception (as revealed by the
Test), is it the case that the only necessary change was to lower the gain
associated with the error in this control loop?

Best Offer

[R_Marken]

[From Rick Marken (980428.0800)]

Me:

Just because people are controlling for being respected doesn't mean
that there is no conflict. My point was that a _side effect_ of the
actions used to control a conflicted controlled variable may be a
disturbance to a perception of "being respected"....The disturbance
to "being respected" isn't the problem here; the problem is that two
groups of people want the same variable (living on _this_ piece of
land) in two mutually exclusive states.

Bruce Gregory (980428.0925 EDT)

What sort of evidence might lead you to question the adequacy of
this story?

I don't know what you mean by "adequacy of this story". What I
have proposed is a model of an interaction between an Israeli and
a Palestinean. The way to test the adequacy of this model is with
The Test. That is, the evidence that might lead me to question the
adequacy of this model would come from the results of Testing for
Controlled Variables.

Here is the model (it looks right in Courier 10 or 12 font):

       Israeli Palestinean
    r2 r1 r1' r2'
     > > > >
  -->C-- -- C<-- -->C-- -->C--
> > > > > > > >
p2 o2 o1 p1 p1' o1' p2' o2' system

···

__________________________________________________________
> > > > > > > > environment
v2<---- | <---- v1----> | v2'<---|
^ | ^ ^ | ^
> > > > > >
> ------------- ----------- |
> > > >
-----------(-------------------------- |
            > >
             -------------------------------

p1 and p1' are perceptions of the state of v1', which is the
variable "who lives on land". r1 is "Israeli lives on land";
r1' is "Palestinean lives on land"; v1 is the controlled variable
that is in conflict; both parties are controlling v1 relative to
different references. In order to get v1 to reference state r1
the Israeli generates o1 -- bulldozing homes, expelling people,
etc etc. In order to get v1 to reference state r1' the Palestinean
generates o1' -- throwing stones, terrorism, etc, etc.

Evidence that the Israeli and Palestinian are controlling v1 would
come from The Test. There is already considerable evidence that
both Israeli's and Palestineans will go to considerable lengths
to protect "lives on this land" from disturbance, mainly those
created by the other. (The Israeli's are currently the stronger
party to this conflict so they are actually living on the land
and the Palestineans, by and large, are not). But more Testing
would be needed to determine what variable(s) is actually being
controlled by both parties. There are probably a number of them.

p2 and p2' are perceptions of v2 and v2', respectively, which
can verbally be called "degree to which I am respected by the
other person". The outputs used to control v1 are a _disturbance_
to the v2 and v2' controlled by the opposite party. For example,
the outputs (bulldozing, expelling), o1, used by the Israeli to
keep v1 under control are a disturbance to the state of v2',
the "degree to which the Palestinean is respected by the Israeli".
Each party has means (o2 and o2') for resisting the disturbances to
their perception of being respected -- things they can say and
do that restore, to the extent possible, the feeling of being
respected to its reference state.

Again, the test of this aspect of the model is to Test to determine
whether or not the parties involved in the conflict are controlling
for a variable that could be called "being respected by the other
party".

I have drawn this model as though the "feeling of being respected"
variables (v2 and v2') are not in conflict. They would be in conflict
if the outputs used to control v2 were also a disturbance to v2' and
if the outputs used to control v2' were also a disturbance to v2;
that is, if what the Palestinean did to restore his feeling of being
respected by the Israeli was a disturbance to the Israeli's feeling
of being respected, and vice versa. This aspect of the model could
also be tested using The Test. The tester would have to determine
whether the outputs that restore one party's feeling of being
respected are a disturbance to the feeling of being respected that
is controlled by the other party, and vice versa.

So, after all that, the answer to your question is the one Bill
[Bill Powers (980428.0300 MDT)] gave to your earlier question about
you could tell whether the Catholics in Northern Ireland are
controlling for a perceived lack of respect:

The way we can tell is by doing the Test on the people involved.

Best

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

[Bruce_Gregory]

[From Bruce Gregory 9980428.1403 EDT)]

Rick Marken (980428.1040)

Thanks.

There are several ways to change a control system so that it appears
to have stopped controlling a perception. Lowering the gain is
certainly one way. Another is to set the reference for the perception
to zero. Still another is to change the perceptual function so that
the system controls a different perceptual variable. Another is
to "dismantle" the control system -- removing its connections to
other control systems and to the environment. There are probably
other ways, too.

One more question: I'm not sure how "setting the reference for the
perception to zero" works. What is "zero" in this case? For example, how
would I set the reference for the thermostat in my living room to zero?

Best Offer

[R_Marken]

[From Rick Marken (980428.1040)]

Bruce Gregory (980428.1140 EDT) --

Is there a name for the aggregate of an individual's controlled
perceptions?

I think it's called the "hierarchy of controlled variables" in PCT.

When the individual stops controlling a perception (as revealed by
the Test), is it the case that the only necessary change was to
lower the gain associated with the error in this control loop?

There are several ways to change a control system so that it appears
to have stopped controlling a perception. Lowering the gain is
certainly one way. Another is to set the reference for the perception
to zero. Still another is to change the perceptual function so that
the system controls a different perceptual variable. Another is
to "dismantle" the control system -- removing its connections to
other control systems and to the environment. There are probably
other ways, too.

Of course, all this changing would have to be done by something
outside of the control system itself, such as another control system.

Best

Rick

···

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

[R_Marken]

[From Rick Marken (980428.1230)]

Bruce Gregory 9980428.1403 EDT) --

One more question: I'm not sure how "setting the reference for the
perception to zero" works. What is "zero" in this case?

If the reference signal is a neural current (impulses/sec) then
a zero reference signal is 0 impulses/sec. This reference signal
specifies that the perceptual signal be 0 impulses/sec. Note that
zero perceptual signal does not necesarily correspond to zero
amount of the controlled variable (whatever that might mean). It
all depends on how the perceptual function converts values of
controlled variable into values (impulses/sec) of perceptual signal.

Setting the reference signal to zero only appears to stop control
in control systems and it does this only in systems where the
perceptual signal cannot go negative. A zero reference really
only _appears_ to stop control and it appears to do this _only_ when
the controlled variable is in states that produce zero perceptual
signal values. Of course, control is not really stopped when the
reference signal is set to zero; the control system just does nothing
when the controlled variable is in states that produce a perceptual
signal of zero. When the controlled variable changes so that the
perceptual signal goes above zero, the control systems swings into
action.

If the reference signal were _above_ (rather than at) zero the
control system would act on the controlled variable, even of it
were in states that produces perceptual values of zero, to bring
the perceptual signal _up_ to the reference level.

Here's a possible example of controlling with a reference level
of zero. Suppose you decide to stop controlling for being with
that pretty girl (let's call her Carmen) who used to put out so
nicely for you -- and all your friends;-) So you set the reference
for the perceptual signal that represents the sight of Carmen
to zero. Now as long as Carmen is nowhere to be seen it looks like
you are no longer controlling for seeing Carmen; the perceptual
signal is at zero, just where you want it. But one day the Carmen
perceptual signal goes above zero because someone similar in
appearance to Carmen steps into view. So you turn and duck into
the crowd to avoid seeing Carmen (actually, to bring the perceptual
signal back to zero, regardless of whether it was Carmen or her
doppleganger that brought it above zero), bringing the Carmen
perceptual signal back to zero.

Best

Rick

···

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

[Bill_Powers1]

[From Bill Powers (980428.1506 MDT)]

Bruce Gregory (980428.0950 EDT)--

How do we deal with people on CSGnet who are certain that they
_know_ the answer to the question without employing the test. So certain
that they know everyone else must be wrong, in fact.

Well, what do you mean by "dealing" with them? Shutting them up? Making
them admit they're wrong? You can try to control them in some way, but I
would predict from PCT that this will not get you what you want.

Maybe what you need to deal with is your own reaction to such people. If
you didn't want them to change their behavior, you wouldn't have to do
anything about them, would you?

Best,

Bill P.

[Bill_Powers1]

[From Bill Powers (980428.1513 MDT)]

Bruce Gregory (980428.1140 EDT)--

Is there a name for the aggregate of an individual's controlled perceptions?

Not that I know of, other than the phrase you just used.

When the individual stops controlling a perception (as revealed by the
Test), is it the case that the only necessary change was to lower the gain
associated with the error in this control loop?

That's one way. Another way is to dismantle the input, comparator, or
output functions through reorganization, or to change them so they are
controlling something else, or by a different means. If the only change is
to lower the gain, then reinstating tbe control system requires only
raising the reference level again. But if the components of the system have
been altered, the system is now a different system; to restore the orginal
control system would require relearning it from scratch, and it would
probably never come out exactly the same as it was before.

Best,

Bill P.

[Fred_Nickols3]

From Fred Nickols (980428.1700 EDT)] --

Bruce Gregory (980427.1406 EDT)] replying to Rick Marken (980427.0930)

Rick:

It does seem that many conflicts result from the fact that the
parties to the conflict perceive the same situation differently.
We argue, for example, about whether or not some rat shock
experiments done many years ago involved the Test for the
Controlled Variable.

<snip>

The variable controlled by both parties is not really their perception
of _the rat shock experiments_; that variable cannot be controlled
because it cannot vary; the experiments are a fait accompli; nothing
about them can change. The variable that is actually in contention
in this situation is, I think, something more like "how these rat
shock experiments are described".

Bruce G:

Isn't it stretching things to call "description of the rat shock
experiments" a controlled environmental variable? I don't say this
sarcastically, but simply to point that we are _way_ beyond the rubber band
and JAVA demos.

Two comments from one of the least informed list members (me):

  1) I'm not sure what Bruce G is driving at. Is it that something
     like a "description" can't be treated as a controlled variable?
     If so, why not? We control for things like pronunciation and
     enunciation (in relation to a reference signal that might be
     stated as "speaking properly"), so why not a description, and
     whether it is a generally accepted one or one that is being
     contended?

  2) The slant I place on Rick's notion that the controlled variable
     is "how these rat shock experiments are described" is apparently
     different from the one taken by Bruce G. For me, Rick's notion
     doesn't tie so much to the description per se but, rather, to the
interpretations placed on it.

Thus, from where I sit, the variable that's being controlled in such
discussions is the "purity" of PCT. A PCT "purist" (e.g., Bill P or
Rick M -- and I do not mean "purist" as a pejorative term) cannot let
stand unchallenged any descriptions of rat shock experiments that are
or appear to be attempts to show how the underlying theory relates to
PCT unless such descriptions are couched in proper PCT terms (which is
probably impossible unless PCT was the underlying theory which is very
unlikely).

I am not quarreling here with purists or the attempt to maintain purity;
I'm simply offering a different interpretation of the arguments and an
alternative as the controlled variable, namely, the purity of PCT
discussions.

Regards,

Fred Nickols
The Distance Consulting Company
nickols@worldnet.att.net
http://home.att.net/~nickols/distance.htm

        "The Internet offers the best graduate-level education
         to be found anywhere."

[Bill_Powers1]

[From Bill Powers (980428.1356 MDT)]

Bruce Gregory 9980428.1403 EDT)--

I'm not sure how "setting the reference for the
perception to zero" works. What is "zero" in this case? For example, how
would I set the reference for the thermostat in my living room to zero?

Set it to the lowest possible temperature. If that temperature is lower
than the lowest temperature the room air will ever attain, the themostat is
effectively turned off: the furnace will never turn on.

Best,

Bill P.

[Bruce_Gregory]

[From Bruce Gregory (980428.2120 EDT)]

Bill Powers (980428.1506 MDT)]

Maybe what you need to deal with is your own reaction to such people. If
you didn't want them to change their behavior, you wouldn't have to do
anything about them, would you?

Hear that Rick? I think Bill's talking to you

Best Offer

[R_Marken]

[From Rick Marken (980428.1910)]

Bruce Gregory (980428.0950 EDT)--

How do we deal with people on CSGnet who are certain that they
_know_ the answer to the question without employing the test. So
certain that they know everyone else must be wrong, in fact.

Bill Powers (980428.1506 MDT) --

Maybe what you need to deal with is your own reaction to such
people. If you didn't want them to change their behavior, you
wouldn't have to do anything about them, would you?

Bruce Gregory (980428.2120 EDT) --

Hear that Rick? I think Bill's talking to you

Maybe. But that would be rather odd since I never expressed
any interest in suggestions about how to "deal with" anyone.
I already know how to deal with people who are "certain"
without testing (observing); I ignore _most_ of what they say.
"Most" because sometimes these people have good ideas for
models that can be tested against observation. What I don't
want (or what I try not to want) is for these people to change
their behavior. When I do start wanting a change in their
behavior (and sometimes I do) I have fallen into the trap of
inter-personal control: conflict.

Best

Rick

···

--

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

[Bill_Powers1]

[From Bill Powers (980429.0300 MDT)]

Bruce Gregory (980428.2120 EDT)--

Maybe what you need to deal with is your own reaction to such people. If
you didn't want them to change their behavior, you wouldn't have to do
anything about them, would you?

Hear that Rick? I think Bill's talking to you

Actually, I thought I was answering your question about what to do about
"such people." Do you want to fix your problem by making Rick behave
differently?

Best,

Bill P.