Control of Perception

[From Rick Marken (2007.04.02.0920)]

A small complaint. In MSOB, p. 25, after an excellent discussion of the nature of perception from a PCT perspective, comes the following:

Half of the jokes in the world are about one person assuming that everyone else sees the world the same way. The husband hangs his new moose head over the fireplace in the living room and says to his aghast wife, “There, doesn’t that look great?”. The other half of the jokes are about people who are unable to see what everyone else (supposedly) can see – that the wife looking at the moose head is trying to remember the phone number of her lawyer.

My complaint is this. Although it is never stated explicitly, this passage leaves me (anyway) with the impression that disagreements result when different people perceiving the same things differently.

I can see two possible implications people might pick up from what is being said here. One implication is that the wife is calling her lawyer because the moose head that looks great to her husband looks terrible to her; the husband and wife have different perceptions of the quality of the moose head. I think someone new to PCT could pick up this implication and conclude that PCT views disagreements as a result of people seeing the quality of same thing (the moose head) differently. I think this is misleading because, from a PCT point of view, perceptions are just perceptions. The quality of a perception (like the perception of a moose head) is determined by the reference specification for that perception; a high quality perception is one that matches the reference and a low quality perception is one that deviates from the reference. Assuming that the wife and husband both perceive a moose head, the reason why the wife is calling the lawyer about it is not because she perceives the moose head as low quality but, rather, because she has a different reference for “things over fireplace” than does her husband.

A second possible implication is that the wife and husband are actually perceiving two different things; the husband is perceiving a moose head and the wife is perceiving some kind of monstrous rat, say. Again, however, the implication is that it is the difference in perception that is cause of the disagreement. But this, like the implication that it is a difference in the perception of quality that is the reason for the disagreement, ignores the fact that it is references – not the perceptions themselves – that determine whether a perception is disliked or liked. Even if the wife perceives as a monstrous rat what the husband perceives as a moose head, there will be no disagreement unless the rat perception is not wanted by the wife while the moose head perception is wanted by the husband.

The way the situation is described in the story quoted above, one could walk away (I think) with a cause-effect notion of how disagreement works. It sounds like different perceptions of the same things cause different actions; the husband’s perception of the thing over the fireplace as a nice moose head causes his joy while the wife’s perception of the same thing as either an ugly moose head or some other awful thing causes her to call the lawyer. In fact, it’s not the perceptions that cause the different behaviors of wife and husband. Rather, it’s the perceptions relative to each of their references for the state of those perceptions that lead to actions aimed at bringing those perceptions to their reference levels.

I am working on a new approach to explaining the role of perception in PCT. Such an explanation will include a discussion (as in the first part of the chapter on perception in MSOB) of the epistemology of perception; that all we know is perception. But I will emphasize the fact that what is important about seeing control in terms of control of perception (rather than control of environmental events) is that there are different ways to perceive the same environmental circumstance. So you can control different perceptual aspects of what is in fact the same physical situation. I will demonstrate this using my “Control of Perception” demo (
http://www.mindreadings.com/ControlDemo/ControlP.html), which is a clunkier (though earlier) version of the demo that Bill Powers showed at the Conference in China last year. In my demo, you can control any of three aspects of a square with a line through it. In Bill’s you can control any of three aspects of a rotating spheroid.

What I will emphasize about control of perception is that we often have many different perceptions of the same situation to pick from to try to control. Controlling some of these perceptions may prove more effective at getting us what we want (at a higher level of control) than controlling others. For example, in playing tennis, controlling for a certain wrist configuration perception may be more effective as a means of hitting a consistent forehand than is controlling of other perceptions of one’s connection to the racquet. In driving, controlling certain perceptions of the relationship between the car and the road may prove more effective (in terms of avoiding collisions) than others.

I’m still working on this idea. If anyone has any suggestions – such as examples of where control of several different perceptions of the same situation can produce the same higher level results, though not all as effectively – they would be greatly appreciated.

Best

Rick

···


Richard S. Marken
rsmarken@gmail.com

marken@mindreadings.com

A small complaint. In MSOB, p.
25, after an excellent discussion of the nature of perception from a PCT
perspective, comes the following:

[From Bill Powers (2007.04.02.1035 MDT)]

Rick Marken (2007.04.02.0920)

Half of the jokes in the world are about one person assuming that
everyone else sees the world the same way. The husband hangs his new
moose head over the fireplace in the living room and says to his aghast
wife, “There, doesn’t that look great?”. The other half of the
jokes are about people who are unable to see what everyone else
(supposedly) can see – that the wife looking at the moose head is trying
to remember the phone number of her lawyer.

My complaint is this. Although it is never stated explicitly, this
passage leaves me (anyway) with the impression that disagreements result
when different people perceiving the same things differently.

I can see that this little argument is going to go on for a while. I
think it’s largely a matter of oversimplification. Yes, if people have
different reference levels for almost the same perceptions of the same
thing, conflict will likely result except for certain pairings of
reference levels. But it is not possible to prevent conflict just by
having people agree on the same reference level for a perception that is
truly the same in both of them. That’s because higher systems are also
involved, and in fact are the true cause of the conflict.

The conflict is not caused by liking or disliking of the moose head; it
is caused (up a level) by her desire to perceive a neat living room and
his desire to display a status symbol. The moosehead disturbs her
perception of a neat living room because it’s hairy and smelly, not
because it’s a moose head. And the moosehead satisfies the husband’s
status yearnings not because it’s an example of (latin name that I don’t
know), but because it shows that he is (barely) clever enough and
stealthy enough and well enough armed to kill a moose. He and his wife
are controlling for different perceptions, and to do so they have to put
the physical moosehead into two different states, present and
absent.

“Present” and “absent” are two different reference
levels for a single variable. So at that level, the level where we
observe the conflict being played out, it is as you say: different
intended reference levels for the same variable. But if we try to resolve
the conflict, we find we have to go to the next level up, to see why
these two people are trying to do this impossible thing together. And now
we find out that the real reason has nothing to do with moose-heads per
se. It has to do with different perceptions in which the moosehead plays
different roles. To the husband it’s a desirable boost to his status
among hunters; to his wife, it’s a yucky piece of junk in her living
room. That is why she sets her reference level for the moosehead to
“absent” and he sets his to “present.” If they didn’t
perceive different effects of the same moosehead’s presence, there would
be no problem.

What I showed with the ThreeSys demo is that the only way to avoid
conflict is for the two parties to perceive the same thing in ways that
are sufficiently different to make the perceptions independent – that
is, either perception can be maintained at any reference level without
interfering with control of the other at any other reference level. If
the two people are perceiving the same thing the same way, then there is
no way to avoid conflict, even by making the reference levels the same.
There might be a moment in which the actions of the two systems are not
opposed to each other, but the slightest disturbance of either system at
the next level up will require a different setting for the common
perception, and that will immediately bring back the conflict. So
conflicts are not cured by bringing different reference levels for the
same perception to the same values. They are cured by having the two
sides control independent perceptual variables.

I will temper that by reminding myself that there is true conflict only
if one or both controllers reaches a limit. So a second way to avoid a
disabling degree of conflict is to lower the loop gain enough that the
error experienced by both sides produces enough less output that the
extra effort can be tolerated.

A second possible
implication is that the wife and husband are actually perceiving two
different things; the husband is perceiving a moose head and the wife is
perceiving some kind of monstrous rat, say. Again, however, the
implication is that it is the difference in perception that is cause of
the disagreement.

This is certainly possible; the moosehead reminds the wife of a monstrous
rat, and that would explain the conflict very well, since she doesn’t
like monstrous rats in her house. The husband, who just sees a trophy (it
could be any animal), doesn’t know what she is perceiving. But this again
puts the cause of the conflict at the next level up, not at the level
where they disagree about whether the moosehead itself should be where it
is. Neither of them has any basic interest in the moosehead; they’re just
using it for controlling different (and incompatible)
perceptions.

The way the
situation is described in the story quoted above, one could walk away (I
think) with a cause-effect notion of how disagreement works. It sounds
like different perceptions of the same things cause different actions;
the husband’s perception of the thing over the fireplace as a nice moose
head causes his joy while the wife’s perception of the same thing as
either an ugly moose head or some other awful thing causes her to call
the lawyer. In fact, it’s not the perceptions that cause the different
behaviors of wife and husband. Rather, it’s the perceptions relative to
each of their references for the state of those perceptions that lead to
actions aimed at bringing those perceptions to their reference
levels.

But that’s not the right story either. To unravel this you have to
think of more than one level of control at a time. The conflict is
expressed or manifested in terms of sending two contradictory or
partially-cancelling reference signals to the same lower system, which
results in a virtual reference level that the lower system achieves
without difficulty. However, the perceptual signal at the lower level
which matches the virtual reference level is not what either system at
the next level up requires, so both of those systems contain errors.
Whether this results in a disabling conflict depends mostly on the loop
gains in the upper-level systems, and on whether other contributing
lower-level perceptions can be adjusted to compensate. And it is not the
system where the conflict is expressed that is disabled, but the two
higher systems.

I refer to this in MSOB, though I could now express it more
clearly.

I am working on a
new approach to explaining the role of perception in PCT. Such an
explanation will include a discussion (as in the first part of the
chapter on perception in MSOB) of the epistemology of perception; that
all we know is perception. But I will emphasize the fact that what is
important about seeing control in terms of control of perception (rather
than control of environmental events) is that there are different ways to
perceive the same environmental circumstance. So you can control
different perceptual aspects of what is in fact the same physical
situation.

This is getting closer to my view. However, it still doesn’t quite
capture the multi-level aspect of conflict. “Different ways of
perceiving” refers to a higher level of perception, “the same
environmental circumstance” refers to the lower-level view where
there is only one perception. The conflict is disabling only at the
higher level; the lower-level system simply controls relative to a single
virtual reference level, and is not disabled.

I will
demonstrate this using my “Control of Perception” demo (

http://www.mindreadings.com/ControlDemo/ControlP.html
), which is a
clunkier (though earlier) version of the demo that Bill Powers showed at
the Conference in China last year. In my demo, you can control any of
three aspects of a square with a line through it. In Bill’s you can
control any of three aspects of a rotating spheroid.

I really like your version a lot. Both are versions of the old Turbo
Pascal program for controlling three cursors with one mouse that I showed
at Haimowoods. It’s part of Demo 1.

In both of our latest versions, it is impossible to control more than one
aspect of the display at a time. There are only a few brief moments when
the action needed to correct the error in one aspect of the display will
also correct the error in another aspect. This is actually a three-way
conflict, caused by trying to use the same mouse-controlling action to
control three different perceptions at the next level up. Even without
the disturbances, there would be only one reference level for each aspect
that could be achieved simultaneously, and disturbing any one of the
aspects would lead to immediate conflict.

Best,

Bill P.

[From Rick Marken (2007.04.02.1230)]

Bill Powers (2007.04.02.1035 MDT)--

>Rick Marken (2007.04.02.0920)

A small complaint. In MSOB, p. 25, after an excellent discussion of

the nature of >>perception from a PCT perspective, comes the
following:

>Half of the jokes in the world are about one person assuming

that everyone else sees the world the same way....

>My complaint is this. Although it is never stated explicitly, this passage leaves

me (anyway) with the impression that disagreements result when different
people perceiving the same things differently.

I can see that this little argument is going to go on for a while. I think it's largely
a matter of oversimplification.

I think it's just a matter of communication. I completely agree with
what you say about conflict. It's just that, in the paragraph I quote,
what you say implies that it is the difference in the way the same
situation is perceived that is the problem.

The conflict is not caused by liking or disliking of the moose head; it is caused
(up a level) by her desire to perceive a neat living room and his desire to display
a status symbol.

Right. But this is not clear in the paragraph I quote. In that
paragraph you are talking about how people perceive the world
differently. The example you give is of a moose head on the mantle.
The implication is that the wife perceives the moose head differently
than than does the husband and that that is the basis of their
differing actions-- joy for the husband, divorce for the wife.

The moosehead disturbs her perception of a neat living room because it's hairy
and smelly, not because it's a moose head.

Right. And even her perception of the neatness of the living room is
not the problem; it's the neatness relative to her reference for
neatness. The reference for a particular level of neatness is what
makes the moose head a disturbance, not the moose head per se.

He and his wife are controlling for different perceptions, and to do so they
have to put the physical moose head into two different states, present and absent.

Right. It's not the different perceptions (neatness and status) that
are the cause of the conflict; it's that the wife and husband are
_controlling_ for these different perceptions that causes them to set
incompatible references for the state of the moose head. What they are
actually perceiving (in terms of neatness or status) is only half the
problem; the other half is that they are controlling for these
perceptions, trying to bring them to reference states that imply
incompatible states of the moose head (present vs absent). I agree
with all this, of course.

All I am saying is that, in the quoted paragraph from MSOB, what you
say can be taken (and, I think, often is taken) to mean that conflict
results from people perceiving the same situation in different ways. I
know that you know that that's not how it works. I'm just saying that
your choice of ways to point out the fact that people probably often
_do_ perceive the same situations differently can lead to this kind of
misinterpretation (I'm seeing it in some of my student papers).

I think it is important for people to understand that people can and
do perceive the same situation differently. What I want to do is find
a way to show why this fact is important, without leading people to
come away with the notion that it implies what is basically a
cause-effect explanation of conflict.

So conflicts are not cured by bringing different reference levels for the
same perception to the same values. They are cured by having the two sides
control independent perceptual variables.

No argument there. In fact. I have been trying to make precisely this point

In fact, it's not the perceptions that cause the different behaviors of wife
and husband. Rather, it's the perceptions relative to each of their

references

for the state of those perceptions that lead to actions aimed at bringing those

perceptions to their reference levels.

But that's not the right story either. To unravel this you have to think of more
than one level of control at a time.

Completely agree. All I'm saying is that the way you describe the
conflict between the wife and husband in terms of their perception of
the moose head could suggest that it is the different ways of
perceiving the moose head that is the cause of the conflict. I agree
that the conflict results from the fact that the wife and husband are
controlling different higher level perceptions and that successful
control of these perceptions requires incompatible states for the
moose head. But this was not clear (to me) in the context of your
example in MSOB. In that example you are talking about the fact that
people can perceive the same situation differently. Then you talk
about a wife and husband in conflict over a moose head, implying that
they perceive the moose head differently. You may have been thinking
that the example illustrates the fact that the wife and husband are
perceiving (and controlling for) two different things: status and
neatness. And that it's that difference in perception that is the
cause of the conflict (which it is). All I'm saying is that that fact
is not communicated (to me anyway) by your example. The example
suggests that the husband and wife perceive the same thing -- the
moose head -- differently and that this is why they are in conflict.

I refer to this in MSOB, though I could now express it more clearly.

I know; you do a great job of explaining conflict in the conflict
chapter. The except I am talking about was in the beginning of the
book, where you are explaining the PCT view of perception and why it
makes a difference. I think the implication of that section is that
perception makes a difference because differences in the way people
perceive the world lead to conflict. I know that you know that this is
not the case; you are very clear (when you talk about conflict) that
conflict actually results when people _control_ for different higher
level perceptions that imply inconsistent states for lower order
perceptions. I just think that what you said about people perceiving
things differently (in this part of MSOB) can be confusing.

I will certainly teach my students that different people's perceptions
of the same situation can be quite different. I'm just going to
develop the importance of this fact -- the fact that we control
perceptions rather than physical variables -- in terms of the fact
that we have _options_. There are many different perceptions that we
can control when we try to accomplish higher level goals (like
catching balls or doing rituals) and controlling some lower level
perceptions may produce better higher level results than others. I
think this is what a lot of teaching/coaching is about; the
teacher/coach tries to point the student to the perceptions that, when
controlled, produce the best results.

> I will demonstrate this using my "Control of Perception" demo (

http://www.mindreadings.com/ControlDemo/ControlP.html)…

I really like your version a lot. Both are versions of the old Turbo Pascal program for controlling three cursors with one mouse that I showed at Haimowoods. It's part of Demo 1.

Yes! You hold the patent;-)

Best regards

Rick

···

--
Richard S. Marken
rsmarken@gmail.com
marken@mindreadings.com

I will certainly teach my
students that different people’s perceptions

of the same situation can be quite different. I’m just going to

develop the importance of this fact – the fact that we control

perceptions rather than physical variables – in terms of the fact

that we have options. There are many different perceptions that we

can control when we try to accomplish higher level goals (like

catching balls or doing rituals) and controlling some lower level

perceptions may produce better higher level results than others. I

think this is what a lot of teaching/coaching is about; the

teacher/coach tries to point the student to the perceptions that,
when

controlled, produce the best results.
[From Bill Powers (2007.04.02.1435 MDT)]

Rick Marken (2007.04.02.1230)–

Fine. I thought we’d eventually converge, never having been very far
apart.

Best,

Bill P.

[From Bjorn Simonsen (2007.04.14,11:05 EUST)]
from Rick Marken (2007.04.02.0920)

I'm still working on this idea. If anyone has any suggestions --
such as examples of where control of several different perceptions
of the same situation can produce the same higher level results,
though not all as effectively -- they would be greatly appreciated.

Since you are still working on the idea, I'll give you two suggestions. You
know them better than me, but all the same.
1. Give three children a dollar note and ask them to make use of it to cause
a joyful feeling the next ten minutes.

One will maybe by an ice cream, the other will maybe by a cartoon and the
third will save the dollar. They will control different perceptions in the
same situation to produce the same higher level results.

If you give the same children a dollar note and ask them to buy an ice
cream, maybe two of them will experience a conflict.

2. Give three female students a ten dollar note each and ask them to spend
the note to obtain a feeling well emotion. Also they will control different
perceptions in the same situation to produce the same higher level results.

If you give the same students a ten dollar note and ask them to buy a
lipstick and use it to feel well, maybe all of them will experience a
conflict.

bjorn

PS. Did you receive the CD I posted you?

I’m still working on this idea.
If anyone has any suggestions –

such as examples of where control of several different
perceptions

of the same situation can produce the same higher level results,

though not all as effectively – they would be greatly
appreciated.
[From Bill Powers (2007.04.14.0730 MDT)]

Rick Marken (2007.04.02.0920)

If you’re referring to conflict, I can help out with a real example. At
the first MOL workshop in Vancouver put on by Tim Carey and me, we had a
volunteer who worked out a conflict and kindly gave permission to use his
session for teaching. I probably have it a bit garbled but it will come
out all right.

The conflict was: sell my house and don’t sell my house. He was moving
elsewhere, but his sister who needed a place to live would have been
helped by using the house. On the other hand, the market was uncertain
and he thought that selling the house before the market went down would
be financially prudent. From one standpoint, selling the house was part
of a perception of financial prudence; from the other standpoint, keeping
the house contributed to a perception of family loyalty. And of course,
at still a higher level, both famility loyalty and financial prudence
contributed toward his sense of being a good person. So it ended up that
in order to be a good person, he had to both sell and not sell the
house.

He resolved the conflict after going back and forth (with a little
pushing) between loyalty and prudence. The critical moment came when he
was saying “On the one hand, I want to take care of my sister”
(gesturing with left hand), “and on the other hand, I want to
realize the value of the house” (gesturing with right hand). He
looked at his hands, and suddenly said “I’m weighing my sister
against money. I don’t want to do that.” And that was the moment of
resolution.

Others who were there can correct my errors, but I think that’s the
structure of the conflict as we finally understood it.

So the same perception, selling the house, was seen as part of two
different perceptions at the next level up: caring for his sister and
being financially prudent. And both of those were perceived as examples
of the same thing at a still higher level: being a good person. The
result was that the net reference level for the selling perception was
set neither at “sell” nor at “don’t sell” but
somewhere in between, or wavering in one direction and then the other,
which did not serve the purpose either of taking care of his sister or
being financially prudent.

A principle that may help with this and other problems involving levels
of perception is that the different levels do not perceive the same
variables. A higher-level variable is not simply a higher-level version
of a lower-level variable: it is something entirely different, a
different category of experience. At one level, selling the house is
someone one can perceive, and it exists on a scale from 0 (not
selling) to 1 (selling). It never changes into a different kind of
perception. For this particular individual, at the next level up (or some
higher level) the perceptions do not concern selling the house any more.
They aren’t even the same kind of perception as selling. They have to do
with a relationship with a sister and a relationship with the financial
world (I don’t mean to pick especially on the “relationship
level” – maybe these are principles). And for this individual,
neither loyalty nor prudence have anything in common with personal
goodness, except that they contributed (along with a lot of other
lower-order perceptions) to the sense of how good one is, at a still
higher level.

I would guess that this person ended up reorganizing at the third
relative level in this conflict. Weighing his sister against money
violated his sense of being a good person, so he stopped perceiving
financial prudence as being independent of caring for his sister, or
something like that. The new perception (that he was weighing his sister
against money) made him feel as if he were not a good person. So first
the conflict shifted to a higher level, and then he quickly reorganized
at that level to remove the conflict. He never did explain exactly how
this resolved it, but he said that the problem was solved as far as he
was concerned.

Best,

Bill P.

[From Bjorn Simonsen (2007.04.14,17:40 EUST)]
from Bill Powers (2007.04.14.0730 MDT)

The conflict was: sell my house and don't sell my house.

Tim Carey also described the example in his "The Method of Levels" chapter
one. I think he describes in convincing way what happens when a person
starts controlling at a level above. The conflict crumbles away and it is
unintelligible that the same person experienced a conflict half an hour
earlier.

I look forward to hear more from Tim Carey in November.

bjorn

[From Rick Marken (2007.04.14.1220)]

Bjorn Simonsen (2007.04.14,11:05 EUST)--
From Rick Marken (2007.04.02.0920)

>I'm still working on this idea. If anyone has any suggestions --
>such as examples of where control of several different perceptions
>of the same situation can produce the same higher level results,
>though not all as effectively -- they would be greatly appreciated.

Since you are still working on the idea, I'll give you two suggestions. You
know them better than me, but all the same.
1. Give three children a dollar note and ask them to make use of it to cause
a joyful feeling the next ten minutes.

Thanks. But this sort of the opposite of what I'm looking for. I want
examples where controlling any one of several different perceptions
could be the means of producing the same higher order result; but
controlling one of these perceptions works better than controlling
others. For example, I think that when shooting a rifle there are
several options regarding the visual variable that can be controlled
when aiming at a moving target. I'd like to know what those
possibilities are and be able to explain why controlling one
perception might work out better (in terms of producing the same
result: hitting the target) then would controlling others.

PS. Did you receive the CD I posted you?

I did, thanks.

Best

Rick

···

--
Richard S. Marken
rsmarken@gmail.com
marken@mindreadings.com

Thanks. But this sort of the
opposite of what I’m looking for. I want

examples where controlling any one of several different perceptions

could be the means of producing the same higher order result; but

controlling one of these perceptions works better than controlling

others.
[From Bill Power (2007.04.14.1409 MDT)]

Rick Marken (2007.04.14.1220)]

OK, how about throwing a ball at a target using your left hand or your
right hand?

Or how about teaching a five-year-old to draw a horse by saying

(a) That’s a terrible horse, it doesn’t even have four legs.

(b) Good start, now try drawing one with four legs and let’s see how that
looks.

Or how about curing your headache by

(a) Taking two pain pills.

(b) Rubbing a crystal and chanting “pain gone, pain
gone…”

[From Rick Marken (2007.04.14.1330)]

Bill Powers (2007.04.14.1409 MDT)--

OK, how about throwing a ball at a target using your left hand or your
right hand?

Yes! Perfect. Of course. The throwing one especially. Why can't I
think of these things?

More. More.

Best

Rick

[From Bjorn Simonsen (2007.04.14,22:50 EUST)]
from Rick Marken (2007.04.14.1330)

More. More.

1. A 5000 meter runner may control for the two last lap spurt or a last lap
spurt.

2. Read the book; Alexandre Dumas: The three Musketeers" or look at the Walt
Disney picture with the same name.

3. Think things yourself or ask Bill for more examples.

bjorn

[From Rick Marken (2007.04.14.1410)]

Bjorn Simonsen (2007.04.14,22:50 EUST)]
>Rick Marken (2007.04.14.1330)
>More. More.
1. A 5000 meter runner may control for the two last lap spurt or a last lap
spurt.

Great!

2. Read the book; Alexandre Dumas: The three Musketeers" or look at the Walt
Disney picture with the same name.

Aw, come on. Give me an example. I rarely read books (obviously;-))

3. Think things yourself or ask Bill for more examples.

OK, I though things out for myself and came up with the idea of asking
Bill for some more examples;-)

Thanks!

Rick

···

--
Richard S. Marken
rsmarken@gmail.com
marken@mindreadings.com

[From Rick Marken (2007.04.14.1500)]

Rick Marken (2007.04.14.1410)--

> Bjorn Simonsen (2007.04.14,22:50 EUST)--
> >Rick Marken (2007.04.14.1330)
> >More. More.
> 1. A 5000 meter runner may control for the two last lap spurt or a last lap
> spurt.

Great!

Actually, I hate to be a pain (really!) but what I would like is an
example where different aspects of the same situation can be perceived
in different ways to produce the same result. For example, in a
tracking task one could control the different in position of target
and cursor or the difference in the distance of the target and cursor
from the left edge of the screen (controlling what are basically two
imaginary bars that start at the left edge of the screen and end at
the current positions of target and cursor, respectively). That may
seems a bit far fetched, perhaps (and probably wouldn't result in any
noticeable difference in tracking accuracy), but it's the kind of
thing I'm looking for. Another example is from catching fly balls; one
could control of LOT or optical acceleration/ lateral angle or optical
velocity/ lateral angle - three different perceptions of the same
situation. But that's a little too esoteric. I'd like something very
easy for a person to try in a "portable" demo.

Best

Rick

···

---
Richard S. Marken
rsmarken@gmail.com
marken@mindreadings.com

OK, I though things out for
myself and came up with the idea of asking

Bill for some more examples;-)
[From Bill Powers (2007.04.14.1905 MDT)]

Rick Marken (2007.04.14.1410)–

All right, here’s another:

Say “Mama”.

Now find a low stool, sit at a table and put your chin on the table, and
say “Mama.”

Bill

[From Rick Marken (2007.04.14.1940)]

Bill Powers (2007.04.14.1905 MDT)--

Rick Marken (2007.04.14.1410)--

OK, I though things out for myself and came up with the idea of asking
Bill for some more examples;-)
All right, here's another:

Say "Mama".

Now find a low stool, sit at a table and put your chin on the table, and
say "Mama."

Not quite what I had in mind. See my next post [Rick Marken
(2007.04.14.1500)] for a description.

Thanks

Rick

···

--
Richard S. Marken
rsmarken@gmail.com
marken@mindreadings.com

Now find a low stool, sit
at a table and put your chin on the table, and

say “Mama.”

Not quite what I had in mind.
[From Bill Powers (2007.04.14.2030 MDT)]

Rick Marken (2007.04.14.1940) –

Did you try it? In the first case, you say Mama by raising and
lowering your jaw to close and open your lips. In the second case your
jaw is against the table so you have to move your whole head up and down
to make the same sound.

Best,

Bill P.

Actually, I hate to be a pain
(really!) but what I would like is an

example where different aspects of the same situation can be
perceived

in different ways to produce the same result.
[From Bill Powers (2007.04.14.2230 MST)]

Rick Marken (2007.04.14.1500) –

When you say “produce the same result,” do you mean
qualitatively (the same kind of result) or quantitatively (the same
precision of result, in detail)? You can judge your car’s speed by
watching the relationship of your car to other cars going the same way,
or the relationship of your car to the road and nearby stationary
objects. Qualitatively, you’re producing the same result, control of
perceived speed, but quantitatively there’s a big difference, especially
when you have to hit the brakes.

Also, are you speaking of producing the same result at the same level as
the different perceptions, or at a higher level? I gave you an
example in which a conflict manifested at level 1 was produced because
the level 1 variable was an element in two different level 2 perceptions,
which in turn were elements of the same level 3 perception. In general a
perception at level n is a function of several perceptions at level
n-1, and often there is enough redundancy so controlling any subset of
the level n-1 perceptions is enough to generate the required level n
perception.

Your example of using different frames of reference for position
perceptions really concerns different ways of deriving the same
perceptual signal from the next level down, so the same functional
relationship results (the zero point cancels out of the equations, so it
doesn’t matter which zero you use). In that case there is no difference
in the perceptions at all, and we’re speaking entirely within one level
of organization. But if we introduce a higher level of perceptions, there
is an infinity of ways you can get the same higher-level perception out
of a collection of lower-level ones. You just plot the surface of
indifference in the lower-level space.

Simple example: p2 = p1a + p1b. There is an infinity of different pairs
of p1a and p1b that will give the same value of p2. And if p2 = p1a + p1b

  • p1c, then you can set any two of the level 1 variables to zero and find
    the value of the remaining one that will produce the desired value of p2.
    Or you can perceive the “aspect” made of of p1 and p2, or the
    aspect made up of p2 and p3, and get the same p2.

The only question remaining is how easy it is to affect p1a, p1b, and p1c
– I think that given the input weights, the output weights have to be
adjusted to be the transpose of the input matrix, but that’s not what
you’re looking for.

I’m too sleepy to get any farther with this.

Best,

Bill P.

···

Content-type: text/plain; charset=Windows-1252
X-MIME-Autoconverted: from 8bit to quoted-printable by vacuum.cso.uiuc.edu id l3DFEB3j028172

[From Bjorn Simonsen (2007.04.15,13:20 EUST)]

from Rick Marken (2007.04.14.1500)

Actually, I hate to be a pain (really!) but what I would like is an
example where different aspects of the same situation can be perceived
in different ways to produce the same result.

I understand the word aspect as a way in which a thing may be viewed or
regarded. Different aspects tell us how a verbal action passes; if it is
permanent, instantaneous, ingressive or iterative.

I will give you an example where different aspects can be perceived in
different ways to produce the same result.

Tell me how the value y is changing by analyzing Y, Y' and Y''

Y = x3 – 7*x + 6

Y’ = 3*x2 – 7

Y’’ = 6*x

I know you do this better than me, therefore I put my further comments in
the enclosed file.

bjorn

funksjon.doc (20.5 KB)

[From Rick Marken (2007.04.15.0900)]

Bill Powers (2007.04.14.2230 MST)--

You can judge your car's speed by watching the relationship of your car to
other cars going the same way, or the relationship of your car to the road
and nearby stationary objects. Qualitatively, you're producing the same
result, control of perceived speed, but quantitatively there's a big
difference, especially when you have to hit the brakes.

Yes, that's a good one.

Also, are you speaking of producing the same result at the same level as
the different perceptions, or at a higher level?

I think what I'm talking about is like your example of controlling
speed; what you control determines how a real variable is influence.
In the case of catching a ball, what you control (LOT or optical
velocity), will influence where you actually end up relative to the
ball "out there"; in the case of driving, what you control (movement
relative to other cars or stationary objects) will influence how fast
the car is actually going (and, thus, when you have to hit the
brakes).

I gave you an example in
which a conflict manifested at level 1 was produced because the level 1
variable was an element in two different level 2 perceptions,

No, I think the car example is the best. A couple more like that and
I'll be happy. It's for my class. I'm teaching it again (practice
makes perfect) and I want to use these to get across the idea of
control of perception.

Your example of using different frames of reference for position
perceptions really concerns different ways of deriving the same perceptual
signal from the next level down, so the same functional relationship results
(the zero point cancels out of the equations, so it doesn't matter which
zero you use).

Yes, I know. Forget that example!

So far I have two examples I like; catching balls by controlling
different aspects of the optical trajectory of the ball and
controlling car speed using different aspects of the relative
movements of car and events outside the car. Just a couple more will
make me happy. How about tuning a guitar by trying to match the pitch
of two simultaneously plucked strings versus by trying to eliminate
the beat modulation envelope? I think that controlling beat modulation
is supposed to produce a better frequency match than pitch matching.
People have to learn to perceive the beats but they are always there
as is the pitch difference.

Best

Rick

···

--
Richard S. Marken
rsmarken@gmail.com
marken@mindreadings.com

[From Rick Marken (2002.06.14.1440)]

Recently, I have read a number of papers that seem to get very close to the
perceptual control theory concept of behavior. But they never really get
there, where "there" is the idea that behavior is the _control of perception_.
Indeed, these papers seem to get everywhere but "there". They get feedback,
they get control, they even get perception (to some extent). But they never
get _control of perception_. It's as though the idea were transparent or were
shielded by a big sign saying "Don't go there".

Reading this stuff makes me realize how lucky I am, not only to have
discovered _Behavior: The control of perception_ but to have taken the time
and effort to _learn it_. What an amazing discovery Bill Powers made. And how
beautifully and clearly he described (and continues to describe) it.

Thanks again, Bill.

Best

Rick

···

--
Richard S. Marken, Ph.D.
The RAND Corporation
PO Box 2138
1700 Main Street
Santa Monica, CA 90407-2138
Tel: 310-393-0411 x7971
Fax: 310-451-7018
E-mail: rmarken@rand.org