More on Disturbances

[From Rick Marken (2006.12.11.0945)]

It seems to me there are two classes of disturbance: (1) those that we are
consciously aware of and attend to as such and (2) those that are simply
compensated for in our behavior with no thought to disturbance as such.

Yes, and there are another two kinds of "disturbances": the change in qi resulting form the effect of a disturbance -- call it delta_qi -- and the disturbance variable itself -- d. In your example of walking on a ship in high seas, delta_qi would be the change in your perception of balance resulting from d, the pitching of the ship. You can be consciously aware or unconscious of delta_qi because delta_qi is a perception. You can also be consciously aware or unconscious of d, too, but only if it exists as a perception. But d is often unperceived. In you ship example, d -- the pitching of the boat -- can certainly be perceived, so you can be aware of it or not. But in many (most) situations d is unperceived, as in the compensatory tracking task. So you can't really be conscious or unconscious of d since it doesn't "exist" as far as you are concerned. All you can perceive (and therefore be aware of or not), in the compensatory tracking task, is delta_qi, which in that case is the perception of a change in the position of the cursor.

Is there any value in conceptualizing two classes of disturbance?

If the two classes are "consciously perceived" versus "not consciously perceived" (regardless of whether you are referring to a disturbance in the delta_qi or d sense) I think, yes, there _may_ be a difference between controlling with and without consciousness. I think both reorganization theory and Zen Buddhism suggest that control would be better when done unconsciously. I think Bill Powers has actually done some experiments that suggest that this is, indeed, the case.

Now, as to Bruce Nevin's distinction: Earlier, I used a sunlight example and
indicated that I had to squint (whether to continue driving safely or simply
to protect my eyes needn't detain us at this point). The sun, if I
understand Bruce Nevin's point, is the "source" of the disturbance. The
"disturbance" itself is the effect of the sun's rays on my retina. Do I
have that right?

Yes, though I think it's better to say that the "disturbance itself is the effect of the sun's rays on the perception under control" rather than on the retina. The perception is a function of aspects of the light on the retina. What aspect of the "sun's rays" constitutes a disturbance depends on what the controlled perception is. In you example, one controlled perception was the perceived overall brightness on the retina. That's qi and it's value varies as a function of the _intensity_ of the sun rays falling on the retina, the intensity of which are d. The value of qi also depends on the level of "squint", which is the output variable, o. So we can say that the magnitude of the controlled perception, qi, which is the brightness of light on the retina, depends on the intensity of the sun's rays, d, as well as on the level of squint, o: qi = d + o.

Best

Rick

···

On Monday, December 11, 2006, at 05:10 AM, Fred Nickols wrote:

[From Rick Marken (2006.12.11.1010)]

Rick Marken (2006.12.11.0945)

It seems to me there are two classes of disturbance: (1) those that we are
consciously aware of and attend to as such and (2) those that are simply
compensated for in our behavior with no thought to disturbance as such.

Yes, and there are another two kinds of "disturbances": the change in qi resulting form the effect of a disturbance -- call it delta_qi -- and the disturbance variable itself -- d. In your example of walking on a ship in high seas, delta_qi would be the change in your perception of balance resulting from d, the pitching of the ship.

Just a small clarification/correction: delta_qi, which is a change on the state of a controlled perception, depends on _both_ d and o. So it was not really correct for me to say that the change in qi results from a change in d. It results from changes in both d and/or o. So delta_qi can be called a "distrubance" to qi but the cause of this disturbance is not just d. I don't want to give the impression that "disturbance" in the delta_qi sense is like a mirror of "disturbance" in the d sense. It's not.

Best

Rick

···

On Monday, December 11, 2006, at 05:10 AM, Fred Nickols wrote:

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

[From Bjorn Simonsen (2006.12.11,23:00
EUST)]

From Bill Powers (2006.12.11.1040
<ST)

No action you produce can alter
the disturbance in any way.

I am working with a Control System
(PCT) in Excel.

I am presented the following
figures, one at the time, in 5 seconds. They are presented in a random order.

image00110.gif

image00211.gif

I am able to extend the presentation
time up to 10 seconds if I press a lengthening button.

I am able to reduce the presentation
time down till 2 seconds if I press a deduction button.

I wish to perceive the figure
showing a green circle as long as possible and the other figures as short as possible.

I sit in front of the PC screen with
my mouse in my right hand, press the start button and move the mouse pointing over
the deduction button.

I shall not mention the reference
value, the different disturbance values and the different conversion constants.

If I am presented a red square, I
press the deduction button and I am presented a new figure. Isn’t this an
example where my actions alter the disturbance?

I know that my actions don’t alter
the random order of figures, but my actions alter the time the figure is
presented. Therefore I think your “in any way” is too dogmatic.

Maybe I think the wrong thoughts,
but I am reading Kent A. McClelland and Thomas J. Fararo’s book “Purpose,
Meaning and Action. Control Systems Theories in Sociology.”

McClelland doesn’t thank you in his
Preface, but I think he makes mention of you in his Introduction to the book. Here
he says that your “B:CP remains the most persuasive account of the explanatory
power of the cybernetic control concept for behavioural science.”

I think he presents PCT and HPCT in
a wonderful way on five pages.

Let me go back to your sentence
above.

McClelland is a sociologist and he is
involved in the Environment and Disturbances. He explains very well the process
of controlling a perceptual signal, but he adds that this controlling
indirectly results in bringing one aspect of the environment under control (the
sociologist). He uses the word indirectly and says that the Negative Feedback
System for controlling Perception is modelled as controlling both an interior
perception and an exterior aspect of the environment from which the perception
is drawn.

My question therefore is: “ Can we alter
the disturbance indirectly when we wish to control a certain perception? If the
answer is “Yes”, the output is free to vary in an unpredictably way to
compensate for randomly occurring disturbances.

PS. Thank you Fred Nickols for your
many questions. Maybe also you should read McClelland’s book. He has a chapter
named “Perceptions of Leadership in Groups: …” (?).

bjorn

oledata12.mso (8.18 KB)

[From Rick Marken (2006.12.11.1500)]

Bjorn Simonsen (2006.12.11,23:00 EUST)--

Bill Powers (2006.12.11.1040 <ST)

No action you produce can alter the disturbance in any way.

I am working with a Control System (PCT) in Excel.

Great!

If I am presented a red square, I press the deduction button and I am presented a new figure. Isn�t this an example where my actions alter the disturbance?

The figures are different values of the controlled perceptual variable, qi. Pressing the button changes the value of qi; it doesn't affect d, which is another cause (besides the pressing) of the value of qi. The variable d, which influences the value of qi presented, is a variable in the Excel program. It's the random variable that determines which of the 9 different display to project. The value of that variable is d and it's not affected by the pressing.

Maybe I think the wrong thoughts, but I am reading Kent A. McClelland and Thomas J. Fararo�s book �Purpose, Meaning and Action. Control Systems Theories in Sociology.�

McClelland doesn�t thank you in his Preface

Hey, I'm not mentioned anywhere in the book at all. So Bill made out OK :wink:

My question therefore is: � Can we alter the disturbance indirectly when we wish to control a certain perception?

I would say "yes" but not by the control system that is controlling the variable affected by the disturbance. It has to be done by a control system that is controlling a perception of the disturbance itself. For example, I could control control for seeing the play (qi) by moving my head (o) every time the person in front shifts in their seat (d). Or I could grab the person in front (now a qi himself) and move him out of my way (o) every time he shifted in their seat (d).

Best

Rick

···

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

[From Bill Powers (2006.12.1615 MST)]

Back home after visit to daughter and grandsons in Durango.

Fred Nickols (2006.12.11.1732 EST}--

The "disturbance of the pyramid" is zero because the pyramid is not trying to control any perceived variables.

A disturbance is a variable that contributes independently of the action of any control system to the state of another variable. The magnitude of the disturbance is the magnitude of the variable that is doing the disturbing. Want to try again?

Hint: You're reading too much into the definition of a disturbance.

Best,

Bill P.

If I am presented a red square,
I press the deduction button and I am presented a new figure. Isn’t this
an example where my actions alter the
disturbance?
[From Bill Powers (2006.12.13.1630 MST)]

Bjorn Simonsen
(2006.12.11,23:00 EUST) –

You do not describe the disturbance here, but only its result. The result
of the disturbance is that some controlled variables for which your
reference level is zero suddenly become nonzero. You then press the
deduction button until they are restored to zero (the redness and the
squareness go away). The red square is clearly a composite controlled
variable which is affected one way by some program in the computer, and
the opposite way by your actions. Your actions alter a perception, not
the disturbance that tends to change the perception.

Perhaps you were thinking that if you want to perceive a green circle,
perceiving a red square instead is a disturbance. But it is not. All that
has happened is that two perceptions have appeared: the color red, and
the shape of a square. The perception of red color fails to match the
reference color, green, and the perception of shape, square, fails to
match the reference shape, a circle. Your program provides no direct way
of affecting either the color or the shape, so you can’t correct the
errors. All you can do is press the deduction button until a new color
and shape appear.

You have no way of affecting the disturbances (there must be two
disturbances since two aspects of the display are being affected). That
is because the disturbances are part of the program you wrote. They are
the program steps that cause a new shape perception or a new color
perception (or both) to appear. Once the program step for selecting a red
color or a square shape has occurred, you can do nothing to change it.
The only thing you can change about any display is how long it remains
visible. So when there is an error, you keep pressing the deduction
button until the display is replaced by a new one, and if there is no
error you press the time-extension button (actually, you should not have
to do anything, but the task is more complex than that – you have to
know that if you do nothing, the display you want to see will disappear
in 5 seconds).

Everyone seems to be ignoring what I’ve been saying about disturbance as
cause and disturbance as effect. Is that because all this confusion is
desired for some reason? Perhaps each of you is reading my words and
thinking “Oh, I already know about that.” If that were true,
Bjorn, Rick, and Fred would not have written what they wrote. All I can
do is say as clearly as I can, that a disturbance is a variable with a
magnitude, which affects the magnitude of another variable according to
the physical linkages that exist. A disturbance is not a perception. A
perception is not a disturbance. If we were to allow speaking of
disturbances as effects as well as causes, we would get – well, we would
get what we have now.

Come on, guys, you’re not finished yet.

Best,

Bill

[From Rick Marken (2006.12.12.1710)]

Bill Powers (2006.12.13.1630 MST)

Unless you're in China again, it's still the 12th, Bill.

Everyone seems to be ignoring what I've been saying about disturbance as cause and disturbance as effect. Is that because all this confusion is desired for some reason? Perhaps each of you is reading my words and thinking "Oh, I already know about that." If that were true, Bjorn, Rick, and Fred would not have written what they wrote.

Actually, I DO already know that, as you could tell form reading my posts on the subject. I know that a disturbance is a causal variable in PCT and I also know that the term "disturbance" is continuously being used to describe the _effect_ of a disturbance on a controlled variable.

The reason I answered your question about the great pyramid incorrectly is because I don't understand physics, not because I don't know the difference between disturbance as cause and effect. I thought the equal and opposite force resulting from the man's 20 pounds of lean force was the force generated by the pyramid that caused the man to stop in the leaning position. Apparently this is wrong but, again, it's wrong because I don't know physics. Perhaps the answer to your question regarding the disturbance of the pyramid is, indeed, zero, because f =ma and the pyramid is not being accelerated into the man, Is that right?

Best

Rick

···

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