RM :
You seem to be saying that it’s not possible to measure the ability to control because controlled variables and the references for the states of these variables are all in the controller’s head and, thus, cannot be measured from outside. Is that what your saying? If so, your going against a lot of empirical data, not the least of which can be easily obtained in the Nature of Control Demo I mentioned above (http://www.mindreadings.com/ControlDemo/BasicTrack.html).
HB :
Problem is that I’m maybe not against your data, but that you have not sufficient theory behind and maybe you are differently inetrpreting “defintion of control,” that could support your interpretation. I suppose that “empirical data” are also obtained on the bases of some theoretical knowledge which could be subjective. But it’s good, we finaly start to talk again.
RM :
While a controlled variable is a perception in the mind of the controller we can get a pretty accurate idea of what that perception is by seeing whether a perception of our own – such as our (quantified) perception of the distance between cursor and target in a tracking task – is kept in some reference state, protected from disturbance,
HB :
The problem I see is that you are putting “controlled variable” into a diagram, although I never saw Bill doing that. If we suppose that whoever has pretty accurate idea of what he is perceiving, it’s perceiving “controled variable” only when he is controlling. But putting “controlled variable” in environment means that is generaly there, although person is not controlling anything in environment.
And please if you could wait with your “protection from disturbances” until we clear problem about it’s meaning.
RM :
This is what the Test for the Controlled Variable (TCV) is about;
HB :
Why not test for the controlled perception, if everything is our minds. Why test for the “controlled variable” ? It’s “perceptual control theory” not “variable control theory”
RM :
… it’s a procedure for identifying in the tester’s perceptions  the perception that the controller is controlling. This TCV can be done even if the controller is not maintaining a constant reference for the controlled variable. This is demonstrated in my Mind Reading demonstration (http://www.mindreadings.com/ControlDemo/Mindread.html).
Of course, we don’t necessarily always get it exactly right; indeed, I have shown that it is probably the angle rather that the distance between cursor and target that is the perceptual variable that corresponds to the one controlled in a tracking task (see Ch. 4 in “Doing Research on Purpose”). But you can measure the ability to control relatively accurately even if your assumption about the perceptual variable that a person is trying to control is not exactly right.
HB :
The relativity of thinking you used here is promising. In your past article about TCV you described also the huge limitations of TCV. The problem is again the simplicity of simlutaions on which data are obtained and interpretation made. It’s a question of whether or not free-living organisms with all the hierarchy working all the time inside organism will act like you predict in testing TCV. People usually hide their real intention with behavior, and it’s very problematic how to determine what they really think. So if I understood right all the “tests for controlled variable” can be tricky. “Real controlling” of free-living organism which is predicted can be false, because data are obtained on wrong theoretical bases. If the test for “controlled perception” is executed, organisms whole control in organism would be considered and maybe real intentions discovered. I hope that we understood, what I wanted to say.
RM:
There may be a slight difference in meaning but I don’t think it has serious implications. I think the difference between the inside and outside meanings of control is alluded to briefly in Powers’ description of “The Coin Game” in B:CP. See in particular the last paragraph on pp. 237-238 of the second edition. Here the experimenter (E) doing The TCV concludes that the subject (S) is controlling for a zigzag pattern of coins. So E might conclude that the subject is controlling for a perception of the letter Z; The S may disagree and say it was an N or something else. But, as BIll notes,such an argument misses the point since “E has discovered what S is controlling”. In other words, E’s perception of the “zigzag” pattern of the coins that S is controlling (the outside view of control) is clearly equivalent to the perception of the “zigzag” pattern of coins that S is controlling (the inside view of control). So the outside view of controlled variables that is provided by The TCV is considered equivalent to the inside view of these variables that are actually being controlled by a controller. I believe this is why Bill said that The Test is  “… the nearest approach I know of to mind reading.”(Powers, BYTE, September, 1979, p. 109). The Test provides an outside view of a controller’s inside (mental) view inasmuch as the the tester’s perception of the controlled variable is equivalent to the perception that is under control. Once you know what perception a controller is controlling you can easily measure how well they are controlling it.
HB: I think this is fine PCT description. You also used “once we know what perception a controller is controlling, not which variable”.
 But I see the problem that nothing what is perceived is exactly what is seen, although maybe it seems to be. People also perceive what they want to perceive and has great imagination when describing things they perceived (think of smokers who permanently lie how many cigarettes they smoke or fisherman and hunter what did they caught :). I want to say that simplified simulations in laboratory conditions can not show the complexity of “real life situations” like Kent noticed. So whatever findings of your experiments were. It doesn’t mean that they will work in everyday life. When Bill analyzed Maturana’s work, he noticed something about “simplicity”, relations between model and experience. Briliant thinking (LCS II, p. 184 - …).
RM:
The standards you are talking about here are not references in the PCT sense.
HB :
Everything what’s “build” in human organism is reference. No matter how we call it. It’s product of internal control - “control of perception”.
RM :
And the measures of inability to control are not based on a determination of whether students are trying to control the variables for which the standard values are given. So while standardized tests can be viewed as measures of students’ ability to control for particular results (the score on the test), because there has been no TCV performed, even informally, the results say little about whether this observed performance results from an inability to control the variable measured (test score) or from the fact that the student wasn’t even trying to control that variable.
HB:
I’m sorry I don’t understand what you wanted to say.
RM:
I agree with much of what follows; I think “stigmatizing” students based on their standardized test scores is not a good thing. Indeed, I think the emphasis in education on standardized testing is unfortunate. But this has nothing to do with how one measures the ability to control a perception using PCT.
HB :
If you put the “references” - standards (tests) into the head of the teachers and students, you will see that they control perception as any other human do. So they try to control with thier own references which are produced inside them. Conflict is inevitable. Try to think in the sense of “Collective control processes” as Kent do. I think that we can’t say that whatever people are doing in school has nothing to do with PCT. It has everything to do. All that people in schools do, is “Control of perception”. All what is happening to people in schools is  included in the “Fact of control”. I think we can not “isolate” “controlled variables” in environment like “standards” or “exames” as something objective that has nothing to do with PCT. All are constructs of people, so we have all the time to think of them as people creations, which are controlled inside them. So PCT is everywhere when social or other living control system are studied.
RM:
I think we should just drop it. You think I have PCT all wrong and I think the same of you. So let’s just go our own ways. People who like your version of PCT will enjoy your discussions of PCT and and people who like my version will like mine.
HB :
I’m sorry Rick. Why should we do that. The only thing we have to do is change our way of discussion. And talk like old friends as we used to do in the beggining. Mutual repect and listening to each other, and myabe we and others will learn something. We just have to let out sarcasm, raillery, offences, …etc. Just try to put our arguments “on the table” as equal “control constructs” and also any others can contribute, although they could have maybe different oppinion or knowledge. It’s not necessary that we all think the same. Diversity is what is “propeling” the world.
RM:
I think that is only true in your version of PCT. I’ve attached a copy of my hierarchy of control spreadsheet simulation. I’ve taken the liberty of setting all the disturbances to 0 – so there are no disturbances to the controlled variables. Press the recalculate key(F9) repeatedly and see what happens:the hierarchy (all control systems in it) controls. It will eventually end up with all control systems keeping the perceptions they control exactly at their references; zero error at all three levels of the hierarchy. So, again, disturbances are not necessary for control in my version of PCT control systems. But maybe they are in yours.
HB :
O.K. I’ll try first to “break” with our tradition of using “You are bad -sentence” as Gordon is proposing. So I’ll tell that I respect your work Rick, although you maybe got an impression that I don’t. But sincerely speaking your tiral (simulation) to simplify understanding of how coordinated work of control units can be done in organisms are  insufficient. They are based on still not general understanding how “Perceptual control” in real organisms work and approximations that PCT is doing is also insufficient, although I think PCT is the best “tool” to explore organisms.
It is a great job that Bill have done, but simluations can be improved if you improve the knowledge about organism on which bases simulations are to be done. So it’s not my or your version of PCT, it’s how near is our thinking about how organisms work, because that is the goal of PCT :
Bill P at all (2011). : Â Perceptual Control Theory (PCT) provides a general theory of functioning for organisms.
So if we start to talk  about internal “controlled variables” in organism, we have to incorporate all the time ongoing disturbances like : continuous heat exchange and gravity disturbances which we can not escape, etc. The most disturbances to internal control in organisms are coming from inside the body beacuse of methabolic processes and so on. I didn’t noticed that Bill included them, but they are very important in understanding why control in organism continuously counteracting. So however you simulation can be good, it’s too simplifyed, to show all the control problems in organism and efects of all disturbances from internal and external environment that are counteracted in organism. Disturbances in outer environment are affecting the whole organism, beside receptors also other structures of organism.
If we want to understand the final “goal” Â of PCT, understanding the organism, then there is quite difficult task in front of all CSGnet group.
All the terms like “inability”, “lack of control” and so on are getting by my oppinion different meaning if observed from the point of internal control in organisms, which primarly is responsible for “achieving and maintaining preselected state…” that is determined by genetics.
Best,
Boris
-----Original Message-----
From: csgnet-request@lists.illinois.edu [mailto:csgnet-request@lists.illinois.edu] On Behalf Of Richard Marken
Sent: Monday, October 27, 2014 12:55 AM
To: csgnet@lists.illinois.edu
Subject: Re: Review of “Reconstructing Your World View� by Bart Madden
[From Rick Marken (2014.10.26.1455)]
On Sat, Oct 25, 2014 at 11:52 PM, Boris Hartman boris.hartman@masicom.net wrote:
HB : I’m wondering what could it mean »inability to control« in PCT.
Can you give some »deeper« PCT explanation of this inability ? Or some example ?
RM: I’m sorry. I assume that readers have a basic understanding of PCT
when I post to CSGNet. But I’m happy to try to explain these things; I
don’t think one can go over the basics too much.
HB :You seemed to become too clever. I thought too, that basic
understanding of PCT is necesary to answer on CSGnet, instead of insulting people.
RM: Sorry you feel insulted.
RM : “Inability to control” means an inability to keep a controlled
variable at its reference….“inability to control” is measuredd as
either RMS error or stability factor (see
http://www.mindreadings.com/ControlDemo/BasicTrack.html). Both are
measures of how much the controlled variable (distance between cursor
and
target) varies around the reference (zero distance between cursor and
target).
HB:  If I understandit right your explanation, there is »controlled
variable« in  outer environment (distance between cursor and target),
which varies around  the reference (zero distance between cursor and
target) also in outer  environment. And this can be meassured like
everything is tried to be  measured in behaviorism and self-regulation, outside.
HB: So it seems that behavior is stimulated in closed loop from
outside (error) and that loop is »controlling itself« implaying
behavior to »control »controled variable«. So zero error outside
(reference) is somehow directing behavior. It is seen as observable
phenomenon which is happening outside. I call it perceptual illusion.
RM: I don’t think you do understand what I am saying. And I find it difficult to understand what you are saying. Maybe it’s a language problem. But my point seems pretty simple: the ability to control is measured in terms of degree of variation of a controlled variable around a presumed reference state (RMS) or as the ratio of observed to expected variance of a controlled variable (stability measure). You seem to be saying that it’s not possible to measure the ability to control because controlled variables and the references for the states of these variables are all in the controller’s head and, thus, cannot be measured from outside. Is that what your saying? If so, your going against a lot of empirical data, not the least of which can be easily obtained in the Nature of Control Demo I mentioned above (http://www.mindreadings.com/ControlDemo/BasicTrack.html).
While a controlled variable is a perception in the mind of the controller we can get a pretty accurate idea of what that perception is by seeing whether a perception of our own – such as our
(quantified) perception of the distance between cursor and target in a tracking task – is kept in some reference state, protected from disturbance, This is what the Test for the Controlled Variable (TCV) is about; it’s a procedure for identifying in the tester’s perceptions  the perception that the controller is controlling. This TCV can be done even if the controller is not maintaining a constant reference for the controlled variable. This is demonstrated in my Mind Reading demonstration (http://www.mindreadings.com/ControlDemo/Mindread.html).
Of course, we don’t necessarily always get it exactly right; indeed, I have shown that it is probably the angle rather that the distance between cursor and target that is the perceptual variable that corresponds to the one controlled in a tracking task (see Ch. 4 in “Doing Research on Purpose”). But you can measure the ability to control relatively accurately even if your assumption about the perceptual variable that a person is trying to control is not exactly right.
HB: So I think that »inability to controol« seen from inside control
perspective has different meaning. It’s not any more difference that
can be seen outside and isn’t telling practicaly anything, except that
people are deviating from something. The main problem I see is, that we can’t see what’s causing it.
The difference between explanation of »inability« outside and inside
has serious implications.
RM: There may be a slight difference in meaning but I don’t think it has serious implications. I think the difference between the inside and outside meanings of control is alluded to briefly in Powers’
description of “The Coin Game” in B:CP. See in particular the last paragraph on pp. 237-238 of the second edition. Here the experimenter
(E) doing The TCV concludes that the subject (S) is controlling for a zigzag pattern of coins. So E might conclude that the subject is controlling for a perception of the letter Z; The S may disagree and say it was an N or something else. But, as BIll notes,such an argument misses the point since “E has discovered what S is controlling”. In other words, E’s perception of the “zigzag” pattern of the coins that S is controlling (the outside view of control) is clearly equivalent to the perception of the “zigzag” pattern of coins that S is controlling (the inside view of control). So the outside view of controlled variables that is provided by The TCV is considered equivalent to the inside view of these variables that are actually being controlled by a controller. I believe this is why Bill said that The Test is  “… the nearest approach I know of to mind reading.”(Powers, BYTE, September, 1979, p. 109). The Test provides an outside view of a controller’s inside (mental) view inasmuch as the the tester’s perception of the controlled variable is equivalent to the perception that is under control. Once you know what perception a controller is controlling you can easily measure how well they are controlling it.
HB: The outside definition of
»inability or ability to control« was very similar to yours and it
said that  population for example children can deviate from standards
(in your case
references) and this deviations can be marked or used for other
purposes. So  children can deviate »over-standard«, »below standard«
and somewhere arround  standard. What means that they by behavioristic
logic they are »un.normal«,  »normal« and more than »normal«, genious.
RM: The standards you are talking about here are not references in the PCT sense. And the measures of inability to control are not based on a determination of whether students are trying to control the variables for which the standard values are given. So while standardized tests can be viewed as measures of students’ ability to control for particular results (the score on the test), because there has been no TCV performed, even informally, the results say little about whether this observed performance results from an inability to control the variable measured (test score) or from the fact that the student wasn’t even trying to control that variable.
HB: In common sense behavioristic logic, based on »obervable facts«,
»deviating« behavior
(inability) below the educational standard (as that difference was
seen outisde) was labeled  as »not appropriate«, »lazy«, »uncapable«,
»dumm«, »stupid«, or in your words, they were  »unable to control« near
reference (standard). In our expert language, we usualy say that  children get »stigma«, just because they were born.
RM: I agree with much of what follows; I think “stigmatizing” students based on their standardized test scores is not a good thing. Indeed, I think the emphasis in education on standardized testing is unfortunate. But this has nothing to do with how one measures the ability to control a perception using PCT.
RM: I hope you now understand what “lack of control” means (the degree
to which one is unable to keep a controlled variable at the reference).
HB : I hope you now understand all the problems you can cause with
your »narrow Mr. Hide« thinking.
RM: I sort of understand why you think there are problems but I think you are misidentifying the source of your concerns. The problem is not with measuring the ability to control. The problem (at least in the
US) seems to come from people who are trying to improve education based on a business model where the aim is to increase “profit”
(standardized test scores) by increasing productivity (getting rid of ineffective teachers – the one’s not increasing standardized test
scores) and reducing costs (lower wages, fewer teachers, cheaper facilities). Don’t blame that on how ability to control is measured in PCT. I think the understanding of ability to control that comes from PCT would lead educators to emphasize reduced class size, broader curriculum and  more discretion about how to teach to teachers who tend to be quite good (when class sizes are manageable) at determining what each child is trying to do (control) and how well they are managing to do it (their ability to control); that is, at informally evaluating studnets’ ability to control.
HB: I think we have to solve the »conflict« about what of your text is
representing PCT and what is not.
RM: I think we should just drop it. You think I have PCT all wrong and I think the same of you. So let’s just go our own ways. People who like your version of PCT will enjoy your discussions of PCT and and people who like my version will like mine.
HB: I just hope that you will not mislead PCT ladies (Bill’s daughters)
into self-regulation  or even behaviorism.
RM: My days of misleading ladies are well behind me but even if they weren’t I know that Bill’s daughters can think for themselves extremely well! I wouldn’t have a chance.
HB: How can LCS control if they are not disturbed ?
RM: The same way they control when they are disturbed.
HB: Â Disturbances are necesary part of control
RM: I think that is only true in your version of PCT. I’ve attached a copy of my hierarchy of control spreadsheet simulation. I’ve taken the liberty of setting all the disturbances to 0 – so there are no disturbances to the controlled variables. Press the recalculate key
(F9) repeatedly and see what happens:the hierarchy (all control systems in it) controls. It will eventually end up with all control systems keeping the perceptions they control exactly at their references; zero error at all three levels of the hierarchy. So, again, disturbances are not necessary for control in my version of PCT control systems. But maybe they are in yours.
Best
Rick
Richard S. Marken, Ph.D.
Author of  Doing Research on Purpose.
Now available from Amazon or Barnes & Noble