VS: To accept a theory (was Re: Control Is)

[Eetu Pikkarainen 2017-02-13]

Hi Warren,

WM: surely the truth is neither that only perception is controlled nor that aspects of the environment are never controlled.

EP: I think this depends on the use of words: What
you mean with “control”…

WM: The view I have been trying to out across is that the control of an aspect/function of the environment is controlled via the control of perception of that aspect of the environment. In instances
where control is working well there will be no differences between them.

EP: If “control” means something like affecting a value of one (“controlled”) variable so that it becomes the same or near to the current value of an other variable (reference value) and / or remains the same or near to it, then anything in the environment
is not controlled. And this has nothing to do with the question how well or badly the control is working. There just are no reference variables and thus reference values in the environment, they exists only inside the control system.

EP: Of course you can you can use words differently, and define “control” as a process where some variable in the environment is more or less stabilized to some value as a consequence of the output of a control-system which is controlling its perception
value to its reference value. But then you use “control” in two different meanings and although it is very usual in colloquial language, it is perhaps not very good practice in science. I can cause unnecessary disputes like these threads. I cannot suggest
what is the good term to that latter phenomenon, I call here its end product stabilization but itself just affecting.

(EP: because control system is a loop, you can say BOTH that a percetion is controlled by affecting + stabilization of a certain environmental variable (cev) AND that a certain environmental aspect is
stabilized by controlling the corresponding perception. One is a consequence or other and visa versa - but one is not the same as the other.)

WM: But I am always fascinated in the exceptions - the instances where control of perception can be shown to be partially independent of control of an aspect of the environment, and whether this might
be revealed by experimentation. Maybe this only occurs in the imagination mode, or in virtual reality (for example where depth information is fully synthesised and doesn’t exist in a real environment)? Or maybe the kind of experiment in which different participants
have different perceptions of the same environmental variable, like the parallax issue we discussed?

···

EP: I aggree this is interesting question, if I understand you right. I think an important and not at all exeptional area of this kind of phenomena is connected to the so called psychical defense mechanisms (repression) and turning the blind eye when the perception
would be too far from reference. An innocent example is shutting your eyes in horror movies or in coaster roller. In these cases the person can effectively control her perceptions without affecting at all to the original environmental variables which are the
source of these perceptions (in principle). Even though, as I believe, we cannot at the moment measure the neural perception and reference signals and then compare them to the stabilization of environmental variables, those kind of “quasi-control” of perceptions
could perhaps be researched somehow.

Eetu

(Sorry for broken English)

Thanks Eetu!

ET: "EP: If "control" means something like affecting a value of one ("controlled") variable so that it becomes the same or near to the current value of an other variable (reference value) and / or remains the same or near to it, then anything in the environment is not controlled. And this has nothing to do with the question how well or badly the control is working. There just are no reference variables and thus reference values in the environment, they exists only inside the control system."

WM: Just like most people on this thread, of course I agree with you in the sense that you are articulating PCT. I am getting to see what this debate is getting at; Rick wants to identify control in the environment, separately from PCT, just like Bill at times, so that PCT has a phenomenon to explain. The problem is, once we unpack the nature of control, we see that it can only be defined accurately from an observer's perspective. I think that Rick has been arguing that the Qi exists in the environment, but would he also propose that the reference value could exist in the environment? Presumably not; in fact of course, like you say Eetu, it doesn't because if it did there would be no point in the organism controlling the Qi. So the question is whether one can truly say that the Qi is controlled if this is only the case because it is transformed into P which is compared with an internal RV continuously.
Maybe the fact that we can't actually see control without taking a perceptual perspective is exactly why so many people find PCT hard to adopt. A part of the loop of control can be identified in the environment but it can't be judged as controlled without taking a perceiver's perspective. Like Rick I am not sure if using the word 'stabilised' helps at all - because often these components will not be stable if there is a changing reference value for its perception within the organism. Could we say the Qi is typically controlled but one can only make this judgement when the observer has an internal reference point to do so? Or is that a weak hybrid of the two positions?
Warren

···

On 13 Feb 2017, at 07:42, Eetu Pikkarainen <eetu.pikkarainen@oulu.fi> wrote:

EP: If "control" means something like affecting a value of one ("controlled") variable so that it becomes the same or near to the current value of an other variable (reference value) and / or remains the same or near to it, then anything in the environment is not controlled. And this has nothing to do with the question how well or badly the control is working. There just are no reference variables and thus reference values in the environment, they exists only inside the control system

Hi Eetu, this is an interesting development: EP: “Perhaps we could talk also about a quasi-reference value which is the Qi in the situation where no error value would be produced. The observer can no way know it beforehand (neither the controller) but it will unveil when control works well.” This is a bit like the ‘knot over the dot’ which can be specified in the environment but it is the specification of it within the perceptual memory of the controller that makes the control happen…
Warren

···

On Mon, Feb 13, 2017 at 10:26 AM, Eetu Pikkarainen eetu.pikkarainen@oulu.fi wrote:

[Eetu Pikkarainen 2017-02-13 2]

Thank you Warren for your kind reply!

I would also be interested what you thought about the suggestion in the end of my message :slight_smile:

I am not sure did I understood all, so I’ll check:

WM: Just like most people on this thread, of course I agree with you in the sense that you are articulating PCT. I am getting to see what this debate is getting at; Rick wants to identify control in the environment, separately from PCT, just like Bill at times, so that PCT has a phenomenon to explain.

EP: I see this so that the “stabilization” (or something like that) is the phenomenon which is explained by control. Like for example the behavior of heavenly bodies can be explained either by some (as such) unperceivable natural laws or by as unperceivable vehicles of gods.

WM: The problem is, once we unpack the nature of control, we see that it can only be defined accurately from an observer’s perspective.

EP: I think it can only be defined accurately from the perspective of an observer who has access to PCT as a theory?

WM: I think that Rick has been arguing that the Qi exists in the environment, but would he also propose that the reference value could exist in the environment? Presumably not; in fact of course, like you say Eetu, it doesn’t because if it did there would be no point in the organism controlling the Qi.

EP: First, the concept of Qi, if it is a quantity, it must refer to a value of some environmental variable? It is it’s current value, isn’t it? Perhaps we could talk also about a quasi-reference value which is the Qi in the situation where no error value would be produced. The observer can no way know it beforehand (neither the controller) but it will unveil when control works well.

WM: So the question is whether one can truly say that the Qi is controlled if this is only the case because it is transformed into P which is compared with an internal RV continuously.

EP: I am not quite sure about that sentence. But I would say that, in the name of conceptual coherence, we should call only EITHER the “stabilization” of Qi-variable to quasi-reference value OR the “stabilization” of P to RV as “control”. The other one is cause and the other is result (or visa versa depending how you read the loop) and one can explain the other. They are not the same, anyway.

WM: Maybe the fact that we can’t actually see control without taking a perceptual perspective is exactly why so many people find PCT hard to adopt.

EP: Yes, I believe so too, and we neither FEEL that we are controlling. That is natural if control is a theoretical concept and thus unperceivable.

WM: A part of the loop of control can be identified in the environment but it can’t be judged as controlled without taking a perceiver’s perspective.

EP: Do you mean controller’s or observerver’s perspective (by taking a perceiver’s perspective)?

WM: Like Rick I am not sure if using the word ‘stabilised’ helps at all - because often these components will not be stable if there is a changing reference value for its perception within the organism.

EP: That is certainly a problem, because the reference values of living organisms are typically changeable! I do not know what would be a better word. But isn’t this just to same problem for observing the control happening? If we cannot perceive the reference, then we can infer control taking place only if the Qi-variable seems to be stabilised against disturbations? Only in simulations were we can handle both the reference (by asking) and the disturbations (by causing them all) we can claim to be “perceiving” control happening, even if the reference would be changing.

Perhaps there were better terms for stabilisation like handling, regulating, adjusting, managing, constraining, affecting etc.

WM: Could we say the Qi is typically controlled but one can only make this judgement when the observer has an internal reference point to do so? Or is that a weak hybrid of the two positions?

EP: Weak and cheap :slight_smile:

Warren

On 13 Feb 2017, at 07:42, Eetu Pikkarainen eetu.pikkarainen@oulu.fi wrote:

EP: If “control” means something like affecting a value of one (“controlled”) variable so that it becomes the same or near to the current value of an other variable (reference value) and / or remains the same or near to it, then anything in the environment is not controlled. And this has nothing to do with the question how well or badly the control is working. There just are no reference variables and thus reference values in the environment, they exists only inside the control system

Dr Warren Mansell
Reader in Clinical Psychology

School of Health Sciences
2nd Floor Zochonis Building
University of Manchester
Oxford Road
Manchester M13 9PL
Email: warren.mansell@manchester.ac.uk

Tel: +44 (0) 161 275 8589

Website: http://www.psych-sci.manchester.ac.uk/staff/131406

Advanced notice of a new transdiagnostic therapy manual, authored by Carey, Mansell & Tai - Principles-Based Counselling and Psychotherapy: A Method of Levels Approach

Available Now

Check www.pctweb.org for further information on Perceptual Control Theory

[From Rick Marken (2017.02.13.1000)]

···

On Mon, Feb 13, 2017 at 12:44 AM, Warren Mansell wmansell@gmail.com wrote:

WM: …I am getting to see what this debate is getting at; Rick wants to identify control in the environment, separately from PCT, just like Bill at times,

RM: Actually, all the time.

Â

WM: Â so that PCT has a phenomenon to explain.

RM: You make it sound like PCT was just sitting around looking for something to explain. In fact, the phenomenon of control was sitting around waiting for PCT to explain it. Again, you betray your “theory first” approach to PCT.

Â

WM: The problem is, once we unpack the nature of control, we see that it can only be defined accurately from an observer’s perspective.

RM: Doesn’t sound like a problem to me. Control is defined as the observation that a variable (such as the position of a cursor) is maintained in a reference state protected from disturbance. You can observe it for yourself in my “Nature of Control” demo (http://www.mindreadings.com/ControlDemo/BasicTrack.html)

WM: I think that Rick has been arguing that the Qi exists in the environment, but would he also propose that the reference value could exist in the environment?

RM: The reference state of Qi exists in the environment (actually, like Qi, it exists as a function of physical variables in the environment; as a perception in the observer); the term “reference value” refers to the value of the theoretical reference signal that is presumed to specify the reference state of the controlled variable.

Â

WM: Presumably not;

RM: Right. The reference value is a theoretical construct that explains the observed reference state of the controlled quantity, Qi.

WM: in fact of course, like you say Eetu, it doesn’t because if it did there would be no point in the organism controlling the Qi. So the question is whether one can truly say that the Qi is controlled if this is only the case because it is transformed into P which is compared with an internal RV continuously.

 RM: You lost me.Â

WM: Maybe the fact that we can’t actually see control without taking a perceptual perspective is exactly why so many people find PCT hard to adopt.

RM: I don’t understand what you mean by “we can’t actually see control without taking a perceptual perspective”. It seems like I can see control by just looking to see whether a variable is maintained in a reference state, protected from disturbance. Is that the “perceptual perspective” of which you speak?

Â

WM: Could we say the Qi is typically controlled but one can only make this judgement when the observer has an internal reference point to do so?

RM: No, we say that Qi is controlled when disturbances have far less of an effect on that variable than would be expected if the variable we not under control; that is, when Qi is kept in a constant or variable reference state, protected from disturbance. There is no theory involved in this determination. There is no need to know that this control occurs because there is a system present that is organized around control of the perception of the variable you see being controlled. This atheoretical approach to determining what variable is controlled is demonstrated in my Mind Reading demo (http://www.mindreadings.com/ControlDemo/Mindread.html), which is not only extremely entertaining but also highly educational.Â

BestÂ

Rick

Â

On 13 Feb 2017, at 07:42, Eetu Pikkarainen eetu.pikkarainen@oulu.fi wrote:

EP: If “control” means something like affecting a value of one (“controlled”) variable so that it becomes the same or near to the current value of an other variable (reference value) and / or remains the same or near to it, then anything in the environment is not controlled. And this has nothing to do with the question how well or badly the control is working. There just are no reference variables and thus reference values in the environment, they exists only inside the control system


Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery

On Mon, Feb 13, 2017 at 12:44 AM, Warren Mansell wmansell@gmail.com wrote:

Thanks Eetu!

ET: “EP: If “control” means something like affecting a value of one (“controlled”) variable so that it becomes the same or near to the current value of an other variable (reference value) and / or remains the same or near to it, then anything in the environment is not controlled. And this has nothing to do with the question how well or badly the control is working. There just are no reference variables and thus reference values in the environment, they exists only inside the control system.”

WM: Just like most people on this thread, of course I agree with you in the sense that you are articulating PCT. I am getting to see what this debate is getting at; Rick wants to identify control in the environment, separately from PCT, just like Bill at times, so that PCT has a phenomenon to explain. The problem is, once we unpack the nature of control, we see that it can only be defined accurately from an observer’s perspective. I think that Rick has been arguing that the Qi exists in the environment, but would he also propose that the reference value could exist in the environment? Presumably not; in fact of course, like you say Eetu, it doesn’t because if it did there would be no point in the organism controlling the Qi. So the question is whether one can truly say that the Qi is controlled if this is only the case because it is transformed into P which is compared with an internal RV continuously.

Maybe the fact that we can’t actually see control without taking a perceptual perspective is exactly why so many people find PCT hard to adopt. A part of the loop of control can be identified in the environment but it can’t be judged as controlled without taking a perceiver’s perspective. Like Rick I am not sure if using the word ‘stabilised’ helps at all - because often these components will not be stable if there is a changing reference value for its perception within the organism. Could we say the Qi is typically controlled but one can only make this judgement when the observer has an internal reference point to do so? Or is that a weak hybrid of the two positions?

Warren

On 13 Feb 2017, at 07:42, Eetu Pikkarainen eetu.pikkarainen@oulu.fi wrote:

EP: If “control” means something like affecting a value of one (“controlled”) variable so that it becomes the same or near to the current value of an other variable (reference value) and / or remains the same or near to it, then anything in the environment is not controlled. And this has nothing to do with the question how well or badly the control is working. There just are no reference variables and thus reference values in the environment, they exists only inside the control system


Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery

Hi Rick, thanks, I understand you and it is all getting clearer. I toyed with leaving it there, but I guess it’s best to check my understanding by trying to put it into my own words.

I can now see how a function of elements of the environment - the Qi - can be controlled. As humans we happen to extract this function through perception, but I can appreciate that the function and its reference state still ‘exists’ in a mathematical sense, that does not require a perceiver. It is required to be perceived to actually implement the control which is why we look for theory in the form of PCT to explain how living things do it. Another theorist might insist that this function is calculated in advance and reacted to in an open loop fashion, for example.

It just so happens that if PCT is true then we are perceivers just as we are control systems doing all this scientific work and discussion. For this reason I can see why there is a resistance to ‘the fact of control’ from many CSGers who have taken PCT on fully and are maybe not trying to test it against ‘facts’ like yourself. I am on the fence.

Could we turn this into an advantage and actually appreciate how one theory can traverse the realist empirical approach and the relativist, subjective approach to reality? The purposes of these two camps seem different to me, but they share more in common than proponents of other theories of behaviour I know of. Sorry if that last comment seems to let the heat out of the debate.

I’ve got you on this one Rick. Maybe I’ll take on your view of the power law next!

I’m off to Narnia now… or maybe Andy’s house to hang out with Buzz…

Warren

···

On Mon, Feb 13, 2017 at 12:44 AM, Warren Mansell wmansell@gmail.com wrote:

WM: …I am getting to see what this debate is getting at; Rick wants to identify control in the environment, separately from PCT, just like Bill at times,

RM: Actually, all the time.

WM: so that PCT has a phenomenon to explain.

RM: You make it sound like PCT was just sitting around looking for something to explain. In fact, the phenomenon of control was sitting around waiting for PCT to explain it. Again, you betray your “theory first” approach to PCT.

WM: The problem is, once we unpack the nature of control, we see that it can only be defined accurately from an observer’s perspective.

RM: Doesn’t sound like a problem to me. Control is defined as the observation that a variable (such as the position of a cursor) is maintained in a reference state protected from disturbance. You can observe it for yourself in my “Nature of Control” demo (http://www.mindreadings.com/ControlDemo/BasicTrack.html)

WM: I think that Rick has been arguing that the Qi exists in the environment, but would he also propose that the reference value could exist in the environment?

RM: The reference state of Qi exists in the environment (actually, like Qi, it exists as a function of physical variables in the environment; as a perception in the observer); the term “reference value” refers to the value of the theoretical reference signal that is presumed to specify the reference state of the controlled variable.

WM: Presumably not;

RM: Right. The reference value is a theoretical construct that explains the observed reference state of the controlled quantity, Qi.

WM: in fact of course, like you say Eetu, it doesn’t because if it did there would be no point in the organism controlling the Qi. So the question is whether one can truly say that the Qi is controlled if this is only the case because it is transformed into P which is compared with an internal RV continuously.

RM: You lost me.

WM: Maybe the fact that we can’t actually see control without taking a perceptual perspective is exactly why so many people find PCT hard to adopt.

RM: I don’t understand what you mean by “we can’t actually see control without taking a perceptual perspective”. It seems like I can see control by just looking to see whether a variable is maintained in a reference state, protected from disturbance. Is that the “perceptual perspective” of which you speak?

WM: Could we say the Qi is typically controlled but one can only make this judgement when the observer has an internal reference point to do so?

RM: No, we say that Qi is controlled when disturbances have far less of an effect on that variable than would be expected if the variable we not under control; that is, when Qi is kept in a constant or variable reference state, protected from disturbance. There is no theory involved in this determination. There is no need to know that this control occurs because there is a system present that is organized around control of the perception of the variable you see being controlled. This atheoretical approach to determining what variable is controlled is demonstrated in my Mind Reading demo (http://www.mindreadings.com/ControlDemo/Mindread.html), which is not only extremely entertaining but also highly educational.

Best

Rick

On 13 Feb 2017, at 07:42, Eetu Pikkarainen eetu.pikkarainen@oulu.fi wrote:

EP: If “control” means something like affecting a value of one (“controlled”) variable so that it becomes the same or near to the current value of an other variable (reference value) and / or remains the same or near to it, then anything in the environment is not controlled. And this has nothing to do with the question how well or badly the control is working. There just are no reference variables and thus reference values in the environment, they exists only inside the control system


Richard S. Marken

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
–Antoine de Saint-Exupery

On Mon, Feb 13, 2017 at 12:44 AM, Warren Mansell wmansell@gmail.com wrote:

Thanks Eetu!

ET: “EP: If “control” means something like affecting a value of one (“controlled”) variable so that it becomes the same or near to the current value of an other variable (reference value) and / or remains the same or near to it, then anything in the environment is not controlled. And this has nothing to do with the question how well or badly the control is working. There just are no reference variables and thus reference values in the environment, they exists only inside the control system.”

WM: Just like most people on this thread, of course I agree with you in the sense that you are articulating PCT. I am getting to see what this debate is getting at; Rick wants to identify control in the environment, separately from PCT, just like Bill at times, so that PCT has a phenomenon to explain. The problem is, once we unpack the nature of control, we see that it can only be defined accurately from an observer’s perspective. I think that Rick has been arguing that the Qi exists in the environment, but would he also propose that the reference value could exist in the environment? Presumably not; in fact of course, like you say Eetu, it doesn’t because if it did there would be no point in the organism controlling the Qi. So the question is whether one can truly say that the Qi is controlled if this is only the case because it is transformed into P which is compared with an internal RV continuously.

Maybe the fact that we can’t actually see control without taking a perceptual perspective is exactly why so many people find PCT hard to adopt. A part of the loop of control can be identified in the environment but it can’t be judged as controlled without taking a perceiver’s perspective. Like Rick I am not sure if using the word ‘stabilised’ helps at all - because often these components will not be stable if there is a changing reference value for its perception within the organism. Could we say the Qi is typically controlled but one can only make this judgement when the observer has an internal reference point to do so? Or is that a weak hybrid of the two positions?

Warren

On 13 Feb 2017, at 07:42, Eetu Pikkarainen eetu.pikkarainen@oulu.fi wrote:

EP: If “control” means something like affecting a value of one (“controlled”) variable so that it becomes the same or near to the current value of an other variable (reference value) and / or remains the same or near to it, then anything in the environment is not controlled. And this has nothing to do with the question how well or badly the control is working. There just are no reference variables and thus reference values in the environment, they exists only inside the control system


Richard S. Marken

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
–Antoine de Saint-Exupery

[From Rick Marken (2017.02.13.1820)]

···

On Mon, Feb 13, 2017 at 1:09 PM, Warren Mansell wmansell@gmail.com wrote:

WM: Hi Rick, thanks, I understand you and it is all getting clearer. I toyed with leaving it there, but I guess it’s best to check my understanding by trying to put it into my own words.Â

RM: Great.Â

WM: I can now see how a function of elements of the environment - the Qi - can be controlled.

RM: Qi is, itself, a function of the elements, which are the actual physical variables in the environment – the v’s that are usually not shown in diagrams of the PCT model, unfortunately. But that may be what you meant. I’m just clarifying.Â

WM: As humans we happen to extract this function through perception, but I can appreciate that the function and its reference state still ‘exists’ in a mathematical sense, that does not require a perceiver.

RM: I would rather say that the basis of this function exists; the variables from which the function can be computed. For example, in the “taste of lemonade” example, the combination of physical variables – amount of sugar (s), acid (a) and oil (o) – provides the possibility of the perception of  the taste of lemonade via the perceptual function b.1s+b.2a+b.3*o. It’s these variables – s,a and o – that exist in the environment; the mathematical function of these variables that is the perception of the taste of lemonade doesn’t ever exist in the environment, either as a possibility or as a reality.Â

WM: It is required to be perceived to actually implement the control

RM: This implies that perception is a process of perceiving an existing mathematical relationship between physical variables. But that’s not how perception works. Perception is the process of computing the mathematical function of the physical variables that is the perceptual variable.

RM: I think you may have read too much JJ Gibson, who talked about  perception being a process of “detecting invariants in the environment”. Gibson made some great observations but you learn that his theory of perception doesn’t really work as soon as you try to build a system that can do, for example, speaker independent speech recognition. You can’t build a speech recognition system that “detects” the “invariant” formant changes without building a pretty complex perceptual function that compute functions of the the changing sound amplitudes at different frequencies.

WM: which is why we look for theory in the form of PCT to explain how living things do it. Another theorist might insist that this function is calculated in advance and reacted to in an open loop fashion, for example.

RM: No, the PCT model of perception is required for both open and closed loop models. Any person who tries to build a model whose actions depend on perception will have to build perceptual functions like those described by PCT. The PCT model is required to account for the observation that organisms keep variable aspects of the environment in reference states. This phenomenon – the phenomenon of control – has simply not been noticed in conventional psychology, so these psychologists have not tried to develop models to account for it.Â

RM: PCT accounts for this phenomenon of control by assuming that organisms are in a closed negative feedback loop relationship with respect to perceptual variables that correspond to the aspects of the environment that are seen to be controlled. The behavior of such closed loop negative feedback systems  can be shown to be organized around control of these perceptual variables, keeping them equal to internal specifications (references) for the state of these perceptions, protected from the effects of disturbance.

Â

WM: It just so happens that if PCT is true then we are perceivers just as we are control systems doing all this scientific work and discussion.

RM: The fact that we are perceivers is certainly not unique to PCT. What is unique to PCT is that we control our perceptions, our perceptions don’t control us. What is unique about PCT is that it says behavior is controlled input, not caused output.Â

WM: For this reason I can see why there is a resistance to ‘the fact of control’ from many CSGers who have taken PCT on fully and are maybe not trying to test it against ‘facts’ like yourself. I am on the fence.

RM: Yes, I know. A lot of people come to PCT because they are basically post modernists who like the idea that “it’s all perception” which, to them, means “it’s all just a matter of opinion”. So my facts are as good as your facts. Sad;-)

WM: Could we turn this into an advantage and actually appreciate how one theory can traverse the realist empirical approach and the relativist, subjective approach to reality?

 RM: You can do whatever you like. But I’m not a fan of what you call the “subjective” approach to reality. I like the scientific approach, which recognizes that “i’s all perception” and realizes that the way to find out about the nature of the reality that is presumably the basis of our perceptual experience is to test theories of what that reality is like and then testing those theories by seeing if, when we manipulate various perceptions in certain ways, we perceive what the theories say we will perceive. PCT is one such theory of the nature of the reality the produces our perceptions of the controlling (purposeful behavior) done by living systems.

WM: The purposes of these two camps seem different to me, but they share more in common than proponents of other theories of behaviour I know of. Sorry if that last comment seems to let the heat out of the debate.

RM: Not at all. I’m happy to continue arguing that my camp – the scientific camp – is the only one that can make any progress in our understanding of reality;in our case, the reality that accounts for out perception of purposeful behavior. To me, the subjective approach is simply mental masturbation (I admit that masturbation can be fun but I prefer the real thing, at least as I recall; I’m no spring chicken;-)

WM: I’ve got you on this one Rick. Maybe I’ll take on your view of the power law next!Â

RM: I look forward to it!

Â

WM: I’m off to Narnia now… or maybe Andy’s house to hang out with Buzz…

RM: I never read the Naria books; too much religion for me. But I do like Andy and Buzz! And since I live in LaLa Land I had the pleasure of meeting Andy (Tom Hanks) a few years ago. A real mensch (him, of course, obviously not me;-).Â

Best

RickÂ

Â

Warren

On 13 Feb 2017, at 18:00, Richard Marken rsmarken@gmail.com wrote:

[From Rick Marken (2017.02.13.1000)]


Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery

On Mon, Feb 13, 2017 at 12:44 AM, Warren Mansell wmansell@gmail.com wrote:

WM: …I am getting to see what this debate is getting at; Rick wants to identify control in the environment, separately from PCT, just like Bill at times,

RM: Actually, all the time.

Â

WM: Â so that PCT has a phenomenon to explain.

RM: You make it sound like PCT was just sitting around looking for something to explain. In fact, the phenomenon of control was sitting around waiting for PCT to explain it. Again, you betray your “theory first” approach to PCT.

Â

WM: The problem is, once we unpack the nature of control, we see that it can only be defined accurately from an observer’s perspective.

RM: Doesn’t sound like a problem to me. Control is defined as the observation that a variable (such as the position of a cursor) is maintained in a reference state protected from disturbance. You can observe it for yourself in my “Nature of Control” demo (http://www.mindreadings.com/ControlDemo/BasicTrack.html)

WM: I think that Rick has been arguing that the Qi exists in the environment, but would he also propose that the reference value could exist in the environment?

RM: The reference state of Qi exists in the environment (actually, like Qi, it exists as a function of physical variables in the environment; as a perception in the observer); the term “reference value” refers to the value of the theoretical reference signal that is presumed to specify the reference state of the controlled variable.

Â

WM: Presumably not;

RM: Right. The reference value is a theoretical construct that explains the observed reference state of the controlled quantity, Qi.

WM: in fact of course, like you say Eetu, it doesn’t because if it did there would be no point in the organism controlling the Qi. So the question is whether one can truly say that the Qi is controlled if this is only the case because it is transformed into P which is compared with an internal RV continuously.

 RM: You lost me.Â

WM: Maybe the fact that we can’t actually see control without taking a perceptual perspective is exactly why so many people find PCT hard to adopt.

RM: I don’t understand what you mean by “we can’t actually see control without taking a perceptual perspective”. It seems like I can see control by just looking to see whether a variable is maintained in a reference state, protected from disturbance. Is that the “perceptual perspective” of which you speak?

Â

WM: Could we say the Qi is typically controlled but one can only make this judgement when the observer has an internal reference point to do so?

RM: No, we say that Qi is controlled when disturbances have far less of an effect on that variable than would be expected if the variable we not under control; that is, when Qi is kept in a constant or variable reference state, protected from disturbance. There is no theory involved in this determination. There is no need to know that this control occurs because there is a system present that is organized around control of the perception of the variable you see being controlled. This atheoretical approach to determining what variable is controlled is demonstrated in my Mind Reading demo (http://www.mindreadings.com/ControlDemo/Mindread.html), which is not only extremely entertaining but also highly educational.Â

BestÂ

Rick

Â

On 13 Feb 2017, at 07:42, Eetu Pikkarainen eetu.pikkarainen@oulu.fi wrote:

EP: If “control” means something like affecting a value of one (“controlled”) variable so that it becomes the same or near to the current value of an other variable (reference value) and / or remains the same or near to it, then anything in the environment is not controlled. And this has nothing to do with the question how well or badly the control is working. There just are no reference variables and thus reference values in the environment, they exists only inside the control system


Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery

On Mon, Feb 13, 2017 at 12:44 AM, Warren Mansell wmansell@gmail.com wrote:

Thanks Eetu!

ET: “EP: If “control” means something like affecting a value of one (“controlled”) variable so that it becomes the same or near to the current value of an other variable (reference value) and / or remains the same or near to it, then anything in the environment is not controlled. And this has nothing to do with the question how well or badly the control is working. There just are no reference variables and thus reference values in the environment, they exists only inside the control system.”

WM: Just like most people on this thread, of course I agree with you in the sense that you are articulating PCT. I am getting to see what this debate is getting at; Rick wants to identify control in the environment, separately from PCT, just like Bill at times, so that PCT has a phenomenon to explain. The problem is, once we unpack the nature of control, we see that it can only be defined accurately from an observer’s perspective. I think that Rick has been arguing that the Qi exists in the environment, but would he also propose that the reference value could exist in the environment? Presumably not; in fact of course, like you say Eetu, it doesn’t because if it did there would be no point in the organism controlling the Qi. So the question is whether one can truly say that the Qi is controlled if this is only the case because it is transformed into P which is compared with an internal RV continuously.

Maybe the fact that we can’t actually see control without taking a perceptual perspective is exactly why so many people find PCT hard to adopt. A part of the loop of control can be identified in the environment but it can’t be judged as controlled without taking a perceiver’s perspective. Like Rick I am not sure if using the word ‘stabilised’ helps at all - because often these components will not be stable if there is a changing reference value for its perception within the organism. Could we say the Qi is typically controlled but one can only make this judgement when the observer has an internal reference point to do so? Or is that a weak hybrid of the two positions?

Warren

On 13 Feb 2017, at 07:42, Eetu Pikkarainen eetu.pikkarainen@oulu.fi wrote:

EP: If “control” means something like affecting a value of one (“controlled”) variable so that it becomes the same or near to the current value of an other variable (reference value) and / or remains the same or near to it, then anything in the environment is not controlled. And this has nothing to do with the question how well or badly the control is working. There just are no reference variables and thus reference values in the environment, they exists only inside the control system


Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery

Hi Warren…

···

From: Richard Marken [mailto:rsmarken@gmail.com]
Sent: Tuesday, February 14, 2017 3:23 AM
To: csgnet@lists.illinois.edu
Subject: Re: VS: To accept a theory (was Re: Control Is)

[From Rick Marken (2017.02.13.1820)]

On Mon, Feb 13, 2017 at 1:09 PM, Warren Mansell wmansell@gmail.com wrote:

WM: Hi Rick, thanks, I understand you and it is all getting clearer. I toyed with leaving it there, but I guess it’s best to check my understanding by trying to put it into my own words.

HB : Yes. Put in your words not Ricks. If you’ll listen to Rick nothing will get more clear. You’ll just get dipper in the mud.

RM: Great.

WM: I can now see how a function of elements of the environment - the Qi - can be controlled.

RM: Qi is, itself, a function of the elements, which are the actual physical variables in the environment – the v’s that are usually not shown in diagrams of the PCT model, unfortunately. But that may be what you meant. I’m just clarifying.

HBÂ : Rick what can I say. I’ve never saw so much bullshit on one place. You don’t understand PCT. It would be better if you don’t clarify.

WM: As humans we happen to extract this function through perception, but I can appreciate that the function and its reference state still ‘exists’ in a mathematical sense, that does not require a perceiver.

RM: I would rather say that the basis of this function exists; the variables from which the function can be computed. For example, in the “taste of lemonade” example, the combination of physical variables – amount of sugar (s), acid (a) and oil (o) – provides the possibility of the perception of the taste of lemonade via the perceptual function b.1s+b.2a+b.3*o. It’s these variables – s,a and o – that exist in the environment; the mathematical function of these variables that is the perception of the taste of lemonade doesn’t ever exist in the environment, either as a possibility or as a reality.

HB : And where do you suppose is this »mathematical funtion« hidden in perceptual signal ? Provides a possibility ?? On which level of hierarchy ???

WM: It is required to be perceived to actually implement the control

RM: This implies that perception is a process of perceiving an existing mathematical relationship between physical variables.

HB : All the mathematical relatioship are just in your head. There is nothing outside in mathematical form. Can you imagine snake or elephant perceiving »mathematical relationship«….???

RM : But that’s not how perception works. Perception is the process of computing the mathematical function of the physical variables that is the perceptual variable.

HB : What a mess…. Perception is the process of computing … ha,haâ€a… you ara a clown Rick… Get on real work and read B:CP what is peerceptual signal and how it looks like. There is much more nonsense in continuing parts so I’ll finnish. here. Incredible. What an imagination….

Best,

Boris

RM: I think you may have read too much JJ Gibson, who talked about perception being a process of “detecting invariants in the environment”. Gibson made some great observations but you learn that his theory of perception doesn’t really work as soon as you try to build a system that can do, for example, speaker independent speech recognition. You can’t build a speech recognition system that “detects” the “invariant” formant changes without building a pretty complex perceptual function that compute functions of the the changing sound amplitudes at different frequencies.

WM: which is why we look for theory in the form of PCT to explain how living things do it. Another theorist might insist that this function is calculated in advance and reacted to in an open loop fashion, for example.

RM: No, the PCT model of perception is required for both open and closed loop models. Any person who tries to build a model whose actions depend on perception will have to build perceptual functions like those described by PCT. The PCT model is required to account for the observation that organisms keep variable aspects of the environment in reference states. This phenomenon – the phenomenon of control – has simply not been noticed in conventional psychology, so these psychologists have not tried to develop models to account for it.

RM: PCT accounts for this phenomenon of control by assuming that organisms are in a closed negative feedback loop relationship with respect to perceptual variables that correspond to the aspects of the environment that are seen to be controlled. The behavior of such closed loop negative feedback systems can be shown to be organized around control of these perceptual variables, keeping them equal to internal specifications (references) for the state of these perceptions, protected from the effects of disturbance.

WM: It just so happens that if PCT is true then we are perceivers just as we are control systems doing all this scientific work and discussion.

RM: The fact that we are perceivers is certainly not unique to PCT. What is unique to PCT is that we control our perceptions, our perceptions don’t control us. What is unique about PCT is that it says behavior is controlled input, not caused output.

WM: For this reason I can see why there is a resistance to ‘the fact of control’ from many CSGers who have taken PCT on fully and are maybe not trying to test it against ‘facts’ like yourself. I am on the fence.

RM: Yes, I know. A lot of people come to PCT because they are basically post modernists who like the idea that “it’s all perception” which, to them, means “it’s all just a matter of opinion”. So my facts are as good as your facts. Sad;-)

WM: Could we turn this into an advantage and actually appreciate how one theory can traverse the realist empirical approach and the relativist, subjective approach to reality?

RM: You can do whatever you like. But I’m not a fan of what you call the “subjective” approach to reality. I like the scientific approach, which recognizes that “i’s all perception” and realizes that the way to find out about the nature of the reality that is presumably the basis of our perceptual experience is to test theories of what that reality is like and then testing those theories by seeing if, when we manipulate various perceptions in certain ways, we perceive what the theories say we will perceive. PCT is one such theory of the nature of the reality the produces our perceptions of the controlling (purposeful behavior) done by living systems.

WM: The purposes of these two camps seem different to me, but they share more in common than proponents of other theories of behaviour I know of. Sorry if that last comment seems to let the heat out of the debate.

RM: Not at all. I’m happy to continue arguing that my camp – the scientific camp – is the only one that can make any progress in our understanding of reality;in our case, the reality that accounts for out perception of purposeful behavior. To me, the subjective approach is simply mental masturbation (I admit that masturbation can be fun but I prefer the real thing, at least as I recall; I’m no spring chicken;-)

WM: I’ve got you on this one Rick. Maybe I’ll take on your view of the power law next!

RM: I look forward to it!

WM: I’m off to Narnia now… or maybe Andy’s house to hang out with Buzz…

RM: I never read the Naria books; too much religion for me. But I do like Andy and Buzz! And since I live in LaLa Land I had the pleasure of meeting Andy (Tom Hanks) a few years ago. A real mensch (him, of course, obviously not me;-).

Best

Rick

Warren

On 13 Feb 2017, at 18:00, Richard Marken rsmarken@gmail.com wrote:

[From Rick Marken (2017.02.13.1000)]

On Mon, Feb 13, 2017 at 12:44 AM, Warren Mansell wmansell@gmail.com wrote:

WM: …I am getting to see what this debate is getting at; Rick wants to identify control in the environment, separately from PCT, just like Bill at times,

RM: Actually, all the time.

WM: so that PCT has a phenomenon to explain.

RM: You make it sound like PCT was just sitting around looking for something to explain. In fact, the phenomenon of control was sitting around waiting for PCT to explain it. Again, you betray your “theory first” approach to PCT.

WM: The problem is, once we unpack the nature of control, we see that it can only be defined accurately from an observer’s perspective.

RM: Doesn’t sound like a problem to me. Control is defined as the observation that a variable (such as the position of a cursor) is maintained in a reference state protected from disturbance. You can observe it for yourself in my “Nature of Control” demo (http://www.mindreadings.com/ControlDemo/BasicTrack.html)

WM: I think that Rick has been arguing that the Qi exists in the environment, but would he also propose that the reference value could exist in the environment?

RM: The reference state of Qi exists in the environment (actually, like Qi, it exists as a function of physical variables in the environment; as a perception in the observer); the term “reference value” refers to the value of the theoretical reference signal that is presumed to specify the reference state of the controlled variable.

WM: Presumably not;

RM: Right. The reference value is a theoretical construct that explains the observed reference state of the controlled quantity, Qi.

WM: in fact of course, like you say Eetu, it doesn’t because if it did there would be no point in the organism controlling the Qi. So the question is whether one can truly say that the Qi is controlled if this is only the case because it is transformed into P which is compared with an internal RV continuously.

RM: You lost me.

WM: Maybe the fact that we can’t actually see control without taking a perceptual perspective is exactly why so many people find PCT hard to adopt.

RM: I don’t understand what you mean by “we can’t actually see control without taking a perceptual perspective”. It seems like I can see control by just looking to see whether a variable is maintained in a reference state, protected from disturbance. Is that the “perceptual perspective” of which you speak?

WM: Could we say the Qi is typically controlled but one can only make this judgement when the observer has an internal reference point to do so?

RM: No, we say that Qi is controlled when disturbances have far less of an effect on that variable than would be expected if the variable we not under control; that is, when Qi is kept in a constant or variable reference state, protected from disturbance. There is no theory involved in this determination. There is no need to know that this control occurs because there is a system present that is organized around control of the perception of the variable you see being controlled. This atheoretical approach to determining what variable is controlled is demonstrated in my Mind Reading demo (http://www.mindreadings.com/ControlDemo/Mindread.html), which is not only extremely entertaining but also highly educational.

Best

Rick

On 13 Feb 2017, at 07:42, Eetu Pikkarainen eetu.pikkarainen@oulu.fi wrote:

EP: If “control” means something like affecting a value of one (“controlled”) variable so that it becomes the same or near to the current value of an other variable (reference value) and / or remains the same or near to it, then anything in the environment is not controlled. And this has nothing to do with the question how well or badly the control is working. There just are no reference variables and thus reference values in the environment, they exists only inside the control system

Richard S. Marken

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
–Antoine de Saint-Exupery

On Mon, Feb 13, 2017 at 12:44 AM, Warren Mansell wmansell@gmail.com wrote:

Thanks Eetu!

ET: “EP: If “control” means something like affecting a value of one (“controlled”) variable so that it becomes the same or near to the current value of an other variable (reference value) and / or remains the same or near to it, then anything in the environment is not controlled. And this has nothing to do with the question how well or badly the control is working. There just are no reference variables and thus reference values in the environment, they exists only inside the control system.”

WM: Just like most people on this thread, of course I agree with you in the sense that you are articulating PCT. I am getting to see what this debate is getting at; Rick wants to identify control in the environment, separately from PCT, just like Bill at times, so that PCT has a phenomenon to explain. The problem is, once we unpack the nature of control, we see that it can only be defined accurately from an observer’s perspective. I think that Rick has been arguing that the Qi exists in the environment, but would he also propose that the reference value could exist in the environment? Presumably not; in fact of course, like you say Eetu, it doesn’t because if it did there would be no point in the organism controlling the Qi. So the question is whether one can truly say that the Qi is controlled if this is only the case because it is transformed into P which is compared with an internal RV continuously.
Maybe the fact that we can’t actually see control without taking a perceptual perspective is exactly why so many people find PCT hard to adopt. A part of the loop of control can be identified in the environment but it can’t be judged as controlled without taking a perceiver’s perspective. Like Rick I am not sure if using the word ‘stabilised’ helps at all - because often these components will not be stable if there is a changing reference value for its perception within the organism. Could we say the Qi is typically controlled but one can only make this judgement when the observer has an internal reference point to do so? Or is that a weak hybrid of the two positions?
Warren

On 13 Feb 2017, at 07:42, Eetu Pikkarainen eetu.pikkarainen@oulu.fi wrote:

EP: If “control” means something like affecting a value of one (“controlled”) variable so that it becomes the same or near to the current value of an other variable (reference value) and / or remains the same or near to it, then anything in the environment is not controlled. And this has nothing to do with the question how well or badly the control is working. There just are no reference variables and thus reference values in the environment, they exists only inside the control system

Richard S. Marken

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
–Antoine de Saint-Exupery

Richard S. Marken

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
–Antoine de Saint-Exupery

Hi Warren…

/p>

···

[From Rick Marken (2017.02.13.1820)]

On Mon, Feb 13, 2017 at 1:09 PM, Warren Mansell wmansell@gmail.com wrote:

WM: Hi Rick, thanks, I understand you and it is all getting clearer. I toyed with leaving it there, but I guess it’s best to check my understanding by trying to put it into my own words.

HB : Yes. Put in your words not Ricks. If you’ll listen to Rick nothing will get more clear. You’ll just get dipper in the mud.

RM: Great.

WM: I can now see how a function of elements of the environment - the Qi - can be controlled.

RM: Qi is, itself, a function of the elements, which are the actual physical variables in the environment – the v’s that are usually not shown in diagrams of the PCT model, unfortunately. But that may be what you meant. I’m just clarifying.

HB : Rick what can I say. I’ve never saw so much bullshit on one place. You don’t understand PCT. It would be better if you don’t clarify.

WM: As humans we happen to extract this function through perception, but I can appreciate that the function and its reference state still ‘exists’ in a mathematical sense, that does not require a perceiver.

RM: I would rather say that the basis of this function exists; the variables from which the function can be computed. For example, in the “taste of lemonade” example, the combination of physical variables – amount of sugar (s), acid (a) and oil (o) – provides the possibility of the perception of the taste of lemonade via the perceptual function b.1s+b.2a+b.3*o. It’s these variables – s,a and o – that exist in the environment; the mathematical function of these variables that is the perception of the taste of lemonade doesn’t ever exist in the environment, either as a possibility or as a reality.

HB : And where do you suppose is this »mathematical funtion« hidden in perceptual signal ? Provides a possibility ?? On which level of hierarchy ???

WM: It is required to be perceived to actually implement the control

RM: This implies that perception is a process of perceiving an existing mathematical relationship between physical variables.

HB : All the mathematical relatioship are just in your head. There is nothing outside in mathematical form. Can you imagine snake or elephant perceiving »mathematical relationship«….???

RM : But that’s not how perception works. Perception is the process of computing the mathematical function of the physical variables that is the perceptual variable.

HB : What a mess…. Perception is the process of computing … ha,ha… you ara a clown Rick… Get on rean real work and read B:CP what is perceptual signal and how it looks like. There is much more nonsense in continuing parts so I’ll finnish. here. Incredible. What an imagination….

Best,

Boris

RM: I think you may have read too much JJ Gibson, who talked about perception being a process of “detecting invariants in the environment”. Gibson made some great observations but you learn that his theory of perception doesn’t really work as soon as you try to build a system that can do, for example, speaker independent speech recognition. You can’t build a speech recognition system that “detects” the “invariant” formant changes without building a pretty complex perceptual function that compute functions of the the changing sound amplitudes at different frequencies.

WM: which is why we look for theory in the form of PCT to explain how living things do it. Another theorist might insist that this function is calculated in advance and reacted to in an open loop fashion, for example.

RM: No, the PCT model of perception is required for both open and closed loop models. Any person who tries to build a model whose actions depend on perception will have to build perceptual functions like those described by PCT. The PCT model is required to account for the observation that organisms keep variable aspects of the environment in reference states. This phenomenon – the phenomenon of control – has simply not been noticed in conventional psychology, so these psychologists have not tried to develop models to account for it.

RM: PCT accounts for this phenomenon of control by assuming that organisms are in a closed negative feedback loop relationship with respect to perceptual variables that correspond to the aspects of the environment that are seen to be controlled. The behavior of such closed loop negative feedback systems can be shown to be organized around control of these perceptual variables, keeping them equal to internal specifications (references) for the state of these perceptions, protected from the effects of disturbance.

WM: It just so happens that if PCT is true then we are perceivers just as we are control systems doing all this scientific work and discussion.

RM: The fact that we are perceivers is certainly not unique to PCT. What is unique to PCT is that we control our perceptions, our perceptions don’t control us. What is unique about PCT is that it says behavior is controlled input, not caused output.

WM: For this reason I can see why there is a resistance to ‘the fact of control’ from many CSGers who have taken PCT on fully and are maybe not trying to test it against ‘facts’ like yourself. I am on the fence.

RM: Yes, I know. A lot of people come to PCT because they are basically post modernists who like the idea that “it’s all perception” which, to them, means “it’s all just a matter of opinion”. So my facts are as good as your facts. Sad;-)

WM: Could we turn this into an advantage and actually appreciate how one theory can traverse the realist empirical approach and the relativist, subjective approach to reality?

RM: You can do whatever you like. But I’m not a fan of what you call the “subjective” approach to reality. I like the scientific approach, which recognizes that “i’s all perception” and realizes that the way to find out about the nature of the reality that is presumably the basis of our perceptual experience is to test theories of what that reality is like and then testing those theories by seeing if, when we manipulate various perceptions in certain ways, we perceive what the theories say we will perceive. PCT is one such theory of the nature of the reality the produces our perceptions of the controlling (purposeful behavior) done by living systems.

WM: The purposes of these two camps seem different to me, but they share more in common than proponents of other theories of behaviour I know of. Sorry if that last comment seems to let the heat out of the debate.

RM: Not at all. I’m happy to continue arguing that my camp – the scientific camp – is the only one that can make any progress in our understanding of reality;in our case, the reality that accounts for out perception of purposeful behavior. To me, the subjective approach is simply mental masturbation (I admit that masturbation can be fun but I prefer the real thing, at least as I recall; I’m no spring chicken;-)

WM: I’ve got you on this one Rick. Maybe I’ll take on your view of the power law next!

RM: I look forward to it!

WM: I’m off to Narnia now… or maybe Andy’s house to hang out with Buzz…

RM: I never read the Naria books; too much religion for me. But I do like Andy and Buzz! And since I live in LaLa Land I had the pleasure of meeting Andy (Tom Hanks) a few years ago. A real mensch (him, of course, obviously not me;-).

Best

Rick

Warren

On 13 Feb 2017, at 18:00, Richard Marken rsmarken@gmail.com wrote:

[From Rick Marken (2017.02.13.1000)]

On Mon, Feb 13, 2017 at 12:44 AM, Warren Mansell wmansell@gmail.com wrote:

WM: …I am getting to see what this debate is getting at; Rick wants to identify control in the environment, separately from PCT, just like Bill at times,

RM: Actually, all the time.

WM: so that PCT has a phenomenon to explain.

RM: You make it sound like PCT was just sitting around looking for something to explain. In fact, the phenomenon of control was sitting around waiting for PCT to explain it. Again, you betray your “theory first” approach to PCT.

WM: The problem is, once we unpack the nature of control, we see that it can only be defined accurately from an observer’s perspective.

RM: Doesn’t sound like a problem to me. Control is defined as the observation that a variable (such as the position of a cursor) is maintained in a reference state protected from disturbance. You can observe it for yourself in my “Nature of Control” demo (http://www.mindreadings.com/ControlDemo/BasicTrack.html)

WM: I think that Rick has been arguing that the Qi exists in the environment, but would he also propose that the reference value could exist in the environment?

RM: The reference state of Qi exists in the environment (actually, like Qi, it exists as a function of physical variables in the environment; as a perception in the observer); the term “reference value” refers to the value of the theoretical reference signal that is presumed to specify the reference state of the controlled variable.

WM: Presumably not;

RM: Right. The reference value is a theoretical construct that explains the observed reference state of the controlled quantity, Qi.

WM: in fact of course, like you say Eetu, it doesn’t because if it did there would be no point in the organism controlling the Qi. So the question is whether one can truly say that the Qi is controlled if this is only the case because it is transformed into P which is compared with an internal RV continuously.

RM: You lost me.

WM: Maybe the fact that we can’t actually see control without taking a perceptual perspective is exactly why so many people find PCT hard to adopt.

RM: I don’t understand what you mean by “we can’t actually see control without taking a perceptual perspective”. It seems like I can see control by just looking to see whether a variable is maintained in a reference state, protected from disturbance. Is that the “perceptual perspective” of which you speak?

WM: Could we say the Qi is typically controlled but one can only make this judgement when the observer has an internal reference point to do so?

RM: No, we say that Qi is controlled when disturbances have far less of an effect on that variable than would be expected if the variable we not under control; that is, when Qi is kept in a constant or variable reference state, protected from disturbance. There is no theory involved in this determination. There is no need to know that this control occurs because there is a system present that is organized around control of the perception of the variable you see being controlled. This atheoretical approach to determining what variable is controlled is demonstrated in my Mind Reading demo (http://www.mindreadings.com/ControlDemo/Mindread.html), which is not only extremely entertaining but also highly educational.

Best

Rick

On 13 Feb 2017, at 07:42, Eetu Pikkarainen eetu.pikkarainen@oulu.fi wrote:

EP: If “control” means something like affecting a value of one (“controlled”) variable so that it becomes the same or near to the current value of an other variable (reference value) and / or remains the same or near to it, then anything in the environment is not controlled. And this has nothing to do with the question how well or badly the control is working. There just are no reference variables and thus reference values in the environment, they exists only inside the control system

Richard S. Marken

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
–Antoine de Saint-Exupery

On Mon, Feb 13, 2017 at 12:44 AM, Warren Mansell wmansell@gmail.com wrote:

Thanks Eetu!

ET: “EP: If “control” means something like affecting a value of one (“controlled”) variable so that it becomes the same or near to the current value of an other variable (reference value) and / or remains the same or near to it, then anything in the environment is not controlled. And this has nothing to do with the question how well or badly the control is working. There just are no reference variables and thus reference values in the environment, they exists only inside the control system.”

WM: Just like most people on this thread, of course I agree with you in the sense that you are articulating PCT. I am getting to see what this debate is getting at; Rick wants to identify control in the environment, separately from PCT, just like Bill at times, so that PCT has a phenomenon to explain. The problem is, once we unpack the nature of control, we see that it can only be defined accurately from an observer’s perspective. I think that Rick has been arguing that the Qi exists in the environment, but would he also propose that the reference value could exist in the environment? Presumably not; in fact of course, like you say Eetu, it doesn’t because if it did there would be no point in the organism controlling the Qi. So the question is whether one can truly say that the Qi is controlled if this is only the case because it is transformed into P which is compared with an internal RV continuously.
Maybe the fact that we can’t actually see control without taking a perceptual perspective is exactly why so many people find PCT hard to adopt. A part of the loop of control can be identified in the environment but it can’t be judged as controlled without taking a perceiver’s perspective. Like Rick I am not sure if using the word ‘stabilised’ helps at all - because often these components will not be stable if there is a changing reference value for its perception within the organism. Could we say the Qi is typically controlled but one can only make this judgement when the observer has an internal reference point to do so? Or is that a weak hybrid of the two positions?
Warren

On 13 Feb 2017, at 07:42, Eetu Pikkarainen eetu.pikkarainen@oulu.fi wrote:

EP: If “control” means something like affecting a value of one (“controlled”) variable so that it becomes the same or near to the current value of an other variable (reference value) and / or remains the same or near to it, then anything in the environment is not controlled. And this has nothing to do with the question how well or badly the control is working. There just are no reference variables and thus reference values in the environment, they exists only inside the control system

Richard S. Marken

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
–Antoine de Saint-Exupery

Richard S. Marken

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
–Antoine de Saint-Exupery

Hi Warren…

/p>

···

[From Rick Marken (2017.02.13.1820)]

On Mon, Feb 13, 2017 at 1:09 PM, Warren Mansell wmansell@gmail.com wrote:

WM: Hi Rick, thanks, I understand you and it is all getting clearer. I toyed with leaving it there, but I guess it’s best to check my understanding by trying to put it into my own words.

HB : Yes. Put in your words not Ricks. If you’ll listen to Rick nothing will get more clear. You’ll just get dipper in the mud.

RM: Great.

WM: I can now see how a function of elements of the environment - the Qi - can be controlled.

RM: Qi is, itself, a function of the elements, which are the actual physical variables in the environment – the v’s that are usually not shown in diagrams of the PCT model, unfortunately. But that may be what you meant. I’m just clarifying.

HB : Rick what can I say. I’ve never saw so much bullshit on one place. You don’t understand PCT. It would be better if you don’t clarify.

WM: As humans we happen to extract this function through perception, but I can appreciate that the function and its reference state still ‘exists’ in a mathematical sense, that does not require a perceiver.

RM: I would rather say that the basis of this function exists; the variables from which the function can be computed. For example, in the “taste of lemonade” example, the combination of physical variables – amount of sugar (s), acid (a) and oil (o) – provides the possibility of the perception of the taste of lemonade via the perceptual function b.1s+b.2a+b.3*o. It’s these variables – s,a and o – that exist in the environment; the mathematical function of these variables that is the perception of the taste of lemonade doesn’t ever exist in the environment, either as a possibility or as a reality.

HB : And where do you suppose is this »mathematical funtion« hidden in perceptual signal ? Provides a possibility ?? On which level of hierarchy ???

WM: It is required to be perceived to actually implement the control

RM: This implies that perception is a process of perceiving an existing mathematical relationship between physical variables.

HB : All the mathematical relatioship are just in your head. There is nothing outside in mathematical form. Can you imagine snake or elephant perceiving »mathematical relationship«….???

RM : But that’s not how perception works. Perception is the process of computing the mathematical function of the physical variables that is the perceptual variable.

HB : What a mess…. Perception is the process of computing … ha,ha… you ara a clown Rick… Get on rean real work and read B:CP what is perceptual signal and how it looks like. There is much more nonsense in continuing parts so I’ll finnish. here. Incredible. What an imagination….

Best,

Boris

RM: I think you may have read too much JJ Gibson, who talked about perception being a process of “detecting invariants in the environment”. Gibson made some great observations but you learn that his theory of perception doesn’t really work as soon as you try to build a system that can do, for example, speaker independent speech recognition. You can’t build a speech recognition system that “detects” the “invariant” formant changes without building a pretty complex perceptual function that compute functions of the the changing sound amplitudes at different frequencies.

WM: which is why we look for theory in the form of PCT to explain how living things do it. Another theorist might insist that this function is calculated in advance and reacted to in an open loop fashion, for example.

RM: No, the PCT model of perception is required for both open and closed loop models. Any person who tries to build a model whose actions depend on perception will have to build perceptual functions like those described by PCT. The PCT model is required to account for the observation that organisms keep variable aspects of the environment in reference states. This phenomenon – the phenomenon of control – has simply not been noticed in conventional psychology, so these psychologists have not tried to develop models to account for it.

RM: PCT accounts for this phenomenon of control by assuming that organisms are in a closed negative feedback loop relationship with respect to perceptual variables that correspond to the aspects of the environment that are seen to be controlled. The behavior of such closed loop negative feedback systems can be shown to be organized around control of these perceptual variables, keeping them equal to internal specifications (references) for the state of these perceptions, protected from the effects of disturbance.

WM: It just so happens that if PCT is true then we are perceivers just as we are control systems doing all this scientific work and discussion.

RM: The fact that we are perceivers is certainly not unique to PCT. What is unique to PCT is that we control our perceptions, our perceptions don’t control us. What is unique about PCT is that it says behavior is controlled input, not caused output.

WM: For this reason I can see why there is a resistance to ‘the fact of control’ from many CSGers who have taken PCT on fully and are maybe not trying to test it against ‘facts’ like yourself. I am on the fence.

RM: Yes, I know. A lot of people come to PCT because they are basically post modernists who like the idea that “it’s all perception” which, to them, means “it’s all just a matter of opinion”. So my facts are as good as your facts. Sad;-)

WM: Could we turn this into an advantage and actually appreciate how one theory can traverse the realist empirical approach and the relativist, subjective approach to reality?

RM: You can do whatever you like. But I’m not a fan of what you call the “subjective” approach to reality. I like the scientific approach, which recognizes that “i’s all perception” and realizes that the way to find out about the nature of the reality that is presumably the basis of our perceptual experience is to test theories of what that reality is like and then testing those theories by seeing if, when we manipulate various perceptions in certain ways, we perceive what the theories say we will perceive. PCT is one such theory of the nature of the reality the produces our perceptions of the controlling (purposeful behavior) done by living systems.

WM: The purposes of these two camps seem different to me, but they share more in common than proponents of other theories of behaviour I know of. Sorry if that last comment seems to let the heat out of the debate.

RM: Not at all. I’m happy to continue arguing that my camp – the scientific camp – is the only one that can make any progress in our understanding of reality;in our case, the reality that accounts for out perception of purposeful behavior. To me, the subjective approach is simply mental masturbation (I admit that masturbation can be fun but I prefer the real thing, at least as I recall; I’m no spring chicken;-)

WM: I’ve got you on this one Rick. Maybe I’ll take on your view of the power law next!

RM: I look forward to it!

WM: I’m off to Narnia now… or maybe Andy’s house to hang out with Buzz…

RM: I never read the Naria books; too much religion for me. But I do like Andy and Buzz! And since I live in LaLa Land I had the pleasure of meeting Andy (Tom Hanks) a few years ago. A real mensch (him, of course, obviously not me;-).

Best

Rick

Warren

On 13 Feb 2017, at 18:00, Richard Marken rsmarken@gmail.com wrote:

[From Rick Marken (2017.02.13.1000)]

On Mon, Feb 13, 2017 at 12:44 AM, Warren Mansell wmansell@gmail.com wrote:

WM: …I am getting to see what this debate is getting at; Rick wants to identify control in the environment, separately from PCT, just like Bill at times,

RM: Actually, all the time.

WM: so that PCT has a phenomenon to explain.

RM: You make it sound like PCT was just sitting around looking for something to explain. In fact, the phenomenon of control was sitting around waiting for PCT to explain it. Again, you betray your “theory first” approach to PCT.

WM: The problem is, once we unpack the nature of control, we see that it can only be defined accurately from an observer’s perspective.

RM: Doesn’t sound like a problem to me. Control is defined as the observation that a variable (such as the position of a cursor) is maintained in a reference state protected from disturbance. You can observe it for yourself in my “Nature of Control” demo (http://www.mindreadings.com/ControlDemo/BasicTrack.html)

WM: I think that Rick has been arguing that the Qi exists in the environment, but would he also propose that the reference value could exist in the environment?

RM: The reference state of Qi exists in the environment (actually, like Qi, it exists as a function of physical variables in the environment; as a perception in the observer); the term “reference value” refers to the value of the theoretical reference signal that is presumed to specify the reference state of the controlled variable.

WM: Presumably not;

RM: Right. The reference value is a theoretical construct that explains the observed reference state of the controlled quantity, Qi.

WM: in fact of course, like you say Eetu, it doesn’t because if it did there would be no point in the organism controlling the Qi. So the question is whether one can truly say that the Qi is controlled if this is only the case because it is transformed into P which is compared with an internal RV continuously.

RM: You lost me.

WM: Maybe the fact that we can’t actually see control without taking a perceptual perspective is exactly why so many people find PCT hard to adopt.

RM: I don’t understand what you mean by “we can’t actually see control without taking a perceptual perspective”. It seems like I can see control by just looking to see whether a variable is maintained in a reference state, protected from disturbance. Is that the “perceptual perspective” of which you speak?

WM: Could we say the Qi is typically controlled but one can only make this judgement when the observer has an internal reference point to do so?

RM: No, we say that Qi is controlled when disturbances have far less of an effect on that variable than would be expected if the variable we not under control; that is, when Qi is kept in a constant or variable reference state, protected from disturbance. There is no theory involved in this determination. There is no need to know that this control occurs because there is a system present that is organized around control of the perception of the variable you see being controlled. This atheoretical approach to determining what variable is controlled is demonstrated in my Mind Reading demo (http://www.mindreadings.com/ControlDemo/Mindread.html), which is not only extremely entertaining but also highly educational.

Best

Rick

On 13 Feb 2017, at 07:42, Eetu Pikkarainen eetu.pikkarainen@oulu.fi wrote:

EP: If “control” means something like affecting a value of one (“controlled”) variable so that it becomes the same or near to the current value of an other variable (reference value) and / or remains the same or near to it, then anything in the environment is not controlled. And this has nothing to do with the question how well or badly the control is working. There just are no reference variables and thus reference values in the environment, they exists only inside the control system

Richard S. Marken

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
–Antoine de Saint-Exupery

On Mon, Feb 13, 2017 at 12:44 AM, Warren Mansell wmansell@gmail.com wrote:

Thanks Eetu!

ET: “EP: If “control” means something like affecting a value of one (“controlled”) variable so that it becomes the same or near to the current value of an other variable (reference value) and / or remains the same or near to it, then anything in the environment is not controlled. And this has nothing to do with the question how well or badly the control is working. There just are no reference variables and thus reference values in the environment, they exists only inside the control system.”

WM: Just like most people on this thread, of course I agree with you in the sense that you are articulating PCT. I am getting to see what this debate is getting at; Rick wants to identify control in the environment, separately from PCT, just like Bill at times, so that PCT has a phenomenon to explain. The problem is, once we unpack the nature of control, we see that it can only be defined accurately from an observer’s perspective. I think that Rick has been arguing that the Qi exists in the environment, but would he also propose that the reference value could exist in the environment? Presumably not; in fact of course, like you say Eetu, it doesn’t because if it did there would be no point in the organism controlling the Qi. So the question is whether one can truly say that the Qi is controlled if this is only the case because it is transformed into P which is compared with an internal RV continuously.
Maybe the fact that we can’t actually see control without taking a perceptual perspective is exactly why so many people find PCT hard to adopt. A part of the loop of control can be identified in the environment but it can’t be judged as controlled without taking a perceiver’s perspective. Like Rick I am not sure if using the word ‘stabilised’ helps at all - because often these components will not be stable if there is a changing reference value for its perception within the organism. Could we say the Qi is typically controlled but one can only make this judgement when the observer has an internal reference point to do so? Or is that a weak hybrid of the two positions?
Warren

On 13 Feb 2017, at 07:42, Eetu Pikkarainen eetu.pikkarainen@oulu.fi wrote:

EP: If “control” means something like affecting a value of one (“controlled”) variable so that it becomes the same or near to the current value of an other variable (reference value) and / or remains the same or near to it, then anything in the environment is not controlled. And this has nothing to do with the question how well or badly the control is working. There just are no reference variables and thus reference values in the environment, they exists only inside the control system

Richard S. Marken

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
–Antoine de Saint-Exupery

Richard S. Marken

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
–Antoine de Saint-Exupery

[From Rick Marken (2017.02.14.0910)]

···

On Tue, Feb 14, 2017 at 1:42 AM, Warren Mansell wmansell@gmail.com wrote:

WM: So Rick, step at a time, you are saying that the Qi is the function of environmental variables that is being controlled. It is a represents a possibly complex relationship between physical variables that we note is held close to a reference state. All of this is in the environment and is controlled simultaneously as perception corresponding to this function is controlled within the organism…

RM: Yes. Â

PS Tom Hanks played Woody…

RM: Well, at least I knew he played someone;-)

BestÂ

RickÂ


Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery

[From Rick Marken (2017.02.14.1010)]

···

On Tue, Feb 14, 2017 at 1:57 AM, Warren Mansell wmansell@gmail.com wrote:

WM: Sorry Rick, I needed a follow up question.

RM: No need to apologize. Questions are always welcome.

Â

WM: For your elegant study on catching baseballs  you found that the retinal vertical optical velocity was controlled. Does this count as a perception or a physical variable across the retina of the ‘body’.

RM: Vertical optical velocity counts as a perceptual variable, p, when it is a variable in the model that simulates the behavior of the actual pursuer, and it counts as Qi when it is in my spreadsheet and I am looking at its value varying over time. The physical variable in the studies of object interception is the actual trajectory of the pursued object in 3 space. This is the data I got from Dennis Shaffer who carried out the research. What he sent me was the varying x,y,z coordinates of the trajectory of the pursued object during each pursuit trial. These varying x,y,z coordinates are the physical variables that are the basis of the computation of the possible controlled variables in this study. We actually tested several possibilities among which were optical acceleration of the tangent of optical angle, optical velocity of the tangent of optical angle, and optical velocity of optical angle. Using the latter as the variable p in the model gave the best fit of the model to pursuer behavior.Â

Â

WM: Moreover what is the function that might correspond to the Qi in the outside world?

RM: It’s the function that produces a value of p which, when used in the model, gives the best fit to the data. In our case, it was the function of x,y,z that produced a measure of velocity of optical angle. That is:

optical velocity = f(x,y,z)

RM: The function f() of the physical variables x,y,z defines the perceptual variable, p, when that function is used as the perceptual function in the model, and it defines Qi when it is seen by an observer (me) as a list of values in the spreadsheet. Qi is just the view of the perceptual variable from the point of view of an observer.Â

RM: This is much easier to understand in a tracking experiment, like the one in my Nature of Control demo. In that demo, the plotted distance from cursor to the target (c-t) is Qi – it is the observer’s view of the controlled perception. The same measure of distance between cursor target (c-t) is the controlled perceptual variable, p, in the subject who did the controlling and in the model of that controlling.Â

Â

WM: Presumably quite a complex equation is it requires the relative motion of the ball, body, head and eyes to calculate? I don’t recall that this Qi is ever reported in the paper?

RM: There is really no need to report Qi in PCT research because it is the same as the perceptual variable, p, used in the model to fit the data. However, Bill does refer to Qi (as q.i) in his paper “A feedback model for behavior” in LCS I. There he tests two different hypotheses about the definition of the input quantity, q.i, controlled by the rat – probability of shock, p.s, and rate of shock, r.s. These two variables are used as perceptual variables in a control model and it is found that p.s gives a better fit to the data than r.s. So p.s is “closer to the actual nature of q.i” Â than r.s. (p. 58).

RM: So in my object interception modeling I could have said that “optical velocity is closer to the actual nature of the controlled input quantity than is optical acceleration”.Â

Best

Rick

Â

All the best,

WarrenÂ


Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery

Hurray!

···

On Tue, Feb 14, 2017 at 1:42 AM, Warren Mansell wmansell@gmail.com wrote:

WM: So Rick, step at a time, you are saying that the Qi is the function of environmental variables that is being controlled. It is a represents a possibly complex relationship between physical variables that we note is held close to a reference state. All of this is in the environment and is controlled simultaneously as perception corresponding to this function is controlled within the organism…

RM: Yes.

PS Tom Hanks played Woody…

RM: Well, at least I knew he played someone;-)

Best

Rick


Richard S. Marken

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
–Antoine de Saint-Exupery

Thanks Rick!

···

On Tue, Feb 14, 2017 at 1:57 AM, Warren Mansell wmansell@gmail.com wrote:

WM: Sorry Rick, I needed a follow up question.

RM: No need to apologize. Questions are always welcome.

WM: For your elegant study on catching baseballs you found that the retinal vertical optical velocity was controlled. Does this count as a perception or a physical variable across the retina of the ‘body’.

RM: Vertical optical velocity counts as a perceptual variable, p, when it is a variable in the model that simulates the behavior of the actual pursuer, and it counts as Qi when it is in my spreadsheet and I am looking at its value varying over time. The physical variable in the studies of object interception is the actual trajectory of the pursued object in 3 space. This is the data I got from Dennis Shaffer who carried out the research. What he sent me was the varying x,y,z coordinates of the trajectory of the pursued object during each pursuit trial. These varying x,y,z coordinates are the physical variables that are the basis of the computation of the possible controlled variables in this study. We actually tested several possibilities among which were optical acceleration of the tangent of optical angle, optical velocity of the tangent of optical angle, and optical velocity of optical angle. Using the latter as the variable p in the model gave the best fit of the model to pursuer behavior.

WM: Moreover what is the function that might correspond to the Qi in the outside world?

RM: It’s the function that produces a value of p which, when used in the model, gives the best fit to the data. In our case, it was the function of x,y,z that produced a measure of velocity of optical angle. That is:

optical velocity = f(x,y,z)

RM: The function f() of the physical variables x,y,z defines the perceptual variable, p, when that function is used as the perceptual function in the model, and it defines Qi when it is seen by an observer (me) as a list of values in the spreadsheet. Qi is just the view of the perceptual variable from the point of view of an observer.

RM: This is much easier to understand in a tracking experiment, like the one in my Nature of Control demo. In that demo, the plotted distance from cursor to the target (c-t) is Qi – it is the observer’s view of the controlled perception. The same measure of distance between cursor target (c-t) is the controlled perceptual variable, p, in the subject who did the controlling and in the model of that controlling.

WM: Presumably quite a complex equation is it requires the relative motion of the ball, body, head and eyes to calculate? I don’t recall that this Qi is ever reported in the paper?

RM: There is really no need to report Qi in PCT research because it is the same as the perceptual variable, p, used in the model to fit the data. However, Bill does refer to Qi (as q.i) in his paper “A feedback model for behavior” in LCS I. There he tests two different hypotheses about the definition of the input quantity, q.i, controlled by the rat – probability of shock, p.s, and rate of shock, r.s. These two variables are used as perceptual variables in a control model and it is found that p.s gives a better fit to the data than r.s. So p.s is “closer to the actual nature of q.i” than r.s. (p. 58).

RM: So in my object interception modeling I could have said that “optical velocity is closer to the actual nature of the controlled input quantity than is optical acceleration”.

Best

Rick

All the best,

Warren


Richard S. Marken

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
–Antoine de Saint-Exupery

Hi Warren

···

From: Warren Mansell [mailto:wmansell@gmail.com]
Sent: Tuesday, February 14, 2017 10:42 AM
To: csgnet@lists.illinois.edu
Subject: Re: VS: To accept a theory (was Re: Control Is)

So Rick, step at a time, you are saying that the Qi is the function of environmental variables that is being controlled. It is a represents a possibly complex relationship between physical variables that we note is held close to a reference state. All of this is in the environment and is controlled simultaneously as perception corresponding to this function is controlled within the organism…

HB : I lookeed in the vocabulary and »simoultaneously« means »at the same time«.

Warren think a little. How can be something simultaneously controlled in organism and environment. You are approaching Ricks’ RCT where everything in the loop happens »at the same time«. Be reasonable. Â

And I also think that you are missing something. Rick seems to be slowly abbandoning his RCT. He became smarter and his terminology more close to PCT. It seems that all these conversation afterall are influencing his theory.

But you are still behind. It seems that he is not telling you eveyrthing about his plans. Rick is a fox. He’ll think first of himself. He is perfect LCS….

Best,

Boris

PS Tom Hanks played Woody…

On 14 Feb 2017, at 08:03, Boris Hartman boris.hartman@masicom.net wrote:

Hi Warren…
<

From: Richard Marken [mailto:rsmarken@gmail.com]
Sent: Tuesday, February 14, 2017 3:23 AM
To: csgnet@lists.illinois.edu
Subject: Re: VS: To accept a theory (was Re: Control Is)

[From Rick Marken (2017.02.13.1820)]

On Mon, Feb 13, 2017 at 1:09 PM, Warren Mansell wmansell@gmail.com wrote:

WM: Hi Rick, thanks, I understand you and it is all getting clearer. I toyed with leaving it there, but I guess it’s best to check my understanding by trying to put it into my own words.

HB : Yes. Put in your words not Ricks. If you’ll listen to Rick nothing will get more clear. You’ll just get dipper in the mud.

RM: Great.

WM: I can now see how a function of elements of the environment - the Qi - can be controlled.

RM: Qi is, itself, a function of the elements, which are the actual physical variables in the environment – the v’s that are usually not shown in diagrams of the PCT model, unfortunately. But that may be what you meant. I’m just clarifying.

HB : Rick what can I say. I’ve never saw so much bullshit on one place. You don’t understand PCT. It would be better if you don’t clarify.

WM: As humans we happen to extract this function through perception, but I can appreciate that the function and its reference state still ‘exists’ in a mathematical sense, that does not require a perceiver.

RM: I would rather say that the basis of this function exists; the variables from which the function can be computed. For example, in the “taste of lemonade” example, the combination of physical variables – amount of sugar (s), acid (a) and oil (o) – provides the possibility of the perception of the taste of lemonade via the perceptual function b.1s+b.2a+b.3*o. It’s these variables – s,a and o – that exist in the environment; the mathematical function of these variables that is the perception of the taste of lemonade doesn’t ever exist in the environment, either as a possibility or as a reality.

HB : And where do you suppose is this »mathematical funtion« hidden in perceptual signal ? Provides a possibility ?? On which level of hierarchy ???

WM: It is required to be perceived to actually implement the control

RM: This implies that perception is a process of perceiving an existing mathematical relationship between physical variables.

HB : All the mathematical relatioship are just in your head. There is nothing outside in mathematical form. Can you imagine snake or elephant perceiving »mathematical relationship«….???

RM : But that’s not how perception works. Perception is the process of computing the mathematical function of the physical variables that is the perceptual variable.

HB : What a mess…. Perception is the process of computing â… ha,ha… you ara a clown Rick… Get on real work and re read B:CP what is perceptual signal and how it looks like. There is much more nonsense in continuing parts so I’ll finnish. here. Incredible. What an imagination….

Best,

Boris

RM: I think you may have read too much JJ Gibson, who talked about perception being a process of “detecting invariants in the environment”. Gibson made some great observations but you learn that his theory of perception doesn’t really work as soon as you try to build a system that can do, for example, speaker independent speech recognition. You can’t build a speech recognition system that “detects” the “invariant” formant changes without building a pretty complex perceptual function that compute functions of the the changing sound amplitudes at different frequencies.

WM: which is why we look for theory in the form of PCT to explain how living things do it. Another theorist might insist that this function is calculated in advance and reacted to in an open loop fashion, for example.

RM: No, the PCT model of perception is required for both open and closed loop models. Any person who tries to build a model whose actions depend on perception will have to build perceptual functions like those described by PCT. The PCT model is required to account for the observation that organisms keep variable aspects of the environment in reference states. This phenomenon – the phenomenon of control – has simply not been noticed in conventional psychology, so these psychologists have not tried to develop models to account for it.

RM: PCT accounts for this phenomenon of control by assuming that organisms are in a closed negative feedback loop relationship with respect to perceptual variables that correspond to the aspects of the environment that are seen to be controlled. The behavior of such closed loop negative feedback systems can be shown to be organized around control of these perceptual variables, keeping them equal to internal specifications (references) for the state of these perceptions, protected from the effects of disturbance.

WM: It just so happens that if PCT is true then we are perceivers just as we are control systems doing all this scientific work and discussion.

RM: The fact that we are perceivers is certainly not unique to PCT. What is unique to PCT is that we control our perceptions, our perceptions don’t control us. What is unique about PCT is that it says behavior is controlled input, not caused output.

WM: For this reason I can see why there is a resistance to ‘the fact of control’ from many CSGers who have taken PCT on fully and are maybe not trying to test it against ‘facts’ like yourself. I am on the fence.

RM: Yes, I know. A lot of people come to PCT because they are basically post modernists who like the idea that “it’s all perception” which, to them, means “it’s all just a matter of opinion”. So my facts are as good as your facts. Sad;-)

WM: Could we turn this into an advantage and actually appreciate how one theory can traverse the realist empirical approach and the relativist, subjective approach to reality?

RM: You can do whatever you like. But I’m not a fan of what you call the “subjective” approach to reality. I like the scientific approach, which recognizes that “i’s all perception” and realizes that the way to find out about the nature of the reality that is presumably the basis of our perceptual experience is to test theories of what that reality is like and then testing those theories by seeing if, when we manipulate various perceptions in certain ways, we perceive what the theories say we will perceive. PCT is one such theory of the nature of the reality the produces our perceptions of the controlling (purposeful behavior) done by living systems.

WM: The purposes of these two camps seem different to me, but they share more in common than proponents of other theories of behaviour I know of. Sorry if that last comment seems to let the heat out of the debate.

RM: Not at all. I’m happy to continue arguing that my camp – the scientific camp – is the only one that can make any progress in our understanding of reality;in our case, the reality that accounts for out perception of purposeful behavior. To me, the subjective approach is simply mental masturbation (I admit that masturbation can be fun but I prefer the real thing, at least as I recall; I’m no spring chicken;-)

WM: I’ve got you on this one Rick. Maybe I’ll take on your view of the power law next!

RM: I look forward to it!

WM: I’m off to Narnia now… or maybe Andy’s house to hang out with Buzz…

RM: I never read the Naria books; too much religion for me. But I do like Andy and Buzz! And since I live in LaLa Land I had the pleasure of meeting Andy (Tom Hanks) a few years ago. A real mensch (him, of course, obviously not me;-).

Best

Rick

Warren

On 13 Feb 2017, at 18:00, Richard Marken rsmarken@gmail.com wrote:

[From Rick Marken (2017.02.13.1000)]

On Mon, Feb 13, 2017 at 12:44 AM, Warren Mansell wmansell@gmail.com wrote:

WM: …I am getting to see what this debate is getting at; Rick wants to identify control in the environment, separately from PCT, just like Bill at times,

RM: Actually, all the time.

WM: so that PCT has a phenomenon to explain.

RM: You make it sound like PCT was just sitting around looking for something to explain. In fact, the phenomenon of control was sitting around waiting for PCT to explain it. Again, you betray your “theory first” approach to PCT.

WM: The problem is, once we unpack the nature of control, we see that it can only be defined accurately from an observer’s perspective.

RM: Doesn’t sound like a problem to me. Control is defined as the observation that a variable (such as the position of a cursor) is maintained in a reference state protected from disturbance. You can observe it for yourself in my “Nature of Control” demo (http://www.mindreadings.com/ControlDemo/BasicTrack.html)

WM: I think that Rick has been arguing that the Qi exists in the environment, but would he also propose that the reference value could exist in the environment?

RM: The reference state of Qi exists in the environment (actually, like Qi, it exists as a function of physical variables in the environment; as a perception in the observer); the term “reference value” refers to the value of the theoretical reference signal that is presumed to specify the reference state of the controlled variable.

WM: Presumably not;

RM: Right. The reference value is a theoretical construct that explains the observed reference state of the controlled quantity, Qi.

WM: in fact of course, like you say Eetu, it doesn’t because if it did there would be no point in the organism controlling the Qi. So the question is whether one can truly say that the Qi is controlled if this is only the case because it is transformed into P which is compared with an internal RV continuously.

RM: You lost me.

WM: Maybe the fact that we can’t actually see control without taking a perceptual perspective is exactly why so many people find PCT hard to adopt.

RM: I don’t understand what you mean by “we can’t actually see control without taking a perceptual perspective”. It seems like I can see control by just looking to see whether a variable is maintained in a reference state, protected from disturbance. Is that the “perceptual perspective” of which you speak?

WM: Could we say the Qi is typically controlled but one can only make this judgement when the observer has an internal reference point to do so?

RM: No, we say that Qi is controlled when disturbances have far less of an effect on that variable than would be expected if the variable we not under control; that is, when Qi is kept in a constant or variable reference state, protected from disturbance. There is no theory involved in this determination. There is no need to know that this control occurs because there is a system present that is organized around control of the perception of the variable you see being controlled. This atheoretical approach to determining what variable is controlled is demonstrated in my Mind Reading demo (http://www.mindreadings.com/ControlDemo/Mindread.html), which is not only extremely entertaining but also highly educational.

Best

Rick

On 13 Feb 2017, at 07:42, Eetu Pikkarainen eetu.pikkarainen@oulu.fi wrote:

EP: If “control” means something like affecting a value of one (“controlled”) variable so that it becomes the same or near to the current value of an other variable (reference value) and / or remains the same or near to it, then anything in the environment is not controlled. And this has nothing to do with the question how well or badly the control is working. There just are no reference variables and thus reference values in the environment, they exists only inside the control system

Richard S. Marken

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
–Antoine de Saint-Exupery

On Mon, Feb 13, 2017 at 12:44 AM, Warren Mansell wmansell@gmail.com wrote:

Thanks Eetu!

ET: “EP: If “control” means something like affecting a value of one (“controlled”) variable so that it becomes the same or near to the current value of an other variable (reference value) and / or remains the same or near to it, then anything in the environment is not controlled. And this has nothing to do with the question how well or badly the control is working. There just are no reference variables and thus reference values in the environment, they exists only inside the control system.”

WM: Just like most people on this thread, of course I agree with you in the sense that you are articulating PCT. I am getting to see what this debate is getting at; Rick wants to identify control in the environment, separately from PCT, just like Bill at times, so that PCT has a phenomenon to explain. The problem is, once we unpack the nature of control, we see that it can only be defined accurately from an observer’s perspective. I think that Rick has been arguing that the Qi exists in the environment, but would he also propose that the reference value could exist in the environment? Presumably not; in fact of course, like you say Eetu, it doesn’t because if it did there would be no point in the organism controlling the Qi. So the question is whether one can truly say that the Qi is controlled if this is only the case because it is transformed into P which is compared with an internal RV continuously.
Maybe the fact that we can’t actually see control without taking a perceptual perspective is exactly why so many people find PCT hard to adopt. A part of the loop of control can be identified in the environment but it can’t be judged as controlled without taking a perceiver’s perspective. Like Rick I am not sure if using the word ‘stabilised’ helps at all - because often these components will not be stable if there is a changing reference value for its perception within the organism. Could we say the Qi is typically controlled but one can only make this judgement when the observer has an internal reference point to do so? Or is that a weak hybrid of the two positions?
Warren

On 13 Feb 2017, at 07:42, Eetu Pikkarainen eetu.pikkarainen@oulu.fi wrote:

EP: If “control” means something like affecting a value of one (“controlled”) variable so that it becomes the same or near to the current value of an other variable (reference value) and / or remains the same or near to it, then anything in the environment is not controlled. And this has nothing to do with the question how well or badly the control is working. There just are no reference variables and thus reference values in the environment, they exists only inside the control system

Richard S. Marken

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
–Antoine de Saint-Exupery

Richard S. Marken

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
–Antoine de Saint-Exupery

Whatever Rick is explaining about »baseball catch« is behaviorisitc explanation which was uncovered in direct conversation between me, Bill and Rick. You can all see it on CSGnet.

I think that in conversation about »baseball catch« (which was long for couple years) Bill was a little angry about Rick who was hidding some »facts« about how he made »baseball catch« working.

All in all Ricks’ RCT can’t explain what is happening in the controller when he is running to some »catch point« without seeing a ball. See the best baseball catch ever. The fielder is »blindly« catching the ball, jumping on the fence.

PCT has to explain all behaviors in the same manner. It is general model about how organism function. No matter what kind of organism is (animal or plant).

But Rcik has for every example his own explanation…. Which is usually in accordance to his RCT.

Boris

···

From: Richard Marken [mailto:rsmarken@gmail.com]
Sent: Tuesday, February 14, 2017 7:08 PM
To: csgnet@lists.illinois.edu
Subject: Re: VS: To accept a theory (was Re: Control Is)

[From Rick Marken (2017.02.14.1010)]

On Tue, Feb 14, 2017 at 1:57 AM, Warren Mansell wmansell@gmail.com wrote:

WM: Sorry Rick, I needed a follow up question.

RM: No need to apologize. Questions are always welcome.

WM: For your elegant study on catching baseballs you found that the retinal vertical optical velocity was controlled. Does this count as a perception or a physical variable across the retina of the ‘body’.

RM: Vertical optical velocity counts as a perceptual variable, p, when it is a variable in the model that simulates the behavior of the actual pursuer, and it counts as Qi when it is in my spreadsheet and I am looking at its value varying over time. The physical variable in the studies of object interception is the actual trajectory of the pursued object in 3 space. This is the data I got from Dennis Shaffer who carried out the research. What he sent me was the varying x,y,z coordinates of the trajectory of the pursued object during each pursuit trial. These varying x,y,z coordinates are the physical variables that are the basis of the computation of the possible controlled variables in this study. We actually tested several possibilities among which were optical acceleration of the tangent of optical angle, optical velocity of the tangent of optical angle, and optical velocity of optical angle. Using the latter as the variable p in the model gave the best fit of the model to pursuer behavior.

WM: Moreover what is the function that might correspond to the Qi in the outside world?

RM: It’s the function that produces a value of p which, when used in the model, gives the best fit to the data. In our case, it was the function of x,y,z that produced a measure of velocity of optical angle. That is:

optical velocity = f(x,y,z)

RM: The function f() of the physical variables x,y,z defines the perceptual variable, p, when that function is used as the perceptual function in the model, and it defines Qi when it is seen by an observer (me) as a list of values in the spreadsheet. Qi is just the view of the perceptual variable from the point of view of an observer.

RM: This is much easier to understand in a tracking experiment, like the one in my Nature of Control demo. In that demo, the plotted distance from cursor to the target (c-t) is Qi – it is the observer’s view of the controlled perception. The same measure of distance between cursor target (c-t) is the controlled perceptual variable, p, in the subject who did the controlling and in the model of that controlling.

WM: Presumably quite a complex equation is it requires the relative motion of the ball, body, head and eyes to calculate? I don’t recall that this Qi is ever reported in the paper?

RM: There is really no need to report Qi in PCT research because it is the same as the perceptual variable, p, used in the model to fit the data. However, Bill does refer to Qi (as q.i) in his paper “A feedback model for behavior” in LCS I. There he tests two different hypotheses about the definition of the input quantity, q.i, controlled by the rat – probability of shock, p.s, and rate of shock, r.s. These two variables are used as perceptual variables in a control model and it is found that p.s gives a better fit to the data than r.s. So p.s is “closer to the actual nature of q.i” than r.s. (p. 58).

RM: So in my object interception modeling I could have said that “optical velocity is closer to the actual nature of the controlled input quantity than is optical acceleration”.

Best

Rick

All the best,

Warren

Richard S. Marken

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
–Antoine de Saint-Exupery