VS: Watch Your p's and Qi's

Thanks Bruce again, this is very important note and point!

This and your next message about adaptation are very hepful and illuminating.

What I mean is that the “control strictly speaking” or “controlling directly” on the one hand and the “control loosely speaking” or “controlling indirectly” are TWO important things that belong together (as thick as thieves - hopefully right saying in English),
like all of the other parts of the loop. Only the whole loop can control anything. But still I think that it is intellectually corrupting to call two different (however similar) concepts with the ​same term. It causes unnecessary fights in areas like science
where strictness is exceptionally important.

One thing is a consequence of the other - they cannot be the same.

I doesn’t sound very good if you have to say that “PCT is a theory of control, loosely speaking”, I think?

Loosely speaking we can say that in the feedback path of the control loop, the controller controls B by controlling A, D by controlling C and so on and finally that controllers perception is controlled by its controlling X. But strictly speaking - I think

  • there should be (beforehand) their own reference level for all of these claimed controls?

Loosely speaking, Fred was quite right with his suggestion [From Fred Nickols (2017.01.29.0950 ET)] that we control
in the every phase of the loop.

Of course it is necessary for the organism that it can affect, stabilize and adjust its environment so that it can stay alive and keep its error signals near zero. But this ability has not developed to affect the environment but to keep the organism itself alive
i.e. its internal stucture and state in the reference limits.

Eetu

···

Lähettäjä: Bruce Abbott bbabbott@frontier.com
Lähetetty: 24. helmikuuta 2017 17:16
Vastaanottaja: csgnet@lists.illinois.edu
Aihe: RE: Watch Your p’s and Qi’s

[From Bruce Abbott (2017.02.24.1015 EST)]

[Eetu Pikkarainen 2017-02-24] –

EP: Thank you Rick for this long and detailed explanation, and thank you Bruce for your clear minded comments.

BA: And thank you, Eetu, for your clear exposition. I have only one correction to suggest. You state that “one of the main strength and originality in Powers insight is that control system is strictly speaking controlling (only and
just) its perceptions.� You seem to be saying that the control system does not control Qi, although I doubt that this was your intention. As you say, “strictly speaking,� the control system directly controls
only that which it perceives, p. However, I will argue that controlling p is only the means by which the system controls Qi, and that the control of Qi is the reason why controlling p is useful to the individual.

BA: By controlling its perceptions, the system also controls the environmental correlates of those controlled perceptions. Controlling p requires taking actions in the environment that affect Qi, and it is those effects on Qi, out there
in the real world, that allow the individual to function successfully. A person who is able to control her hunger (keeping it near a zero reference level) by eating only imaginary food soon will perish. The usual case, of course, is that keeping the perception
of hunger (p) near a reference value of zero entails consuming real food, which has the effect of nourishing the body while also reducing p. It is this effect of keeping the body nourished that makes controlling hunger worthwhile.

BA: Consider what happens when the link between p and Qi is lost. A rat whose lateral hypothalamus has been destroyed no longer shows any interest in food, and would starve to death amid plenty if the experimenter did not keep it alive
through force feeding. One way to account for this finding is to assume that the lateral hypothalamus (or nerve fibers passing through it) is an essential part of a brain circuit that transduces the bodily state of nutrition (Qi) into a perception of hunger-level
(p). If destroying the lateral hypothalamus causes p to be equal to the reference value of zero no matter what the state of Qi, the rat will never eat, as in fact is observed to happen. If p no longer relates to Qi, it may still be possible to control p,
but doing so will no longer confer its essential benefit. We control p in order to control Qi. Conversely, because it has only p to go on, a control system can control Qi
only by controlling p.

Bruce

[eetu pikkarainen 2017-02-26 2]

[From Rick Marken (2017.02.24.2300 PST)]

RM: Control always requires affecting the aspect of the environment that is controlled; the aspect of the environment that corresponds to the controlled perception.

EP: The truth of this statement depends on the two vague concepts in it. A control unit always affects its immediate environment, but these local environments can be inside the the organism. Secondly what means "aspect" here? (Continues below...)

RM: The only time control doesn't involve affecting the environment is when controlling is done in imagination.

EP: Perhaps imagination is (must be) mixed in all control because memory is used there too? This is not however so central here, but what do you think about the idea of "turning the blind eye" type of control? This means for example a situation in horror movie or in traffic accident where the person does not want to see what she sees (because it causes too strong error), and closes her eyes. This is effective control of perception but it does not affect the source of the perception in the environment. I would think that the at least is some cases the sensory adaptation described by Bruce could be this kind of control of perception. Think about the way how the iris of the eye the tries to keep the amount of light, and thus the brightness of the perception, in suitable limits, control it to the reference.

Sorry the clumsy explanation...

RM: Yes, but I think it's better to say that we study control by determining the aspect (function) of the environment that is under control.

EP: This sounds that we agree that the variable Qi, or the state of affairs like I called it (or rather the one aspect of it) really is in the environment and it is a different thing than the perception to which it corresponds? (It would not be reasonable to say that something corresponds to itself?) The disagreement is that I still stand pat that in the control loop, only and just the perception is controlled and all the other processes in the loop are causes and results of this control. In science generally the perceivable causes and results are used to study and explain the studied subject matters, which often are unperceivable themselves. For example the perceivable tracks in the bubble cabin (hopefully right term in English) are used as signs of as such unperceivable elementary particles. In similar way the perceivable stabilization of something in the environment (as a consequence of the subject’s behavior and against the disturbances) is used to explain or point or prove the unperceivable phenomenon of the control of perception.

RM: When we have identified that aspect of the environment, we have identified the controlled variable, Qi, that corresponds to the perception that is being controlled.

EP: OK, but strictly speaking not "controlled variable", but just the variable, which corresponds the controlled perception.

RM: I don't think I am dealing in metaphysics; just plain old PCT. ...

EP: Every scientific approach (and every cognizing person) has that what is often called ontological commitments. Typically they are not discussed explicitly. I think it is good to discuss about them, because they can cause confusion when they are only implicit.

RM: The relationship between Qi and p is really not as metaphysical as it may seem. I suggest that you read the section on the "Coin Game" in B:CP and then play the game with a friend, with you as E and your friends as S. When you see the aspect of the arrangement of the coins that your friend is trying to maintain (control) you are seeing Qi. In the example in B:CP Bill describes the observer (E) discovering that the controller (S) is trying to maintain a "zig zag" pattern. E's perception of that "zig zag" pattern is the reference state of Qi -- the pattern of coins. S's perception of that "zig zag" pattern is the reference state of p -- also the pattern of coins.

EP: Yes, I agree this: “When you see the aspect of the arrangement of the coins that your friend is trying to maintain (control) you are seeing Qi.” except of course the bracketed “control” My friend is controlling his perception of the arrangement, his perception of the Qi – and yes tries to maintain it in the environment. But because Qi is in environment it has no reference state, strictly speaking. I think that E’s reference (as purpose) is not to see the “zig zag” pattern, but infer from the pattern of the coins (the Qi) what the reference of S is?

RM: All I meant to say is that the "environment" (or reality) in PCT is what our current best models of physics and chemistry say it is. Our perceptions are presumed to be constructed from the sensory effects of this environment. But the fact is that our models of the perceptual functions that produce the perceptual variables that are (in theory) what is being controlled never really describe these functions in terms environmental variables or the sensory effects thereof. The input to perceptual functions in our models are always variables that are themselves presumably functions of the sensory effects of environmental variables. For example, in the "What is Size" demo, the area variable is defined as Qi = h * w, where h and w are the height and width of the rectangle. But h and w are not really environmental or sensory variables; they are perceptions themselves; functions of the state of the receptors in an area of the retina.

EP: I think I understand what you mean: We never know what are the variables and their values in the environment. We only have their corresponding variables and values in our own perceptions. The observer does not know the variables and their values in the environment i.e. she never knows what the Qi or the state of affairs is. But she knows what the variables and values are in her own perception (Po = perception of the observer). And because she is similar human as the controller, she can infer that she knows also the Pc (controller’s perception). Thus the environmental correlate of the perceptions is irrelevant and Po can be substituted for Qi?

EP: I think that for empirical research this can be a rational procedure, but it is confusing to change the terms. Qi is (df) the environmental input quantity to the sensory functions of the controller. If you instead want to talk about observer’s perception, it is OK, but I think you should call it Po or something which is better explicitly understandable.

EP (earlier): As for the diagram, those which look like mouse and laptop are not my perceptions, they are your drawings. I did not put them there, but you did J

RM: They are your perceptions because they are being constructed by the perceptual functions in your nervous system, just as they are in mine.

EP: Oh no, take it literally. What in your diagram looks like mouse, is an image you drew or pasted there. It is not constructed by my perceptual functions but rather by your output via drawing program. You did that with the purpose that I would perceive there something looking like a mouse. You could attain your purpose only because the image you selected looks like a mouse. So that which look like a mouse is your image (which looks like a mouse). It cannot be my perception, because my perceptions do not look like a mouse. I have never seen my perceptions directly, but inferring from the diagrams and descriptions it looks like a neural current. What I perceive in my environment are not my (neither your) perceptions. What I get in me as a result of perceiving is the perception. Perception and that which is perceived are different things.

RM: Maybe what is hard to understand is that all those things out there that seem so real are your perceptions. The people, the computers, the tables and chairs, the racquetball that hit me squarely in the ear today and hurt like hell -- all are perceptions constructed by your perceptual functions. They are constructed from real physical variables -- the fact that everything you experience is perception doesn't mean everything is make-believe or an illusion; There is really stuff out there -- the world of physical variables -- that is the basis of the look and feel and heft of the table I'm working on, for example. What our perceptual functions do is organize this reality into the variables that we perceive and can control. I'm sure we perceive that reality as we do -- in terms of objects with shape, weight, movement, relationships to other objects, etc, for example -- because this is the adaptive way to perceive the reality on the other side of our senses.

EP: I agree, but it but why is it adaptive? The most natural answer to this is that it is adaptive to perceive something as say dangerous predator because that something may be just like the dangerous predators are. (And it is adaptive to perceive something which look like a mouse as something which looks like a mouse.)

RM: ... The way you state it, it sounds like you are saying what I am saying -- that there is a real world of physical variables that are the basis of our perceptions.

EP: Yes clearly so far we agree!

RM: The only thing that makes empirical research more difficult -- actually impossible -- is the idea that Qi, the researcher's perception of the controlled variable, is an object in the environment that is perceived, with varying levels of accuracy and controlled by the control system.

EP: Yes it is certainly true, “that the idea that the researcher's perception of the controlled variable, is an object in the environment that is perceived” would be a fatal error. We do not perceive other’s perceptions. But as fatal to common understanding is the idea that environmental input quantity (=Qi) is the same thing as the researcher's perception. We don’t (now loosely speaking) control or try to control the perceptions of other people. We are strictly speaking controlling only our own perceptions and loosely speaking we are trying to control (strictly speaking maintain, affect, stabilize etc.) the corresponding environmental variable (CEV) or input quantity (Qi) – of which we have no other knowledge than our own perception, but which we still can affect by our output functions.

RM: I believe it is this idea that has resulted in so few researchers doing the kind of research Powers hoped would be done to test and develop his model of purposeful behavior; research aimed at discovering the kinds of variables -- the kinds of Qi -- that organisms control.

EP: As I said earlier I don’t believe that that kind of research requires those problematic ontological and conceptual solutions. And in addition I think that developing the theoretical model may require also something else than just discovering different kinds of variables whose functional positions is still the same in the model. But it is of course important anyway!

Best

Eetu

Dear Boris,

Thank you for your “reinforcing” comment.

I agree quite whole heartedly. I said that the control, as the core idea of PCT, is the control of just and only the perception. Nothing else is controlled. The idea, as I understand it, is NOT that ALSO perception is controlled, but that ONLY AND JUST perceptiont
is controlled.

This control, to take place, requires the whole loop but all else that happens along the loop is only consequences (and prerequisites) of control.

The confusing problem is that a) we empirically perceive only the consequences of control in environment and b) in everyday and loose talk we naturally tend to say that we control something in the environment when we should say that we control our perception
of that something - and if that something changes it is a consequence of our control. There seems to be (at least) these two strong pressures that cause many in this list say something like “(of course) also the environment is controlled when perception is
controlled”. Even though they certainly do know that this is loose talk. In strict talk environmental change or stabilization is just a consequence of control. But if we allow that loose talk then we have no reason why not say that all consequences of control
are control.

So I suggest that we should say clearly when we talk in loose and when in strict sense, and perhaps avoid loose talk in the list. When we talk strictly we should use a certain term, different from control, when we refer to environmental consequences of control.
I do not know which is best: affect, constrain, manipulate, stabilize etc. Perhaps it doesn’t even matter, if it is just not control.

The a) pressure above: empirical perceivability of the control only in its environmental consequences is the reason for Rick to keep it central. But even he seems to say only in a loose meaning that environment is controlled, because the Qi discussion showed
that actually for him it is the observer’s – his – percepteptions, which are controlled by the test subject’s action. At least so I have now understood it.

Eetu

···

Lähettäjä: Boris Hartman boris.hartman@masicom.net
Lähetetty: 2. maaliskuuta 2017 16:11
Vastaanottaja: csgnet@lists.illinois.edu
Aihe: RE: Watch Your p’s and Qi’s

Dear Eetu. I wouldn’t beleive everything that can be »heard« on this forum. I think it’s better to check it in PCT definitions which also explain the loop in LCS III.

From: Eetu Pikkarainen [mailto:eetu.pikkarainen@oulu.fi]
Sent: Friday, February 24, 2017 9:51 PM
To: csgnet@lists.illinois.edu
Subject: VS: Watch Your p’s and Qi’s

Thanks Bruce again, this is very important note and point!

This and your next message about adaptation are very hepful and illuminating.

What I mean is that the “control strictly speaking” or “controlling directly” on the one hand and the “control loosely speaking” or “controlling indirectly” are TWO important things that belong
together (as thick as thieves - hopefully right saying in English), like all of the other parts of the loop. Only the whole loop can control anything. But still I think that it is intellectually corrupting to call two different (however similar) concepts with
the ​same term. It causes unnecessary fights in areas like science where strictness is exceptionally important.

One thing is a consequence of the other - they cannot be the same.

I doesn’t sound very good if you have to say that “PCT is a theory of control, loosely speaking”, I think?

Loosely speaking we can say that in the feedback path of the control loop, the controller controls B by controlling A, D by controlling C and so on and finally that controllers perception is controlled
by its controlling X. But strictly speaking - I think - there should be (beforehand) their own reference level for all of these claimed controls?

Loosely speaking, Fred was quite right with his suggestion [From Fred Nickols (2017.01.29.0950
ET)] that we control in the every phase of the loop.

HB : No Eetu. The control is not happening in the whole loop. If you think like this you’ll have to support this statement with evidence
how control is transffered to environment via neuromuscular connection . How ? And how control enters the organism through perceptual signal ? How control is distributed in the perceptual signal ? This is the problem Rick
has from the beggining of our conversation some years ago. He is misleading forum with his RCT, wrong interpretation of PCT which includes the »control happening in the whole loop«. The easiest way to check this assumption is to make physiological tour arround
the loop, which was already done by Powers in B:CP. And that’s how we got PCT.

If you look at the definitons of loop »elements« you’ll see that there is no control in all of them (post to Rick). So also from PCT definitions we can conclude that control
is not present in the whole control loop. It works in PCT sense something like this :

Bill P and others
(2011) Perceptual Control Theory (PCT) provides a general theory of functioning for organisms. At the core concept of the theory is the observation that living things control perceived environment by means
of their behavior. Consequently, the phenomen of control takes center stage in PCT, with observable behavior playing an important but supporting role.

HB : Behavior is playing a supporting role in control loop. It doesn’t carry control.

Bill P :

Using the internal point of view, we can understand many aspects of behavior by seeing control as control of perception rather than of an objective world.

HB : We can see that there is no control in »objective« world. So there is no »controlled variable« in outer environment or the correlate to »Controlled perception« through
»Control of behavior« as Rick is writing.

Bill P :

A control system controls what it senses, and what it senses is the result of applying a continuous transformaton process to the elementary sensory inputs to the nervous system.

HB : This is feed-back. The control system controls perception with adding effects to disturbances. Comparator is the only place in organism where control happens. No control
is happening outside. When percpetion is controlled there is no other »simultaneous control in outer environment« as Rick and Warren are trying to present. These explanations don’t follow PCT definitions. If you do this you’ll have to introduce Telekinesis
to transfer control to outer environment as Rick is doing.

Applying continuos process in environment means effects of output to the sensors. Feed-back is »effects of output on input«. There is no control in output effects. Otherwise
I beleive that Bill would define »controlled effects to environment« what would mean that behavior is control. But this is not so. See diagrma in LCS III.

And of course there can’t be any control in sensor effects. So there is no »Controlled Perceptual variable« or »Controlled Perception«. Anyway there is no such term in PCT.
Rick inventied them to wrongly support his RCT with wrong definitions (please see post to Rick)

Dear Eetu. If you want to follow the control loop in PCT sense you just have to follow defintions or PCT and physiological evidences that show how control loop works in organism.
And I beleive you will have no problem determining what PCT is about and how control is distributed in »control loop«.

EP : Of course it is necessary for the organism that it can affect, stabilize and adjust its environment so that it can stay
alive and keep its error signals near zero. But this ability has not developed to affect the environment but to keep the organism itself alive i.e. its internal stucture and state in the reference limits.

HB : Right. The cells developed into structures that can achieve and maintain their »internal structrue« in the predefined state (genetically defined).

Boris

Eetu


Lähettäjä: Bruce Abbott bbabbott@frontier.com
Lähetetty: 24. helmikuuta 2017 17:16
Vastaanottaja: csgnet@lists.illinois.edu
Aihe: RE: Watch Your p’s and Qi’s

[From Bruce Abbott (2017.02.24.1015 EST)]

[Eetu Pikkarainen 2017-02-24] –>

EP: Thank you Rick for this long and detailed explanation, and thank you Bruce for your clear minded comments.

BA: And thank you, Eetu, for your clear exposition. I have only one correction to suggest. You state that “one of the main strength and originality in Powers insight is that control system is strictly speaking
controlling (only and just) its perceptions.� You seem to be saying that the control system does not control Qi, although I doubt that this was your intention. As you say, “strictly speaking,� the control system directly controls
only that which it perceives, p. However, I will argue that controlling p is only the means by which the system controls Qi, and that the control of Qi is the reason why controlling p is useful to the individual.

BA: By controlling its perceptions, the system also controls the environmental correlates of those controlled perceptions. Controlling p requires taking actions in the environment that affect Qi, and it is
those effects on Qi, out there in the real world, that allow the individual to function successfully. A person who is able to control her hunger (keeping it near a zero reference level) by eating only imaginary food soon will perish. The usual case, of course,
is that keeping the perception of hunger § near a reference value of zero entails consuming real food, which has the effect of nourishing the body while also reducing p. It is this effect of keeping the body nourished that makes controlling hunger worthwhile.

BA: Consider what happens when the link between p and Qi is lost. A rat whose lateral hypothalamus has been destroyed no longer shows any interest in food, and would starve to death amid plenty if the experimenter
did not keep it alive through force feeding. One way to account for this finding is to assume that the lateral hypothalamus (or nerve fibers passing through it) is an essential part of a brain circuit that transduces the bodily state of nutrition (Qi) into
a perception of hunger-level §. If destroying the lateral hypothalamus causes p to be equal to the reference value of zero no matter what the state of Qi, the rat will never eat, as in fact is observed to happen. If p no longer relates to Qi, it may still
be possible to control p, but doing so will no longer confer its essential benefit. We control p
in order to control Qi. Conversely, because it has only p to go on, a control system can control Qi
only by controlling p.

Bruce

Warren, I partially agree with your diagnosis.

I see there at least two issues which make Rick’s PCT seem like RTC. First is his empirism. He is not interested in speculations about how the inside of a controlling organism works, but only in what an observer can perceive in his own environment (and additionally
what the computer of the observer can register). The second is that he uses the PCT conceptions so that they are applied to this empirism: what others think that must be inside the organism he thinks it must be outside and perceivable. This is a problem in
use of terms and concepts and I think this is the original contradiction.

Then one more problem is the radical empirism and its ontological commitments, which require Rick to situate the input quantity inside the observer while others think it is in the environment.

What makes Boris to react so surprisingly harshly and violently, that it is difficult to understand for newcomers, must be the history of frustrations in the discussions. This same history, however, seems
to make those reactions futile, because Rick’s reaction is to dismiss them.

A somewhat sad situation.

Eetu

···

Lähettäjä: Warren Mansell wmansell@gmail.com
Lähetetty: 2. maaliskuuta 2017 16:52
Vastaanottaja: csgnet@lists.illinois.edu
Aihe: Re: Watch Your p’s and Qi’s

Boris, Rick knows everything you are stating about PCT and agrees with it. He doesn’t have an RCT. He is trying to describe the control of elements and functions of the environment that provide us with evidence for PCT independently from PCT, and explain
to us why these need to be stated separately from PCT itself from a scientific, empirical standpoint.

Yet, because PCT has the benefit of also being used to explain the experimenter’s motives and perception, a second stage can be carried out, which I think you are describing, where we try to explain not just what we observe in the experiment but we model
the experimenter controlling their perception too. I think this is also a worthwhile enterprise but for different reasons.

I have a suspicion both you and Rick will disagree with my representation of both your views. I could have got them slightly wrong. But I wonder whether the main reason is that you are using PCT for different explanatory purposes. I am rather expecting
you will both try to tell me they are fundamentally incompatible with one another and with Bill’s vision of PCT, but I see them as complementary.

All the best,

Warren

On 2 Mar 2017, at 14:11, Boris Hartman boris.hartman@masicom.net wrote:

Dear Eetu. I wouldn’t beleive everything that can be »heard« on this forum. I think it’s better to check it in PCT definitions which also explain the loop in LCS III.

From: Eetu Pikkarainen [mailto:eetu.pikkarainen@oulu.fi]
Sent: Friday, February 24, 2017 9:51 PM
To: csgnet@lists.illinois.edu
Subject: VS: Watch Your p’s and Qi’s

Thanks Bruce again, this is very important note and point!

This and your next message about adaptation are very hepful and illuminating.

What I mean is that the “control strictly speaking” or “controlling directly” on the one hand and the “control loosely speaking” or “controlling indirectly” are TWO important things that belong
together (as thick as thieves - hopefully right saying in English), like all of the other parts of the loop. Only the whole loop can control anything. But still I think that it is intellectually corrupting to call two different (however similar) concepts with
the ​same term. It causes unnecessary fights in areas like science where strictness is exceptionally important.

One thing is a consequence of the other - they cannot be the same.

I doesn’t sound very good if you have to say that “PCT is a theory of control, loosely speaking”, I think?

Loosely speaking we can say that in the feedback path of the control loop, the controller controls B by controlling A, D by controlling C and so on and finally that controllers perception is controlled
by its controlling X. But strictly speaking - I think - there should be (beforehand) their own reference level for all of these claimed controls?

Loosely speaking, Fred was quite right with his suggestion [From Fred Nickols (2017.01.29.0950
ET)] that we control in the every phase of the loop.

HB : No Eetu. The control is not happening in the whole loop. If you think like this you’ll have to support this statement with evidence
how control is transffered to environment via neuromuscular connection . How ? And how control enters the organism through perceptual signal ? How control is distributed in the perceptual signal ? This is the problem Rick
has from the beggining of our conversation some years ago. He is misleading forum with his RCT, wrong interpretation of PCT which includes the »control happening in the whole loop«. The easiest way to check this assumption is to make physiological tour arround
the loop, which was already done by Powers in B:CP. And that’s how we got PCT.

If you look at the definitons of loop »elements« you’ll see that there is no control in all of them (post to Rick). So also from PCT definitions we can conclude that control
is not present in the whole control loop. It works in PCT sense something like this :

Bill P and others
(2011) Perceptual Control Theory (PCT) provides a general theory of functioning for organisms. At the core concept of the theory is the observation that living things control perceived environment by means
of their behavior. Consequently, the phenomen of control takes center stage in PCT, with observable behavior playing an important but supporting role.

HB : Behavior is playing a supporting role in control loop. It doesn’t carry control.

Bill P :

Using the internal point of view, we can understand many aspects of behavior by seeing control as control of perception rather than of an objective world.

HB : We can see that there is no control in »objective« world. So there is no »controlled variable« in outer environment or the correlate to »Controlled perception« through
»Control of behavior« as Rick is writing.

Bill P :

A control system controls what it senses, and what it senses is the result of applying a continuous transformaton process to the elementary sensory inputs to the nervous system.

HB : This is feed-back. The control system controls perception with adding effects to disturbances. Comparator is the only place in organism where control happens. No control
is happening outside. When percpetion is controlled there is no other »simultaneous control in outer environment« as Rick and Warren are trying to present. These explanations don’t follow PCT definitions. If you do this you’ll have to introduce Telekinesis
to transfer control to outer environment as Rick is doing.

Applying continuos process in environment means effects of output to the sensors. Feed-back is »effects of output on input«. There is no control in output effects. Otherwise
I beleive that Bill would define »controlled effects to environment« what would mean that behavior is control. But this is not so. See diagrma in LCS III.

And of course there can’t be any control in sensor effects. So there is no »Controlled Perceptual variable« or »Controlled Perception«. Anyway there is no such term in PCT.
Rick inventied them to wrongly support his RCT with wrong definitions (please see post to Rick)

Dear Eetu. If you want to follow the control loop in PCT sense you just have to follow defintions or PCT and physiological evidences that show how control loop works in organism.
And I beleive you will have no problem determining what PCT is about and how control is distributed in »control loop«.

EP : Of course it is necessary for the organism that it can affect, stabilize and adjust its environment so that it can stay
alive and keep its error signals near zero. But this ability has not developed to affect the environment but to keep the organism itself alive i.e. its internal stucture and state in the reference limits.

HB : Right. The cells developed into structures that can achieve and maintain their »internal structrue« in the predefined state (genetically defined).

Boris

Eetu


Lähettäjä: Bruce Abbott bbabbott@frontier.com
Lähetetty: 24. helmikuuta 2017 17:16
Vastaanottaja: csgnet@lists.illinois.edu
Aihe: RE: Watch Your p’s and Qi’s

[From Bruce Abbott (2017.02.24.1015 EST)]

[Eetu Pikkarainen 2017-02-24] –

EP: Thank you Rick for this long and detailed explanation, and thank you Bruce for your clear minded comments.

BA: And thank you, Eetu, for your clear exposition. I have only one correction to suggest. You state that “one of the main strength and originality in Powers insight is that control system is strictly speaking
controlling (only and just) its perceptions.� You seem to be saying that the control system does not control Qi, although I doubt that this was your intention. As you say, “strictly speaking,� the control system directly controls
only that which it perceives, p. However, I will argue that controlling p is only the means by which the system controls Qi, and that the control of Qi is the reason why controlling p is useful to the individual.

BA: By controlling its perceptions, the system also controls the environmental correlates of those controlled perceptions. Controlling p requires taking actions in the environment that affect Qi, and it is
those effects on Qi, out there in the real world, that allow the individual to function successfully. A person who is able to control her hunger (keeping it near a zero reference level) by eating only imaginary food soon will perish. The usual case, of course,
is that keeping the perception of hunger (p) near a reference value of zero entails consuming real food, which has the effect of nourishing the body while also reducing p. It is this effect of keeping the body nourished that makes controlling hunger worthwhile.

BA: Consider what happens when the link between p and Qi is lost. A rat whose lateral hypothalamus has been destroyed no longer shows any interest in food, and would starve to death amid plenty if the experimenter
did not keep it alive through force feeding. One way to account for this finding is to assume that the lateral hypothalamus (or nerve fibers passing through it) is an essential part of a brain circuit that transduces the bodily state of nutrition (Qi) into
a perception of hunger-level (p). If destroying the lateral hypothalamus causes p to be equal to the reference value of zero no matter what the state of Qi, the rat will never eat, as in fact is observed to happen. If p no longer relates to Qi, it may still
be possible to control p, but doing so will no longer confer its essential benefit. We control p
in order to control Qi. Conversely, because it has only p to go on, a control system can control Qi
only by controlling p.

Bruce