# Kenneth J. W. Craik on levels of perception and control

**URL:** <http://discourse.iapct.org/t/kenneth-j-w-craik-on-levels-of-perception-and-control/6035>\
**Category:** CSG2017\
**Created:** [December 7, 2017, 3:26pm UTC](http://discourse.iapct.org/t/kenneth-j-w-craik-on-levels-of-perception-and-control/6035 "2017-12-07T15:26:54Z")\
**Posts on this page:** 20\
**Page:** 1

<div class="post-metadata">

**Author:** ![Abbott](http://discourse.iapct.org/user_avatar/discourse.iapct.org/abbott/32/2879_2.png) [@Abbott](http://discourse.iapct.org/u/Abbott)\
**Post date:** [December 7, 2017, 3:26pm UTC](http://discourse.iapct.org/t/kenneth-j-w-craik-on-levels-of-perception-and-control/6035/1 "2017-12-07T15:26:54Z")

</div>

[From Bruce Abbott (2017.12.07.1025 EST)]

Continuing my series of posts on Kenneth J. W. Craik’s unfinished book, _The Mechanism of Human Action_, Chapter 4 was to be entitled “Brain Mechanisms: Levels.” The chapter begins as follows:

One of the ways of approaching the mechanism of the brain is to think how we should design a machine to do what it does, and to see how far we are driven to the same devices as those used in the brain. One typical example is the segregation into levels, both of patterns of activity and of cerebral structure. It is necessary in the administrative sphere that information should be selected and classified for consideration by the highest officers, and similarly that the commands of those officers should relate to units of activity which are worked out by subordinates. For instance, the C.-in-C., Fighter Command, presumably says: ‘We want a sweep carried out over such and such an area’; he does not have to add, ‘This means that Spitfire No. so-and-so on such and such a station must have so many gallons of petrol in its tanks and care must be taken that its plugs are clean and its guns loaded.’ These latter details are delegated to subordinates. In just the same way, for rapidity and certainty in action, it is essential that certain units of activity, such as looking at an object, walking, grasping, using words, or balancing one’s body, should be delegated to lower levels, in which they are organized as units which can be rapidly and certainly turned on or off at the command of these higher centres. Conversely, on the sensory side, it is clear that it is desirable that we should react sometimes to objects in a simple and unanalytic way—for instance by dodging an approaching motor-car—but that at other times a more closely analytical attack is requisite; the lower levels can react in a simple way to outer objects while leaving the higher levels free for the more analytical treatment.

This is not a description of the levels of perception/control found in Powers’ PCT, but it does show that Craik was thinking along similar lines: The brain is organized into levels of perception and control, with goals at higher levels being pursued by setting the goals (references) to be pursued by the levels below them, and with the lower levels able to deal with simple disturbances without awaiting orders from the higher levels.

It would be interesting to know whether Bill Powers was aware of Craik’s work along these lines. A pair of Craik’s papers was published posthumously (Craik, 1947, 1948) entitled “Theory of the Human Operator in Control Systems” (Parts 1 and 2) that describes some of the ideas that appeared in Craik’s unfinished book, and the book itself, in so far as it was written, appeared in 1966 as part of a book of selected writings. During WW II, Craik was part of a group of individuals whose work contributed to the founding of cybernetics (including Warren McCulloch of “Pitts-McCulloch neuron” fame), and it is possible (although perhaps unlikely) that Bill may have learned of Craik’s work during his contact with the cybernetics movement in the early 1950s.

Comments?

Bruce

---

<div class="post-metadata">

**Author:** ![MartinT](http://discourse.iapct.org/letter_avatar_proxy/v4/letter/m/ee7513/32.png) [@MartinT](http://discourse.iapct.org/u/MartinT)\
**Post date:** [December 7, 2017, 4:24pm UTC](http://discourse.iapct.org/t/kenneth-j-w-craik-on-levels-of-perception-and-control/6035/2 "2017-12-07T16:24:02Z")

</div>

[From Bruce Abbott (2017.12.07.1025 EST)]

```
      Continuing my series of posts on Kenneth J.

```

W. Craik’s unfinished book, _The Mechanism of Human Action_ ,  
Chapter 4 was to be entitled “Brain Mechanisms: Levels.” The  
chapter begins as follows:

```
      One of the ways of

```

approaching the mechanism of the brain is to think how we  
should design a machine to do what it does, and to see how far  
we are driven to the same devices as those used in the brain.  
One typical example is the segregation into levels, both of  
patterns of activity and of cerebral structure. It is  
necessary in the administrative sphere that information should  
be selected and classified for consideration by the highest  
officers, and similarly that the commands of those officers  
should relate to units of activity which are worked out by  
subordinates. For instance, the C.-in-C., Fighter Command,  
presumably says: ‘We want a sweep carried out over such and  
such an area’; he does not have to add, ‘This means that  
Spitfire No. so-and-so on such and such a station must have so  
many gallons of petrol in its tanks and care must be taken  
that its plugs are clean and its guns loaded.’ These latter  
details are delegated to subordinates. In just the same way,  
for rapidity and certainty in action, it is essential that  
certain units of activity, such as looking at an object,  
walking, grasping, using words, or balancing one’s body,  
should be delegated to lower levels, in which they are  
organized as units which can be rapidly and certainly turned  
on or off at the command of these higher centres. Conversely,  
on the sensory side, it is clear that it is desirable that we  
should react sometimes to objects in a simple and unanalytic  
way—for instance by dodging an approaching motor-car—but that  
at other times a more closely analytical attack is requisite;  
the lower levels can react in a simple way to outer objects  
while leaving the higher levels free for the more analytical  
treatment.

```
      This is not a description of the levels of

```

perception/control found in Powers’ PCT, but it does show that  
Craik was thinking along similar lines: The brain is  
organized into levels of perception and control, with goals at  
higher levels being pursued by setting the goals (references)  
to be pursued by the levels below them, and with the lower  
levels able to deal with simple disturbances without awaiting  
orders from the higher levels.

```
      It would be interesting to know whether

```

Bill Powers was aware of Craik’s work along these lines. A  
pair of Craik’s papers was published posthumously (Craik,  
1947, 1948) entitled “Theory of the Human Operator in Control  
Systems” (Parts 1 and 2) that describes some of the ideas that  
appeared in Craik’s unfinished book, and the book itself, in  
so far as it was written, appeared in 1966 as part of a book  
of selected writings. During WW II, Craik was part of a group  
of individuals whose work contributed to the founding of  
cybernetics (including Warren McCulloch of “Pitts-McCulloch  
neuron” fame), and it is possible (although perhaps unlikely)  
that Bill may have learned of Craik’s work during his contact  
with the cybernetics movement in the early 1950s.

Comments?

---

<div class="post-metadata">

**Author:** ![rsmarken](http://discourse.iapct.org/user_avatar/discourse.iapct.org/rsmarken/32/3553_2.png) [@rsmarken](http://discourse.iapct.org/u/rsmarken)\
**Post date:** [December 7, 2017, 11:49pm UTC](http://discourse.iapct.org/t/kenneth-j-w-craik-on-levels-of-perception-and-control/6035/3 "2017-12-07T23:49:13Z")

</div>

[From Rick Marken (2017.12.07.1550)]

> **···**
>
> > Bruce Abbott (2017.12.07.1025 EST)–
> 
> > BA: Continuing my series of posts on Kenneth J. W. Craikâs unfinished book, _The Mechanism of Human Action_, Chapter 4 was to be entitled âBrain Mechanisms: Levels.â?Â The chapter begins as follows:…
> 
> > BA: It would be interesting to know whether Bill Powers was aware of Craikâs work along these lines.Â A pair of Craikâs papers was published posthumously (Craik, 1947, 1948) entitled âTheory of the Human Operator in Control Systemsâ? (Parts 1 and 2) that describes some of the ideas that appeared in Craikâs unfinished book, and the book itself, in so far as it was written, appeared in 1966 as part of a book of selected writings.Â During WW II, Craik was part of a group of individuals whose work contributed to the founding of cybernetics (including Warren McCulloch of âPitts-McCulloch neuronâ? fame), and it is possible (although perhaps unlikely) that Bill may have learned of Craikâs work during his contact with the cybernetics movement in the early 1950s.
> 
> > BA: Comments?
> 
> RM: Sure. Here’s two. First, while it’s certainly interesting to look for possible historical precursors to Powers’ Control Theory (PCT) (one of the best being William James’ description of how to discriminate purposive from non-purposive behavior, which I found in an old copy of his _Principles of Psychology_ while wandering through a used book store) I think it can lead to a mild version of the “nothing but” syndrome that Bill mentions as early as B:CP. That is, it can lead one to think that PCT is “nothing but” a natural evolution from existing ideas when, in fact, PCT is a completely revolutionary model of behavior, based on the never before recognized fact that behavior _is_ control. And second, I highly recommend that people who are interested in developing PCT science put the search for possible historical precursors to PCT on hold until they have put considerable effort into doing research to test and extend the PCT model – research that mainly involves determining the types of perceptual variables organisms control and how they control them.Â
> 
> Best
> 
> Rick
> 
> –  
> Richard S. MarkenÂ
> 
> "Perfection is achieved not when you have nothing more to add, but when you  
> have nothing left to take away.â?  
> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â --Antoine de Saint-Exupery

---

<div class="post-metadata">

**Author:** ![Abbott](http://discourse.iapct.org/user_avatar/discourse.iapct.org/abbott/32/2879_2.png) [@Abbott](http://discourse.iapct.org/u/Abbott)\
**Post date:** [December 9, 2017, 4:28pm UTC](http://discourse.iapct.org/t/kenneth-j-w-craik-on-levels-of-perception-and-control/6035/4 "2017-12-09T16:28:59Z")

</div>

[From Bruce Abbott (2017.12.09.1125 EST)]

Rick Marken (2017.12.07.1550)]

> Bruce Abbott (2017.12.07.1025 EST)–

> BA: Continuing my series of posts on Kenneth J. W. Craikâ€™s unfinished book, _The Mechanism of Human Action_, Chapter 4 was to be entitled â€œBrain Mechanisms: Levels.â€? The chapter begins as follows:…

> BA: It would be interesting to know whether Bill Powers was aware of Craikâ€™s work along these lines. A pair of Craikâ€™s papers was published posthumously (Craik, 1947, 1948) entitled â€œTheory of the Human Operator in Control Systemsâ€? (Parts 1 and 2) that describes some of the ideas that appeared in Craikâ€™s unfinished book, and the book itself, in so far as it was written, appeared in 1966 as part of a book of selected writings. During WW II, Craik was part of a group of individuals whose work contributed to the founding of cybernetics (including Warren McCulloch of â€œPitts-McCulloch neuronâ€? fame), and it is possible (although perhaps unlikely) that Bill may have learned of Craikâ€™s work during his contact with the cybernetics movement in the early 1950s.

> BA: Comments?

RM: Sure. Here’s two. First, while it’s certainly interesting to look for possible historical precursors to Powers’ Control Theory (PCT) (one of the best being William James’ description of how to discriminate purposive from non-purposive behavior, which I found in an old copy of his _Principles of Psychology_ while wandering through a used book store) I think it can lead to a mild version of the “nothing but” syndrome that Bill mentions as early as B:CP. That is, it can lead one to think that PCT is “nothing but” a natural evolution from existing ideas when, in fact, PCT is a completely revolutionary model of behavior, based on the never before recognized fact that behavior _is_ control.

But Rick, PCT _ **is** _ â€œa natural evolution from existing ideasâ€?! (Note that I did not say â€œnothing but.â€?)Â The recognition that behavior is a means by which humans and other animals exert control over certain variables was not new when Bill and colleagues first published the theory that eventually would be called perceptual control theory.

Which leads me to ask, have you read _any_ of the Pre-PCT literature on this topic prior to Powers?Â If so, please indicate which books/articles you have read.

Bruce

---

<div class="post-metadata">

**Author:** ![rsmarken](http://discourse.iapct.org/user_avatar/discourse.iapct.org/rsmarken/32/3553_2.png) [@rsmarken](http://discourse.iapct.org/u/rsmarken)\
**Post date:** [December 10, 2017, 12:19am UTC](http://discourse.iapct.org/t/kenneth-j-w-craik-on-levels-of-perception-and-control/6035/5 "2017-12-10T00:19:39Z")

</div>

[From Rick Marken (2017.12.09.1620)]

> **···**
>
> > Bruce Abbott (2017.12.09.1125 EST)
> 
> > > BA: Comments?
> 
> > RM: Sure. Here’s two. First, while it’s certainly interesting to look for possible historical precursors to Powers’ Control Theory (PCT) …Â I think it can lead to a mild version of the “nothing but” syndrome that Bill mentions as early as B:CP. That is, it can lead one to think that PCT is “nothing but” a natural evolution from existing ideas when, in fact, PCT is a completely revolutionary model of behavior, based on the never before recognized fact that behavior _is_ control.
> 
> > BA: But Rick, PCT _ **is** _ âa natural evolution from existing ideasâ?! (Note that I did not say ânothing but.â?)Â The recognition that behavior is a means by which humans and other animals exert control over certain variables was not new when Bill and colleagues first published the theory that eventually would be called perceptual control theory.
> 
> RM: What they didn’t recognize is that all behavior IS control. They didn’t see that even the behaviors that are the “means by which humans and other animals exert control over certain variables” – behaviors such as moving a handle to keep a cursor on target – are themselves controlled variables. This was a truly revolutionary observation. Other evidence that PCT is revolutionary include the following: 1) PCT shows that the general linear model, the basic paradigm on which all psychological research based, is wrong when applied to the behavior of living systems (Powers, 1978)Â 2) PCT requires a new approach to research, one that is aimed at determining the perceptual variables that organisms control (ibid)Â 3) this approach to methodology has been ignored or rejected by the scientific psychology establishment and 4) PCT itself has been dismissed, ignored or rejected by the scientific psychology establishment.Â
> 
> > BA: Which leads me to ask, have you read _any_ of the Pre-PCT literature on this topic prior to Powers?Â If so, please indicate which books/articles you have read.
> 
> RM: I read Weiner’s books, several books on manual control (Wickens, Poulton, McFarland are names that come to mind) and some other books on Cybernetics (I forget the authors; one was possibly by Ashby). And I had several conversations about manual control with Henry Jex, who was very big in the field of manual control. All of this stuff just seemed like the rest of psychology at the time – disconnected findings that didn’t speak to a general concept of how organisms “work” (although later I realized that they did all speak to one general concept of how organisms work: input causes output).
> 
> BestÂ
> 
> Rick
> 
> Â
> 
> > Â
> 
> > Bruce
> 
> > Â
> 
> > Â
> 
> –
> 
> Richard S. MarkenÂ
> 
> "Perfection is achieved not when you have nothing more to add, but when you  
> have nothing left to take away.â?  
> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â --Antoine de Saint-Exupery

---

<div class="post-metadata">

**Author:** ![Abbott](http://discourse.iapct.org/user_avatar/discourse.iapct.org/abbott/32/2879_2.png) [@Abbott](http://discourse.iapct.org/u/Abbott)\
**Post date:** [December 10, 2017, 4:22pm UTC](http://discourse.iapct.org/t/kenneth-j-w-craik-on-levels-of-perception-and-control/6035/6 "2017-12-10T16:22:17Z")

</div>

[From Bruce Abbott (2017.12.10.1120 EST)]

Rick Marken (2017.12.09.1620)]

> Bruce Abbott (2017.12.09.1125 EST)

> > BA: Comments?

> RM: Sure Here’s two. First, while it’s certainly interesting to look for possible historical precursors to Powers’ Control Theory (PCT) … I think it can lead to a mild version of the “nothing but” syndrome that Bill mentions as early as B:CP. That is, it can lead one to think that PCT is “nothing but” a natural evolution from existing ideas when, in fact, PCT is a completely revolutionary model of behavior, based on the never before recognized fact that behavior _is_ control.

> BA: But Rick, PCT _ **is** _ â€œa natural evolution from existing ideasâ€?! (Note that I did not say â€œnothing but.â€?) The recognition that behavior is a means by which humans and other animals exert control over certain variables was not new when Bill and colleagues first published the theory that eventually would be called perceptual control theory.

RM: What they didn’t recognize is that all behavior IS control. They didn’t see that even the behaviors that are the “means by which humans and other animals exert control over certain variables” – behaviors such as moving a handle to keep a cursor on target – are themselves controlled variables. This was a truly revolutionary observation.

I have elsewhere referred to this assertion as the â€œPowers conjecture.â€?Â It is an assumption to be tested empirically.Â As the field of cybernetics emerged, its developers saw clearly that purposive behavior was carried out via control-systems, but it was not clear to them whether other behaviors might occur via open-loop mechanisms.

RM: Other evidence that PCT is revolutionary include the following: 1) PCT shows that the general linear model, the basic paradigm on which all psychological research based, is wrong when applied to the behavior of living systems (Powers, 1978)

You are conflating the general linear model with lineal causality.Â The general linear model is the model on which statistical analyses such as ANOVA and linear regression are based.Â Your mistaken â€œhidden biasâ€? analysis of the power law makes use of the general linear model to find the so-called â€œhidden bias.â€?Â Lineal causality refers to open-loop cause and effect.Â As Iâ€™ve shown fairly recently on CSGnet, one can find numerous examples of research findings that are not invalidated by the presence of control systems.

1. PCT requires a new approach to research, one that is aimed at determining the perceptual variables that organisms control (ibid) 3) this approach to methodology has been ignored or rejected by the scientific psychology establishment and 4) PCT itself has been dismissed, ignored or rejected by the scientific psychology establishment.

> BA: Which leads me to ask, have you read _any_ of the Pre-PCT literature on this topic prior to Powers? If so, please indicate which books/articles you have read.

RM: I read Weiner’s books, several books on manual control (Wickens, Poulton, McFarland are names that come to mind) and some other books on Cybernetics (I forget the authors; one was possibly by Ashby). And I had several conversations about manual control with Henry Jex, who was very big in the field of manual control. All of this stuff just seemed like the rest of psychology at the time – disconnected findings that didn’t speak to a general concept of how organisms “work” (although later I Â realized that they did all speak to one general concept of how organisms work: input causes output).

I am aware only of one book by Wiener (not Weiner): _Cybernetics, or control and communication in the animal and the machine_, first published in 1948.Â This is the book that Bill P. credited with starting him on the path to PCT. Â Are there others by Wiener that you read?Â Ashby wrote two books, _Design for a brain_ (first published in 1952) and _Introduction to cybernetics_ (first published in 1956).Â The first presented a method of reorganization based on so-called â€œessential variablesâ€? that Bill P. incorporated into PCT (while changing â€œessentialâ€? to â€œintrinsicâ€?).Â The second provides an introduction to cybernetics that even a mathematically challenged person like me could understand.

As for your conclusion that these sources â€œdid all speak to one general concept of how organisms work: input causes output,â€?Â it is a good thing that Bill P. found in at least some of them valuable ideas that led ultimately to his insight that the function of behavior is to control oneâ€™s perceptions, and that the control systems through which such control is achieved might naturally emerge through the operation of a reorganizing system.

Bruce

---

<div class="post-metadata">

**Author:** ![MartinT](http://discourse.iapct.org/letter_avatar_proxy/v4/letter/m/ee7513/32.png) [@MartinT](http://discourse.iapct.org/u/MartinT)\
**Post date:** [December 10, 2017, 4:33pm UTC](http://discourse.iapct.org/t/kenneth-j-w-craik-on-levels-of-perception-and-control/6035/7 "2017-12-10T16:33:29Z")

</div>

[Martin Taylor 2017.12.10.11.31]

> [From Rick Marken (2017.12.09.1620)]

```
Rick, what perception are you controlling that is so disturbed by

```

the idea that Bill Powers was a _human_ scientist rather than  
a God who could create something from nothing?

```
Martin

```

> **···**
>
> > ```
> > Bruce Abbott
> > 
> > ```
> > 
> > (2017.12.09.1125 EST)
> 
> > > BA: Comments?
> 
> > ```
> > RM: Sure. Here's two.
> > 
> > ```
> > 
> > First, while it’s certainly interesting to look  
> > for possible historical precursors to Powers’  
> > Control Theory (PCT) …Â I think it can lead to  
> > a mild version of the “nothing but” syndrome  
> > that Bill mentions as early as B:CP. That is, it  
> > can lead one to think that PCT is “nothing but”  
> > a natural evolution from existing ideas when, in  
> > fact, PCT is a completely revolutionary model of  
> > behavior, based on the never before recognized  
> > fact that behavior _is_ control.
> 
> > BA: But Rick, PCT _ **is** _  
> > â€œa natural evolution from existing ideasâ€?! (Note  
> > that I did not say â€œnothing but.â€?)Â The  
> > recognition that behavior is a means by which  
> > humans and other animals exert control over  
> > certain variables was not new when Bill and  
> > colleagues first published the theory that  
> > eventually would be called perceptual control  
> > theory.
> 
> ```
> RM: What they didn't recognize is that all behavior IS
> 
> ```
> 
> control. They didn’t see that even the behaviors that are  
> the “means by which humans and other animals exert control  
> over certain variables” – behaviors such as moving a  
> handle to keep a cursor on target – are themselves  
> controlled variables. This was a truly revolutionary  
> observation. Other evidence that PCT is revolutionary …

---

<div class="post-metadata">

**Author:** ![rsmarken](http://discourse.iapct.org/user_avatar/discourse.iapct.org/rsmarken/32/3553_2.png) [@rsmarken](http://discourse.iapct.org/u/rsmarken)\
**Post date:** [December 10, 2017, 8:16pm UTC](http://discourse.iapct.org/t/kenneth-j-w-craik-on-levels-of-perception-and-control/6035/8 "2017-12-10T20:16:21Z")

</div>

[From Rick Marken (2017.12.10.1215)]

> **···**
>
> > Â Bruce Abbott (2017.12.10.1120 EST)
> 
> > RM: What they didn’t recognize is that all behavior IS control. They didn’t see that even the behaviors that are the “means by which humans and other animals exert control over certain variables” – behaviors such as moving a handle to keep a cursor on target – are themselves controlled variables. This was a truly revolutionary observation.
> 
> > BA: I have elsewhere referred to this assertion as the âPowers conjecture.â?Â It is an assumption to be tested empirically.Â
> 
> RM: You betcha. You should try it sometime.Â
> 
> > RM: Other evidence that PCT is revolutionary include the following: 1) PCT shows that the general linear model, the basic paradigm on which all psychological research based, is wrong when applied to the behavior of living systems (Powers, 1978)Â
> 
> > BA: You are conflating the general linear model with lineal causality.Â
> 
> RM: Which is appropriate (according to the conventional paradigm in psychology) when the predictor variables are the independent variables and the criterion variable is the dependent variable in an experimental study.Â
> 
> > BA: Your mistaken âhidden biasâ? analysis of the power law makes use of the general linear model to find the so-called âhidden bias.â?Â
> 
> RM:Â As does your _truly_ mistaken power law as well; mistaken in the sense that this “law” is taken to reflect something about the “constraints” on the physiological processes that produce the movement when, of course, it doesn’t. And there is no mistake about my “hidden bias” analysis (actually, it’s calledÂ “omitted variable bias” (OVB) analysis; nothing is hidden); it accounts for the observed deviations from the 1/3 (or 2/3) power law perfectly.Â
> 
> > BA: Lineal causality refers to open-loop cause and effect.Â As Iâve shown fairly recently on CSGnet, one can find numerous examples of research findings that are not invalidated by the presence of control systems.
> 
> RM: PCT doesn’t invalidate findings; it invalidates the interpretation of the findings. The finding that rats press a lever more often when the press is followed by food is not invalidated by PCT. What is invalidated is the idea that it’s the food selecting the press. PCT shows that it is actually the press that is being used to select the food.
> 
> > BA: I am aware only of one book by Wiener (not Weiner): _Cybernetics, or control and communication in the animal and the machine_, first published in 1948. This is the book that Bill P. credited with starting him on the path to PCT.Â Â
> 
> RM: When I say that PCT was revolutionary I didn’t mean that it had nothing to do with the many achievements that went before. Control theory itself was developed well before PCT and Bill certainly was influenced by that. But PCT represented a revolutionary break from these influences. That doesn’t mean he discarded everything that went before; he just applied these ideas in a revolutionary new way. If you read the first few sections of Bill’s 1978 _Psych Review_ paper you will see that Bill pays homage to his predecessors in cybernetics, control theory and man-machine systems development but points out where they went wrong before launching into a description of his revolutionary re-conceptualization of control theory as it should be applied to the behavior of living systems – aÂ re-conceptualization based on his rather well supported “conjecture” that the behavior of living systems IS control.Â
> 
> > BA: Are there others by Wiener that you read?Â
> 
> RM: Yes, one called “The Human Use of Human Beings” which is even more relevant today than when he wrote it.Â
> 
> > BA: Ashby wrote two books, _Design for a brain_ (first published in 1952) and _Introduction to cybernetics_ (first published in 1956).Â
> 
> RM: Yes, I think I read (or skimmed) both.Â Â
> 
> > BA: The first presented a method of reorganization based on so-called âessential variablesâ? that Bill P. incorporated into PCT (while changing âessentialâ? to âintrinsicâ?).Â The second provides an introduction to cybernetics that even a mathematically challenged person like me could understand.
> 
> RM: I encountered those books before I found B:CP. And B:CP might have just rolled off me like those books did. But I had the good fortune of getting a personal computer in 1978, finding Bill’s 1978 _Psych Review_ article and 1979Â _Byte_ articles, programmed up the demos and models, which blew my mind, did some experiments and I was off to the races. It was the hands-on experience with the demos and models that helped me understand Powers’ control theory and see that it was the real deal, revolution-wise.
> 
> > BA: As for your conclusion that these sources âdid all speak to one general concept of how organisms work: input causes output,â?Â it is a good thing that Bill P. found in at least some of them valuable ideas that led ultimately to his insight that the function of behavior is to control oneâs perceptions, and that the control systems through which such control is achieved might naturally emerge through the operation of a reorganizing system.
> 
> RM: There is no question that Bill found some valuable ideas from these sources; Bill certainly acknowledges his intellectual debt to these sources. But where Bill went with these ideas was truly revolutionary. All of these sources assumed that input causes output. Based on his understanding of behavior as control, Bill showed that what is actually happening is output controlling input.
> 
> BestÂ
> 
> Rick
> 
> > Â
> 
> > Bruce
> 
> –  
> Richard S. MarkenÂ
> 
> "Perfection is achieved not when you have nothing more to add, but when you  
> have nothing left to take away.â?  
> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â --Antoine de Saint-Exupery

---

<div class="post-metadata">

**Author:** ![\_Boris\_Hartman](http://discourse.iapct.org/letter_avatar_proxy/v4/letter/_/e8c25b/32.png) [@\_Boris\_Hartman](http://discourse.iapct.org/u/_Boris_Hartman)\
**Post date:** [December 18, 2017, 7:04pm UTC](http://discourse.iapct.org/t/kenneth-j-w-craik-on-levels-of-perception-and-control/6035/9 "2017-12-18T19:04:15Z")

</div>

Just in shortâ€¦

RM : ….but points out where they went wrong before launchiing into a description of his revolutionary re-conceptualization of control theory as it should be applied to the behavior of living systems – a re-conceptualization based on his rather well supported “conjecture” that the behavior of living systems IS control.

RM : But where Bill went with these ideas was truly revolutionary. All of these sources assumed that input causes output. Based on his understanding of behavior as control, Bill showed that what is actually happening is output controlling input.

HB : Rick. Your nonsense thinking that Â»Behavior can be controlÂ« or that Â»Behavior is controlling inputÂ« was proved as wrong many times. Behavior among other effects also affect input.

The best description of your wrong understanding was by my oppnion given by Rupert :

RY earlier : Sure, a perceptual signal (q.i\*g) may correspond to, or be a function of, variable aspects of the environment (q.i) but it is the perceptual signal that is controlled not the variable aspects of the environment.

HB : Rupert assumes right that only perceptual signal is controlled in comparator. This is only control in the whole loop. It’s nervous system that is controlling.

Bill P (LCS I):

The Living Control System of this kind must sense the controlled quantity in each dimenssion **in which the quantity is to be controlled** ; this implies the inner model of the quantity in the form of a signal or set of signals.

HB : In both cases you can see that perceptual signal (quantity) is to be controlled in comparatorÂ (it’s future tense). Â»Perceptual signal is only controlled variable there is in control loop. At least in PCT.

Bill P :

FEED-BACK FUNCTION : The box represents the set of physical laws, properties, arrangements, linkages, by which the action of this system feeds-back to affect its own input, the controlled variable. That’s what feed-back means : it’s an effect of a system’s output on it’s own input.

There are many other evidences that your statements are worng,so I don’t understand why are you doing this. To save your worthless books and articles where you are promoting wrong ideas about PCT or better where you promote your RCT ?

Let us finnish your sharade once for all.

1. 

```
 All you have to prove is that you can control your behavior (output) ? Bill proved that you can't. Let me see your evidences thatyou canâ€¦

```

2. 

```
 And you have to prove that there is some Â»Controlled Perceptual VariableÂ« in PCT ?

```

If you prove these Â»factsÂ« we’ll have to beleive that you are right ?

But if you’ll not prove it I expect that you’ll shut up and stop confussing people on CSGnet and go somewhere else where you could sell your bullshitt.

You are contradicting the main Bills’ finding that Â»Perception is controlled, not behaviorÂ«. Do you understand this.

Why don’t you beleive Bill and Mary Powers :

Mary Powers :

PCT requires a major shift in thinking from the traditional approach : that what is controlled is not behavior, but perception.

HB : I donâ€™t understand that after all these years you canâ€™t make major shift in your mind from â€œBehavior is controlâ€? to â€œPerception is controlledâ€?. And why you donâ€™t believe them ???

Bill P : (LCS III) : Note that we classify the controlled variable as an input variable, not an output variable.

Bill P : OUTPUT FUNCTION : The portion of a system that converts the magnitude or state of a signal inside the system into corresponding set of effects on the immediate environment of the system.

HB : So you can see that behavior is not controlled variable,It’s just effects to immediate environment.

HB : There is no controlled effects to environmentâ€¦There is nothing controlled in outer environment. It’s just your imagination…

So will you except my proposal for IAPCT that Bills’ diagram (LCS III)and definitions (B:CP) are acccepted for the bases of it’s representation of Bills theory to the world ?Â

Boris

> **···**
>
> **From:** Richard Marken [[mailto:rsmarken@gmail.com](mailto:rsmarken@gmail.com)]  
> **Sent:** Sunday, December 10, 2017 9:16 PM  
> **To:** csgnet@lists.illinois.edu  
> **Subject:** Re: Kenneth J. W. Craik on levels of perception and control
> 
> [From Rick Marken (2017.12.10.1215)]
> 
> > Bruce Abbott (2017.12.10.1120 EST)
> 
> > RM: What they didn’t recognize is that all behavior IS control. They didn’t see that even the behaviors that are the “means by which humans and other animals exert control over certain variables” – behaviors such as moving a handle to keep a cursor on target – are themselves controlled variables. This was a truly revolutionary observation.
> 
> > BA: I have elsewhere referred to this assertion as the â€œPowers conjecture.â€? It is an assumption to be tested empirically.
> 
> RM: You betcha. You should try it sometime.
> 
> > RM: Other evidence that PCT is revolutionary include the following: 1) PCT shows that the general linear model, the basic paradigm on which all psychological research based, is wrong when applied to the behavior of living systems (Powers, 1978)
> 
> > BA: You are conflating the general linear model with lineal causality.
> 
> RM: Which is appropriate (according to the conventional paradigm in psychology) when the predictor variables are the independent variables and the criterion variable is the dependent variable in an experimental study.
> 
> > BA: Your mistaken â€œhidden biasâ€? analysis of the power law makes use of the general linear model to find the so-called â€œhidden bias.â€?
> 
> RM: As does your _truly_ mistaken power law as well; mistaken in the sense that this “law” is taken to reflect something about the “constraints” on the physiological processes that produce the movement when, of course, it doesn’t. And there is no mistake about my “hidden bias” analysis (actually, it’s called “omitted variable bias” (OVB) analysis; nothing is hidden); it accounts for the observed deviations from the 1/3 (or 2/3) power law perfectly.
> 
> > BA: Lineal causality refers to open-loop cause and effect. As Iâ€™ve shown fairly recently on CSGnet, one can find numerous examples of research findings that are not invalidated by the presence of control systems.
> 
> RM: PCT doesn’t invalidate findings; it invalidates the interpretation of the findings. The finding that rats press a lever more often when the press is followed by food is not invalidated by PCT. What is invalidated is the idea that it’s the food selecting the press. PCT shows that it is actually the press that is being used to select the food.
> 
> > BA: I am aware only of one book by Wiener (not Weiner): _Cybernetics, or control and communication in the animal and the machine_, first published in 1948. This is the book that Bill P. credited with starting him on the path to PCT.
> 
> RM: When I say that PCT was revolutionary I didn’t mean that it had nothing to do with the many achievements that went before. Control theory itself was developed well before PCT and Bill certainly was influenced by that. But PCT represented a revolutionary break from these influences. That doesn’t mean he discarded everything that went before; he just applied these ideas in a revolutionary new way. If you read the first few sections of Bill’s 1978 _Psych Review_ paper you will see that Bill pays homage to his predecessors in cybernetics, control theory and man-machine systems development but points out where they went wrong before launching into a description of his revolutionary re-conceptualization of control theory as it should be applied to the behavior of living systems – a re-conceptualization based on his rather well supported “conjecture” that the behavior of living systems IS control.
> 
> > BA: Are there others by Wiener that you read?
> 
> RM: Yes, one called “The Human Use of Human Beings” which is even more relevant today than when he wrote it.
> 
> > BA: Ashby wrote two books, _Design for a brain_ (first published in 1952) and _Introduction to cybernetics_ (first published in 1956).
> 
> RM: Yes, I think I read (or skimmed) both.
> 
> > BA: The first presented a method of reorganization based on so-called â€œessential variablesâ€? that Bill P. incorporated into PCT (while changing â€œessentialâ€? to â€œintrinsicâ€?). The second provides an introduction to cybernetics that even a mathematically challenged person like me could understand.
> 
> RM: I encountered those books before I found B:CP. And B:CP might have just rolled off me like those books did. But I had the good fortune of getting a personal computer in 1978, finding Bill’s 1978 _Psych Review_ article and 1979 _Byte_ articles, programmed up the demos and models, which blew my mind, did some experiments and I was off to the races. It was the hands-on experience with the demos and models that helped me understand Powers’ control theory and see that it was the real deal, revolution-wise.
> 
> > BA: As for your conclusion that these sources â€œdid all speak to one general concept of how organisms work: input causes output,â€? it is a good thing that Bill P. found in at least some of them valuable ideas that led ultimately to his insight that the function of behavior is to control oneâ€™s perceptions, and that the control systems through which such control is achieved might naturally emerge through the operation of a reorganizing system.
> 
> RM: There is no question that Bill found some valuable ideas from these sources; Bill certainly acknowledges his intellectual debt to these sources. But where Bill went with these ideas was truly revolutionary. All of these sources assumed that input causes output. Based on his understanding of behavior as control, Bill showed that what is actually happening is output controlling input.
> 
> Best
> 
> Rick
> 
> > Bruce
> 
> –
> 
> Richard S. Marken
> 
> "Perfection is achieved not when you have nothing more to add, but when you  
> have nothing left to take away.â€?  
> –Antoine de Saint-Exupery

---

<div class="post-metadata">

**Author:** ![rsmarken](http://discourse.iapct.org/user_avatar/discourse.iapct.org/rsmarken/32/3553_2.png) [@rsmarken](http://discourse.iapct.org/u/rsmarken)\
**Post date:** [December 20, 2017, 1:03am UTC](http://discourse.iapct.org/t/kenneth-j-w-craik-on-levels-of-perception-and-control/6035/10 "2017-12-20T01:03:52Z")

</div>

[From Rick Marken (2017.12.19.1700)]

[sizeareasqtestwdisttop.xlsx](http://discourse.iapct.org/uploads/short-url/yPUwTxkV21BRCF5lVKTWpqx9d3a.xlsx) (30.6 KB)

> **···**
>
> On Mon, Dec 18, 2017 at 11:04 AM, Boris Hartman [boris.hartman@masicom.net](mailto:boris.hartman@masicom.net) wrote:
> 
> > HB : Rick. Your nonsense thinking that Â»Behavior can be controlÂ« or that Â»Behavior is controlling inputÂ« was proved as wrong many times. Behavior among other effects also affect input.
> 
> > The best description of your wrong understanding was by my oppnion given by Rupert :
> 
> > RY earlier : Sure, a perceptual signal (q.i\*g) may correspond to, or be a function of, variable aspects of the environment (q.i) but it is the perceptual signal that is controlled not the variable aspects of the environment.
> 
> > HB : Rupert assumes right that only perceptual signal is controlled in comparator. This is only control in the whole loop. It’s nervous system that is controlling.
> 
> RM: It is true that only the perceptual signal is controlled by a control system. But this doesn’t mean that the aspect of the environment that corresponds to that signal is not controlled. All it means is that a disturbance to the perceptual signal itself will be compensated for by the actions of the system, bringing the perceptual signal back into a match with the reference signal. Of course, a disturbance to a perceptual signal is a rather invasive process, involving adding a stream of electricalÂ impulses directly into the normal firing rate of the neuron(s) carrying the perceptual signal. But even if such disturbances were introduced directly into a perceptual signal, PCT predicts that the aspect of the environment that corresponds to that perceptual signal will still be controlled; all that will be changed is the reference state in which that aspect of the environment is maintained. This is demonstrated in the attached spreadsheet.Â
> 
> RM: This spreadsheet implements a control system that controls either the area or perimeter of a rectangle, depending on the nature of its perceptual function. Which perceptual function the control system uses is determined by the value entered into cell N1. If you enter 1, the system perceives and controls the perimeter of the rectangle; if you enter 0 the system perceives and controls the area of the rectangle. The graph shows how well the system is controlling perimeter and area. The spreadsheet comes as an area controller and the graph shows that area is well controlled because the area of the rectangleÂ --Â the aspect of the environment under control-- stays constant (the flat purple line) over time despite disturbances (the wavy blue line), which are variations in the height of the rectangle . The area line remains flat because the perceptual signal that corresponds to area is being kept matching a constant reference signal (value = 400); so when the perceptual signal is kept matching the constant reference signal the aspect of the environment that corresponds to the perceptual signal remains in a constant reference state.Â The fact that perimeter is not controlled is shown by the fact that perimeter varies over time (the wavy green line) due to the effects of the disturbance and the system’s own outputs, which are acting to control area.
> 
> RM: You can test the effect of directly disturbing the perceptual signal by entering a number greater than 0 and less than 50 into cell K1. Try 50 for starters. The control model should automatically run when you hit “Enter” after entering this disturbance but you can also force the model to run by pressing the F9 key. What you will see is that, despite the disturbance to the perceptual signal, the area of the rectangle remains under control; the purple line stays flat. But the area is now being maintained in a different reference state. Indeed, if you entered a disturbance of 50 to the perceptual signalÂ then you will see that the purple line has moved from 400 area units to 350. This is because the disturbance (of 50) to the perceptual signal has made it so that an area of 350, which would result in a perceptual signal value of 350 if there were no disturbance to the perceptual signal, now results in a perceptual signal of 400, which matches the reference signal value.Â
> 
> RM: Now go back to a perceptual signal disturbance of 0 (no disturbance) and change the system from an area to a perimeter controller by changing the value in cell N1 from a 0 to a 1. What you will see after the model runs (automatically or by pressing F9) is that now the perimeter of the rectangle is held constant (flat green line) in a reference state of 80 units. And the area is no longer controlled, as indicated by the now wavy purple line. SoÂ the aspect of the environment that is now being controlled is the perimeter rather than the area of the rectangle. If you now introduce a disturbance of 50 to the perceptual signal you will see that perimeter remains under control but at a new reference level, 30 rather than 80. Again, this is because the disturbance (of 50) to the perceptual signal has made it so that a perimeter of 30, which would result in a perceptual signal value of 30 if there were no disturbance to the perceptual signal, now results in a perceptual signal of 80, which matches the reference signal value.Â
> 
> RM: The point of this little demo is that, while it is true that the perceptual signal, p, is what is controlled in a control system, the aspect of the environment that corresponds to that signal, called the controlled quantity or q.i in PCT, is also controlled. That is, in PCT p = q.i.Â A disturbance to the perceptual signal doesn’t change what is being controlled; it just afects the reference state in which is is controlled. And this is true because p is a theoretical construct that is part of the explanation of how q.i is controlled; q.i and the reference state thereof is a fact; p is part of the explanation of that fact. The explanation, by the way, is called “control theory”
> 
> Â
> 
> > Â Bill P (LCS I):
> 
> > The Living Control System of this kind must sense the controlled quantity in each dimenssion **in which the quantity is to be controlled** ; this implies the inner model of the quantity in the form of a signal or set of signals.
> 
> > HB : n both cases you can see that perceptual signal (quantity) is to be controlled in comparatorÂ (it’s future tense). Â»Perceptual signal is only controlled variable there is in control loop. At least in PCT.
> 
> RM:Â The fact that the controlled quantity (q.i)Â _must be sensed_Â means that it is external to the control system. The “inner model” of that controlled quantity is i\*n the form of a signal,Â \*which is the perceptual signal (p). So Bill is making the point here that I make with the attached spreadsheet; the controlled quantity, q.i, is represented in the system in the form of a signal. p: in other words, p=q.i.Â
> 
> BestÂ
> 
> Rick
> 
> Â
> 
> > 
> 
> > Â Bill P :
> 
> > FEED-BACK FUNCTION : The box represents the set of physical laws, properties, arrangements, linkages, by which the action of this system feeds-back to affect its own input, the controlled variable. That’s what feed-back means : it’s an effect of a system’s output on it’s own input.
> 
> > Â
> 
> > Â
> 
> > There are many other evidences that your statements are worng,so I don’t understand why are you doing this. To save your worthless books and articles where you are promoting wrong ideas about PCT or better where you promote your RCT ?
> 
> > Â
> 
> > Let us finnish your sharade once for all.
> 
> > Â
> 
> > 1.Â Â Â Â Â All you have to prove is that you can control your behavior (output) ? Bill proved that you can’t. Let me see your evidences thatyou can…
> 
> > Â
> 
> > 2.Â Â Â Â Â And you have to prove that there is some Â»Controlled Perceptual VariableÂ« in PCT ?
> 
> > Â
> 
> > If you prove these Â»factsÂ« we’ll have to beleive that you are right ?
> 
> > Â
> 
> > But if you’ll not prove it I expect that you’ll shut up and stop confussing people on CSGnet and go somewhere else where you could sell your bullshitt.
> 
> > Â
> 
> > You are contradicting the main Bills’ finding that Â»Perception is controlled, not behaviorÂ«. Do you understand this.
> 
> > Â
> 
> > Why don’t you beleive Bill and Mary Powers :
> 
> > Â
> 
> > Mary Powers :
> 
> > PCT requires a major shift in thinking from the traditional approach : that what is controlled is not behavior, but perception.
> 
> > Â
> 
> > HB : I donâ€™t understand that after all these years you canâ€™t make major shift in your mind from â€œBehavior is controlâ€? to â€œPerception is controlledâ€?. And why you donâ€™t believe them ???
> 
> > Â
> 
> > Bill P : (LCS III) : Note that we classify the controlled variable as an input variable, not an output variable.
> 
> > Â
> 
> > Bill P : OUTPUT FUNCTION : The portion of a system that converts the magnitude or state of a signal inside the system into corresponding set of effects on the immediate environment of the system.
> 
> > Â
> 
> > HB : So you can see that behavior is not controlled variable,It’s just effects to immediate environment.
> 
> > Â
> 
> > Â
> 
> > HB : There is no controlled effects to environment…There is nothing controlled in outer environment. It’s just yyour imagination…
> 
> > Â
> 
> > So will you except my proposal for IAPCT that Bills’ diagram (LCS III)and definitions (B:CP) are acccepted for the bases of it’s representation of Bills theory to the world ?Â
> 
> > Â
> 
> > Boris
> 
> > Â
> 
> > Â
> 
> > Â
> 
> > Â
> 
> > Â
> 
> > Â
> 
> > **From:** Richard Marken [[mailto:rsmarken@gmail.com](mailto:rsmarken@gmail.com)]  
> > **Sent:** Sunday, December 10, 2017 9:16 PM  
> > **To:** csgnet@lists.illinois.edu  
> > **Subject:** Re: Kenneth J. W. Craik on levels of perception and control
> 
> > Â
> 
> > [From Rick Marken (2017.12.10.1215)]
> 
> > > Â Bruce Abbott (2017.12.10.1120 EST)
> 
> > > RM: What they didn’t recognize is that all behavior IS control. They didn’t see that even the behaviors that are the “means by which humans and other animals exert control over certain variables” – behaviors such as moving a handle to keep a cursor on target – are themselves controlled variables. This was a truly revolutionary observation.
> 
> > > BA: I have elsewhere referred to this assertion as the â€œPowers conjecture.â€?Â It is an assumption to be tested empirically.Â
> 
> > RM: You betcha. You should try it sometime.Â
> 
> > > RM: Other evidence that PCT is revolutionary include the following: 1) PCT shows that the general linear model, the basic paradigm on which all psychological research based, is wrong when applied to the behavior of living systems (Powers, 1978)Â
> 
> > > BA: You are conflating the general linear model with lineal causality.Â
> 
> > RM: Which is appropriate (according to the conventional paradigm in psychology) when the predictor variables are the independent variables and the criterion variable is the dependent variable in an experimental study.Â
> 
> > > BA: Your mistaken â€œhidden biasâ€? analysis of the power law makes use of the general linear model to find the so-called â€œhidden bias.â€?Â
> 
> > RM:Â As does your _truly_ mistaken power law as well; mistaken in the sense that this “law” is taken to reflect something about the “constraints” on the physiological processes that produce the movement when, of course, it doesn’t. And there is no mistake about my “hidden bias” analysis (actually, it’s calledÂ “omitted variable bias” (OVB) analysis; nothing is hidden); it accounts for the observed deviations from the 1/3 (or 2/3) power law perfectly.Â
> 
> > > BA: Lineal causality refers to open-loop cause and effect.Â As Iâ€™ve shown fairly recently on CSGnet, one can find numerous examples of research findings that are not invalidated by the presence of control systems.
> 
> > RM: PCT doesn’t invalidate findings; it invalidates the interpretation of the findings. The finding that rats press a lever more often when the press is followed by food is not invalidated by PCT. What is invalidated is the idea that it’s the food selecting the press. PCT shows that it is actually the press that is being used to select the food.
> 
> > > BA: I am aware only of one book by Wiener (not Weiner): _Cybernetics, or control and communication in the animal and the machine_, first published in 1948. This is the book that Bill P. credited with starting him on the path to PCT.Â Â
> 
> > RM: When I say that PCT was revolutionary I didn’t mean that it had nothing to do with the many achievements that went before. Control theory itself was developed well before PCT and Bill certainly was influenced by that. But PCT represented a revolutionary break from these influences. That doesn’t mean he discarded everything that went before; he just applied these ideas in a revolutionary new way. If you read the first few sections of Bill’s 1978 _Psych Review_ paper you will see that Bill pays homage to his predecessors in cybernetics, control theory and man-machine systems development but points out where they went wrong before launching into a description of his revolutionary re-conceptualization of control theory as it should be applied to the behavior of living systems – aÂ re-conceptualization based on his rather well supported “conjecture” that the behavior of living systems IS control.Â
> 
> > > BA: Are there others by Wiener that you read?Â
> 
> > RM: Yes, one called “The Human Use of Human Beings” which is even more relevant today than when he wrote it.Â
> 
> > > BA: Ashby wrote two books, _Design for a brain_ (first published in 1952) and _Introduction to cybernetics_ (first published in 1956).Â
> 
> > RM: Yes, I think I read (or skimmed) both.Â Â
> 
> > > BA: The first presented a method of reorganization based on so-called â€œessential variablesâ€? that Bill P. incorporated into PCT (while changing â€œessentialâ€? to â€œintrinsicâ€?).Â The second provides an introduction to cybernetics that even a mathematically challenged person like me could understand.
> 
> > RM: I encountered those books before I found B:CP. And B:CP might have just rolled off me like those books did. But I had the good fortune of getting a personal computer in 1978, finding Bill’s 1978 _Psych Review_ article and 1979Â _Byte_ articles, programmed up the demos and models, which blew my mind, did some experiments and I was off to the races. It was the hands-on experience with the demos and models that helped me understand Powers’ control theory and see that it was the real deal, revolution-wise.
> 
> > > BA: As for your conclusion that these sources â€œdid all speak to one general concept of how organisms work: input causes output,â€?Â it is a good thing that Bill P. found in at least some of them valuable ideas that led ultimately to his insight that the function of behavior is to control oneâ€™s perceptions, and that the control systems through which such control is achieved might naturally emerge through the operation of a reorganizing system.
> 
> > RM: There is no question that Bill found some valuable ideas from these sources; Bill certainly acknowledges his intellectual debt to these sources. But where Bill went with these ideas was truly revolutionary. All of these sources assumed that input causes output. Based on his understanding of behavior as control, Bill showed that what is actually happening is output controlling input.
> 
> > Â
> 
> > BestÂ
> 
> > Â
> 
> > Rick
> 
> > > Â
> 
> > > Bruce
> 
> > 
> 
> > Â
> 
> > –
> 
> > Richard S. MarkenÂ
> 
> > "Perfection is achieved not when you have nothing more to add, but when you  
> > have nothing left to take away.â€?  
> > Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â --Antoine de Saint-Exupery
> 
> –  
> Richard S. MarkenÂ
> 
> "Perfection is achieved not when you have nothing more to add, but when you  
> have nothing left to take away.â€?  
> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â --Antoine de Saint-Exupery

---

<div class="post-metadata">

**Author:** ![MartinT](http://discourse.iapct.org/letter_avatar_proxy/v4/letter/m/ee7513/32.png) [@MartinT](http://discourse.iapct.org/u/MartinT)\
**Post date:** [December 20, 2017, 4:50am UTC](http://discourse.iapct.org/t/kenneth-j-w-craik-on-levels-of-perception-and-control/6035/11 "2017-12-20T04:50:47Z")

</div>

[Martin Taylor 2017.12.19.23.28]

```
No. p(t) = f(q.i, t), where f might involve a delay, some

```

time-smearing such as a leaky integral, and/or a nonlinear  
transform. Even that is not strictly accurate, because the signal is  
in the form of neural firings, and is therefore noisy. So it is  
better to put p(t) = f(q.i, t) + n(t), where n(t) represents a noise  
waveform. When the signal are large, n(t) doesn’t matter much, but  
just try controlling a perception of colour in very low light. What  
you see isn’t always what you get. Before the light is so dim that  
everything is grey, you might see the same surface fluctuating in  
colour over quite a wide range if you look carefully. But I agree  
with the general import of what Rick is trying to say. The problem is, as it has been for years, in the precise definition  
of “control”. A disagreement in which each disputant uses a  
different definition of the word in dispute is inherently  
irresoluble. Boris likes the concept of control as in theory  
bringing some variable toward a commensurate reference variable,  
whether or not either variable can be observed. Rick doesn’t dispute  
that as a possibility, but includes as “control” the appearance that  
an observable variable seems to be brought toward and maintained  
near some unobservable virtual reference value, which can be deduced  
only by observing the variable said to be controlled. At least  
that’s my understanding of the bone of contention. It’s not a  
dispute about PCT, but about the correct writing of a dictionary  
definition. It is much the same as would be a dispute about the word  
“perception”, in which one party claimed that a perception is  
everything of which he was consciously aware and nothing else, while  
the other claimed it meant a signal within a control system that  
depends on real or imagined sensory input and is capable of being  
controlled. Such an argument would never be resolved, either.  
Martin

> **···**
>
> On 2017/12/19 8:03 PM, Richard Marken  
> wrote:
> 
> > [From Rick Marken (2017.12.19.1700)]
> > 
> > ```
> > Bill
> > 
> > ```
> > 
> > P (LCS I):
> > 
> > > ```
> > > The
> > > 
> > > ```
> > > 
> > > Living Control System of this kind must sense the  
> > > controlled quantity in each dimenssion \*\* in which  
> > > the quantity is to be controlled\*\* ; this  
> > > implies the inner model of the quantity in the  
> > > form of a signal or set of signals.
> 
> > > ```
> > > HB : n both cases you
> > > 
> > > ```
> > > 
> > > can see that perceptual signal (quantity) is to be controlled  
> > > in comparator (it’s future tense). »Perceptual  
> > > signal is only controlled variable there is in  
> > > control loop. At least in PCT.
> 
> > RM: The fact that the controlled quantity (q.i) \* must  
> > be sensed\* means that it is external to the control  
> > system. The “inner model” of that controlled quantity is i\* n  
> > the form of a signal, \* which is the perceptual signal  
> > (p). So Bill is making the point here that I make with the  
> > attached spreadsheet; the controlled quantity, q.i, is  
> > represented in the system in the form of a signal. p: in  
> > other words, p=q.i.
> 
> > 
> 
> >

---

<div class="post-metadata">

**Author:** ![wmansell](http://discourse.iapct.org/user_avatar/discourse.iapct.org/wmansell/32/36_2.png) [@wmansell](http://discourse.iapct.org/u/wmansell)\
**Post date:** [December 20, 2017, 11:11am UTC](http://discourse.iapct.org/t/kenneth-j-w-craik-on-levels-of-perception-and-control/6035/12 "2017-12-20T11:11:59Z")

</div>

Thanks Martin - this neat analysis gets right into the intersection between Rick’s and Boris’s seemingly endless debate and hopeful will get them both to pause for thought…

Warren

> **···**
>
> On 2017/12/19 8:03 PM, Richard Marken  
> wrote:
> 
> > [From Rick Marken (2017.12.19.1700)]
> > 
> > ```
> > Bill
> > 
> > ```
> > 
> > P (LCS I):
> > 
> > > ```
> > > The
> > > 
> > > ```
> > > 
> > > Living Control System of this kind must sense the  
> > > controlled quantity in each dimenssion \*\* in which  
> > > the quantity is to be controlled\*\* ; this  
> > > implies the inner model of the quantity in the  
> > > form of a signal or set of signals.
> 
> > > ```
> > > HB : n both cases you
> > > 
> > > ```
> > > 
> > > can see that perceptual signal (quantity) is to be controlled  
> > > in comparator (it’s future tense). »Perceptual  
> > > signal is only controlled variable there is in  
> > > control loop. At least in PCT.
> 
> > RM: The fact that the controlled quantity (q.i) \* must  
> > be sensed\* means that it is external to the control  
> > system. The “inner model” of that controlled quantity is i\* n  
> > the form of a signal, \* which is the perceptual signal  
> > (p). So Bill is making the point here that I make with the  
> > attached spreadsheet; the controlled quantity, q.i, is  
> > represented in the system in the form of a signal. p: in  
> > other words, p=q.i.
> 
> > 
> 
> >

---

<div class="post-metadata">

**Author:** ![FredNickols](http://discourse.iapct.org/user_avatar/discourse.iapct.org/frednickols/32/2645_2.png) [@FredNickols](http://discourse.iapct.org/u/FredNickols)\
**Post date:** [December 20, 2017, 11:31am UTC](http://discourse.iapct.org/t/kenneth-j-w-craik-on-levels-of-perception-and-control/6035/13 "2017-12-20T11:31:47Z")

</div>

[From Fred Nickols (2017.12.20.0628 ET)]

I am responding only to the snippet immediately below (which appears near the end of Rickâ€™s response to Boris:

RM: The fact that the controlled quantity (q.i) _must be sensed_ means that it is external to the control system. The “inner model” of that controlled quantity is i\*n the form of a signal, \*which is the perceptual signal (p). So Bill is making the point here that I make with the attached spreadsheet; the controlled quantity, q.i, is represented in the system in the form of a signal. p: in other words, p=q.i.

FN:Â It seems to me that it is incorrect to say that p=q.i.Â P is no doubt some function of q.i. but it is not equal to nor the equivalent of q.i.Â So I would write p=f(q.i.).

Fred Nickols

> **···**
>
> **From:** Richard Marken [[mailto:rsmarken@gmail.com](mailto:rsmarken@gmail.com)]  
> **Sent:** Tuesday, December 19, 2017 8:04 PM  
> **To:** csgnet@lists.illinois.edu  
> **Subject:** Re: Kenneth J. W. Craik on levels of perception and control
> 
> [From Rick Marken (2017.12.19.1700)]
> 
> On Mon, Dec 18, 2017 at 11:04 AM, Boris Hartman [boris.hartman@masicom.net](mailto:boris.hartman@masicom.net) wrote:
> 
> > HB : Rick. Your nonsense thinking that Â»Behavior can be controlÂ« or that Â»Behavior is controlling inputÂ« was proved as wrong many times. Behavior among other effects also affect input.
> 
> > The best description of your wrong understanding was by my oppnion given by Rupert :
> 
> > RY earlier : Sure, a perceptual signal (q.i\*g) may correspond to, or be a function of, variable aspects of the environment (q.i) but it is the perceptual signal that is controlled not the variable aspects of the environment.
> 
> > HB : Rupert assumes right that only perceptual signal is controlled in comparator. This is only control in the whole loop. It’s nervous system that is controlling.
> 
> RM: It is true that only the perceptual signal is controlled by a control system. But this doesn’t mean that the aspect of the environment that corresponds to that signal is not controlled. All it means is that a disturbance to the perceptual signal itself will be compensated for by the actions of the system, bringing the perceptual signal back into a match with the reference signal. Of course, a disturbance to a perceptual signal is a rather invasive process, involving adding a stream of electrical impulses directly into the normal firing rate of the neuron(s) carrying the perceptual signal. But even if such disturbances were introduced directly into a perceptual signal, PCT predicts that the aspect of the environment that corresponds to that perceptual signal will still be controlled; all that will be changed is the reference state in which that aspect of the environment is maintained. This is demonstrated in the attached spreadsheet.
> 
> RM: This spreadsheet implements a control system that controls either the area or perimeter of a rectangle, depending on the nature of its perceptual function. Which perceptual function the control system uses is determined by the value entered into cell N1. If you enter 1, the system perceives and controls the perimeter of the rectangle; if you enter 0 the system perceives and controls the area of the rectangle. The graph shows how well the system is controlling perimeter and area. The spreadsheet comes as an area controller and the graph shows that area is well controlled because the area of the rectangle – the aspect of the environment under control-- stays constant (the flat purple line) over time despite disturbances (the wavy blue line), which are variations in the height of the rectangle . The area line remains flat because the perceptual signal that corresponds to area is being kept matching a constant reference signal (value = 400); so when the perceptual signal is kept matching the constant reference signal the aspect of the environment that corresponds to the perceptual signal remains in a constant reference state. The fact that perimeter is not controlled is shown by the fact that perimeter varies over time (the wavy green line) due to the effects of the disturbance and the system’s own outputs, which are acting to control area.
> 
> RM: You can test the effect of directly disturbing the perceptual signal by entering a number greater than 0 and less than 50 into cell K1. Try 50 for starters. The control model should automatically run when you hit “Enter” after entering this disturbance but you can also force the model to run by pressing the F9 key. What you will see is that, despite the disturbance to the perceptual signal, the area of the rectangle remains under control; the purple line stays flat. But the area is now being maintained in a different reference state. Indeed, if you entered a disturbance of 50 to the perceptual signal then you will see that the purple line has moved from 400 area units to 350. This is because the disturbance (of 50) to the perceptual signal has made it so that an area of 350, which would result in a perceptual signal value of 350 if there were no disturbance to the perceptual signal, now results in a perceptual signal of 400, which matches the reference signal value.
> 
> RM: Now go back to a perceptual signal disturbance of 0 (no disturbance) and change the system from an area to a perimeter controller by changing the value in cell N1 from a 0 to a 1. What you will see after the model runs (automatically or by pressing F9) is that now the perimeter of the rectangle is held constant (flat green line) in a reference state of 80 units. And the area is no longer controlled, as indicated by the now wavy purple line. So the aspect of the environment that is now being controlled is the perimeter rather than the area of the rectangle. If you now introduce a disturbance of 50 to the perceptual signal you will see that perimeter remains under control but at a new reference level, 30 rather than 80. Again, this is because the disturbance (of 50) to the perceptual signal has made it so that a perimeter of 30, which would result in a perceptual signal value of 30 if there were no disturbance to the perceptual signal, now results in a perceptual signal of 80, which matches the reference signal value.
> 
> RM: The point of this little demo is that, while it is true that the perceptual signal, p, is what is controlled in a control system, the aspect of the environment that corresponds to that signal, called the controlled quantity or q.i in PCT, is also controlled. That is, in PCT p = q.i. A disturbance to the perceptual signal doesn’t change what is being controlled; it just afects the reference state in which is is controlled. And this is true because p is a theoretical construct that is part of the explanation of how q.i is controlled; q.i and the reference state thereof is a fact; p is part of the explanation of that fact. The explanation, by the way, is called “control theory”
> 
> > Bill P (LCS I):
> 
> > The Living Control System of this kind must sense the controlled quantity in each dimenssion **in which the quantity is to be controlled** ; this implies the inner model of the quantity in the form of a signal or set of signals.
> 
> > HB : n both cases you can see that perceptual signal (quantity) is to be controlled in comparator (it’s future tense). Â»Perceptual signal is only controlled variable there is in control loop. At least in PCT.
> 
> RM: The fact that the controlled quantity (q.i) _must be sensed_ means that it is external to the control system. The “inner model” of that controlled quantity is i\*n the form of a signal, \*which is the perceptual signal (p). So Bill is making the point here that I make with the attached spreadsheet; the controlled quantity, q.i, is represented in the system in the form of a signal. p: in other words, p=q.i.
> 
> Best
> 
> Rick
> 
> > Bill P :
> 
> > FEED-BACK FUNCTION : The box represents the set of physical laws, properties, arrangements, linkages, by which the action of this system feeds-back to affect its own input, the controlled variable. That’s what feed-back means : it’s an effect of a system’s output on it’s own input.
> 
> > 
> 
> > 
> 
> > There are many other evidences that your statements are worng,so I don’t understand why are you doing this. To save your worthless books and articles where you are promoting wrong ideas about PCT or better where you promote your RCT ?
> 
> > 
> 
> > Let us finnish your sharade once for all.
> 
> > 
> 
> > 1. 
> > 
> > ```
> > All you have to prove is that you can control your behavior (output) ? Bill proved that you can't. Let me see your evidences thatyou can…
> > 
> > ```
> 
> > 
> 
> > 1. 
> > 
> > ```
> > And you have to prove that there is some Â»Controlled Perceptual VariableÂ« in PCT ?
> > 
> > ```
> 
> > 
> 
> > If you prove these Â»factsÂ« we’ll have to beleive that you are right ?
> 
> > 
> 
> > But if you’ll not prove it I expect that you’ll shut up and stop confussing people on CSGnet and go somewhere else where you could sell your bullshitt.
> 
> > 
> 
> > You are contradicting the main Bills’ finding that Â»Perception is controlled, not behaviorÂ«. Do you understand this.
> 
> > 
> 
> > Why don’t you beleive Bill and Mary Powers :
> 
> > 
> 
> > Mary Powers :
> 
> > PCT requires a major shift in thinking from the traditional approach : that what is controlled is not behavior, but perception.
> 
> > 
> 
> > HB : I donâ€™t understand that after all these years you canâ€™t make major shift in your mind from â€œBehavior is controlâ€? to â€œPerception is controlledâ€?. And why you donâ€™t believe them ???
> 
> > 
> 
> > Bill P : (LCS III) : Note that we classify the controlled variable as an input variable, not an output variable.
> 
> > 
> 
> > Bill P : OUTPUT FUNCTION : The portion of a system that converts the magnitude or state of a signal inside the system into corresponding set of effects on the immediate environment of the system.
> 
> > 
> 
> > HB : So you can see that behavior is not controlled variable,It’s just effects to immediate environment.
> 
> > 
> 
> > 
> 
> > HB : There is no controlled effects to environment…There is nothing controllled in outer environment. It’s just your imagination…
> 
> > 
> 
> > So will you except my proposal for IAPCT that Bills’ diagram (LCS III)and definitions (B:CP) are acccepted for the bases of it’s representation of Bills theory to the world ?
> 
> > 
> 
> > Boris
> 
> > 
> 
> > 
> 
> > **From:** Richard Marken [[mailto:rsmarken@gmail.com](mailto:rsmarken@gmail.com)]  
> > **Sent:** Sunday, December 10, 2017 9:16 PM  
> > **To:** csgnet@lists.illinois.edu  
> > **Subject:** Re: Kenneth J. W. Craik on levels of perception and control
> 
> > [From Rick Marken (2017.12.10.1215)]
> 
> > > Bruce Abbott (2017.12.10.1120 EST)
> 
> > > RM: What they didn’t recognize is that all behavior IS control. They didn’t see that even the behaviors that are the “means by which humans and other animals exert control over certain variables” – behaviors such as moving a handle to keep a cursor on target – are themselves controlled variables. This was a truly revolutionary observation.
> 
> > > BA: I have elsewhere referred to this assertion as the â€œPowers conjecture.â€? It is an assumption to be tested empirically.
> 
> > RM: You betcha. You should try it sometime.
> 
> > > RM: Other evidence that PCT is revolutionary include the following: 1) PCT shows that the general linear model, the basic paradigm on which all psychological research based, is wrong when applied to the behavior of living systems (Powers, 1978)
> 
> > > BA: You are conflating the general linear model with lineal causality.
> 
> > RM: Which is appropriate (according to the conventional paradigm in psychology) when the predictor variables are the independent variables and the criterion variable is the dependent variable in an experimental study.
> 
> > > BA: Your mistaken â€œhidden biasâ€? analysis of the power law makes use of the general linear model to find the so-called â€œhidden bias.â€?
> 
> > RM: As does your _truly_ mistaken power law as well; mistaken in the sense that this “law” is taken to reflect something about the “constraints” on the physiological processes that produce the movement when, of course, it doesn’t. And there is no mistake about my “hidden bias” analysis (actually, it’s called “omitted variable bias” (OVB) analysis; nothing is hidden); it accounts for the observed deviations from the 1/3 (or 2/3) power law perfectly.
> 
> > > BA: Lineal causality refers to open-loop cause and effect. As Iâ€™ve shown fairly recently on CSGnet, one can find numerous examples of research findings that are not invalidated by the presence of control systems.
> 
> > RM: PCT doesn’t invalidate findings; it invalidates the interpretation of the findings. The finding that rats press a lever more often when the press is followed by food is not invalidated by PCT. What is invalidated is the idea that it’s the food selecting the press. PCT shows that it is actually the press that is being used to select the food.
> 
> > > BA: I am aware only of one book by Wiener (not Weiner): _Cybernetics, or control and communication in the animal and the machine_, first published in 1948. This is the book that Bill P. credited with starting him on the path to PCT.
> 
> > RM: When I say that PCT was revolutionary I didn’t mean that it had nothing to do with the many achievements that went before. Control theory itself was developed well before PCT and Bill certainly was influenced by that. But PCT represented a revolutionary break from these influences. That doesn’t mean he discarded everything that went before; he just applied these ideas in a revolutionary new way. If you read the first few sections of Bill’s 1978 _Psych Review_ paper you will see that Bill pays homage to his predecessors in cybernetics, control theory and man-machine systems development but points out where they went wrong before launching into a description of his revolutionary re-conceptualization of control theory as it should be applied to the behavior of living systems – a re-conceptualization based on his rather well supported “conjecture” that the behavior of living systems IS control.
> 
> > > BA: Are there others by Wiener that you read?
> 
> > RM: Yes, one called “The Human Use of Human Beings” which is even more relevant today than when he wrote it.
> 
> > > BA: Ashby wrote two books, _Design for a brain_ (first published in 1952) and _Introduction to cybernetics_ (first published in 1956).
> 
> > RM: Yes, I think I read (or skimmed) both.
> 
> > > BA: The first presented a method of reorganization based on so-called â€œessential variablesâ€? that Bill P. incorporated into PCT (while changing â€œessentialâ€? to â€œintrinsicâ€?). The second provides an introduction to cybernetics that even a mathematically challenged person like me could understand.
> 
> > RM: I encountered those books before I found B:CP. And B:CP might have just rolled off me like those books did. But I had the good fortune of getting a personal computer in 1978, finding Bill’s 1978 _Psych Review_ article and 1979 _Byte_ articles, programmed up the demos and models, which blew my mind, did some experiments and I was off to the races. It was the hands-on experience with the demos and models that helped me understand Powers’ control theory and see that it was the real deal, revolution-wise.
> 
> > > BA: As for your conclusion that these sources â€œdid all speak to one general concept of how organisms work: input causes output,â€? it is a good thing that Bill P. found in at least some of them valuable ideas that led ultimately to his insight that the function of behavior is to control oneâ€™s perceptions, and that the control systems through which such control is achieved might naturally emerge through the operation of a reorganizing system.
> 
> > RM: There is no question that Bill found some valuable ideas from these sources; Bill certainly acknowledges his intellectual debt to these sources. But where Bill went with these ideas was truly revolutionary. All of these sources assumed that input causes output. Based on his understanding of behavior as control, Bill showed that what is actually happening is output controlling input.
> 
> > Best
> 
> > Rick
> 
> > >
> 
> > > Bruce
> 
> > 
> 
> > –
> 
> > Richard S. Marken
> 
> > "Perfection is achieved not when you have nothing more to add, but when you  
> > have nothing left to take away.â€?  
> > –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.â€?  
> –Antoine de Saint-Exupery

---

<div class="post-metadata">

**Author:** ![rsmarken](http://discourse.iapct.org/user_avatar/discourse.iapct.org/rsmarken/32/3553_2.png) [@rsmarken](http://discourse.iapct.org/u/rsmarken)\
**Post date:** [December 20, 2017, 9:59pm UTC](http://discourse.iapct.org/t/kenneth-j-w-craik-on-levels-of-perception-and-control/6035/14 "2017-12-20T21:59:48Z")

</div>

[From Rick Marken (2017.12.20.1400)]

> **···**
>
> On Wed, Dec 20, 2017 at 3:11 AM, Warren Mansell [wmansell@gmail.com](mailto:wmansell@gmail.com) wrote:
> 
> > Thanks Martin - this neat analysis gets right into the intersection between Rickâ€™s and Borisâ€™s seemingly endless debate and hopeful will get them both to pause for thought…
> 
> RM: First, the “endless debate” between Boris and I has been rather one-sided. Boris spews his ignorant hatred and I generally ignore it since it is so ridiculous. I didn’t ignore it some of what he said this time because I saw it as an opportunity to explain the relationship between q.i (fact) and p (theory). As far as the “neatness” of Martin’s analysis, I’m afraid that, after pausing to think about it, I have concluded that it is just more of the usual “theory first” (or “theory only”) treatment of PCT. Apparently it rings a lot of people’s bell here but not mine. I’ll stick with Bill Powers. Oh, how I miss him.
> 
> BestÂ
> 
> Rick
> 
> Â
> 
> > Warren
> 
> > 
> 
> > On 20 Dec 2017, at 04:50, Martin Taylor [mmt-csg@mmtaylor.net](mailto:mmt-csg@mmtaylor.net) wrote:
> 
> > > [Martin Taylor 2017.12.19.23.28]
> 
> > > ```
> > > On 2017/12/19 8:03 PM, Richard Marken
> > > 
> > > ```
> > > 
> > > wrote:
> 
> > > > [From Rick Marken (2017.12.19.1700)]
> 
> > > ```
> > > No. p(t) = f(q.i, t), where f might involve a delay, some
> > > 
> > > ```
> > > 
> > > time-smearing such as a leaky integral, and/or a nonlinear  
> > > transform. Even that is not strictly accurate, because the signal is  
> > > in the form of neural firings, and is therefore noisy. So it is  
> > > better to put p(t) = f(q.i, t) + n(t), where n(t) represents a noise  
> > > waveform. When the signal are large, n(t) doesn’t matter much, but  
> > > just try controlling a perception of colour in very low light. What  
> > > you see isn’t always what you get. Before the light is so dim that  
> > > everything is grey, you might see the same surface fluctuating in  
> > > colour over quite a wide range if you look carefully. But I agree  
> > > with the general import of what Rick is trying to say.
> > > 
> > > ```
> > > The problem is, as it has been for years, in the precise definition
> > > 
> > > ```
> > > 
> > > of “control”. A disagreement in which each disputant uses a  
> > > different definition of the word in dispute is inherently  
> > > irresoluble. Boris likes the concept of control as in theory  
> > > bringing some variable toward a commensurate reference variable,  
> > > whether or not either variable can be observed. Rick doesn’t dispute  
> > > that as a possibility, but includes as “control” the appearance that  
> > > an observable variable seems to be brought toward and maintained  
> > > near some unobservable virtual reference value, which can be deduced  
> > > only by observing the variable said to be controlled. At least  
> > > that’s my understanding of the bone of contention. It’s not a  
> > > dispute about PCT, but about the correct writing of a dictionary  
> > > definition.
> > > 
> > > ```
> > > It is much the same as would be a dispute about the word
> > > 
> > > ```
> > > 
> > > “perception”, in which one party claimed that a perception is  
> > > everything of which he was consciously aware and nothing else, while  
> > > the other claimed it meant a signal within a control system that  
> > > depends on real or imagined sensory input and is capable of being  
> > > controlled. Such an argument would never be resolved, either.
> > > 
> > > ```
> > > Martin
> > > 
> > > ```
> 
> –  
> Richard S. MarkenÂ
> 
> "Perfection is achieved not when you have nothing more to add, but when you  
> have nothing left to take away.â€?  
> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â --Antoine de Saint-Exupery
> 
> Â Bill  
> P (LCS I):
> 
> > ```
> > The
> > 
> > ```
> > 
> > Living Control System of this kind must sense the  
> > controlled quantity in each dimenssion \*\* in which  
> > the quantity is to be controlled\*\* ; this  
> > implies the inner model of the quantity in the  
> > form of a signal or set of signals.
> 
> > ```
> > HB : n both cases you
> > 
> > ```
> > 
> > can see that perceptual signal (quantity) is to be controlled  
> > in comparatorÂ (it’s future tense). Â»Perceptual  
> > signal is only controlled variable there is in  
> > control loop. At least in PCT.
> 
> RM:Â The fact that the controlled quantity (q.i)Â \* must  
> be sensed\* Â means that it is external to the control  
> system. The “inner model” of that controlled quantity is i\* n  
> the form of a signal,Â \* which is the perceptual signal  
> (p). So Bill is making the point here that I make with the  
> attached spreadsheet; the controlled quantity, q.i, is  
> represented in the system in the form of a signal. p: in  
> other words, p=q.i.Â

---

<div class="post-metadata">

**Author:** ![MartinT](http://discourse.iapct.org/letter_avatar_proxy/v4/letter/m/ee7513/32.png) [@MartinT](http://discourse.iapct.org/u/MartinT)\
**Post date:** [December 21, 2017, 4:29am UTC](http://discourse.iapct.org/t/kenneth-j-w-craik-on-levels-of-perception-and-control/6035/15 "2017-12-21T04:29:03Z")

</div>

[Martin Taylor 2017.12.20.17.35]

```
There speaks the Ultimate Authority on what other people perceive or

```

believe, and on what they have said and written.  
When we talk about control in a PCT context we ask what the theory  
says, and compare that with what we observe. Or we ask about what we  
observe and judge whether the theory agrees, modifying the theory  
(or so PCT would suggest) if there’s a mismatch. It’s something  
called a “feedback loop”, in which there’s no “theory first” or  
“observations first” – at least not if you have any understanding  
at all of what Bill wrought.  
There was nothing new about the kinds of observations that led to  
PCT when Bill made them, following Aristotle, John Dewey, (and I  
have recently learned, two different Kenneth Craiks). Bill’s genius  
was to understand that the old idea of hierarchic perception and the  
not-so-old concept of the servomechanism, together with the  
recognition that it was sensor input data that servomechanisms  
control, all fitted together perfectly as a foundational theory, not  
only of psychology, but of life itself. There aren’t many geniuses  
who can do things like that. Even fewer, if any, can produce a  
totally new fundamental idea out of nothing, not even Einstein. You don’t have to exalt Bill into a kind of God who could and did  
create something out of nothing, and made it all perfect. Indeed, I  
think it is counterproductive to pretend you can reasonably do so.  
People who might find PCT useful if they were to learn it may be put  
off when they learn that only a demigod can understand it.  
What Bill did, and this is the sign of true genius, was create  
something quite new out of a lot of old bits and pieces that had  
been lying around being used for small things of little moment. Bill  
made something that is useful for many small things that never  
before seemed to be connected, and for some big things, too. PCT has  
an awful lot of unrealized potentiality, way beyond “the search for  
the controlled variable”. On the one hand you exalt Bill into a kind  
of God, and on the other you try to chain him down so as to make it  
seem that he created nothing of any consequence. But you really  
can’t do that with much success, The “Powers of Perceptual Control”  
are stronger than your chains. And they are great indeed.  
Although Bill’s observations may have led him to PCT (how else could  
PCT have happened?), the theory did not lead to the idea that we  
seem to perceive things and states in our environment and act on  
them to make them be as we wish them to be. That, too, is an  
observation we each make independently. Since each of us (I am  
refusing the solipsism theory) does it for ourselves, most theories  
would presume that others do it, too. PCT suggest how.  
Martin

> **···**
>
> On 2017/12/20 4:59 PM, Richard Marken  
> wrote:
> 
> > [From Rick Marken (2017.12.20.1400)]
> > 
> > ```
> > On Wed, Dec 20, 2017 at 3:11 AM,
> > 
> > ```
> > 
> > Warren Mansell [wmansell@gmail.com](mailto:wmansell@gmail.com)  
> > wrote:
> > 
> > > ```
> > > Thanks Martin - this neat analysis gets right into
> > > 
> > > ```
> > > 
> > > the intersection between Rick’s and Boris’s seemingly  
> > > endless debate and hopeful will get them both to pause  
> > > for thought…
> 
> > 
> 
> > ```
> > RM: ...As far as the "neatness" of Martin's analysis,
> > 
> > ```
> > 
> > I’m afraid that, after pausing to think about it, I have  
> > concluded that it is just more of the usual “theory first”  
> > (or “theory only”) treatment of PCT.

---

<div class="post-metadata">

**Author:** ![EetuP](http://discourse.iapct.org/letter_avatar_proxy/v4/letter/e/df705f/32.png) [@EetuP](http://discourse.iapct.org/u/EetuP)\
**Post date:** [December 21, 2017, 10:08am UTC](http://discourse.iapct.org/t/kenneth-j-w-craik-on-levels-of-perception-and-control/6035/16 "2017-12-21T10:08:03Z")

</div>

[Eetu Pikkarainen 2017-12-21]

Yes I agree that Martin condensed the core of the dispute very well. I would suggest that much of the dispute could be solved (or at least calmed down) if the discussants would use the  
terms â€œcontrol(e)â€? for the externally observable phenomenon of environmental control/stabilization and â€œcontrol(p)â€? for the only theoretically understandable phenomenon of control of perception. Or if they could read the term â€œcontrolâ€? used by others as â€œcontrol(e)â€?  
or â€œcontrol(p)â€? respectively according to the context.

But still there were left another problem: is PCT about one or another concepts or both and what is their relationship; and should we consequently concentrate or restrict to one or another  
concept. To this dispute I would suggest a MOL kind of solution: get a higher perception and save both.

> **···**
>
> Eetu
> 
> Please, regard all my statements as questions,
> 
> no matter how they are formulated.
> 
> **From:** Warren Mansell [[mailto:wmansell@gmail.com](mailto:wmansell@gmail.com)]  
> **Sent:** Wednesday, December 20, 2017 1:12 PM  
> **To:** csgnet@lists.illinois.edu  
> **Subject:** Re: Kenneth J. W. Craik on levels of perception and control
> 
> Thanks Martin - this neat analysis gets right into the intersection between Rickâ€™s and Borisâ€™s seemingly endless debate and hopeful will get them both to pause for thought…
> 
> Warren
> 
> On 20 Dec 2017, at 04:50, Martin Taylor [mmt-csg@mmtaylor.net](mailto:mmt-csg@mmtaylor.net) wrote:
> 
> > [Martin Taylor 2017.12.19.23.28]
> 
> > On 2017/12/19 8:03 PM, Richard Marken wrote:
> 
> > > [From Rick Marken (2017.12.19.1700)]
> 
> > > Bill P (LCS I):
> 
> > > > The Living Control System of this kind must sense the controlled quantity in each dimenssion **in which the quantity is to be controlled** ; this implies the inner model of the quantity in the form  
> > > > of a signal or set of signals.
> 
> > > > HB : n both cases you can see that perceptual signal (quantity)  
> > > > is to be controlled in comparator (it’s future tense). Â»Perceptual signal is only controlled variable there is in control loop. At least in PCT.
> 
> > > RM: The fact that the controlled quantity (q.i) _must be sensed_ means that it is external to the control system. The “inner model” of that controlled quantity is i\*n the form of a signal, \* which is the perceptual signal (p).  
> > > So Bill is making the point here that I make with the attached spreadsheet; the controlled quantity, q.i, is represented in the system in the form of a signal. p: in other words, p=q.i.
> 
> > No. p(t) = f(q.i, t), where f might involve a delay, some time-smearing such as a leaky integral, and/or a nonlinear transform. Even that is not strictly accurate, because the signal is in the form of neural  
> > firings, and is therefore noisy. So it is better to put p(t) = f(q.i, t) + n(t), where n(t) represents a noise waveform. When the signal are large, n(t) doesn’t matter much, but just try controlling a perception of colour in very low light. What you see isn’t  
> > always what you get. Before the light is so dim that everything is grey, you might see the same surface fluctuating in colour over quite a wide range if you look carefully. But I agree with the general import of what Rick is trying to say.
> > 
> > The problem is, as it has been for years, in the precise definition of “control”. A disagreement in which each disputant uses a different definition of the word in dispute is inherently irresoluble. Boris likes the concept of control as in theory bringing some  
> > variable toward a commensurate reference variable, whether or not either variable can be observed. Rick doesn’t dispute that as a possibility, but includes as “control” the appearance that an observable variable seems to be brought toward and maintained near  
> > some unobservable virtual reference value, which can be deduced only by observing the variable said to be controlled. At least that’s my understanding of the bone of contention. It’s not a dispute about PCT, but about the correct writing of a dictionary definition.
> > 
> > It is much the same as would be a dispute about the word “perception”, in which one party claimed that a perception is everything of which he was consciously aware and nothing else, while the other claimed it meant a signal within a control system that depends  
> > on real or imagined sensory input and is capable of being controlled. Such an argument would never be resolved, either.
> > 
> > Martin

---

<div class="post-metadata">

**Author:** ![\_Boris\_Hartman](http://discourse.iapct.org/letter_avatar_proxy/v4/letter/_/e8c25b/32.png) [@\_Boris\_Hartman](http://discourse.iapct.org/u/_Boris_Hartman)\
**Post date:** [December 21, 2017, 10:49pm UTC](http://discourse.iapct.org/t/kenneth-j-w-craik-on-levels-of-perception-and-control/6035/17 "2017-12-21T22:49:28Z")

</div>

Downâ€¦

> **···**
>
> **From:** Richard Marken [[mailto:rsmarken@gmail.com](mailto:rsmarken@gmail.com)]  
> **Sent:** Wednesday, December 20, 2017 11:00 PM  
> **To:** csgnet@lists.illinois.edu  
> **Subject:** Re: Kenneth J. W. Craik on levels of perception and control
> 
> [From Rick Marken (2017.12.20.1400)]
> 
> On Wed, Dec 20, 2017 at 3:11 AM, Warren Mansell [wmansell@gmail.com](mailto:wmansell@gmail.com) wrote:
> 
> Thanks Martin - this neat analysis gets right into the intersection between Rickâ€™s and Borisâ€™s seemingly endless debate and hopeful will get them both to pause for thought…
> 
> RM: First, the “endless debate” between Boris and I has been rather one-sided. Boris spews his ignorant hatred and I generally ignore it since it is so ridiculous.
> 
> HB : It’s true that I exhibit hatred because manipulative kind of human deosn’t deserve any other treatment. About who is ridiculus we can talk.
> 
> First you need to explain your ridiculous statements. So I’ll repeat all your nonsense in these last years so that people will understand what to avoid in your manipulations.
> 
> 1. 
> 
> ```
> People control people all the ime
> 
> ```
> 
> 2. 
> 
> ```
> All events in control loop happen at the same time.
> 
> ```
> 
> 3. 
> 
> ```
> There is not only perception but also some Â»extrasensory perceptionÂ«
> 
> ```
> 
> 4. 
> 
> ```
> There is always some Â»controlled variableÂ« in environment of Living control system
> 
> ```
> 
> 5. 
> 
> ```
> Organisms are generaly protected from disturbances
> 
> ```
> 
> 6. 
> 
> ```
> To the extent that LCS controls inside it also controls outside.
> 
> ```
> 
> 7. 
> 
> ```
> Behavior is itself a controlled variable
> 
> ```
> 
> 8. 
> 
> ```
> There is some Â»Controlled Perceptual VariableÂ«….
> 
> ```
> 
> > 
> 
> 1. 
> 
> ```
> There are some Â»side effectsÂ« of output on Â»controlled variableÂ« in external environmentâ€¦ So it seems that are some Â»controlled effectsÂ« of output on CV and some Â»sideÂ« or noncontrolled effects on environment.
> 
> ```
> 
> 2. Rick has some Telepathic ability for reading people mind with TCV
> 
> 3. There seems to be some Â»Third eyeÂ« or third Â»z control unitÂ«, beside x and y (left and right eye), for analyzing objects moving in 3 D space (x,y,z). The case with toy helicopter. It seems that nervous system is functioning with independent control units only in 2-dimensions (x,y).
> 
> And we had to add some new :
> 
> 12 Â That is, in PCT p = q.i.
> 
> 13 In the PCT model, p is a noiseless function of environmental variables.
> 
> 14 The taste of lemonade exists as an _aspect_ (function) of the environment (the “environment” being the variables in the world according to physics)
> 
> RM : I didn’t ignore it some of what he said this time because I saw it as an opportunity to explain the relationship between q.i (fact) and p (theory).
> 
> HB : Ha,ha,ha… ups here I see another great Ricks nonsense. Rick see Â»factsÂ« he sees Â»real realityÂ« … I don’t doubt because of his Â»extrasensory abilitiesÂ«…. Ha,ha,haâ€€¦
> 
> There is no Â»factsÂ«. And yiu didn’t answer till now because you have no arguments for your RCT (Ricks Control Theory). Or you want finally to confirm that you have your own (homemade) theory.
> 
> 15 Explaination of relationship between q.i (fact) and p (theory).
> 
> Rick you are World Champion in outdoing yourself in stupidity
> 
> RM earlier : In my rush to show that this is not the case I came up with what has to be the dumbest rebuttal of all time – outdoing even myself in stupidity;-)
> 
> Boris
> 
> As far as the “neatness” of Martin’s analysis, I’m afraid that, after pausing to think about it, I have concluded that it is just more of the usual “theory first” (or “theory only”) treatment of PCT. Apparently it rings a lot of people’s bell here but not mine. I’ll stick with Bill Powers. Oh, how I miss him.
> 
> Best
> 
> Rick
> 
> > Warren
> 
> > On 20 Dec 2017, at 04:50, Martin Taylor [mmt-csg@mmtaylor.net](mailto:mmt-csg@mmtaylor.net) wrote:
> 
> > > [Martin Taylor 2017.12.19.23.28]
> 
> > > On 2017/12/19 8:03 PM, Richard Marken wrote:
> 
> > > > [From Rick Marken (2017.12.19.1700)]
> 
> > > > Bill P (LCS I):
> 
> > > > The Living Control System of this kind must sense the controlled quantity in each dimenssion **in which the quantity is to be controlled** ; this implies the inner model of the quantity in the form of a signal or set of signals.
> 
> > > > HB : n both cases you can see that perceptual signal (quantity) is to be controlled in comparator (it’s future tense). Â»Perceptual signal is only controlled variable there is in control loop. At least in PCT.
> 
> > > > RM: The fact that the controlled quantity (q.i) _must be sensed_ means that it is external to the control system. The “inner model” of that controlled quantity is i\*n the form of a signal, \*which is the perceptual signal (p). So Bill is making the point here that I make with the attached spreadsheet; the controlled quantity, q.i, is represented in the system in the form of a signal. p: in other words, p=q.i.
> 
> > > No. p(t) = f(q.i, t), where f might involve a delay, some time-smearing such as a leaky integral, and/or a nonlinear transform. Even that is not strictly accurate, because the signal is in the form of neural firings, and is therefore noisy. So it is better to put p(t) = f(q.i, t) + n(t), where n(t) represents a noise waveform. When the signal are large, n(t) doesn’t matter much, but just try controlling a perception of colour in very low light. What you see isn’t always what you get. Before the light is so dim that everything is grey, you might see the same surface fluctuating in colour over quite a wide range if you look carefully. But I agree with the general import of what Rick is trying to say.
> > > 
> > > The problem is, as it has been for years, in the precise definition of “control”. A disagreement in which each disputant uses a different definition of the word in dispute is inherently irresoluble. Boris likes the concept of control as in theory bringing some variable toward a commensurate reference variable, whether or not either variable can be observed. Rick doesn’t dispute that as a possibility, but includes as “control” the appearance that an observable variable seems to be brought toward and maintained near some unobservable virtual reference value, which can be deduced only by observing the variable said to be controlled. At least that’s my understanding of the bone of contention. It’s not a dispute about PCT, but about the correct writing of a dictionary definition.
> > > 
> > > It is much the same as would be a dispute about the word “perception”, in which one party claimed that a perception is everything of which he was consciously aware and nothing else, while the other claimed it meant a signal within a control system that depends on real or imagined sensory input and is capable of being controlled. Such an argument would never be resolved, either.
> > > 
> > > Martin
> 
> –
> 
> Richard S. Marken
> 
> "Perfection is achieved not when you have nothing more to add, but when you  
> have nothing left to take away.â€?  
> –Antoine de Saint-Exupery

---

<div class="post-metadata">

**Author:** ![\_Boris\_Hartman](http://discourse.iapct.org/letter_avatar_proxy/v4/letter/_/e8c25b/32.png) [@\_Boris\_Hartman](http://discourse.iapct.org/u/_Boris_Hartman)\
**Post date:** [December 21, 2017, 10:44pm UTC](http://discourse.iapct.org/t/kenneth-j-w-craik-on-levels-of-perception-and-control/6035/18 "2017-12-21T22:44:58Z")

</div>

Martin

> **···**
>
> **From:** Martin Taylor [[mailto:mmt-csg@mmtaylor.net](mailto:mmt-csg@mmtaylor.net)]  
> **Sent:** Wednesday, December 20, 2017 5:51 AM  
> **To:** csgnet@lists.illinois.edu  
> **Subject:** Re: Kenneth J. W. Craik on levels of perception and control
> 
> [Martin Taylor 2017.12.19.23.28]
> 
> On 2017/12/19 8:03 PM, Richard Marken wrote:
> 
> > [From Rick Marken (2017.12.19.1700)]
> 
> > Bill P (LCS I):
> 
> > The Living Control System of this kind must sense the controlled quantity in each dimenssion **in which the quantity is to be controlled** ; this implies the inner model of the quantity in the form of a signal or set of signals.
> 
> > HB : In both cases you can see that perceptual signal (quantity) is to be controlled in comparator (it’s future tense). Â»Perceptual signal is only controlled variable there is in control loop. At least in PCT.
> 
> > RM: The fact that the controlled quantity (q.i) _must be sensed_ means that it is external to the control system. The “inner model” of that controlled quantity is i\*n the form of a signal, \*which is the perceptual signal (p). So Bill is making the point here that I make with the attached spreadsheet; the controlled quantity, q.i, is represented in the system in the form of a signal. p: in other words, p=q.i.
> 
> > HB : You usual you don’t understand PCT and it’s obviously that you’ll have to go into prelimiary school to rrepeat tenses. The. Â»Inner modelÂ« of the Â»input quantityÂ« is Â»controlled quantiyÂ« becaus ethere is where Â»perceptual signal is controlled. There is no Â»controlled quantityÂ« outside orgaism. Input quantity is Â»added effectsÂ« of feedback and distrubances. No Â»control effects in environment…There is no Â»controolled quantityÂ« outside.
> 
> > **Bill P (B:CP) : Consider once again the meaning of the term _controlled quantity_. A controlled quantity is controlled only because it is detected by a control system, compared with a reference, and affected by outputs based on the error thus detected. The controlled quantity is defined strictly by the behaving system’s perceptual computers**
> 
> HB : Do you understand now what is Â»Controlled QuantityÂ« in PCT ???
> 
> MT : No. p(t) = f(q.i, t), where f might involve a delay, some time-smearing such as a leaky integral, and/or a nonlinear transform. Even that is not strictly accurate, because the signal is in the form of neural firings, and is therefore noisy. So it is better to put p(t) = f(q.i, t) + n(t), where n(t) represents a noise waveform. When the signal are large, n(t) doesn’t matter much, but just try controlling a perception of colour in very low light. What you see isn’t always what you get. Before the light is so dim that everything is grey, you might see the same surface fluctuating in colour over quite a wide range if you look carefully. But I agree with the general import of what Rick is trying to say.
> 
> The problem is, as it has been for years, in the precise definition of “control”. A disagreement in which each disputant uses a different definition of the word in dispute is inherently irresoluble.
> 
> HB : You are right Martin that each disputant uses different denitions not only of the word Â»controlÂ« but also of other elements of control loop. In PCT Â»control is quite clearly defined :
> 
> CONTROL : Achievement and maintenance of a preselected state in the controlling system, through actions on the environment that also cancel the effects of disturbances.
> 
> HB : It’s true that many members don’t understand physiology and how organisms function which is quite necessary to understand this definition
> 
> MT : Boris likes the concept of control as in theory bringing some variable toward a commensurate reference variable, whether or not either variable can be observed.
> 
> HB : Sorry Martin I don’t get it what you meant ? But it seems that you know exatly what I’m thinking. Some kind of Â»TelepathyÂ« ? Well explain to me what I like about concept of control. Maybe you could include all Physiological evidences which I like too ??? Â Observed variables by who ?
> 
> Martin whatever I’m doing on CSGnet is that I’m trying to keep balance so that PCT still exist. I think that Powers ladies should do that, but I doubt that they understand PCT to the level so that they could moderate on CSGnet. Barb was talking about some “core group” which is advising them about PCT. But I’m asking myself who is in that “core group” and why they don’t “protect” or upgrade PCT. Who are members in “core group”. I hope it’s not Rick ???
> 
> I’m here with aproximatelly 50 % of arguments. If I would reveal all arguments we could possiblyÂ finnish diagram on p. 191 (B:CP, 2005) which Bill and me couldn’t finnish. And than maybe you could better understand how PCT organism function.
> 
> I think that for the time being we just have to have some reference so that we could agree about basicaly what PCT is.
> 
> I’ve chosen to propose diagram LCS III and definitions of control loop in B:CP which by my oppinion reflect Title of the book. This seems to represent PCT as Bill wrote it.
> 
> RCT (Rick’s Control Theory) look like behavioristic and self-regulation theory. You menationed once that he is promoting just opposite theory “Perception : The control of Behavior”. Or even worse : “Controlled Perception : The control of Behavior”. Do you think that PCT should be changed into something like that ?
> 
> Aren’t we all interested to find out how organisms function ?
> 
> I expect from Powers ladies and CSgnet forum to decide about my proposal. Yuu can make changes to my proposal or even to PCT itself. But do something, phylosophing and qo0rding will not change anything. I already challenged you once but you didn’t “respond”…\>
> 
> So again : what are you supporting
> 
> 1. 
> 
> ```
> RCT or
> 
> ```
> 
> 2. 
> 
> ```
> PCT or
> 
> ```
> 
> 3. 
> 
> ```
> Something in the middle ?
> 
> ```
> 
> Both theories are listed down. I would like to Â»hearÂ« your answer and if possible from all members and of course Ricks answer as he is manipulativelly avoiding confrontation with my proposal. He is just walking like a cat arround hot milk.
> 
> So suggestions or affirmation of which theory is better representing PCT… No phylosophy anymore.
> 
> RCT (Ricks Control Theory) definition of control loop
> 
> 1. 
> 
> ```
> CONTROL : Keeping of some Â»aspect of outer environmentÂ« in reference state, protected (defended) from disturbances.
> 
> ```
> 
> 2. OUTPUT FUNCTION : controlled effects (control of behavior) to outer environment so to keep some Â»controlled variableÂ« in reference state
> 
> 3. 
> 
> ```
> FEED-BACK FUNCTION : Â»ControlÂ« of some Â»aspect of outer environmentÂ« in reference state.
> 
> ```
> 
> 4. 
> 
> ```
> INPUT FUNCTION : produce Â»Controlled Perceptual VariableÂ« or Â»Controlled PerceptionÂ«, the perceptual correlate of Â»controlled q.i.Â«
> 
> ```
> 
> 5. 
> 
> ```
> COMPARATOR : ????
> 
> ```
> 
> 6. 
> 
> ```
> ERROR SIGNAL : ???
> 
> ```
> 
> Proposal of representation of PCT theory :
> 
> Bill P (B:CP):
> 
> 1. 
> 
> ```
> CONTROL : Achievement and maintenance of a preselected state in the controlling system, through actions on the environment that also cancel the effects of disturbances.
> 
> ```
> 
> Bill P (B:CP):
> 
> 1. 
> 
> ```
> OUTPUT FUNCTION : The portion of a system that converts the magnitude or state of a signal inside the system into a corresponding set of effects on the immediate environment of the system
> 
> ```
> 
> Bill P (LCS III):… **the output function** shown in it’s own box rrepresents the means this system has for causing changes in it’s environment.
> 
> Bill P (LCS III):
> 
> 1. 
> 
> ```
> FEED-BACK FUNCTION : The box represents the set of physical laws, properties, arrangements, linkages, by which the action of this system feeds-back to affect its own input, the controlled variable. That's what feed-back means : it's an effect of a system's output on it's own input.
> 
> ```
> 
> Bill P (B:CP) :
> 
> 1. 
> 
> ```
> INPUT FUNCTION : The portion of a system that receives signals or stimuli from outside the system, and generates a perceptual signal that is some function of the received signals or stimuli.
> 
> ```
> 
> Bill P (B:CP) :
> 
> 1. 
> 
> ```
> COMPARATOR : The portion of control system that computes the magnitude and direction of mismatch between perceptual and reference signal.
> 
> ```
> 
> Bill P (B:CP)
> 
> 1. 
> 
> ```
> : ERROR : The discrepancy between a perceptual signal and a reference signal, which drives a control systemâ€™s output function. The discrepancy between a controlled quantity and itâ€™s present reference level, which causes observable behavior.
> 
> ```
> 
> Bill P (B:CP) :
> 
> 1. 
> 
> ```
> ERROR SIGNAL : A signal indicating the magnitude and direction of error.
> 
> ```
> 
> Boris
> 
> Rick doesn’t dispute that as a possibility, but includes as “control” the appearance that an observable variable seems to be brought toward and maintained near some unobservable virtual reference value, which can be deduced only by observing the variable said to be controlled. At least that’s my understanding of the bone of contention. It’s not a dispute about PCT, but about the correct writing of a dictionary definition.
> 
> It is much the same as would be a dispute about the word “perception”, in which one party claimed that a perception is everything of which he was consciously aware and nothing else, while the other claimed it meant a signal within a control system that depends on real or imagined sensory input and is capable of being controlled. Such an argument would never be resolved, either.
> 
> Martin

---

<div class="post-metadata">

**Author:** ![\_Boris\_Hartman](http://discourse.iapct.org/letter_avatar_proxy/v4/letter/_/e8c25b/32.png) [@\_Boris\_Hartman](http://discourse.iapct.org/u/_Boris_Hartman)\
**Post date:** [December 21, 2017, 10:50pm UTC](http://discourse.iapct.org/t/kenneth-j-w-craik-on-levels-of-perception-and-control/6035/19 "2017-12-21T22:50:47Z")

</div>

Nice trial Rick. So you still persist with your RCT (Ricks’ Control Theory) that Â»Behavior is controlledÂ« and that some Â»controlledÂ aspect of environmentÂ«Â and that there is some Â»Controlled Perceptual VariableÂ« ? As I said before. It’s your Â»half loop theoryÂ«.Â Your are manipulating with informations so drastically that I understand how can you manipulate Powers ladies and most of the CSGnet forum for such a long time.

Rick you never understood what PCT is and you never will. Stop hidding behind PCT and announce your own nonsense theory RCT. You can include all your nonsense statements above.

> **···**
>
> **From:** Richard Marken [[mailto:rsmarken@gmail.com](mailto:rsmarken@gmail.com)]  
> **Sent:** Wednesday, December 20, 2017 2:04 AM  
> **To:** csgnet@lists.illinois.edu  
> **Subject:** Re: Kenneth J. W. Craik on levels of perception and control
> 
> [From Rick Marken (2017.12.19.1700)]
> 
> On Mon, Dec 18, 2017 at 11:04 AM, Boris Hartman [boris.hartman@masicom.net](mailto:boris.hartman@masicom.net) wrote:
> 
> HB : Rick. Your nonsense thinking that Â»Behavior can be controlÂ« or that Â»Behavior is controlling inputÂ« was proved as wrong many times. Behavior among other effects also affect input.
> 
> The best description of your wrong understanding was by my oppnion given by Rupert :
> 
> RY earlier : Sure, a perceptual signal (q.i\*g) may correspond to, or be a function of, variable aspects of the environment (q.i) but it is the perceptual signal that is controlled not the variable aspects of the environment.
> 
> HB : Rupert assumes right that only perceptual signal is controlled in comparator. This is only control in the whole loop. It’s nervous system that is controlling.
> 
> RM: It is true that only the perceptual signal is controlled by a control system. But this doesn’t mean that the aspect of the environment that corresponds to that signal is not controlled.
> 
> HB : So it seems that you don’t agree with Rupert. He is saying that Â»perceptual signal that is controlled not the variable aspects of the environment
> 
> and you are saying that : Â»perceptual signal is controlled and also aspect of environment is controlledÂ«.Â
> 
> HBÂ : The prbolem is maybe not so much in your insinuation that Â»aspect of environment is controlledÂ« but in the way you try to prove it. You try to prove that Â»aspect of enviromentÂ« is controlled by Â»Control of behaviorÂ« and that result is some Â»Controled Perceptual VariableÂ« or as you assume that Â»controlled aspect of environmentÂ« was somehow transformed into Â»Controlled PerceptionÂ« or CPV what is PCT nonsense.
> 
> So your control loop is a PCT disaster.
> 
> So we are waiting for a long time that you prove:
> 
> 1. 
> 
> ```
> how behavior can be controlled
> 
> ```
> 
> 2. 
> 
> ```
> What is exactly Â»Controlled Perceptual VariableÂ«.
> 
> ```
> 
> Do if you are saying that controlled aspect of environment Â»correspondÂ« to something formed by input function it has to be Â»Controlled Perceptual SignalÂ« or CPV ? If Â»controlled aspectÂ« of environment exist than control has to come somehow into perceptual signal.
> 
> The question is with what you are controlling Â»aspect of environmentÂ« if not with Â»control of behavior.Â With Telekinesis.Â Are you controlling outer environment with your thoughts ? We know that Behavior is not controlled in PCT. So what is controlling Â»aspect of environmentÂ« ?
> 
> Despite your sweet talkings it seems that you are still standing on the position that Â»behavior is controlÂ« ???
> 
> Bill P ///Â OUTPUT FUNCTION : The portion of a system that converts the magnitude or state of a signal inside the system into corresponding set of effects on the immediate environment of the system.
> 
> Bill P /// FEED-BACK FUNCTION : The box represents the set of physical laws, properties, arrangements, linkages, by which the action of this system feeds-back to affect its own input, the controlled variable. That’s what feed-back means : it’s an effect of a system’s output on it’s own input.
> 
> HB : We can see that Â»behavior is not controlledÂ«, in PCT soÂ that there is no Â»controlled aspect of environmentÂ« and there is no Â»Controlled Perceptual VariableÂ«.
> 
> As you based most of your statements on the assumption that something is controlled in external environment we have to make it clear how Â»aspect of external environmentÂ« is generally controlled.
> 
> Your proposal for changing PCT in these years was quiteb clear (it can be seen through archives).
> 
> So rearange Bills’ definitions (B:CP) which you wanted to avoid in original post, and expalin to us how how Â»errorÂ« signal is formed and how Â»errorÂ« signal drives output and so on so that something can be controlled in outer environment.
> 
> We need the whole control loop to see how your Â»half-loopÂ« theory feets into whole picture of control in organisms, because you claim that something is controlled in outer environment and you don’t have proofs for it. You way of proving is corrupt.
> 
> RCT (Ricks Control Theory) definition of control loop
> 
> 1. 
> 
> ```
> CONTROL : Keeping of some Â»aspect of outer environmentÂ« in reference state, protected (defended) from disturbances.
> 
> ```
> 
> 2. OUTPUT FUNCTION : controlled effects (control of behavior) to outer environment so to keep some Â»controlled variableÂ« in reference state
> 
> 3. 
> 
> ```
> FEED-BACK FUNCTION : Â»ControlÂ« of some Â»aspect of outer environmentÂ« in reference state.
> 
> ```
> 
> 4. 
> 
> ```
> INPUT FUNCTION : produce Â»Controlled Perceptual VariableÂ« or Â»Controlled PerceptionÂ«, the perceptual correlate of Â»controlled q.i.Â«
> 
> ```
> 
> 5. 
> 
> ```
> COMPARATOR : ????
> 
> ```
> 
> 6. 
> 
> ```
> ERROR SIGNAL : ???
> 
> ```
> 
> Whatever you are trying to prove has to do with RCT thinking. It’s just opposite to PCT thinking.
> 
> With these statements you are proving that Â»some aspect of environmentÂ« is generally controlled even in sleeping, sitting and thinking…. And so on… The concept of Â»controlled aspect of environmentÂ« hass by your insinuations also implication that Â»Behavior is controlÂ« and that there is some Â»Controlled Perceptual VariableÂ« or CPV what is total PCT nonsense.
> 
> We have PCT which is proving that you are wrong and that no aspect of environment is controlled, and that Â»behavior is not controlledÂ« and that there is no Â»Controlled Perceptual VariableÂ«.
> 
> Bill P (B:CP):
> 
> 1. 
> 
> ```
> CONTROL : Achievement and maintenance of a preselected state in the controlling system, through actions on the environment that also cancel the effects of disturbances.
> 
> ```
> 
> Bill P (B:CP):
> 
> 1. 
> 
> ```
> OUTPUT FUNCTION : The portion of a system that converts the magnitude or state of a signal inside the system into a corresponding set of effects on the immediate environment of the system
> 
> ```
> 
> Bill P (LCS III):… **the output funnction** shown in it’s own box represents the means this system has for causing changes in it’s environment.
> 
> Bill P (LCS III):
> 
> 1. 
> 
> ```
> FEED-BACK FUNCTION : The box represents the set of physical laws, properties, arrangements, linkages, by which the action of this system feeds-back to affect its own input, the controlled variable. That's what feed-back means : it's an effect of a system's output on it's own input.
> 
> ```
> 
> Bill P (B:CP) :
> 
> 1. 
> 
> ```
> INPUT FUNCTION : The portion of a system that receives signals or stimuli from outside the system, and generates a perceptual signal that is some function of the received signals or stimuli.
> 
> ```
> 
> Bill P (B:CP) :
> 
> 1. 
> 
> ```
> COMPARATOR : The portion of control system that computes the magnitude and direction of mismatch between perceptual and reference signal.
> 
> ```
> 
> Bill P (B:CP)
> 
> 1. 
> 
> ```
> : ERROR : The discrepancy between a perceptual signal and a reference signal, which drives a control systemâ€™s output function. The discrepancy between a controlled quantity and itâ€™s present reference level, which causes observable behavior.
> 
> ```
> 
> Bill P (B:CP) :
> 
> 1. 
> 
> ```
> ERROR SIGNAL : A signal indicating the magnitude and direction of error.
> 
> ```
> 
> So what is gone be ? RCT or PCT ?
> 
> My proposal still stands. The representative of PCT should be diagram in LCS III and Bills’ definitions of control loop in B:CP.Â Who is against ? Rick ?
> 
> Boris

---

<div class="post-metadata">

**Author:** ![MartinT](http://discourse.iapct.org/letter_avatar_proxy/v4/letter/m/ee7513/32.png) [@MartinT](http://discourse.iapct.org/u/MartinT)\
**Post date:** [December 22, 2017, 4:12am UTC](http://discourse.iapct.org/t/kenneth-j-w-craik-on-levels-of-perception-and-control/6035/20 "2017-12-22T04:12:52Z")

</div>

[Martin Taylor 2017.12.21.17.58]

```
I usually don't respond to your postings, because your perceptions

```

of the world are expressed in language that makes it appear that you  
see things very differently from the way I do. That make useful  
interaction very different, so I just don’t bother. However, on this  
occasion, I just would like to point out that Bill’s definition  
applies equally to controlling systems that are constructed of steel  
and rubber, or of electrons in transistors that allow abstract  
variables to vary in simulations. For the definition of control, the  
manner in which the material substrate exists is totally irrelevant.  
I said I had a hypothesis. Rick said my hypothesis was wrong about  
him. Apparently my hypothesis about you is equally wrong. I had  
understood that your idea was that control meant bringing a  
perception toward its reference value and keeping it there in the  
presence of disturbances to the environmental variable being  
perceived. If I was wrong, you have to forgive my inability to read  
your English.  
Maybe I read or English wrongly in my understanding of what you  
believe Rick believes about PCT, but from what I do understand, I  
would say that you are quite wrong in most of it.  
Language is such a problem. I believe that my understanding of PCT  
is very like Bill’s and like Rick’s. Rick does not. I have found at  
CSGnet meetings (the two that I have attended), that when we meet in  
person, Rick and I are much more in agreement than we are on CSGnet.  
I don’t know why, but at least we use language in much the same way,  
which seems not to be the case when I read your messages. So I  
expect to revert to simply not responding to them.  
Martin

> **···**
>
> On 2017/12/21 5:44 PM, Boris Hartman  
> wrote:
> 
> > Martin
> 
> > **From:**  
> > Martin Taylor Wednesday, December 20, 2017 5:51 AM  
> > Re: Kenneth J. W. Craik on levels of  
> > perception and control
> 
> > [Martin Taylor 2017.12.19.23.28]
> 
> > …
> 
> > ```
> > HB
> > 
> > ```
> > 
> > : You are right Martin that each disputant uses different  
> > denitions not only of the word »control« but also of other  
> > elements of control loop. In PCT »control is quite clearly  
> > defined :
> 
> > ```
> > CONTROL :
> > 
> > ```
> > 
> > Achievement and maintenance of a preselected state in the  
> > controlling system, through actions on the environment that  
> > also cancel the effects of disturbances.
> 
> > ```
> > HB : It's true that many members
> > 
> > ```
> > 
> > don’t understand physiology and how organisms function  
> > which is quite necessary to understand this definition
> 
> > MT : Boris likes the  
> > concept of control as in theory bringing some variable toward  
> > a commensurate reference variable, whether or not either  
> > variable can be observed.
> 
> > ```
> > HB
> > 
> > ```
> > 
> > : Sorry Martin I don’t get it what you meant ? But it seems  
> > that you know exatly what I’m thinking. Some kind of  
> > »Telepathy« ?
> 
> [mailto:mmt-csg@mmtaylor.net](mailto:mmt-csg@mmtaylor.net)  
> **Sent:**  
> \*\*To:\*\*csgnet@lists.illinois.edu  
> **Subject:**

[Next page](http://discourse.iapct.org/t/kenneth-j-w-craik-on-levels-of-perception-and-control/6035.md?page=2)
