TCV and Collective Control ...

[wmansell]

Hi Kent, it’s actually rare to read you do any kind of rant and I don’t regard your eloquent defence of CEVs as a rant at all!

Warren

···

On 10 Dec 2016, at 00:11, McClelland, Kent MCCLEL@Grinnell.EDU wrote:

from Kent McClelland (2016.12.09.1400)

Erling Jorgensen (2016.12.08 1150 EST)

Rick Marken (2016.12.07.1710)

Rick and Rupert have argued in this thread that perceptions constructed from clusters of sensory input variables do not necessarily have any counterparts in the physical environment. Therefore, they argue, in Martin Taylor’s “mirror world� diagram (2016.10.16.10.32),
where dots above the dividing line between organism and environment represent perceptions and dots below the line represent complex environmental variables (CEVs), the dots above the line are real, while those below the line aren’t. Thus, Rick concludes, “there’s
no such thing as a CEVâ€? (2016.10.27.1300). Rick’s most recent post in the thread offers a selection of quotes from Bill Powers in which Bill makes similar arguments.

While I appreciate the logic of this argument, which takes as its frame of reference the relationship between the individual actor/perceiver and the individual’s environment, I still think that it’s short-sighted, because it misses the bigger picture that could
be revealed by adopting a sociological frame of reference, which would pay attention the fact that the individual’s environment contains other people.

Bill Powers provided the classic statement of Rick and Rupert’s argument in his lemonade example in Behavior: The Control of Perception. Here’s what he said on p. 112 of the 2005 edition:

This is a good opportunity to emphasize a “philosophical fact” that emerges from this theory: perceptual signals depend on physical events, but what they represent does not necessarily
have any physical significance…. The taste of fresh lemonade, for example, contains an easily recognizable vector, derived from the intensity signals generated by sugar and acid (together with some oil smells). However unitary and real this vector seems, there
is no physical entity corresponding to it….

This means we would be much safer in general to speak of sensation-creating input functions rather than sensation-recognizing functions. To speak of recognition implies tacitly that the environment contains an entity to be recognized, and that all we have to
do is learn to detect it. It seems far more realistic to me to speak instead of functions that construct perceptions with the question of external counterparts to these perceptions being treated with much skepticism.

Bill’s radical skepticism here is fine as far as it goes. Yes, there’s nothing in the environment corresponding to the taste of lemonade. Lemonade is a mixture, not a chemical compound. But how is it that he can talk about the taste of lemonade and be confident
that his readers will know what he’s talking about? It’s because there IS something in their environment, and quite a lot of that something, that has the precise combination of chemical ingredients to allow almost anyone who samples it to experience the perception
of tasting lemonade.

It’s called “lemonade," and it’s a liquid so ubiquitous in the Western world that most people are exposed to it repeatedly over their lives and thus develop by reorganization a dedicated control system somewhere in their perceptual hierarchy for
“recognizing� lemonade when they drink it and saying unequivocally whether any liquid they drink happens to taste like lemonade or not.

So it seems ridiculous to me to say that there’s nothing in the environment of the individual that can compute from that vector of physical variables the perception of the taste of lemonade (or whatever perception you want to talk about that is
based on widespread cultural patterns), when there are millions upon millions of people in the individual’s environment who can do just that. And these millions of people by their collective efforts to control their perceptions of the taste of lemonade end
up creating and drinking billions of gallons of the stuff. The liquid is available in bottled form in every supermarket or convenience store.

Our collective control of the perception of a liquid called lemonade is a “stabilized niche� in the feedback function, as Erling puts it in his latest post, for any individual who may be trying to decide what to drink with lunch. Or you could
call it a collectively controlled “invariance,� one that “can form the basis for a new perceptual category,� as Bill says in his chapter describing reorganization (BCP 2005, p. 204), confronting any child who grows up in the social environment of the Western
world.

[By the way, Rick has asked me repeatedly for an example of something that is collectively controlled with conflict but yet has longterm stability. I offer the perception of the “taste for lemonade.� Everybody knows what lemonade contains: sugar,
water, and lemon juice. But people have very different reference values for what the drink should taste like. Compare the taste of the potent Italian “limonade� to the sickly sweet and watered down drink often served in the USA, or to the vile chemical concoction
that goes by the name of Country Time Lemonade.]

My feeling is that it’s important for us to be able to talk from the PCT perspective about these “stabilized niches in the feedback function� that serve, as Erling notes, as a kind stable platform for effective control of other perceptions. Coining
the term “atenfel� was one attempt to disaggregate feedback functions and pick out those regions of dependably stable, socially produced invariance that facilitate effective control, so that they can become a focus of PCT-informed investigation. The idea
of CEVs is another way to conceptualize the environmental stabilities that are collectively produced.

I had been wondering why Rick has been so hostile to the notion of a CEV, while to me it makes intuitive sense, and it occurred to me that this disagreement may also be a matter of his individual frame of reference vs. my social frame of reference.
If we’re considering a model of how an individual perceives some unknown thing in the environment, there’s no need to postulate a CEV. The model works fine without it. But as soon as we add a second person to the model who is also controlling the same perception
in the same physical environment, that is, we make it into an interactive model of collective control rather than individual behavior, the CEV becomes necessary. The feedback functions of the two actors must converge in some set of environmental variables
for any interaction to take place.

Consider Tom Bourbon’s experiment in collective control, which Rick holds up as an ideal example of PCT-infomed sociology (Bourbon, W. Thomas, Invitation to the dance: Explaining the variance when control systems interact.

American Behavioral Scientist. September/October 1990, vol. 34 no. 1, 95-105. doi: 10.1177/0002764290034001009).

The task in this experiment was to use two joysticks to keep two cursors on a computer screen in line with each other and with a target. A subject could use one joystick for one cursor and one for the other, but the movements of one joystick also
disturbed the other cursor, and vice versa. Bourbon showed that this task could be equally well done by a single person with a joystick in each hand or by two people each running a single joystick. (From my point of view, the two-person experiment is an example
of cooperative collective control, because both subjects control their perceptions of the cursor positions using the same reference values for alignment.)

Here is the diagram that Tom provided to show the PCT model of the two-person experiment.

Notice that the two PCT models for the two experimental subjects (one on the right and the other on the left) meet in the center of the diagram in a set of environmental variables—the CEV for this modeel. For the experiment to work, the two subjects
have to be looking at the same computer screen. In other words, their feedback functions have to pass through the same “stabilized niche� of computer, screen, and moving cursors. Without that region of invariance in their common environment, the two people
in the experiment would not be interacting at all.

If we want PCT to be perceived as relevant to social sciences like political science, economics, sociology, and anthropology, and to the humanities as well, we need to develop a clear analytic vocabulary for talking about the collectively controlled
stabilities in our common social environments, not just what is going on in people’s heads. Maybe CEV, atenfel, and mirror world aren’t the right words for it, but we need something. To just conceptualize the environment as a pulsating set of unconnected physical
variables or else a mush of undifferentiated individual feedback functions, pretending that every individual constructs from that morass his own independent set of perceptions, will not do.

And even to do the Test for the Controlled Variable with an individual, we need to be able to take seriously the stabilized environments in which social interaction occurs. If it weren’t for the social and cultural stabilities that we’re all exposed
to growing up—the “boss realitiesâ€? that disciplinee our reorganizing sets of perceptual control systems into hierarchies that resemble one another’s—we would never be able to identify tthe perceptions another person is apparently controlling or or talk about
them in a meaningful way. These social realities are built into our individual psychology. We’ve got to pay some serious attention to them.

Enough of my rant for now!

Best to all,

Kent

On Dec 8, 2016, at 12:01 PM, Erling Jorgensen EJorgensen@riverbendcmhc.org wrote:

[From Erling Jorgensen (2016.12.08 1150 EST)]

Rick Marken (2016.12.07.1710)

[RM] (referring to your modification of Martin’s diagram, not reproduced here…) One function that defines the variable q.i is the perceptual function inside the system controlling q.i. Another could be in an observer (the silhouette to the left of q.i)
capable of computing a function of the environmental variables, v’s, that is the same as the perceptual function of the control system.

[EJ] Rick, I appreciate the archival search you did to find some of Bill Powers’ thoughts on this issue. For one thing, it helps me relax about designating the CV term (Controlled Variable) the Observer’s version of the perception seemingly being controlled.
As Bill states:

Bill Powers (961224.1145 MST)

BP: Remember that as far as the observer is concerned, what is controlled is ONLY the CV. The idea that this CV is represented by a perceptual signal inside the other system is theoretical. We can observe CV, but not p. When we apply a disturbance, we apply
it to CV, not to p. The action that opposes the effect of the disturbance acts on CV, not p. The Test does not involve p at all. It involves only observables

[EJ] This says that from the Observer’s point of view, the only thing they have to work with is the CV. In the Test for the Controlled Variable, they build up a hypothesis about the other person’s relation to the CV – attempting to control it or not – based
on what happens following the Observer’s disturbance of that variable. If the CV moves as expected, then try something else, because that’s not the variable the other is controlling for. But if the CV does not move as much as expected from their own disturbance,
that is a significant finding, seemingly because something or someone is keeping that result from happening. When variables are stabilized in that way, we suspect “control”, with a very useful theoretical model for what might be going on. And indeed, you
and Bill developed the notion of a “Stability Factor”, to quantify the degree of that hypothesized control.

[EJ] However, I disagree with an earlier point you (and Bill) make. The relevant context is the two sentences prior to Bill’s quote above:

Bruce Abbott (961224.1310 EST)

BA: Rick’s response was to deny that the distinction Martin was making between “stabilized” and “controlled” was useful. CV, he said, either is controlled or is not controlled.

BP: I would tend to agree with Rick, because of my definition of control given above.

[EJ] I have two reasons for considering the notion of “stabilized-but-not-controlled” to be useful. The first reason derives from Kent McClelland’s modeling of conflictive control situations. When two living control systems are trying to keep the same perceptual
variable in two different reference states, the result is often a “virtual reference level” somewhere in between the two preferred states, roughly proportional to the relative contributions of each party’s output gain. While neither party achieves satisfactory
“control” of the variable, its value is definitely “stabilized” somewhere in the middle, with each party pulling as hard as they can. In fact, this is how Bill Powers used to talk about such situations – when the output of each party is maxed out in this
way, they have lost effective control, and the variable may well drift according to whatever other disturbances are in play.

[EJ] My other reason for considering “stabilized” a useful concept that does not simply overlap with “control” was given in my recent post (Erling Jorgensen (2016.12.01 1430 EST). There I argued that a stabilized niche can be an important part of the Environmental
Feedback Function for controlling other variables. Here, the term stability refers to properties of a given control loop, not the values generated by the loop itself. As I said in more detail there, “Stabilized properties, for more effective control.”

All the best,

Erling

<PastedGraphic-4.tiff>

[MartinT]

[Martin Taylor 2016.10.30.16.48]

[Rupert Young (2016.10.30 15.00)]

I get the impression we are talking at cross-purposes, so I'm not going to try to answer your points here, though I may if we ever get aligned on what we are actually talking about.

Here's what I think you think. I expect (and hope) to be corrected. You think that there is a real reality that we can know, and you think that I think the same. I think you think that in my diagram the dots below the line correspond to real things in that real reality and that the dots signify that we know them. You use the word "objective" in talking about the measurable true distance between objects, as though those objects exist in the real reality, and the distance between them also exists in real reality. You use illusions and misperceptions as examples, as though the illusion is based on a perception that does not agree with what real reality is known to be.

I assume that the following is a joke, but because it allows me to highlight a point, I will nevertheless treat it as though you meant it seriously.

[RY] I am talking about perceptual functions that persist, as with my above examples. E.g. you can control a perceptual function of purple by influencing the input elements to it without there being such a thing as purple in the environment.

[MT] On what basis do you assert so confidently that there is no such thing as purple in the environment?

[RY] Because I created the example scenario.

The joke, I assume, is that it is proper to substitute a fantasy "real reality" you create in your imagination for the "real reality" in which we all seem to live. (Rick does the same thing with his "in fact"s and similar assertions of revealed truth, but in his case I don't perceive the humourous intention). However, I will pretend it was not a joke because it makes a good point.

The point that this brings up is that I think (as opposed to what I believe you think I think) that there is a literally infinite number of possibilities for how the actual real world is put together so as to create any of the perceptions we have, and we cannot know which of these possibilities is "really real". What we do know is that in certain perceptual contexts we can control particular perceptions (influence them in a specific direction) by acting on the environment in certain ways. Our ancestors survived by controlling perceptions as if they represented what was actually in the real world most of the time, and we ourselves have survived long enough to have this conversation under the same assumption.

So the point that derives from your joke is that we can invert it, to say that the world in which we imagine we live (Bill Powers's B:CP Figure 15.3 and associated text) might actually be the real world. For example, since we perceive purple to be an attribute of some things in the world that we perceive to be surfaces of objects, real reality might contain real surfaces that are really purple, "surfaces" and "purple" being defined by the elemental relationships in the perceptual function that produces those perceptions.

As for illusions and misperceptions, I do not think that they are problems of perceptions failing to correspond with real reality. I think the problem with them is that they are not influenced in the way such perceptions ordinarily are influences by certai actions. The Ames room is perceived to be an an illusion not because the room isn't rectangular, but because when you move your viewpoint while looking at a rectangular room, it continues to look rectangular, whereas the Ames room doesn't. But what you perceive at any moment is reality for you at that moment, however well or badly if corresponds to something in unknowable real reality.

I don't know whether this may help towards discovering the "cross-purposes" and help toward beginning a discussion with aligned purposes. I hope it does.

Martin

[rupert]

[From Rupert Young (2016.11.18 16.00)]

Sorry for the delay in replying, I've been away for a couple of weeks and just picking up the CSG messages.

[Martin Taylor 2016.10.30.16.48]
I get the impression we are talking at cross-purposes,

Maybe, though not so sure.

Here's what I think you think. I expect (and hope) to be corrected. You think that there is a real reality that we can know, and you think that I think the same. I think you think that in my diagram the dots below the line correspond to real things in that real reality and that the dots signify that we know them. You use the word "objective" in talking about the measurable true distance between objects, as though those objects exist in the real reality, and the distance between them also exists in real reality. You use illusions and misperceptions as examples, as though the illusion is based on a perception that does not agree with what real reality is known to be.

Well, I do think there is a real world out there. We can "know" it to the extent that we can make physical measurements of it, but that is not what perception is about. Perceptions are combinations of real world physical signals and/or also internal signals (perceptions). Your dots below the line seems to indicate that there are variables in the real world which correspond (1 to 1) to perceptions, and are independent of the perceiving system. Those objects exist in reality (unless we're in the Matrix) and there is an "objective" distance between them, in terms of their positions in a universal frame of reference, independent of perceiving systems. Perceptions are illusions, whether or not some of them correspond to the real world is irrelevant; until you try to control in some cases.

I'm not sure what you think, but to go back to the concept of a CEV this seems to indicate that there are complex variables in the environment, which correspond to perceptions. We generally think of perceptual functions as a many-to-one relationship between raw inputs and the perception. If there were a corresponding complex variable in the environment then there would only need to be a one-to-one relationship. Is that what you mean?

I assume that the following is a joke, but because it allows me to highlight a point, I will nevertheless treat it as though you meant it seriously.

[RY] I am talking about perceptual functions that persist, as with my above examples. E.g. you can control a perceptual function of purple by influencing the input elements to it without there being such a thing as purple in the environment.

[MT] On what basis do you assert so confidently that there is no such thing as purple in the environment?

[RY] Because I created the example scenario.

The joke, I assume, is that it is proper to substitute a fantasy "real reality" you create in your imagination for the "real reality" in which we all seem to live. (Rick does the same thing with his "in fact"s and similar assertions of revealed truth, but in his case I don't perceive the humourous intention). However, I will pretend it was not a joke because it makes a good point.

I've no idea why you think this was a joke. And don't know what you mean by "substitute a fantasy 'real reality' you create in your imagination for the 'real reality' in which we all seem to live", it doesn't sound like anything I meant.

The point that this brings up is that I think (as opposed to what I believe you think I think) that there is a literally infinite number of possibilities for how the actual real world is put together so as to create any of the perceptions we have, and we cannot know which of these possibilities is "really real". What we do know is that in certain perceptual contexts we can control particular perceptions (influence them in a specific direction) by acting on the environment in certain ways. Our ancestors survived by controlling perceptions as if they represented what was actually in the real world most of the time, and we ourselves have survived long enough to have this conversation under the same assumption.

This sounds odd to me, how can there be an infinite number of possibilities for how the actual real world is put together? There is only one way the world is put together, but there can be an infinite number of possibilities for the perceptual signals inside a perceiving system. I would also say it is not valid to consider these possibilities as "really real"; they are perceptions! Although I agree that we can control particular perceptions by acting on the environment in certain ways I think it is mistaken to think of perceptions as "representations" of the real world, even though they may seem like it sometimes. And maybe this is our point on contention. It may seem like a perception of "purple" represents there being something of purple in the real world, but with other perceptions, such as fear or honesty or the RDS dolphin, it doesn't make sense. So, as soon as we can cease to think of any perception as being a "representation" of the real world then I think we can move on.

As for illusions and misperceptions, I do not think that they are problems of perceptions failing to correspond with real reality. I think the problem with them is that they are not influenced in the way such perceptions ordinarily are influences by certai actions. The Ames room is perceived to be an an illusion not because the room isn't rectangular, but because when you move your viewpoint while looking at a rectangular room, it continues to look rectangular, whereas the Ames room doesn't. But what you perceive at any moment is reality for you at that moment, however well or badly if corresponds to something in unknowable real reality.

I don't think there is a problem with illusions, but they indicate that all perceptions are "illusions", in the sense that perceptions are constructed signals which may or may not correspond to some objective reality, as in the case of the Ames room where we cannot tell the difference between a rectangular room and a skewed room. Btw, I distinguish between perception and reality, as the external world. So, I wouldn't say that a perception is "at any moment is reality for you at that moment". This may be a point where we are talking at cross purposes. Perception and reality are two different things.

I don't know whether this may help towards discovering the "cross-purposes" and help toward beginning a discussion with aligned purposes. I hope it does.

I hope so too. Let me know.

Rupert

···

On 02/11/2016 03:10, Martin Taylor wrote:

[rsmarken]

[From Rick Marken (2016.11.18.1020)]

···

Rupert Young (2016.11.18 16.00)

RY: Sorry for the delay in replying, I’ve been away for a couple of weeks and just picking up the CSG messages.

RM: I’m busily working on something else myself (not the least getting myself used to the fact that that a fascist anti-scientist has won the presidency with two million fewer votes than the loser; so much for the “genius” of the founders;-) which has kept me from doing anything on CSGNet. But I just wanted to let you know that I liked this post of your; your ideas on perception are, as Bill said somewhere in B:CP “comfortably compatible” with mine.

RM: One thing you might want to note is that the idea of perception as representation is uncomfortably incompatible with the hierarchical model of perception of PCT. The hierarchy is a constructivist model of perception where higher level perceptions are constructed from lower level ones. I might be possible to conceive of the lowest level perceptions – intensities – as representations of physical variables; but it’s constructed perceptions from there on up. And these constructed perceptions, like purple and lemonade and fascism, don’t necessarily represent entities that are actually “out there” in the real world.

RM: And, again, I should point out that this is not just a “philosophical” disagreement (between representationalists and constructivists). The difference could have real consequences for one’s ability to do PCT science, particularly the Test for the Controlled Variable. A representationalist would be oriented toward finding out whether a person is controlling “correctly” – whether the perception is a correct representation of some entity in the world (this is the approach of conventional psychology). A constructivist would be oriented toward finding out what “constructed” as aspect of physical reality (as we know it via physics and chemistry) is being controlled (this is the approach of PCT).

Best

Rick

On 02/11/2016 03:10, Martin Taylor wrote:

[Martin Taylor 2016.10.30.16.48]

I get the impression we are talking at cross-purposes,

Here’s what I think you think. I expect (and hope) to be corrected. You think that there is a real reality that we can know, and you think that I think the same. I think you think that in my diagram the dots below the line correspond to real things in that real reality and that the dots signify that we know them. You use the word “objective” in talking about the measurable true distance between objects, as though those objects exist in the real reality, and the distance between them also exists in real reality. You use illusions and misperceptions as examples, as though the illusion is based on a perception that does not agree with what real reality is known to be.
I assume that the following is a joke, but because it allows me to highlight a point, I will nevertheless treat it as though you meant it seriously.

[RY] I am talking about perceptual functions that persist, as with my above examples. E.g. you can control a perceptual function of purple by influencing the input elements to it without there being such a thing as purple in the environment.

[MT] On what basis do you assert so confidently that there is no such thing as purple in the environment?

[RY] Because I created the example scenario.

The joke, I assume, is that it is proper to substitute a fantasy “real reality” you create in your imagination for the “real reality” in which we all seem to live. (Rick does the same thing with his "in fact"s and similar assertions of revealed truth, but in his case I don’t perceive the humourous intention). However, I will pretend it was not a joke because it makes a good point.
The point that this brings up is that I think (as opposed to what I believe you think I think) that there is a literally infinite number of possibilities for how the actual real world is put together so as to create any of the perceptions we have, and we cannot know which of these possibilities is “really real”. What we do know is that in certain perceptual contexts we can control particular perceptions (influence them in a specific direction) by acting on the environment in certain ways. Our ancestors survived by controlling perceptions as if they represented what was actually in the real world most of the time, and we ourselves have survived long enough to have this conversation under the same assumption.
As for illusions and misperceptions, I do not think that they are problems of perceptions failing to correspond with real reality. I think the problem with them is that they are not influenced in the way such perceptions ordinarily are influences by certai actions. The Ames room is perceived to be an an illusion not because the room isn’t rectangular, but because when you move your viewpoint while looking at a rectangular room, it continues to look rectangular, whereas the Ames room doesn’t. But what you perceive at any moment is reality for you at that moment, however well or badly if corresponds to something in unknowable real reality.
I don’t know whether this may help towards discovering the “cross-purposes” and help toward beginning a discussion with aligned purposes. I hope it does.

Maybe, though not so sure.

Well, I do think there is a real world out there. We can “know” it to the extent that we can make physical measurements of it, but that is not what perception is about. Perceptions are combinations of real world physical signals and/or also internal signals (perceptions). Your dots below the line seems to indicate that there are variables in the real world which correspond (1 to 1) to perceptions, and are independent of the perceiving system. Those objects exist in reality (unless we’re in the Matrix) and there is an “objective” distance between them, in terms of their positions in a universal frame of reference, independent of perceiving systems. Perceptions are illusions, whether or not some of them correspond to the real world is irrelevant; until you try to control in some cases.

I’m not sure what you think, but to go back to the concept of a CEV this seems to indicate that there are complex variables in the environment, which correspond to perceptions. We generally think of perceptual functions as a many-to-one relationship between raw inputs and the perception. If there were a corresponding complex variable in the environment then there would only need to be a one-to-one relationship. Is that what you mean?

I’ve no idea why you think this was a joke. And don’t know what you mean by “substitute a fantasy ‘real reality’ you create in your imagination for the ‘real reality’ in which we all seem to live”, it doesn’t sound like anything I meant.

This sounds odd to me, how can there be an infinite number of possibilities for how the actual real world is put together? There is only one way the world is put together, but there can be an infinite number of possibilities for the perceptual signals inside a perceiving system. I would also say it is not valid to consider these possibilities as “really real”; they are perceptions! Although I agree that we can control particular perceptions by acting on the environment in certain ways I think it is mistaken to think of perceptions as “representations” of the real world, even though they may seem like it sometimes. And maybe this is our point on contention. It may seem like a perception of “purple” represents there being something of purple in the real world, but with other perceptions, such as fear or honesty or the RDS dolphin, it doesn’t make sense. So, as soon as we can cease to think of any perception as being a “representation” of the real world then I think we can move on.

I don’t think there is a problem with illusions, but they indicate that all perceptions are “illusions”, in the sense that perceptions are constructed signals which may or may not correspond to some objective reality, as in the case of the Ames room where we cannot tell the difference between a rectangular room and a skewed room. Btw, I distinguish between perception and reality, as the external world. So, I wouldn’t say that a perception is “at any moment is reality for you at that moment”. This may be a point where we are talking at cross purposes. Perception and reality are two different things.

I hope so too. Let me know.

Rupert

–
Richard S. Marken

“The childhood of the human race is far from over. We
have a long way to go before most people will understand that what they do for
others is just as important to their well-being as what they do for
themselves.” – William T. Powers

[MartinT]

[Martin Taylor 2016.11.18.17.06]

[From Rupert Young (2016.11.18 16.00)]

Sorry for the delay in replying, I've been away for a couple of weeks and just picking up the CSG messages.

No problem. I sometimes like to delay my responses without being away, to slow the feedback loop.

[Martin Taylor 2016.10.30.16.48]
I get the impression we are talking at cross-purposes,

Maybe, though not so sure.

[MT] I don't know either, now. Instead, I think we start from radically different basic concepts, so if we are to work toward a common view, we have to probe them. I'm going to try to agree or disagree (with rationale) with your statements without deleting anything from your message, which will make this reply longer than I would normally like. You have to look through my interpolations, mostly labelled [MT] for this round.

Here's what I think you think. I expect (and hope) to be corrected. You think that there is a real reality that we can know, and you think that I think the same. I think you think that in my diagram the dots below the line correspond to real things in that real reality and that the dots signify that we know them. You use the word "objective" in talking about the measurable true distance between objects, as though those objects exist in the real reality, and the distance between them also exists in real reality. You use illusions and misperceptions as examples, as though the illusion is based on a perception that does not agree with what real reality is known to be.

Well, I do think there is a real world out there.

[MT] Agreed. But it's only a theory, though in order to communicate at all, we must assume the theory is correct. That's why I agree. I do think there is a real world out there.

We can "know" it to the extent that we can make physical measurements of it,

[MT] What do you think is a "physical measurement"? I think it is a relationship perception, as true of the real world as is any other perception. And as subject to illusion. A "physical measurement" may relate a perceived distance to some marks on a ruler, but what about a physical measurement of weight or of mass? You have to have an apparatus to compare two weights, and that's subject to illusion, even with an ordinary balance scale.

but that is not what perception is about. Perceptions are combinations of real world physical signals and/or also internal signals (perceptions).

[MT] Agreed, because I have the same assumption. That doesn't mean we are both correct. though.

Your dots below the line seems to indicate that there are variables in the real world which correspond (1 to 1) to perceptions,

[MT] Agreed.

and are independent of the perceiving system.

[MT] Very, very, strongly disagree. The dots below the line depend entirely and only on the perceiving system. They are what the perceiving system perceives to be in the external environment, not what really is there, since we can never know what really is there.

Those objects

[MT] "Objects"? I suppose it depends on what you mean by "object". To me, the word refers to something that can be sensed by touch -- a tangible thing. Only some perceptions correspond to objects in the environment, and even that correspondence (according to J. G. Taylor, me, I think Bill P., and the modern enactivists) depends entirely on the ability to influence the environment in a way that changes those perceptions in a consistent manner. Most perceptions are not of tangible things. I give as one example your "physical measurement". The things whose properties are being measured may be objects, or they may not -- a distance, for example, is not.

exist in reality (unless we're in the Matrix) and there is an "objective" distance between them, in terms of their positions in a universal frame of reference, independent of perceiving systems.

Oh really? (pun intended)

Einstein might argue that there is a determinable space-time distance between them, but he would not agree that there is a universal frame of reference. Your word "objective" is slipped into the conversation most conveniently, given that we are discussing the nature of that objectivity, and whether the word "objective" can properly be applied at all, to anything.

Perceptions are illusions, whether or not some of them correspond to the real world is irrelevant; until you try to control in some cases.

Agreed. That is the position I have tried to convince you of. If you come at it despite what seems to be a fundamental disagreement about the knowability of reality, there's some even deeper problem to be discovered.

I'm not sure what you think, but to go back to the concept of a CEV this seems to indicate that there are complex variables in the environment, which correspond to perceptions.

Agreed.

We generally think of perceptual functions as a many-to-one relationship between raw inputs and the perception. If there were a corresponding complex variable in the environment then there would only need to be a one-to-one relationship. Is that what you mean?

That's an ambiguous question. As I see it, there's a one-to-one correspondence between what you perceive and the CEV defined by the perceptual function (taking the perceptual function to include the entire perceptual processing system between sensors and the actual perception). There's a one-to-many relationship between the perception and what might actually be in the real world. There's a many-to-one relationship between the raw input sources and the CEV.

I assume that the following is a joke, but because it allows me to highlight a point, I will nevertheless treat it as though you meant it seriously.

[RY] I am talking about perceptual functions that persist, as with my above examples. E.g. you can control a perceptual function of purple by influencing the input elements to it without there being such a thing as purple in the environment.

[MT] On what basis do you assert so confidently that there is no such thing as purple in the environment?

[RY] Because I created the example scenario.

The joke, I assume, is that it is proper to substitute a fantasy "real reality" you create in your imagination for the "real reality" in which we all seem to live. (Rick does the same thing with his "in fact"s and similar assertions of revealed truth, but in his case I don't perceive the humourous intention). However, I will pretend it was not a joke because it makes a good point.

I've no idea why you think this was a joke. And don't know what you mean by "substitute a fantasy 'real reality' you create in your imagination for the 'real reality' in which we all seem to live", it doesn't sound like anything I meant.

Well, what did you mean when you said you knew that there's no purple in the environment because you created the world in question, and in the world you created there is no purple? I couldn't imagine that it was anything but a joke, but if it wasn't, it must have had a serious point. Were you actually expecting me to take as a real argument the idea that the world you created IS the actual real world in which we both live? It still sounds to me as though you had to have been joking.

The point that this brings up is that I think (as opposed to what I believe you think I think) that there is a literally infinite number of possibilities for how the actual real world is put together so as to create any of the perceptions we have, and we cannot know which of these possibilities is "really real". What we do know is that in certain perceptual contexts we can control particular perceptions (influence them in a specific direction) by acting on the environment in certain ways. Our ancestors survived by controlling perceptions as if they represented what was actually in the real world most of the time, and we ourselves have survived long enough to have this conversation under the same assumption.

This sounds odd to me, how can there be an infinite number of possibilities for how the actual real world is put together?

Take a trivial example, the Ames Room. From the only permitted viewpoint, the top right corner could be anywhere along a line that reaches from your viewport to to infinity, the back wall might be in twenty thousand segments all at different distances, and so on. The actual room was designed and built by people controlling their perceptions, but you, the viewer, can't control any of these possible variables. What you see is what you get. And it's all you get.

Currently running on our TV is a commercial that starts out showing a whole lot of junk on a raised platform in a field. A man comes in and starts putting bits and pieces together, but even when he is finished it still looks like a lot of junk scattered over the platform. Then when it is dark he sets up a light at a carefully defined point and turns it on, at which point you see a well formed piece of text as shadows on a wall behind the platform.

We don't know what changes of viewpoint are available to us when we observe our world, just as the Ames room viewer cannot change viewpoint in x and y, or the reader of the shadow text cannot know the wild and wonderful arrangement of junk that produces what she sees. There's an infinite number of arrangements of opaque pieces that could create the same shadow text, and there's an infinite number of possible reasons why we see what we see (feel, smell, touch, hear). Of course, there's also an in finite number of ways the world could be that would not produce the same set of perceptions. The infinite ways that could form only a small subset, but that doesn't mean that the possibilities are any less infinite.

There is only one way the world is put together, but there can be an infinite number of possibilities for the perceptual signals inside a perceiving system.

I'd say that is backwards. Given the existing perceptual functions and raw inputs, the real world can produce only the perceptions it does, but there is an infinite number of worlds that could produce that same set.

On the other hand, counterfactually the perceiving system could be put together in an infinite number of different ways, each of which would produce its own set of perceptions from the actual, unknowable, real world.

I would also say it is not valid to consider these possibilities as "really real"; they are perceptions!

Agreed, but there's also Bill P's position, that these perceptions are the ONLY things of whose reality we can be sure. Perceptions (mine in my case, yours in your case) are real, but what they represent is anyone's guess.

Although I agree that we can control particular perceptions by acting on the environment in certain ways I think it is mistaken to think of perceptions as "representations" of the real world, even though they may seem like it sometimes.

I don't know what you mean by "represent" that makes it wrong to say that perceptions represent the real world. Whatever the real world might be, our perceptions "re-present" it to us and our control systems. You could use the word "correspond" if you like, but it would mean the same thing. Something in the real world and our history forms a pattern that we perceive. The perception corresponds to that pattern. That's what I mean by "represent". Does it mean something else to you?

And maybe this is our point on contention. It may seem like a perception of "purple" represents there being something of purple in the real world, but with other perceptions, such as fear or honesty or the RDS dolphin, it doesn't make sense.

It makes eminent sense to me. I find it hard to think of a line or argument that can reconcile "That makes no sense" and "Oh yes it does". So I offer nothing here. Maybe you can think of a way of drilling into this disagreement -- moving our two viewpoints onto this Ames Room of reality so that we both see things differently? Think of the random-dot stereogram dolphin. You can't see the dolphin in either monocular dot pattern, but the picture is there in the relationships among the dot positions in the two stereograms, just as, say, a "tree" is there in the relationships among its branches and leaves. They are all perceptions, and whether we see a dolphin or a tree depends on whether we have perceptual functions that put our inputs from the world together in that way.

I'm reminded of an old joke. A botanist is on holiday in the North American prairie. In the village is a tree of a kind unfamiliar o him, so he asks an old-timer sitting on a bench: "What's that". To which the answer is "That's a tree, Son. Ain't you never seen one before?"

So, as soon as we can cease to think of any perception as being a "representation" of the real world then I think we can move on.

Explain, please.

As for illusions and misperceptions, I do not think that they are problems of perceptions failing to correspond with real reality. I think the problem with them is that they are not influenced in the way such perceptions ordinarily are influences by certai actions. The Ames room is perceived to be an an illusion not because the room isn't rectangular, but because when you move your viewpoint while looking at a rectangular room, it continues to look rectangular, whereas the Ames room doesn't. But what you perceive at any moment is reality for you at that moment, however well or badly if corresponds to something in unknowable real reality.

I don't think there is a problem with illusions, but they indicate that all perceptions are "illusions", in the sense that perceptions are constructed signals which may or may not correspond to some objective reality, as in the case of the Ames room where we cannot tell the difference between a rectangular room and a skewed room.

You use my example, but start from the notion that there is an "objective reality" of the Ames room independent of how anyone perceives it. I agree that there presumably IS an objective reality, but ask how anyone knows what it is other than by being able to control some aspects of it. In the viewer's case, you may be told that the room is skewed, but unless you can control the angles the parts make with each other as you move your eyes, how do you know your informant is telling the truth? Suppose they are trying to kid you, and the constructors had actually controlled perceptions of all the relevant angles with references for them to be right angles?

Btw, I distinguish between perception and reality, as the external world. So, I wouldn't say that a perception is "at any moment is reality for you at that moment".

No, but the totality of all your perceptions is. That's all you've got, and no matter how it changes a second, a month, or a decade from now, your perception now is all the reality you can know now. Since we agreed that there is some reality independent of perception, we both distinguish between reality and perceptions.

This may be a point where we are talking at cross purposes. Perception and reality are two different things.

Agreed. That's not the point of disagreement.

I don't know whether this may help towards discovering the "cross-purposes" and help toward beginning a discussion with aligned purposes. I hope it does.

I hope so too. Let me know.

I suspect this reply may be one step back. Let's see if we can work on two steps forward. One place to start might be to seek a common thread in the places we disagree. I get the impression that you think we can know reality in some way other than through our perceptions, whereas I don't. Do you think that's right? If so, does the rest of the disagreement follow only from that?

Martin

···

On 02/11/2016 03:10, Martin Taylor wrote:

[rupert]

[From Rupert Young (2016.11.19 18.30)]

(Rick Marken (2016.11.18.1020)]

Thanks. I am beginning to think we have a problem with the term

“perception” which has representational connotations, whereas what
we, more generally, mean is internal neural signals that,
ultimately, contribute to the organism maintaining energy levels
such that it lives long enough to reproduce.

We must rename the theory from PCT to INSTUCTTOMELSTILLETRCT!

Yes, I recall, at University, being unconvinced with the main

approach to computer vision. David Marr’s approach of a one-way
processing pipeline in order to extract the information from static
images. I preferred Gibson’s dynamic approach, but I guess his is
still one-way processing.

Rupert

···

RM: I’m busily working on something else myself (not
the least getting myself used to the fact that that a
fascist anti-scientist has won the presidency with two
million fewer votes than the loser; so much for the
“genius” of the founders;-) which has kept me from doing
anything on CSGNet. But I just wanted to let you know that
I liked this post of your; your ideas on perception are,
as Bill said somewhere in B:CP “comfortably compatible”
with mine.

          RM: One thing you might want to note is that the idea

of perception as representation is uncomfortably
incompatible with the hierarchical model of perception of
PCT. The hierarchy is a constructivist model of perception
where higher level perceptions are constructed from lower
level ones. I might be possible to conceive of the lowest
level perceptions – intensities – as representations of
physical variables; but it’s constructed perceptions from
there on up. And these constructed perceptions, like
purple and lemonade and fascism, don’t necessarily
represent entities that are actually “out there” in the
real world.

          RM: And, again, I should point out that this is not

just a “philosophical” disagreement (between
representationalists and constructivists). The difference
could have real consequences for one’s ability to do PCT
science, particularly the Test for the Controlled
Variable. A representationalist would be oriented toward
finding out whether a person is controlling “correctly” –
whether the perception is a correct representation of some
entity in the world (this is the approach of conventional
psychology). A constructivist would be oriented toward
finding out what “constructed” as aspect of physical
reality (as we know it via physics and chemistry) is being
controlled (this is the approach of PCT).

[MartinT]

[Martin Taylor 2016.11.19.14.45]

[From Rick Marken (2016.11.18.1020)]

I find doing PCT-related things (not just on CSGnet) is good therapy

for my understanding that the near-term end of civilization has just
become appreciably more likely, and there’s nothing I can do about
it.

You and Rupert must mean something by "representation" that differs

from what I mean by the word.

I agree, but I would say that constructivism has to be taken

seriously, which I think neither of you are doing. You use it when
convenient and deny it at will. You both seem stuck on the idea that
there really are “things” in the environment that you can know in
some way independent of your perceptions. If there are such “things”
knowable outside perception, then it would be OK to think of
“representation” as a “correct” mapping of the “things” onto signals
in the brain. So whence cometh this God-like knowledge that you both
have? If the only way you can know of the “things” is through your
perception of them, you can’t map them onto the perception that
created them. “Representation” has to map in the other direction.
For the perceiver, the “thing” is there because you perceive it, so
the perception inevitably represents the “thing”. It’s all you’ve
got, or ever can have.

"Physisical variables" are constructed perceptions, just like

anything else. You can’t perceive a photon, a force, energy, or
voltage. They are all hard-won scientific concepts based on
perceptions of touch, muscle tension, lightness, sound, etc., with
lots of control in imagination and in experiments needed to
construct them. They have no more “real” existence than the colour
purple.

The way I look at it is that there's no other way of finding out

what’s actually “out there” than by perceiving it. In science, we
prefer experiment to passive observation. Why? Because we can try to
influence our perceptions by experiment, which narrows down the set
of possibilities for what’s “out there” rather dramatically if we
are successful (though they remain infinite). The same applies to
our natural perceptions. If we can act so as to influence them in a
consistent way, our ability to do so narrows down the possibilities
for what may be “out there”. The ability to control a perception P
doesn’t uniquely specify a “thing” in the environment. It only
specifies that when we control a particular function of other
perceptual variables, something out there probably corresponds to
that function of other perceptions. I’d call that “representation”.

Perhaps, depending on what you mean by a "representationalist". You

and Rupert don’t seem to be representative of :constructionivists,
or so I judge because of the self-contradictory approach you both
use, saying that there is/isn’t this and that really in the
environment and then saying that the environment is constructed by
the perceptual functions. I prefer consistency. It makes analysis
easier. By using self-contradiction, you can prove anything you
want. I take my constructivism straight, unwatered and with no ice.

Martin

···

          Rupert Young

(2016.11.18 16.00)

          RY: Sorry for the delay in replying, I've been away for a

couple of weeks and just picking up the CSG messages.

          RM: I'm busily working on something else myself (not

the least getting myself used to the fact that that a
fascist anti-scientist has won the presidency with two
million fewer votes than the loser; so much for the
“genius” of the founders;-) which has kept me from doing
anything on CSGNet.

          But I just wanted to let you know that I liked this

post of your; your ideas on perception are, as Bill said
somewhere in B:CP “comfortably compatible” with mine.

          RM: One thing you might want to note is that the idea

of perception as representation is uncomfortably
incompatible with the hierarchical model of perception of
PCT.

          The hierarchy is a constructivist model of perception

where higher level perceptions are constructed from lower
level ones.

          I might be possible to conceive of the lowest level

perceptions – intensities – as representations of
physical variables; but it’s constructed perceptions from
there on up. And these constructed perceptions, like
purple and lemonade and fascism, don’t necessarily
represent entities that are actually “out there” in the
real world.

          RM: And, again, I should point out that this is not

just a “philosophical” disagreement (between
representationalists and constructivists). The difference
could have real consequences for one’s ability to do PCT
science, particularly the Test for the Controlled
Variable. A representationalist would be oriented toward
finding out whether a person is controlling “correctly” –
whether the perception is a correct representation of some
entity in the world (this is the approach of conventional
psychology). A constructivist would be oriented toward
finding out what “constructed” as aspect of physical
reality (as we know it via physics and chemistry) is being
controlled (this is the approach of PCT).

Best

Rick

          On 02/11/2016 03:10, Martin Taylor wrote:

[Martin Taylor 2016.10.30.16.48]

              I get the impression we are talking at cross-purposes,


              Here's what I think you think. I expect (and hope) to

be corrected. You think that there is a real reality
that we can know, and you think that I think the
same. I think you think that in my diagram the dots
below the line correspond to real things in that real
reality and that the dots signify that we know them.
You use the word “objective” in talking about the
measurable true distance between objects, as though
those objects exist in the real reality, and the
distance between them also exists in real reality. You
use illusions and misperceptions as examples, as
though the illusion is based on a perception that does
not agree with what real reality is known to be.
I assume that the following is a joke, but because it
allows me to highlight a point, I will nevertheless
treat it as though you meant it seriously.

                [RY] I am talking about perceptual functions that

persist, as with my above examples. E.g. you can
control a perceptual function of purple by
influencing the input elements to it without there
being such a thing as purple in the environment.

                [MT] On what basis do you assert so confidently that

there is no such thing as purple in the environment?

                [RY] Because I created the example scenario.

              The joke, I assume, is that it is proper to substitute

a fantasy “real reality” you create in your
imagination for the “real reality” in which we all
seem to live. (Rick does the same thing with his "in
fact"s and similar assertions of revealed truth, but
in his case I don’t perceive the humourous intention).
However, I will pretend it was not a joke because it
makes a good point.
The point that this brings up is that I think (as
opposed to what I believe you think I think) that
there is a literally infinite number of possibilities
for how the actual real world is put together so as to
create any of the perceptions we have, and we cannot
know which of these possibilities is “really real”.
What we do know is that in certain perceptual contexts
we can control particular perceptions (influence them
in a specific direction) by acting on the environment
in certain ways. Our ancestors survived by controlling
perceptions as if they represented what was actually
in the real world most of the time, and we ourselves
have survived long enough to have this conversation
under the same assumption.
As for illusions and misperceptions, I do not think
that they are problems of perceptions failing to
correspond with real reality. I think the problem with
them is that they are not influenced in the way such
perceptions ordinarily are influences by certai
actions. The Ames room is perceived to be an an
illusion not because the room isn’t rectangular, but
because when you move your viewpoint while looking at
a rectangular room, it continues to look rectangular,
whereas the Ames room doesn’t. But what you perceive
at any moment is reality for you at that moment,
however well or badly if corresponds to something in
unknowable real reality.
I don’t know whether this may help towards discovering
the “cross-purposes” and help toward beginning a
discussion with aligned purposes. I hope it does.

          Maybe, though not so sure.





          Well, I do think there is a real world out there. We can

“know” it to the extent that we can make physical
measurements of it, but that is not what perception is
about. Perceptions are combinations of real world physical
signals and/or also internal signals (perceptions). Your
dots below the line seems to indicate that there are
variables in the real world which correspond (1 to 1) to
perceptions, and are independent of the perceiving system.
Those objects exist in reality (unless we’re in the
Matrix) and there is an “objective” distance between them,
in terms of their positions in a universal frame of
reference, independent of perceiving systems. Perceptions
are illusions, whether or not some of them correspond to
the real world is irrelevant; until you try to control in
some cases.

          I'm not sure what you think, but to go back to the concept

of a CEV this seems to indicate that there are complex
variables in the environment, which correspond to
perceptions. We generally think of perceptual functions as
a many-to-one relationship between raw inputs and the
perception. If there were a corresponding complex variable
in the environment then there would only need to be a
one-to-one relationship. Is that what you mean?

          I've no idea why you think this was a joke. And don't know

what you mean by “substitute a fantasy ‘real reality’ you
create in your imagination for the ‘real reality’ in which
we all seem to live”, it doesn’t sound like anything I
meant.

          This sounds odd to me, how can there be an infinite number

of possibilities for how the actual real world is put
together? There is only one way the world is put together,
but there can be an infinite number of possibilities for
the perceptual signals inside a perceiving system. I would
also say it is not valid to consider these possibilities
as “really real”; they are perceptions! Although I agree
that we can control particular perceptions by acting on
the environment in certain ways I think it is mistaken to
think of perceptions as “representations” of the real
world, even though they may seem like it sometimes. And
maybe this is our point on contention. It may seem like a
perception of “purple” represents there being something of
purple in the real world, but with other perceptions, such
as fear or honesty or the RDS dolphin, it doesn’t make
sense. So, as soon as we can cease to think of any
perception as being a “representation” of the real world
then I think we can move on.

          I don't think there is a problem with illusions, but they

indicate that all perceptions are “illusions”, in the
sense that perceptions are constructed signals which may
or may not correspond to some objective reality, as in the
case of the Ames room where we cannot tell the difference
between a rectangular room and a skewed room. Btw, I
distinguish between perception and reality, as the
external world. So, I wouldn’t say that a perception is
“at any moment is reality for you at that moment”. This
may be a point where we are talking at cross purposes.
Perception and reality are two different things.

          I hope so too. Let me know.



              Rupert

–
Richard S. Marken

                                    "The childhood of the human

race is far from over. We
have a long way to go before
most people will understand that
what they do for
others is just as important to
their well-being as what they do
for
themselves." – William T.
Powers

[rsmarken]

[From Rick Marken (2016.11.20.0940)]

···

Rupert Young (2016.11.19 18.30)–

(Rick Marken (2016.11.18.1020)]

RY: Thanks. I am beginning to think we have a problem with the term

“perception”

RM: I agree. I would rather that Powers’ control theory had been called “input control theory” (IPT) but it’s too late now, I guess. The problem with calling it perceptual control theory is that many lay people think of “perception” as meaning something like “opinion”. So many people come to PCT thinking it justifies what is basically a postmodernist approach to understanding the world – an approach that treats ideas as just a matter of opinion and my opinion is just as good as yours. The irony is that this has led many people who are basically anti-science to become fervent supporters of the science that Bill was trying to create. Perhaps everything would have worked out better if Bill had titled his book “Behavior:The Control of Input”. But maybe not.

RM: When I first saw Bill’s book, on that fateful day in 1974 while roaming through the library at UCSB, I found the title shocking for the reason I believe Bill intended it to be shocking: because it was describing behavior in the exact opposite way that psychologists thought of it. I was trained to see behavior as controlled (caused, really) by input; and here was a book that said appeared to be saying the opposite. So I was interested in the book to see how the author could justify such an apparently absurd statement. It wasn’t the word “perception” that caught my attention – that part of the title was uncontroversial to me. There wasn’t a psychologist I know who wouldn’t have said that it is the perception of environment that is the input that causes behavior. What was controversial about the title of Powers’ book is that it was saying that input was what was controlled rather than what was doing the controlling.

RM: So if I had the power to do it, I would change the name of Powers’ theory from PCT to ICT (and, in the mean time, also change the result of the recent election to that of the person who won the popular vote). I think I will be equally successful at both;-)

Best

Rick

which has representational connotations, whereas what

we, more generally, mean is internal neural signals that,
ultimately, contribute to the organism maintaining energy levels
such that it lives long enough to reproduce.

We must rename the theory from PCT to INSTUCTTOMELSTILLETRCT!

Yes, I recall, at University, being unconvinced with the main

approach to computer vision. David Marr’s approach of a one-way
processing pipeline in order to extract the information from static
images. I preferred Gibson’s dynamic approach, but I guess his is
still one-way processing.

Rupert

–
Richard S. Marken

“The childhood of the human race is far from over. We
have a long way to go before most people will understand that what they do for
others is just as important to their well-being as what they do for
themselves.” – William T. Powers

          RM: ...But I just wanted to let you know that

I liked this post of your; your ideas on perception are,
as Bill said somewhere in B:CP “comfortably compatible”
with mine.

          RM: And, again, I should point out that this is not

just a “philosophical” disagreement (between
representationalists and constructivists). The difference
could have real consequences for one’s ability to do PCT
science, particularly the Test for the Controlled
Variable. A representationalist would be oriented toward
finding out whether a person is controlling “correctly” –
whether the perception is a correct representation of some
entity in the world (this is the approach of conventional
psychology). A constructivist would be oriented toward
finding out what “constructed” as aspect of physical
reality (as we know it via physics and chemistry) is being
controlled (this is the approach of PCT).

[rsmarken]

[From Rick Marken (2016.11.20.1010)]

···

Martin Taylor (2016.11.19.14.45)–

MT: I agree, but I would say that constructivism has to be taken

seriously, which I think neither of you are doing. You use it when
convenient and deny it at will. You both seem stuck on the idea that
there really are “things” in the environment that you can know in
some way independent of your perceptions.

RM: Then we haven’t been clear. What I think is in the environment is what physics says is out there. The environment is, itself, a model – the model of physics. In that model there are no trees, houses, people, White socks, racists, hypocrites, bullies, etc. There are no “things” out there, per the models of physics.

MT:: If there are such "things"

knowable outside perception, then it would be OK to think of
“representation” as a “correct” mapping of the “things” onto signals
in the brain. So whence cometh this God-like knowledge that you both
have?

RM: From science, which provides the models of the physical world that is presumably the basis of our perceptual experience. It’s god like knowledge; it’s very human.

MT: If the only way you can know of the "things" is through your

perception of them, you can’t map them onto the perception that
created them. “Representation” has to map in the other direction.
For the perceiver, the “thing” is there because you perceive it, so
the perception inevitably represents the “thing”. It’s all you’ve
got, or ever can have.

RM: Yes, correct. And this is our criticism of the representational view of perception. We are saying that only a constructivist model of perception makes sense.

MT: "Physisical variables" are constructed perceptions, just like

anything else. You can’t perceive a photon, a force, energy, or
voltage. They are all hard-won scientific concepts based on
perceptions of touch, muscle tension, lightness, sound, etc., with
lots of control in imagination and in experiments needed to
construct them. They have no more “real” existence than the colour
purple.

RM: Absolutely!! Physical variables are perceptions, but of a particular kind. They are perceptions that constitute a model of the presumed reality on the other side of our senses that are the ultimate basis of our perceptions.

MT: You

and Rupert don’t seem to be representative of :constructionivists,
or so I judge because of the self-contradictory approach you both
use, saying that there is/isn’t this and that really in the
environment and then saying that the environment is constructed by
the perceptual functions. I prefer consistency. It makes analysis
easier. By using self-contradiction, you can prove anything you
want. I take my constructivism straight, unwatered and with no ice.

RM: Your ideas about perception, as described in this post, are comfortably compatible with my own. Indeed, they are perfectly compatible with my own. So I don’t understand why you seem to think that there exists something called a “controlled environmental variable” (CEV) that is not the same as the controlled perceptual variable (CV). You seem to think of a CEV as an entity in the environment that corresponds to the CV. That’s why we think of your view of perception is representational; you seem to think of the CV as a representation of the CEV, and not always an accurate one. A constructivist view would see what you call the CEV as being a perceptual variable that corresponds exactly to the CV; the difference is that the CV is a perceptual variable constructed by the system controlling it while the CEV is the same perceptual variable construct by someone (or something) other than the system controlling it.

RM: Is my understanding of your concept of a CEV incorrect?

Best

Rick

–
Richard S. Marken

“The childhood of the human race is far from over. We
have a long way to go before most people will understand that what they do for
others is just as important to their well-being as what they do for
themselves.” – William T. Powers

          RM: The hierarchy is a constructivist model of perception

where higher level perceptions are constructed from lower
level ones.

Rick

          On 02/11/2016 03:10, Martin Taylor wrote:

[Martin Taylor 2016.10.30.16.48]

              I get the impression we are talking at cross-purposes,


              Here's what I think you think. I expect (and hope) to

be corrected. You think that there is a real reality
that we can know, and you think that I think the
same. I think you think that in my diagram the dots
below the line correspond to real things in that real
reality and that the dots signify that we know them.
You use the word “objective” in talking about the
measurable true distance between objects, as though
those objects exist in the real reality, and the
distance between them also exists in real reality. You
use illusions and misperceptions as examples, as
though the illusion is based on a perception that does
not agree with what real reality is known to be.
I assume that the following is a joke, but because it
allows me to highlight a point, I will nevertheless
treat it as though you meant it seriously.

                [RY] I am talking about perceptual functions that

persist, as with my above examples. E.g. you can
control a perceptual function of purple by
influencing the input elements to it without there
being such a thing as purple in the environment.

                [MT] On what basis do you assert so confidently that

there is no such thing as purple in the environment?

                [RY] Because I created the example scenario.

              The joke, I assume, is that it is proper to substitute

a fantasy “real reality” you create in your
imagination for the “real reality” in which we all
seem to live. (Rick does the same thing with his "in
fact"s and similar assertions of revealed truth, but
in his case I don’t perceive the humourous intention).
However, I will pretend it was not a joke because it
makes a good point.
The point that this brings up is that I think (as
opposed to what I believe you think I think) that
there is a literally infinite number of possibilities
for how the actual real world is put together so as to
create any of the perceptions we have, and we cannot
know which of these possibilities is “really real”.
What we do know is that in certain perceptual contexts
we can control particular perceptions (influence them
in a specific direction) by acting on the environment
in certain ways. Our ancestors survived by controlling
perceptions as if they represented what was actually
in the real world most of the time, and we ourselves
have survived long enough to have this conversation
under the same assumption.
As for illusions and misperceptions, I do not think
that they are problems of perceptions failing to
correspond with real reality. I think the problem with
them is that they are not influenced in the way such
perceptions ordinarily are influences by certai
actions. The Ames room is perceived to be an an
illusion not because the room isn’t rectangular, but
because when you move your viewpoint while looking at
a rectangular room, it continues to look rectangular,
whereas the Ames room doesn’t. But what you perceive
at any moment is reality for you at that moment,
however well or badly if corresponds to something in
unknowable real reality.
I don’t know whether this may help towards discovering
the “cross-purposes” and help toward beginning a
discussion with aligned purposes. I hope it does.

          Maybe, though not so sure.





          Well, I do think there is a real world out there. We can

“know” it to the extent that we can make physical
measurements of it, but that is not what perception is
about. Perceptions are combinations of real world physical
signals and/or also internal signals (perceptions). Your
dots below the line seems to indicate that there are
variables in the real world which correspond (1 to 1) to
perceptions, and are independent of the perceiving system.
Those objects exist in reality (unless we’re in the
Matrix) and there is an “objective” distance between them,
in terms of their positions in a universal frame of
reference, independent of perceiving systems. Perceptions
are illusions, whether or not some of them correspond to
the real world is irrelevant; until you try to control in
some cases.

          I'm not sure what you think, but to go back to the concept

of a CEV this seems to indicate that there are complex
variables in the environment, which correspond to
perceptions. We generally think of perceptual functions as
a many-to-one relationship between raw inputs and the
perception. If there were a corresponding complex variable
in the environment then there would only need to be a
one-to-one relationship. Is that what you mean?

          I've no idea why you think this was a joke. And don't know

what you mean by “substitute a fantasy ‘real reality’ you
create in your imagination for the ‘real reality’ in which
we all seem to live”, it doesn’t sound like anything I
meant.

          This sounds odd to me, how can there be an infinite number

of possibilities for how the actual real world is put
together? There is only one way the world is put together,
but there can be an infinite number of possibilities for
the perceptual signals inside a perceiving system. I would
also say it is not valid to consider these possibilities
as “really real”; they are perceptions! Although I agree
that we can control particular perceptions by acting on
the environment in certain ways I think it is mistaken to
think of perceptions as “representations” of the real
world, even though they may seem like it sometimes. And
maybe this is our point on contention. It may seem like a
perception of “purple” represents there being something of
purple in the real world, but with other perceptions, such
as fear or honesty or the RDS dolphin, it doesn’t make
sense. So, as soon as we can cease to think of any
perception as being a “representation” of the real world
then I think we can move on.

          I don't think there is a problem with illusions, but they

indicate that all perceptions are “illusions”, in the
sense that perceptions are constructed signals which may
or may not correspond to some objective reality, as in the
case of the Ames room where we cannot tell the difference
between a rectangular room and a skewed room. Btw, I
distinguish between perception and reality, as the
external world. So, I wouldn’t say that a perception is
“at any moment is reality for you at that moment”. This
may be a point where we are talking at cross purposes.
Perception and reality are two different things.

          I hope so too. Let me know.



              Rupert

–
Richard S. Marken

                                    "The childhood of the human

race is far from over. We
have a long way to go before
most people will understand that
what they do for
others is just as important to
their well-being as what they do
for
themselves." – William T.
Powers

[_Boris_Hartman]

In the text bellow….

···

From: Richard Marken [mailto:rsmarken@gmail.com]
Sent: Sunday, November 20, 2016 6:38 PM
To: csgnet@lists.illinois.edu
Subject: Re: TCV and Collective Control …

[From Rick Marken (2016.11.20.0940)]

Rupert Young (2016.11.19 18.30)–

(Rick Marken (2016.11.18.1020)]

RM: …But I just wanted to let you know that I liked this post of your; your ideas on perception are, as Bill said somewhere in B:CP “comfortably compatible” with mine.

RY: Thanks. I am beginning to think we have a problem with the term “perception”

RM: I agree. I would rather that Powers’ control theory had been called “input control theory” (IPT) but it’s too late now, I guess. The problem with calling it perceptual control theory is that many lay people think of “perception” as meaning something like “opinion”. So many people come to PCT thinking it justifies what is basically a postmodernist approach to understanding the world – an approach that treats ideas as just a matter of opinion and my opinion is just as good as yours. The irony is that this has led many people who are basically anti-science to become fervent supporters of the science that Bill was trying to create. Perhaps everything would have worked out better if Bill had titled his book “Behavior:The Control of Input”. But maybe not.

HB : Of course not. That would suit you because you are selling all the time »Behavioral Control theory«. Bill did what he did. Diagram in LCS III clearly shows that behavior is the result of »subtraction« between reference and perceptual signal which causes effector to function and affect environment. See Bills’ diagram.Â

And we established many times that you don’t understand how orgaisms work and you don’t understand how »Perceptual Theory« works so you think that also other people don’t understand because you are egoistically looking only through your point of view.

i doubt that you are the one who can judge Bill’s work. He knew too much in comparison with you. And Barb wrote once how her Dad was precise in choosing terms. I hope she is not taking back her statement. And you Rick if you don’t like his theory what is preventing you from establishing your new forum where »Behavior is Control« and »Behavior is Control of Input«. You are anyway all the time writing about »Behavioral Control Theory« (BCT) which needs the »control« in outer environment what is opposite to »Perceptual Control Theory« (PCT). Why don’t you go your own way ? Why hiding behind PCT ?Â

But I was surprised when you confirmed Eetus’ statements :

EP: Instead perception takes place inside organism, in the input function and after it in perception signal?

RM: Yes, of course.

EP: What we can control is perception inside the organism - the perception signal in comparator (if I at the moment have understood right the concept of control)?

RM: Yes, its the perceptual signal that is controlled.

Hb : So there is nothing more to say. Perception is directly controlled, not output or input or whatever.

RM: When I first saw Bill’s book, on that fateful day in 1974 while roaming through the library at UCSB, I found the title shocking for the reason I believe Bill intended it to be shocking: because it was describing behavior in the exact opposite way that psychologists thought of it. I was trained to see behavior as controlled (caused, really) by input; and here was a book that said appeared to be saying the opposite. So I was interested in the book to see how the author could justify such an apparently absurd statement. It wasn’t the word “perception” that caught my attention – that part of the title was uncontroversial to me. There wasn’t a psychologist I know who wouldn’t have said that it is the perception of environment that is the input that causes behavior.

HB : Your trainig that »Beahavior is Control or sontrolled« left terrible consequences. It doesn’t matter what »controls« behavior. It’s wrong assumption. »Behavior is not control or controlled in any way. As I said before. What Ricky learned, Rick knows. It’s not only other psychoogist who were describing and thought of behavior in the exact opposite way. It’s also you who thinks just in opposite way. Yours RCT where »Behavior is Control« is just opposite to Bills’ where »Perception is controlled«.

The perception of the environment don’t control the behavior and neither »error« signal. If you’ll look again in the Bill’s diagram (LCS III) you will see that the difference between reference and perceptual signal is the cause of behavior which affects environment. Your simple mind will never understand such a great mind as Bill was.

RM : What was controversial about the title of Powers’ book is that it was saying that input was what was controlled rather than what was doing the controlling.

HB : It’s the problem what you think here as »Control of input«. If you think of it as »Input to comparator« than it’s right, because perceptula signal is deirectly conrtrolled. If you think of it as »input function« than it’s wrong.

Percpetual signals are controlled on all levels of hierarchy not »inputs«, specialy if you have in mind only »external receptor« as input. But as I said before, you don’t understand PCT so you are not in the position to judge anything about PCT specialy not about changing it. How can you change something that you don’t understand. Bills’ theroy is showing that »Perception is controlled« and that’s it. He must have a reason to put it this way. And the reason is clearly described in B:CP and his other literature. You just have to read it right.

The members of RT (Glassers Reality Therapy) directed me to your unconvincing definiton of term perception. So they have to search over internet what it means in PCT. And they ran into definiton of Bruce Abbott :

BA : “It is important to note that this use of the term “perception� conveys a somewhat different meaning in PCT than is familiar to sense physiologist and psychologist. As usually defined, a perception is a cognitive process of interpretation of sensory input; in PCT a perception is simply a signal conveying sense-data from sensory receptors or a signal derived from other such signals. A perception so defined need no conscious representation.� (Bruce B. Abbot, Ph.D)

HB : If we compare Bruces’ defintion with Bills’ we will see that they are close :

W.T.Powers (B:CP, 2005)

•      Perception : A perceptual signal (inside a system) that is a continuous analog of a state of affairs outside the system.
  

•      Perceptual signal : The signal emitted by the input function of a system, an internal analog of some aspect of environment.
  

HB : It’s internal analog (not precise »photo« of environment or some precisely defined aspect of environment which could show on presence of »controlled variable«. You got it all wroing, It means that perceptions are organisms »construction« or »model« of environment (state of affairs) some functional transformation of physical variables which affect input function form »endless variables in environment« as Ashby would say. It’s just a small part of environmental variables that are being trasnformed into perceptual signal by human (aspect of environment).

Percpetual signal is controlled in comparator. I could explain to you why and how but I think it’s better for you to go and read B:CP and Henry Yins’ article again instead of watching »football«.

It’s clear from Bills’ diagram in LCS III, that perceptual signal is controlled in comparator and the result is »error« signal and later behavior (output) which is affecting outer environment (see Bills’ diagram). If everything would be so easy as you are showing how nervous system (hierarchy) work why all those books were written ?

RM: So if I had the power to do it, I would change the name of Powers’ theory from PCT to ICT …

HB : Why ? You wrote it for yourself.

EP: What we can control is perception inside the organism - the perception signal in comparator (if I at the moment have understood right the concept of control)?

RM: Yes, its the perceptual signal that is controlled.

HB : I’m interested how would you adapt diagram to your »genious« idea of »Input Control Theory« and how would you adapt also the whole book B:CP with all those physiogical evidences. Would there be a »Controlled Variable« in environment and »Controled Perceptual Variable« ? Would there be »Behavior is Control« ???

I hope that Powers ladies are not so much »hypnotised« by you that they will let you do such stupid thing. Once Barb wrote that changes of Bills’ theory would be the same as to change Einsteins’ »Relativity Theory«. And I agreed with her. So I hope you are right that changing Bills’ theory would be as difficult as to change the result of USA elections J.

Best,

Boris

Best

Rick

which has representational connotations, whereas what we, more generally, mean is internal neural signals that, ultimately, contribute to the organism maintaining energy levels such that it lives long enough to reproduce.

We must rename the theory from PCT to INSTUCTTOMELSTILLETRCT!

RM: And, again, I should point out that this is not just a “philosophical” disagreement (between representationalists and constructivists). The difference could have real consequences for one’s ability to do PCT science, particularly the Test for the Controlled Variable. A representationalist would be oriented toward finding out whether a person is controlling “correctly” – whether the perception is a correct representation of some entity in the world (this is the approach of conventional psychology). A constructivist would be oriented toward finding out what “constructed” as aspect of physical reality (as we know it via physics and chemistry) is being controlled (this is the approach of PCT).

Yes, I recall, at University, being unconvinced with the main approach to computer vision. David Marr’s approach of a one-way processing pipeline in order to extract the information from static images. I preferred Gibson’s dynamic approach, but I guess his is still one-way processing.

Rupert

–

Richard S. Marken

“The childhood of the human race is far from over. We have a long way to go before most people will understand that what they do for others is just as important to their well-being as what they do for themselves.” – William T. Powers

[rupert]

[From Rupert Young (2016.11.23 12.00)]

(Martin Taylor 2016.11.18.17.06]

There's lots of stuff in here, most with which I agree, some not so much; which I may come back to later. But for the moment, due to my short attention span, let's look at this, crucial, point, again.

Your dots below the line seems to indicate that there are variables in the real world which correspond (1 to 1) to perceptions,

[MT] Agreed.

and are independent of the perceiving system.

[MT] Very, very, strongly disagree. The dots below the line depend entirely and only on the perceiving system. They are what the perceiving system perceives to be in the external environment, not what really is there, since we can never know what really is there.

Ok good, that you are saying that the variables are not independent of the perceiving system. But you are agreeing that they are in the real world. This sounds inconsistent to me. How can they in the real world, but not independent of the perceiving system? And if they are environmental variables (CEV) how are they not independent of the perceiving system? And why call them environmental variables? And why show them below the line? Perhaps, you could give some examples.

Your statement "what the perceiving system perceives to be in the external environment, not what really is there" seems to be describing perception, which you are giving, unnecessarily, another name. What does it (CEV) add to the theory, that is not already captured by the current understanding of perceptual control loop and processes?

Btw, I do not agree that there are environmental variables that correspond (1 to 1) to perceptions, at anything except the very lowest level, and probably not even then.

Rupert

[_Erling_Jorgensen]

[From Erling Jorgensen (2016.11.23 1030 EST)]

Rupert Young (2016.11.23 12.00) in his ongoing discussion with –

(Martin Taylor 2016.11.18.17.06], including previous exchanges (initials added):

[RY] Your dots below the line seems to indicate that there are variables
in the real world which correspond (1 to 1) to perceptions,

[MT] Agreed.

[RY] and are independent of the perceiving system.

[MT] Very, very, strongly disagree. The dots below the line depend
entirely and only on the perceiving system. They are what the
perceiving system perceives to be in the external environment, not
what really is there, since we can never know what really is there.

[RY] Ok good, that you are saying that the variables are not independent of
the perceiving system. But you are agreeing that they are in the real
world. This sounds inconsistent to me.

[EJ] In following this exchange, I am reminded of a passage from Powers’ B:CP that was very formative for me in turning from a realist to a constructivist epistemology. I don’t have my copy of B:CP with me right now, so I’ll have to reconstruct this from memory.

I think it was in the section where he talked about Relationships, giving examples of perceived relationships that are controlled. Then Powers said something like, “We could ask which relationships are really there in the environment, but that is a trivial question. All relationships are really there, even ones that are meaningless.” This is definitely my paraphrase of his point!

The “ah ha” part for me was to realize that the environment holds innumerable possibilities for perceptual regularities, but the key is always which (small) subset of them will matter enough to me to control. In other words, the PCT mantra that “It’s all perception.”

Yes, there seems to be an environmental substrate for perception, but the only interface we have is through the perceptions we construct, so the environment remains unknowable in itself. The best we can hope for is that it may be infer-able. And that occurs through the “reality test” (can I call it that?) built into every act of control. If the perceptions we’ve constructed are too far unhinged from the reality that is supposedly out there in that environment, then our efforts at control will not be very good. We’ll keep getting frustrated because the environment doesn’t seem to be cooperating. In such circumstances, it is likely our constructed perceptions that are not cooperating. They are too much at odds from the (unknowable) substrate out there.

So in terms of Martin’s diagram (I think it was in [Martin Taylor 2016.10.16.10.32]), I think I would pretty much cover the environmental portion below the line with even dimmer dots and lines, and then allow certain dots and connections to be slightly bolder (ala Martin’s scheme), corresponding with what is above the line in the organism’s particular set of constructed perceptions. That would suggest that for this organism, certain environmental regularities are more “useful” than others, while still conveying that there remains a universe of untapped possibilities left to others to construct and perceive.

I basically agree with Rupert’s query about the concept of a Complex Environmental Variable, to whit:

[RY] What does it (CEV) add to the theory, that is not already captured by the current
understanding of perceptual control loop and processes?

All the best,

Erling

  NOTICE: This e-mail communication (including any attachments) is CONFIDENTIAL and the materials contained herein are PRIVILEGED and intended only for disclosure to or use by the person(s) listed above. If you are neither the intended recipient(s), nor a person responsible for the delivery of this communication to the intended recipient(s), you are hereby notified that any retention, dissemination, distribution or copying of this communication is strictly prohibited. If you have received this communication in error, please notify me immediately by using the "reply" feature or by calling me at the number listed above, and then immediately delete this message and all attachments from your computer. Thank you.

[MartinT]

[Martin Taylor 2016.11.23.11.16]

[From Rupert Young (2016.11.23 12.00)]

(Martin Taylor 2016.11.18.17.06]

There's lots of stuff in here, most with which I agree, some not so much; which I may come back to later. But for the moment, due to my short attention span, let's look at this, crucial, point, again.

Your dots below the line seems to indicate that there are variables in the real world which correspond (1 to 1) to perceptions,

[MT] Agreed.

and are independent of the perceiving system.

[MT] Very, very, strongly disagree. The dots below the line depend entirely and only on the perceiving system. They are what the perceiving system perceives to be in the external environment, not what really is there, since we can never know what really is there.

Ok good, that you are saying that the variables are not independent of the perceiving system. But you are agreeing that they are in the real world. This sounds inconsistent to me. How can they in the real world, but not independent of the perceiving system? And if they are environmental variables (CEV) how are they not independent of the perceiving system? And why call them environmental variables? And why show them below the line? Perhaps, you could give some examples.

It seems that we may be getting closer to mutual understanding, but there's still a long way to go.

[RY copied] Ok good, that you are saying that the variables are not independent of the perceiving system. But you are agreeing that they are in the real world. This sounds inconsistent to me.

It's not at all inconsistent to me. So I'll try to answer your contingent questions. The Ames Room remains a good analogue for the much larger question, so I'll use it.

I start with the basic point that we can never know what is "really" in the real world any more than the peephole viewer of the Ames room can know what is the actual configuration of its parts, or even what parts it has. But the viewer of the Ames room can know that her perceptions are not what they would be if she were looking at a real giant ice-cream cone or a real flower bed. Something in the real world that is not an ice-cream cone or a flower bed is responsible for her perceptions when she looks through the peephole. Almost everything that could be imagined in the environment would not give her the same perceptions as does the Ames Room.

But this still leaves an infinite range of real-world configurations that would indeed give her exactly the perceptions she does have. Cantor proved that there are more points on a line than there are rational numbers, and I remember reading that there are more functions of N variables than there are points on a line. Since every element of the Ames room could be at any distance from the peephole, the number of possibilities for that specific element of the room is the second order of infinity (called aleph-one). The room has aleph-one such elements, each at aleph-one possible distances from the viewer, ignoring the fact that the viewer's input systems would be able to resolve only a finite number of them as being different. The "real world" knows no such limitations. So there are at least aleph-one configurations of the real world that would produce the same perceptions in the viewer, who has no way of distinguishing which are closer to the truth.

And yet, the number of configurations of the real world that are NOT going to give the same perception as would any of these possible Ames Rooms is vastly greater than the number that are, in the same sense that the number of integers is greater than the number of integers evenly divisible by 2, 53, or 200 million, even though all of these have the same cardinality of aleph-null. Choose an intgeger at random, and it probably is not divisible exactly by 3,176,454,913,545. So there's no inconsistency in saying that the possible Ames Rooms are in the real world AND are determined by the perceiving system. The perceiving system "claims" that in the visible real world is a room, not an ice-cream cone or a flower bed. The perceiving system does not claim that one particular version of the possible Ames Rooms is the one that is really real.

So why do we look through the peephole and (for most of us) perceive only a rectangularly structured room? I argue that the answer is that through a lifetime (and an evolutionary history of ancestors), when we have controlled perceptions with similar inputs, control has worked when we perceive those visual configurations at "corners" to be right-angles. While walking, hanging pictures, setting furniture, and so forth, control has worked if we have developed perceptual functions that produce a perception of a rectangularly structured room. The Ames Room inputs are like all those other inputs from rectangularly structured rooms, so that's how we perceive the Ames room, rather than as a fluctuating or fuzzy perception of an arbitrary one of the infinite number of possible Ames rooms.

Your statement "what the perceiving system perceives to be in the external environment, not what really is there" seems to be describing perception, which you are giving, unnecessarily, another name. What does it (CEV) add to the theory, that is not already captured by the current understanding of perceptual control loop and processes?

The CEV is already part of the "current understanding of the perceptual control loop and processes". It doesn't add or subtract anything to the theory, but I find it adds to the ability to talk about control loops, particular about systems of interacting control loops that interact only through their different influences on the external environment. The numerical value of the CEV is often labelled "qi", and diagrammatically it is usually shown as a small circle without a name where the output and the disturbance have their influences mixed. I just give it a name that makes talking about it easier. The fact that it is defined by the perceptual function is not new either. It just is a variable that is complex (formed from potentially many interacting lower-level variables) and is environmental (in what the perceiver perceives to be outside the perceiver's body, usually, and always between the output and the perceptual input of the control loop). Hence Complex Environmental Variable, or CEV.

Naming an important point in the control loop is hardly an addition to the theory, but it does make discussions of the theory easier. For example, if I want to explain a part of the TCV, I might say something like "in the TCV, the tester influences the value of the [???]". It's not "the value of the qi" because qi is itself a value, a value of the [???]. Or when discussing collective control (and this thread is called "TCV and collective control") I find it easier to say that the collective controllers all influence the CEV in the directions of their reference values for the corresponding perceptions rather than to waffle around with saying that they influence each other's perceptual values, to which they have no direct access according to the theory.

Btw, I do not agree that there are environmental variables that correspond (1 to 1) to perceptions, at anything except the very lowest level, and probably not even then.

Rupert

I'm not clear what or who you are disagreeing with here. I'll assume it is with me, and that your problem is that the value of X+Y is not uniquely determined by the value of X alone or of Y alone. I assume that the problem is that a perceptual function that produces the value of X+Y thus corresponds to a whole continuum of possibilities for X and Y, which is true. But it's not a "real" problem. To see those possibilities, you have to have at least one other viewpoint besides the one that produces X+Y, just as to see that the Ames Room is not rectangularly structured you have to find another place to view it from.

Just as I argue that the Ames Room is one of an infinite number of possible configurations that would produce the same perception, so I would argue that the CEV of a controlled perception (which the Ames room is not) is one of an infinite number of possibilities that would lead to the same perception. Of this infinity, only a very small proportion would respond to action influences in ways that allow effective control. But that proportion is still infinite. From my point of view, not only is the CEV removed in time from the perception, it represents a class of possibilities, all of which have in common that the derived perceptions respond the same way to control actions that influence them. That's a one-to-many relationship, but the "many" is irrelevant for the purpose of analyzing a control loop.

The "dot below the line" represents any real-world function that behaves exactly as does the perceptual function when influenced by the output of the internal portion of the control loop. It's the "rectangularly structured" possibility of the Ames Room, and in that it is unique. It represents our perceived version of what the real world was at some past time, regardless of how much integration (time-binding or time-smearing) is involved in the perceiving system.

Martin

[rsmarken]

[From Rick Marken (2016.11.23.1120)]

···

Erling Jorgensen (2016.11.23 1030 EST)–

[EJ] In following this exchange, I am reminded of a passage from Powers’ B:CP that was very formative for me in turning from a realist to a constructivist epistemology. I don’t have my copy of B:CP with me right now, so I’ll have to reconstruct this from memory.

I think it was in the section where he talked about Relationships, giving examples of perceived relationships that are controlled. Then Powers said something like, “We could ask which relationships are really there in the environment, but that is a trivial question. All relationships are really there, even ones that are meaningless.” This is definitely my paraphrase of his point!

RM: This is a great point, Erling, and one I’ve tried to make in a couple of my demos. The point is that the same environmental variables can be perceived in many different ways. It’s the perceptual functions that determine how we experience the environment, not the environment itself. There is nothing in the environment that necessarily corresponds to what we perceive; but what is out in the environment is the basis for what we perceive.

RM: For example, in my “Control of Perception” demo (http://www.mindreadings.com/ControlDemo/ControlOfPerception.html) the same environmental variables – the display on the screen – can be perceived in at least three different ways – shape, size (area) and angle. And each of these perceptual variables can be controlled. But none of them are in the environment! Only the possibility of these perceptions is out there. And each possibility is made a “reality” when you perceive things that way using your perceptual functions.

RM: When I say that only the possibility of a shape, size or angle perception is out there I mean that all that’s out there (presumably) is electromagnetic energy of different frequencies, phases and intensities distributed over the space of the display. Call these spatially distributed environmental variables x.i. The perceptions of shape, size and angle are then functions of the sensory effects of these variables: so shape = f1(s(x.1), s(x.2)…s(x.n)), size = f2(s(x.1), s(x.2)…s(x.n)), and angle = f3(s(x.1), s(x.2)…s(x.n)), where the s()'s are the sensory effects of the environmental variables and the f()'s are perceptual functions of those sensory effects – the functions that produce what we experience as the shape, size or angle of the displayed variables.

RM: Hopefully, this shows what I think is the problem with the concept of a CEV. The idea of a CEV implies that there is some variable in the environment that corresponds to the perception that is controlled. But there is no variable in the environment that corresponds to, say, shape, size or angle. Size (area) is a function of environmental variables: say size = x.1*x.2. The variables x.1 and x.2 are, indeed, variables in the environment but size – the product of x.1 and x.2 – is not. The perception of size exists only because there is a perceptual function that can compute it.

RM: The fact that size is not in the environment – but the environmental variables that are the basis of size, are – can, perhaps, be seen more clearly in the fact that the same environmental variables are also the basis of shape and angle. Shape, for example, might be defined as x.1/x.2 so you are perceiving a square when x.1/x.2 = 1. But the square is not an environmental variable any more than size was. The concept of a CEV suggests that x.1 and x.2 can’t be both a size and a shape in the environment at the same time. But actually all that x.1 and x.2 can be is x.1 and x.2; size and shape exist only as perceptions, not as environmental variables.

RM; So I prefer to stick with the term “controlled quantity”, q.o, to refer to the perception of the controlled perceptual variable, p, from the point of view of an observer. It makes it clear that the controlled perceptual variable and the controlled quantity are both perceptions, but one (the controlled perception or controlled variable) is the perception in the head of the person doing the controlling and the other (controlled quantity) is the same perception in the head of the person observing, studying or trying to understand the controller.

RM: Another way I thought of to keep this straight is to think of environmental variables as the argument of a function that results in what we experience as perception. The flaw in the concept of the CEV is that it mistakes the argument for the function.

Best regards

Rick

The “ah ha” part for me was to realize that the environment holds innumerable possibilities for perceptual regularities, but the key is always which (small) subset of them will matter enough to me to control. In other words, the PCT mantra that “It’s all perception.”

Yes, there seems to be an environmental substrate for perception, but the only interface we have is through the perceptions we construct, so the environment remains unknowable in itself. The best we can hope for is that it may be infer-able. And that occurs through the “reality test” (can I call it that?) built into every act of control. If the perceptions we’ve constructed are too far unhinged from the reality that is supposedly out there in that environment, then our efforts at control will not be very good. We’ll keep getting frustrated because the environment doesn’t seem to be cooperating. In such circumstances, it is likely our constructed perceptions that are not cooperating. They are too much at odds from the (unknowable) substrate out there.

So in terms of Martin’s diagram (I think it was in [Martin Taylor 2016.10.16.10.32]), I think I would pretty much cover the environmental portion below the line with even dimmer dots and lines, and then allow certain dots and connections to be slightly bolder (ala Martin’s scheme), corresponding with what is above the line in the organism’s particular set of constructed perceptions. That would suggest that for this organism, certain environmental regularities are more “useful” than others, while still conveying that there remains a universe of untapped possibilities left to others to construct and perceive.

I basically agree with Rupert’s query about the concept of a Complex Environmental Variable, to whit:

[RY] What does it (CEV) add to the theory, that is not already captured by the current
understanding of perceptual control loop and processes?

All the best,

Erling

  NOTICE: This e-mail communication (including any attachments) is CONFIDENTIAL and the materials contained herein are PRIVILEGED and intended only for disclosure to or use by the person(s) listed above. If you are neither the intended recipient(s), nor a person responsible for the delivery of this communication to the intended recipient(s), you are hereby notified that any retention, dissemination, distribution or copying of this communication is strictly prohibited. If you have received this communication in error, please notify me immediately by using the "reply" feature or by calling me at the number listed above, and then immediately delete this message and all attachments from your computer. Thank you.

–

Richard S. Marken

“The childhood of the human race is far from over. We
have a long way to go before most people will understand that what they do for
others is just as important to their well-being as what they do for
themselves.” – William T. Powers

[rsmarken]

[From Rick Marken (2016.11.23.1130)]

RM: I’ve got to proof more carefully:

···

On Wed, Nov 23, 2016 at 11:21 AM, Richard Marken rsmarken@gmail.com wrote:

[From Rick Marken (2016.11.23.1120)]

I’ve got to proof read more carefully. I said:

What I should have said was: The concept of a CEV suggests that x.1 and x.2 CAN be both a size and a shape in the environment at the same time.

Best

Rick

–

The concept of a CEV suggests that x.1 and x.2 can’t be both a size and a shape in the environment at the same time.

Richard S. Marken

“The childhood of the human race is far from over. We
have a long way to go before most people will understand that what they do for
others is just as important to their well-being as what they do for
themselves.” – William T. Powers

[_Erling_Jorgensen]

[From Erling Jorgensen (2016.11.23 1445 EST)]

Rick Marken (2016.11.23.1120)]

RM: The point is that the same environmental variables can be perceived in many different ways. It’s the perceptual functions that determine how we experience the environment, not the environment itself. There is nothing in the environment that necessarily corresponds to what we perceive; but what is out in the environment is the basis for what we perceive.

[EJ] Hi Rick. My sense is that we have very similar epistemologies here, & I fully agree with the way you state this. This is also why I disagree with a formulation you sometimes make, that when we control certain perceptions we are thereby controlling the environmental aspects that correspond to them. I understand the logic of what you’re trying to get at – after all, the perceptions are functions of those environmental substrates. I just think we should reserve the word “control” for perceptions themselves. That’s why I lean toward a word like “stabilize,” in the sense that something in the environment gets stabilized by our actions to actually control our own perceptions. It’s a subtle difference between us, & maybe just a judgment call as to wording.

RM: So I prefer to stick with the term “controlled quantity”, q.o, to refer to the perception of the controlled perceptual variable, p, from the point of view of an observer. It makes it clear that the controlled perceptual variable and the controlled quantity are both perceptions, but one (the controlled perception or controlled variable) is the perception in the head of the person doing the controlling and the other (controlled quantity) is the same perception in the head of the person observing, studying or trying to understand the controller.

[EJ] Here, too, I understand what you are getting at, but the formulation still feels a little klunky to me. Mind you, my own way of thinking of it gets pretty klunky, too. 'Cause I lean toward Martin’s “layered protocol” language about ‘my belief about what you are perceiving, & your belief about what I am perceiving,’ & with all those recursive loops it becomes a wonder that we can come to believe that we are jointly talking about the same thing at all!

[EJ] I don’t have a good solution for how we should talk about these interactions, but I think we share a good PCT epistemology that we are always talking about our own perceptions of things, even when we do not explicitly remember to make that point.

All the best,

Erling

  NOTICE: This e-mail communication (including any attachments) is CONFIDENTIAL and the materials contained herein are PRIVILEGED and intended only for disclosure to or use by the person(s) listed above. If you are neither the intended recipient(s), nor a person responsible for the delivery of this communication to the intended recipient(s), you are hereby notified that any retention, dissemination, distribution or copying of this communication is strictly prohibited. If you have received this communication in error, please notify me immediately by using the "reply" feature or by calling me at the number listed above, and then immediately delete this message and all attachments from your computer. Thank you.

[rupert]

[From Rupert Young (2016.11.23 21.00)]

(Martin Taylor 2016.11.23.11.16]

[From Rupert Young (2016.11.23 12.00)]

Your dots below the line seems to indicate that there are variables in the real world which correspond (1 to 1) to perceptions,

[MT] Agreed.

and are independent of the perceiving system.

[MT] Very, very, strongly disagree. The dots below the line depend entirely and only on the perceiving system. They are what the perceiving system perceives to be in the external environment, not what really is there, since we can never know what really is there.

Ok good, that you are saying that the variables are not independent of the perceiving system. But you are agreeing that they are in the real world. This sounds inconsistent to me. How can they in the real world, but not independent of the perceiving system? And if they are environmental variables (CEV) how are they not independent of the perceiving system? And why call them environmental variables? And why show them below the line? Perhaps, you could give some examples.

It seems that we may be getting closer to mutual understanding, but there's still a long way to go.

[RY copied] Ok good, that you are saying that the variables are not independent of the perceiving system. But you are agreeing that they are in the real world. This sounds inconsistent to me.

It's not at all inconsistent to me. So I'll try to answer your contingent questions. The Ames Room remains a good analogue for the much larger question, so I'll use it.

I start with the basic point that we can never know what is "really" in the real world any more than the peephole viewer of the Ames room can know what is the actual configuration of its parts, or even what parts it has. But the viewer of the Ames room can know that her perceptions are not what they would be if she were looking at a real giant ice-cream cone or a real flower bed. Something in the real world that is not an ice-cream cone or a flower bed is responsible for her perceptions when she looks through the peephole. Almost everything that could be imagined in the environment would not give her the same perceptions as does the Ames Room.

But this still leaves an infinite range of real-world configurations that would indeed give her exactly the perceptions she does have. Cantor proved that there are more points on a line than there are rational numbers, and I remember reading that there are more functions of N variables than there are points on a line. Since every element of the Ames room could be at any distance from the peephole, the number of possibilities for that specific element of the room is the second order of infinity (called aleph-one). The room has aleph-one such elements, each at aleph-one possible distances from the viewer, ignoring the fact that the viewer's input systems would be able to resolve only a finite number of them as being different. The "real world" knows no such limitations. So there are at least aleph-one configurations of the real world that would produce the same perceptions in the viewer, who has no way of distinguishing which are closer to the truth.

And yet, the number of configurations of the real world that are NOT going to give the same perception as would any of these possible Ames Rooms is vastly greater than the number that are, in the same sense that the number of integers is greater than the number of integers evenly divisible by 2, 53, or 200 million, even though all of these have the same cardinality of aleph-null. Choose an intgeger at random, and it probably is not divisible exactly by 3,176,454,913,545. So there's no inconsistency in saying that the possible Ames Rooms are in the real world AND are determined by the perceiving system. The perceiving system "claims" that in the visible real world is a room, not an ice-cream cone or a flower bed. The perceiving system does not claim that one particular version of the possible Ames Rooms is the one that is really real.

So why do we look through the peephole and (for most of us) perceive only a rectangularly structured room? I argue that the answer is that through a lifetime (and an evolutionary history of ancestors), when we have controlled perceptions with similar inputs, control has worked when we perceive those visual configurations at "corners" to be right-angles. While walking, hanging pictures, setting furniture, and so forth, control has worked if we have developed perceptual functions that produce a perception of a rectangularly structured room. The Ames Room inputs are like all those other inputs from rectangularly structured rooms, so that's how we perceive the Ames room, rather than as a fluctuating or fuzzy perception of an arbitrary one of the infinite number of possible Ames rooms.

Sure, but I don't see what this has got to do with the question, so may be where we are talking at cross purposes. And, actually, it aligns with my point that there is not a 1-to-1 correspondence between the perception and the real world variable. That is, there is not a 1-to-1 correspondence between the perception of a rectangular room and there actually being a rectangular room. As you have said there are many possible configurations of the real world variables that can give rise to the perception of a rectangular room. But, in the case of the Ames room, there isn't actually a rectangular room.

Your statement "what the perceiving system perceives to be in the external environment, not what really is there" seems to be describing perception, which you are giving, unnecessarily, another name. What does it (CEV) add to the theory, that is not already captured by the current understanding of perceptual control loop and processes?

The CEV is already part of the "current understanding of the perceptual control loop and processes". It doesn't add or subtract anything to the theory, but I find it adds to the ability to talk about control loops, particular about systems of interacting control loops that interact only through their different influences on the external environment. The numerical value of the CEV is often labelled "qi", and diagrammatically it is usually shown as a small circle without a name where the output and the disturbance have their influences mixed. I just give it a name that makes talking about it easier. The fact that it is defined by the perceptual function is not new either. It just is a variable that is complex (formed from potentially many interacting lower-level variables) and is environmental (in what the perceiver perceives to be outside the perceiver's body, usually, and always between the output and the perceptual input of the control loop). Hence Complex Environmental Variable, or CEV.

Again, these "defined by the perceptual function", "formed from potentially many interacting lower-level variables" and "what the perceiver perceives to be outside the perceiver's body" sound like perception. And why call it Environmental if it is not in the environment; the real world?

Naming an important point in the control loop is hardly an addition to the theory, but it does make discussions of the theory easier. For example, if I want to explain a part of the TCV, I might say something like "in the TCV, the tester influences the value of the [???]". It's not "the value of the qi" because qi is itself a value, a value of the [???]. Or when discussing collective control (and this thread is called "TCV and collective control") I find it easier to say that the collective controllers all influence the CEV in the directions of their reference values for the corresponding perceptions rather than to waffle around with saying that they influence each other's perceptual values, to which they have no direct access according to the theory.

In the TCV the tester influences the value of the perception; the neural signal. Though what we usually mean when we talk about the TCV is the TPCV. Where P = proxy, in that, as we can't actually observe the perception (without brain surgery) we use a proxy of that perception, as perceived by the tester.

Btw, I do not agree that there are environmental variables that correspond (1 to 1) to perceptions, at anything except the very lowest level, and probably not even then.

I'm not clear what or who you are disagreeing with here. I'll assume it is with me, and that your problem is that the value of X+Y is not uniquely determined by the value of X alone or of Y alone. I assume that the problem is that a perceptual function that produces the value of X+Y thus corresponds to a whole continuum of possibilities for X and Y, which is true. But it's not a "real" problem. To see those possibilities, you have to have at least one other viewpoint besides the one that produces X+Y, just as to see that the Ames Room is not rectangularly structured you have to find another place to view it from.

I am thinking of something like fear, where there is no such thing in the real-world, external to the perceiving system.

Just as I argue that the Ames Room is one of an infinite number of possible configurations that would produce the same perception, so I would argue that the CEV of a controlled perception (which the Ames room is not) is one of an infinite number of possibilities that would lead to the same perception. Of this infinity, only a very small proportion would respond to action influences in ways that allow effective control. But that proportion is still infinite. From my point of view, not only is the CEV removed in time from the perception, it represents a class of possibilities, all of which have in common that the derived perceptions respond the same way to control actions that influence them. That's a one-to-many relationship, but the "many" is irrelevant for the purpose of analyzing a control loop.

The "dot below the line" represents any real-world function that behaves exactly as does the perceptual function when influenced by the output of the internal portion of the control loop.

What do you mean by real-world function? How is it implemented? How can it be both real-world and not independent of the perceiving system?

Rupert

[MartinT]

[Martin Taylor 2016.11.23.16.50]

[From Rupert Young (2016.11.23 21.00)]

(Martin Taylor 2016.11.23.11.16]

[From Rupert Young (2016.11.23 12.00)]

Your dots below the line seems to indicate that there are variables in the real world which correspond (1 to 1) to perceptions,

[MT] Agreed.

and are independent of the perceiving system.

[MT] Very, very, strongly disagree. The dots below the line depend entirely and only on the perceiving system. They are what the perceiving system perceives to be in the external environment, not what really is there, since we can never know what really is there.

Ok good, that you are saying that the variables are not independent of the perceiving system. But you are agreeing that they are in the real world. This sounds inconsistent to me. How can they in the real world, but not independent of the perceiving system? And if they are environmental variables (CEV) how are they not independent of the perceiving system? And why call them environmental variables? And why show them below the line? Perhaps, you could give some examples.

It seems that we may be getting closer to mutual understanding, but there's still a long way to go.

[RY copied] Ok good, that you are saying that the variables are not independent of the perceiving system. But you are agreeing that they are in the real world. This sounds inconsistent to me.

It's not at all inconsistent to me. So I'll try to answer your contingent questions. The Ames Room remains a good analogue for the much larger question, so I'll use it.

I start with the basic point that we can never know what is "really" in the real world any more than the peephole viewer of the Ames room can know what is the actual configuration of its parts, or even what parts it has. But the viewer of the Ames room can know that her perceptions are not what they would be if she were looking at a real giant ice-cream cone or a real flower bed. Something in the real world that is not an ice-cream cone or a flower bed is responsible for her perceptions when she looks through the peephole. Almost everything that could be imagined in the environment would not give her the same perceptions as does the Ames Room.

But this still leaves an infinite range of real-world configurations that would indeed give her exactly the perceptions she does have. Cantor proved that there are more points on a line than there are rational numbers, and I remember reading that there are more functions of N variables than there are points on a line. Since every element of the Ames room could be at any distance from the peephole, the number of possibilities for that specific element of the room is the second order of infinity (called aleph-one). The room has aleph-one such elements, each at aleph-one possible distances from the viewer, ignoring the fact that the viewer's input systems would be able to resolve only a finite number of them as being different. The "real world" knows no such limitations. So there are at least aleph-one configurations of the real world that would produce the same perceptions in the viewer, who has no way of distinguishing which are closer to the truth.

And yet, the number of configurations of the real world that are NOT going to give the same perception as would any of these possible Ames Rooms is vastly greater than the number that are, in the same sense that the number of integers is greater than the number of integers evenly divisible by 2, 53, or 200 million, even though all of these have the same cardinality of aleph-null. Choose an intgeger at random, and it probably is not divisible exactly by 3,176,454,913,545. So there's no inconsistency in saying that the possible Ames Rooms are in the real world AND are determined by the perceiving system. The perceiving system "claims" that in the visible real world is a room, not an ice-cream cone or a flower bed. The perceiving system does not claim that one particular version of the possible Ames Rooms is the one that is really real.

So why do we look through the peephole and (for most of us) perceive only a rectangularly structured room? I argue that the answer is that through a lifetime (and an evolutionary history of ancestors), when we have controlled perceptions with similar inputs, control has worked when we perceive those visual configurations at "corners" to be right-angles. While walking, hanging pictures, setting furniture, and so forth, control has worked if we have developed perceptual functions that produce a perception of a rectangularly structured room. The Ames Room inputs are like all those other inputs from rectangularly structured rooms, so that's how we perceive the Ames room, rather than as a fluctuating or fuzzy perception of an arbitrary one of the infinite number of possible Ames rooms.

Sure, but I don't see what this has got to do with the question, so may be where we are talking at cross purposes. And, actually, it aligns with my point that there is not a 1-to-1 correspondence between the perception and the real world variable. That is, there is not a 1-to-1 correspondence between the perception of a rectangular room and there actually being a rectangular room.

That's my point, too.

As you have said there are many possible configurations of the real world variables that can give rise to the perception of a rectangular room. But, in the case of the Ames room, there isn't actually a rectangular room.

My point is that you don't and can't know that there isn't actually a rectangular room. It seems to me that you are still working on the principle that there's a way other than through perception that you can know what's really out there.

Your statement "what the perceiving system perceives to be in the external environment, not what really is there" seems to be describing perception, which you are giving, unnecessarily, another name. What does it (CEV) add to the theory, that is not already captured by the current understanding of perceptual control loop and processes?

The CEV is already part of the "current understanding of the perceptual control loop and processes". It doesn't add or subtract anything to the theory, but I find it adds to the ability to talk about control loops, particular about systems of interacting control loops that interact only through their different influences on the external environment. The numerical value of the CEV is often labelled "qi", and diagrammatically it is usually shown as a small circle without a name where the output and the disturbance have their influences mixed. I just give it a name that makes talking about it easier. The fact that it is defined by the perceptual function is not new either. It just is a variable that is complex (formed from potentially many interacting lower-level variables) and is environmental (in what the perceiver perceives to be outside the perceiver's body, usually, and always between the output and the perceptual input of the control loop). Hence Complex Environmental Variable, or CEV.

Again, these "defined by the perceptual function", "formed from potentially many interacting lower-level variables" and "what the perceiver perceives to be outside the perceiver's body" sound like perception. And why call it Environmental if it is not in the environment; the real world?

The CEV is a component of a theory that proposes that there exists one real world. In that theory, the CEV IS in the environment. If you want to be a solipsist, of course, it isn't, because there is no environment. But the theory we are working with (I hope both of us) is not solipsistic, and in that environment there is an infinity of possible causes for any set of perceptions. That infinity is infinitesimal compared to the possible realities that do not cause those perceptions. The small (yet infinite) class of possbilities is selected by a perceptual function (according to the theory). That set of possibilities includes possibilities that are not in the real world (you may be hallucinating, for example, with all the input coming from your imagination), but excluding those still leaves an infinity of possibilities that ARE in the environment.

Naming an important point in the control loop is hardly an addition to the theory, but it does make discussions of the theory easier. For example, if I want to explain a part of the TCV, I might say something like "in the TCV, the tester influences the value of the [???]". It's not "the value of the qi" because qi is itself a value, a value of the [???]. Or when discussing collective control (and this thread is called "TCV and collective control") I find it easier to say that the collective controllers all influence the CEV in the directions of their reference values for the corresponding perceptions rather than to waffle around with saying that they influence each other's perceptual values, to which they have no direct access according to the theory.

In the TCV the tester influences the value of the perception; the neural signal.

No, or rather that influence is only indirect, by influencing the environment that contributes to the perceptual value.

Though what we usually mean when we talk about the TCV is the TPCV. Where P = proxy, in that, as we can't actually observe the perception (without brain surgery) we use a proxy of that perception, as perceived by the tester.

So why use the name "TCV" at all? It's more misleading than "CEV". Yes, the controlled variable is perception, not something in the environment, but the aim of the exercise is to try to influence just that function of the environment that is determined by the perceptual function that interests you. It works only because you are working on (or trying to find) a "dot below the line".

Btw, I do not agree that there are environmental variables that correspond (1 to 1) to perceptions, at anything except the very lowest level, and probably not even then.

I'm not clear what or who you are disagreeing with here. I'll assume it is with me, and that your problem is that the value of X+Y is not uniquely determined by the value of X alone or of Y alone. I assume that the problem is that a perceptual function that produces the value of X+Y thus corresponds to a whole continuum of possibilities for X and Y, which is true. But it's not a "real" problem. To see those possibilities, you have to have at least one other viewpoint besides the one that produces X+Y, just as to see that the Ames Room is not rectangularly structured you have to find another place to view it from.

I am thinking of something like fear, where there is no such thing in the real-world, external to the perceiving system.

When do you perceive "fear"? Are those occasions entirely independent of what is happening in the external environment?

I have two comments on this sentence itself. (1) It once again asserts that you have some privileged access to what is and what isn't in the external world, and (2) Many, if not most, perceptions include inputs from inside the body, but all of the inputs (according to PCT) are external to the perceiving system. If you perceive something you call "fear", you must have a means of perceiving it -- a perceptual function for it (again working within a particular theoretical framework we call Perceptual Control Theory). Some particular pattern of external and internal states leads to a perception of "fear". But what's the variable defined by this function whose perceptual output has one value called "fear"? Is "unease" another value of the same variable? Is "confidence"?

The point is that the environment of every control unit includes all the inputs to its perceptual function (and all the places that receive its output signal).

Just as I argue that the Ames Room is one of an infinite number of possible configurations that would produce the same perception, so I would argue that the CEV of a controlled perception (which the Ames room is not) is one of an infinite number of possibilities that would lead to the same perception. Of this infinity, only a very small proportion would respond to action influences in ways that allow effective control. But that proportion is still infinite. From my point of view, not only is the CEV removed in time from the perception, it represents a class of possibilities, all of which have in common that the derived perceptions respond the same way to control actions that influence them. That's a one-to-many relationship, but the "many" is irrelevant for the purpose of analyzing a control loop.

The "dot below the line" represents any real-world function that behaves exactly as does the perceptual function when influenced by the output of the internal portion of the control loop.

What do you mean by real-world function? How is it implemented? How can it be both real-world and not independent of the perceiving system?

I thought I offered an overly long and perhaps overly detailed explanation of this in the message for which I am now replying to your response. I have not deleted anything you quoted. What more do you need? Repeating and expanding part of my previous answer: If you perceive X+Y, that creates a presumption that the environment contains something we can call EX that creates a perceptual value X and something that we can call EY that creates a perceptual value Y. When whatever it is in the environment creates those values, then the environment necessarily permits any function f(EX, EY), which creates some function f(X,Y) of the corresponding perceptions. But f(X,Y) is defined by a perceptual function to be X+Y, nothing else. All functions of EX and EY that do not create the perception X+Y are excluded as possibilities.

f(EX, EY) need not be EX+EY. It might, for example be EX*EY, if X and Y are logarithmic functions of EX and EY. But there must exist a function of the environmental variables that maps through the intervening channels and processes to produce X+Y. As we have agreed, this is not unique, but it is a very small, infinitesimally small, subset of the possible functions of the unknown EX and EY, each of which has an infinite number of possibilities that might generate X and Y.

The basic argument, going back to J.G. Taylor (1962) is that by reorganization that permits good control, we can refined that small subset further, creating an infinitesimally smaller (but still infinite) set of possibilities in the real world that create perceptions that we can usefully control. Erling got it right [From Erling Jorgensen (2016.11.23 1030 EST)] :

[EJ] "Powers said something like, "We could ask which relationships are really there in the environment, but that is a trivial question. _All_ relationships are really there, even ones that are meaningless." This is definitely my paraphrase of his point!
[EJ, continued] The "ah ha" part for me was to realize that the environment holds innumerable possibilities for perceptual regularities, but the key is always which (small) subset of them will matter enough to me to control. In other words, the PCT mantra that "It's all perception."

Martin

···

Rupert

[rsmarken]

[From Rick Marken (2016.11.23.1530)]

···

Erling Jorgensen (2016.11.23 1445 EST)–

RM: I think what I usually say is that when we are controlling perceptions we are controlled aspects of the environment. By “aspect of the environment” I mean “functions of physical variables”, which is equivalent to “perceptions”. Perceptions are functions of physical (environmental) variables so when we control perceptions we are controlling functions (aspects) of environmental variables.

RM: I agree. And that’s what I’m reserving it for. We control functions of environmental variables which is the same as saying that we control perceptions. We don’t control the environmental variables themselves, just functions of those variables. So when we control the area of a rectangle, for example, we are controlling hw were h and w are the physical (environmental) variables of which area is a function (the product of those variables). We don’t control the physical variables themselves that are the arguments of that perception. We control an aspect of the physical variables – a perception. When we do that, we are controlling that aspect of the environment that hw. The environment itself – h and w – can compute that function; but that function of h*w – that aspect of the physical variables, h and y, is being controlled – staying equal to some value, protected from disturbances (such as variations in h and/or w).

RM: I don’t think we stabilize physical variables either. We control aspects (functions or perceptions) of physical variables, not necessarily the physical variables themselves – except in the case where the perception is a function only of an actual physical variable. This would apply only to intensity perceptions. So to the extent that, say, loudness a function of only the physical amplitude to acoustic vibration (which it actually is not) then when we control loudness we would be controlling an environmental variable.

RM: But, again, I think getting this right is only relevant to researchers; ordinary people can think of controlling perception as the same as controlling reality and I don’t think there is anything wrong with that because for most normal human beings perception IS reality, except in those rare cases when what we perceive conflicts with what we know, on other grounds, to be the case; illusions, for example.

Best

Rick

  NOTICE: This e-mail communication (including any attachments) is CONFIDENTIAL and the materials contained herein are PRIVILEGED and intended only for disclosure to or use by the person(s) listed above. If you are neither the intended recipient(s), nor a person responsible for the delivery of this communication to the intended recipient(s), you are hereby notified that any retention, dissemination, distribution or copying of this communication is strictly prohibited. If you have received this communication in error, please notify me immediately by using the "reply" feature or by calling me at the number listed above, and then immediately delete this message and all attachments from your computer. Thank you.

–
Richard S. Marken

“The childhood of the human race is far from over. We
have a long way to go before most people will understand that what they do for
others is just as important to their well-being as what they do for
themselves.” – William T. Powers

[EJ] Hi Rick. My sense is that we have very similar epistemologies here, & I fully agree with the way you state this. This is also why I disagree with a formulation you sometimes make, that when we control certain perceptions we are thereby controlling the environmental aspects that correspond to them.

EJ: I understand the logic of what you’re trying to get at – after all, the perceptions are functions of those environmental substrates. I just think we should reserve the word “control” for perceptions themselves.

EJ: That’s why I lean toward a word like “stabilize,” in the sense that something in the environment gets stabilized by our actions to actually control our own perceptions. It’s a subtle difference between us, & maybe just a judgment call as to wording.

RM: So I prefer to stick with the term “controlled quantity”, q.o, to refer to the perception of the controlled perceptual variable, p, from the point of view of an observer. It makes it clear that the controlled perceptual variable and the controlled quantity are both perceptions, but one (the controlled perception or controlled variable) is the perception in the head of the person doing the controlling and the other (controlled quantity) is the same perception in the head of the person observing, studying or trying to understand the controller.

[EJ] Here, too, I understand what you are getting at, but the formulation still feels a little klunky to me. Mind you, my own way of thinking of it gets pretty klunky, too. 'Cause I lean toward Martin’s “layered protocol” language about ‘my belief about what you are perceiving, & your belief about what I am perceiving,’ & with all those recursive loops it becomes a wonder that we can come to believe that we are jointly talking about the same thing at all!

[EJ] I don’t have a good solution for how we should talk about these interactions, but I think we share a good PCT epistemology that we are always talking about our own perceptions of things, even when we do not explicitly remember to make that point.

All the best,

Erling

[rsmarken]

[From Rick Marken (2016.11.23. 1545)]

Damn, did it again. Made a critical typo. I’ve just got to remember that IU jsut can do this and listen to Izack Perlman play the Tchaikovsky violin concerto at the same time!! Here’s a corrected version of:

···

Back to Izaak.

Best

Rick

–

RM: We control an aspect of the physical variables – a perception. When we do that, we are controlling that aspect of the environment that IS h*w. The environment itself – h and w – CAN’T compute that function; but that function – THE aspect of the physical variables, h and y, THAT is being controlled – staying equal to some value, protected from disturbances (such as variations in h and/or w).

Richard S. Marken

“The childhood of the human race is far from over. We
have a long way to go before most people will understand that what they do for
others is just as important to their well-being as what they do for
themselves.” – William T. Powers