[From Fred Nickols (2017.12.04.0645 ET)]
Well, FWIW, I wrote that paragraph on the IAPCT page. Not sure who put it on the web site. I look through my materials and see exactly from whence it came.
Also, FWIW, I stand by it. Makes sense to me.
Fred Nickols
···
From: Bruce Nevin [mailto:bnhpct@gmail.com]
Sent: Sunday, December 3, 2017 11:25 PM
To: CSG csgnet@lists.illinois.edu
Subject: “:” and what is perceived
[From Bruce Nevin (2017.12.03.23:00 ET)]
Boris Hartman (Fri, Dec 1, 2017 at 11:49 PM) –
I’ve been putting some effort into understanding what Boris is trying to say.
Referring to a passage quoted from the IAPCT front page, Boris says:
HB : Problem I see here is that PCT is described as »control of variables« and of course that »behavior serves to control our perceptions« what seems to mean that »Behavior is control«.
He identifies what he sees as three problems.
- He sees that the colon in the title Behavior: the control of perception is being interpreted or paraphrased as “is” and claims that this is wrong.
HB : Ttitle of the book is not : »Behavior is control of perception«, because somebody could think that »Behavior is control«. Titile is »Behavior : Control of Perception«. It’s a problem when many interpretation of Bills’ book Title are present.
HB : So Title can be explained in many ways , but with the explanation »Behavior is control of perception« the interpretation somehow determine the way it has to be understand (what is mostly Ricks merit). And it seems that this interpretation offers »Behavior is control« although also other mening as »control of perception« can be in the game.
HB : So in which way the sentence »Behavior is control of perception« has to be understand ?
A colon divides a sentence into two parts, such that the second part defines the first part by paraphrasing it, elaborating on it, stating essential inferences, or summarizing such elaborations of its meaning. To confirm this understanding of the function of a colon as a punctuation mark in a sentence, after I wrote the above I looked up the word ‘colon’ in the Random House dictionary of English, where I find the following definition of ‘colon’ as a punctuation mark:
the sign ( used to mark a major division in a sentence, to indicate that what follows is an elaboration, summation, implication, etc., of what precedes
A colon is analogous to the equals sign (=) in mathematics, though of course not all mathematical usages apply to language. (All analogies leak.) Maybe the colon has a different significance in some other language, but Bill was writing in English.
There can be no doubt that the colon in the title Behavior: the control of perception asserts that the phrase “the control of perception” is Bill’s definition of the word “behavior”. It is also the accepted PCT definition of the word “behavior”. Therefore the sentence “behavior is the control of perception” is an accurate and adequate paraphrase of the title and (like the title) states what Bill’s book is about.
Boris’s further complaint seems to be about how this equivalence of “behavior” and “the control of perception” is interpreted.
HB: it seems that this interpretation offers »Behavior is control« although also other mening as »control of perception« can be in the game.
Why divide “control” from “control of perception”? What distinction is being made? This seems to have to do with the question what is controlled. There was some controversy on CSGnet not too long ago whether only perceptions controlled or whether variables in the environment are also controlled. Boris has expressed distaste and disinterest in philosophical questions about the epistemology of PCT, but of course that is exactly what this is, an epistemological question about the veridicality of perceptions.
In my view, the answer to the question is “both”. More on this presently.
So far as I can find, Boris does not offer an alternative paraphrase in place of “behavior is the control of perception”. He just thinks its wrong. So why does he think it’s wrong?
Is he invoking a more behaviorist sense of the word “behavior”, meaning “observable actions” or “behavioral outputs”? “Behavior” in this sense is not the control of perceptions, it is the observable means of controlling perceptions. Behavior in the sense of “observable actions” or “behavioral outputs” is represented in a control diagram by the output function. Behavior as the control of perception is represented in a control diagram by the entire control loop.
Farther on, he quotes the 2011 jointly authored paper. (I wonder if he might think less highly of that paper and this quote if he fully apprehended that Rick, Martin, and others were co-authors, and and that they may well have written or modified this very passage.)
Here is the quote:
living things control the perceived environment by means of their behavior. Consequently, the phenomenon of control takes center stage in PCT, with observable behavior playing an important but supporting role.
The phrase “control the perceived environment” neatly encapsulates the “both” answer that I offered above. There can be no controlled perception in the absence of that which is perceived. (Bear in mind that in the PCT model imagination is not in control mode, it is in imagination mode.) During that argument about environmental vs. perceptual variables we found several passages, authored by Bill alone, that say much the same thing
Boris asserts that the above 2011 passage contrasts with a paragraph on the front page of the IAPCT website. Here is that paragraph for reference:
PCT views people as purposeful, living control systems, whose behavior shapes its consequences instead of the other way around. PCT is a feedback-governed view of human behavior. It holds that we target certain variables for control and we compare our perceptions of the current state of those variables with our goal state or reference condition for those variables. If unacceptable gaps exist, we behave in ways that serve to close those gaps. Thus it is that our behavior serves to control our perceptions. There are, however, other actors and factors at work that influence the same variables we are trying to control. Ordinarily these disturbances as they are known in PCT are compensated for and pose no problem. On occasion they can prove overwhelming. Our control is far from perfect.
I don’t know who wrote this, but it does not matter. Almost any piece of writing can be improved in one way or another, and as we begin reworking the website we are open to suggestions. Let’s review his objections.
-
PCT is general theory about how all organisms function
This is evidently a sin of omission. He quotes from B:CP about the scope of PCT extending to all living things.
-
Behavior is means of control with supporting role to »Control of perception«
I believe I can paraphrase this more clearly as
Behavior is [the or a?] means of control of environmental variables, with a supporting role from control of perception.
I believe he gets this from the IAPCT website sentence
we target certain variables for control and we compare our perceptions of the current state of those variables with our goal state or reference condition for those variables.
He does not like the idea that we control perceived variables in the environment. But this assertion that we control environmental variables does not relegate the control of perception to a “supporting role”. Control of a neural signal generated by environmental input to a sensor is inseparable from control of whatever it is in the environment that is being sensed. You can’t have one without the other, because the control loop is closed through the environment. The distinction between environmental variable and perceptual variable is analytical but not actual. The perceptual variable is all that we can possess of the environmental variable. There is a well-known process of refining our perceptions and becoming more sure of their veridicality by testing them and by freeing them from a clutter of imagined perceptions. In its most careful and rigorous form, this process is called science. Less formally, it’s called learning from experience. To the extent that this process gives us perceptions of which we can be more confident, it simultaneously gives us that which is perceived. It gives us that which is perceived precisely because (and to the extent that) we can control.
Going back now to the paragraph from Boris’s post that I quoted at the outset:
HB : Problem I see here is that PCT is described as »control of variables« and of course that »behavior serves to control our perceptions« what seems to mean that »Behavior is control«.
He objects to saying “behavior is the control of perception”, evidently because someone might read only “behavior is control” and think that it means “behavior is control of environmental variables”. But no, it clearly says “control of perception.” Anyway, I do not understand what problem he finds with saying that behavior is control, unless he is referring to observable actions, behavioral outputs that are represented in a control diagram by the output function. Behavior, understood as that which is represented in a control diagram by the entire control loop, is control.
His third objection to the above paragraph quoted from the IAPCT front page seems to be another sin of omission:
- Phenomen of Control … involve also »control of variables« in organism not outside it. Because in organisms there are also processes which don’t look like a control loop but are essential for organisms functioning and they support and even enable control.
I think everyone here agrees that there are variables within the body of the organism which are controlled variables in the sensed environment of the nervous system, and that there are also many biochemical control systems (homeostatic systems) which interact with the neurological control hierarchy only indirectly. And I think we probably all agree that this needs more research. The paragraph on the IAPCT front page does not mention this. It does not contradict it either. And there are many other aspects of PCT that it does not mention. As we begin to reorganize this website we can consider adding something about that, always bearing in mind that this is an introductory paragraph for a lay audience.
Going farther down, after some quotations snipped from some of Bill’s writings, things get murkier.
HB : From Bills’ literature we can see also that »control in organism« is quite hard concept to understand. It’s not just about »Controlling variables in organism« and controlling with behavior, but it involves the whole functioning of the organism, where »control of variables« in environment outside organism is not included as Rick is proposing. There is no such a thing in PCT.
I’m having trouble understanding the assertion that control in the organism is not just about controlling variables in the organism. This seems to refer to control of variables within the body such as I just mentioned. Homeostatic systems do not directly concern variables outside the body, but that is not an argument about the existence of environmental variables.
Studies of collective control pretty conclusively demonstrate the reality of controlled environmental variables. That’s an important part of PCT.
Finally, Bill strongly disapproved of any sort of PCT fundamentalism referring to what he said or wrote as authoritative gospel. It’s antithetical to the proper character of science, which guards against such natural human proclivities as argument from authority, ad hominem argument, confirmation bias, and so on. Of course, scientists do frequently fall into these traps, but the stance of science is to be alert to them and guard against them. Any kind of authoritarian cult will interfere with the acceptance and establishment of PCT that we seek.
As I said, I’ve been putting some effort into understanding what Boris is trying to say. I’m not sure the effort has been much rewarded.
I will not respond to vituperation, ad hominem attack, or just plain incoherence.
/Bruce
On Fri, Dec 1, 2017 at 11:49 PM, Boris Hartman boris.hartman@masicom.net wrote:
Hi all,…
From: Bruce Nevin [mailto:bnhpct@gmail.com]
Sent: Sunday, November 26, 2017 6:31 PM
To: Control Systems Group Network (CSGnet)
Subject: Re: source code
We just need to visit https://sites.google.com/site/perceptualcontroldemos/ more often, to boost it in the Google ranking.
HB : Well I accidentaly saw this :
This website serves as a repository for computer programs that were developed to highlight various aspects of Perceptual Control Theory (PCT), a theory of human and animal behavior developed by William T. Powers. According to PCT (and the title of Bill’s seminal 1973 book), behavior is the control of perception. To control a perception is to take actions that tend to bring the perception toward a given state (reference value) and keep it there by opposing the effects of any disturbances to that perception, as in steering a car so as to keep it on the road and heading where we want it to go.
HB : Ttitle of the book is not : »Behavior is control of perception«, because somebody could think that »Behavior is control«. Titile is »Behavior : Control of Perception«. It’s a problem when many interpretation of Bills’ book Title are present.
So Title can be explained in many ways , but with the explanation »Behavior is control of perception« the interpretation somehow determine the way it has to be understand (what is mostly Ricks merit). And it seems that this interpretation offers »Behavior is control« although also other mening as »control of perception« can be in the game.
So in which way the sentence »Behavior is control of perception« has to be understand ?
The second problem which is much the same as upper interpretation I see in IAPCT interpretation of what PCT means :
IAPCT : Essentially, PCT views people as purposeful, living control systems, whose behavior shapes its consequences instead of the other way around. PCT is a feedback-governed view of human behavior. It holds that we target certain variables for control and we compare our perceptions of the current state of those variables with our goal state or reference condition for those variables. If unacceptable gaps exist, we behave in ways that serve to close those gaps. Thus it is that our behavior serves to control our perceptions. There are, however, other actors and factors at work that influence the same variables we are trying to control. Ordinarily these disturbances as they are known in PCT are compensated for and pose no problem. On occasion they can prove overwhelming. Our control is far from perfect.
HB : Problem I see here is that PCT is described as »control of variables« and of course that »behavior serves to control our perceptions« what seems to mean that »Behavior is control«.
Both versions of interpreting Powers work seems to me of »lower level« understanding not worth of Powers real intelectual power.
In my version of PCT interpretation (as Rick emphasized) I wrote many times that William T. Powers was great genius and such interpretations of PCT as we see above is by my oppinion degradation of his great mind. PCT is much more then :
»Control of variables« (what is probably Ricks' version of control as he is the one who is emphasizing Control of variables in LCS environment)
»Behavior serves to control perception« or »Behavior is Control of Perception« what could mean that »Behavior is control«.
Both statements can be seen often in RCT (Ricks Control Theory). So we are back to basic problem what PCT is about. My oppinion is that IAPCT should citate William T.Powers thoughts about PCT (it’s his Theory) not inventing some Ricks’ Control imaginational constructs.
So my proposal is that first page of IAPCT should contain great citations of William T.Powers and it should serve as memorial to his Theory.
My proposal is that IAPCT should start with citation which shows the real nature of PCT and that is :
Bill P. at all (50th Anniversary, 2011) :
Perceptual Control Theory (PCT) provides a general theory of functioning for organisms. At the conceptual core of the theory is the observation that living things control the perceived environment by means of their behavior. Consequently, the phenomenon of control takes center stage in PCT, with observable behavior playing an important but supporting role.
Bill P (B:CP) :
PCT…«can explain a fundamental aspect of how every living thing works, form the tiniest amoeba to the being who is reading these words.«
HB : This is significant difference to what was proposed above. It says that :
PCT is general theory about how all organisms function
Behavior is means of control with supporting role to »Control of perception«
Phenomen of Control takes central stage in PCT what involve also »control of variables« in organism not outside it. Because in organisms there are also processes which don't look like a control loop but are essential for organisms functioning and they support and even enable control.
So definition of control in PCT is (B:CP) :
CONTROL : Achievement and maintenance of a preselected state in the controlling system, through actions on the environment that also cancel the effects of disturbances.
HB : Achievement and maintainance of »preselected state« in organism is also enabled by »control« of subsequent processes to support actual control loop processes.
Bill P :
Obviously not every variable ….can be involved in this wrongness-detection. Some processes are burried deep in the details of organ function and cellular function.
Bill P :
For control of most of the variables in the physiological-biochemical system we rely entirely on the inherited system to work right.
HB : Bills’ legacy show that interpretation of PCT should involve also physiological and biological knowledge not just psychological and algebraic if we want to understand the phenomenon of how references are produced and realized :
Bill P (LCS I) : Reference state can not exist under the old cause-effect model. They refer, as far as external observations are concerned only to future states of the organism or it’s environment. They cannot affect present behavior, and they must be treated simply as outcomes of events caused by prior events. The flaw of this reasoning is hard to understand if one does not know (as the founders of scientific psychology did not know) of organizations capable of complex internal activities that are essentially independent of current external events.
HB : From Bills’ literature we can see also that »control in organism« is quite hard concept to understand. It’s not just about »Controlling variables in organism« and controlling with behavior, but it involves the whole functioning of the organism, where »control of variables« in environment outside organism is not included as Rick is proposing. There is no such a thing in PCT.
Bill P :
All sensory endings act to convert the magnitude of some physical interaction into the magnitude of a neural current (with or without significant emphasis of rates of change). Coverversely, all sensory information available to more central parts of the brain must first exist in the form of these primary neural currents.
The organism acts to bring under control, in relation to some reference state, the sensed perceptions.
HB :It’s quite clear that time line shows that perception will be controlled after it is sensed not before so to be »controlled perception«. So how »Control of Perception« really function :
Bill P (1998) : Our only view of the real world is our view of the neural signals that represent it inside our own brains. When we act to make a perception change to our more desireble state – when wwe make the perception of the glass change from »on the table« to »near the mouth« - we have no direct knowledge of what we are doing to the reality that is the origin of our neural signal; we know only the final result, how the result looks, feels, smells, sounds, tastes, and so forth…It means that we produce actions that alter the world of percception…
HB : It’s obviously that »Behavior is not controlling perception«, but it’s used just for changing the world of perception. And it’s obviously that we change perception to our more desirable state not about how we »Control behavior« or some »Controlled variable« in environment or that we even control perception with behavior. Theory is about »Control of perception« inside organism. Everything is grasped in perception.
Bill P : Briefly, then: what I call the hierarchy of perceptions is the model. When you open your eyes and look around, what you see – and feel, smell, hear, and taste – is the model. In fact we never experience ANYTHING BUT the model. The model is composed of perceptions of all kinds from intensities on up.
HB : As I proposed many times in these years in my version of PCT understanding PCT definitions should be respected which by my oppinion represent PCT control loop which can be confirmed by biological and physiological evidences :
Bill P (B:CP):
OUTPUT FUNCTION : The portion of a system that converts the magnitude or state of a signal inside the system into a corresponding set of effects on the immediate environment of the system
Bill P (LCS III)::…the output function shown in it’s own box represents the means this system has for causing changes in it’s environment.
Bill P (LCS III):
FEED-BACK FUNCTION : The box represents the set of physical laws, properties, arrangements, linkages, by which the action of this system feeds-back to affect its own input, the controlled variable. That's what feed-back means : it's an effect of a system's output on it's own input.
Bill P (B:CP) :
INPUT FUNCTION : The portion of a system that receives signals or stimuli from outside the system, and generates a perceptual signal that is some function of the received signals or stimuli.
Bill P (B:CP) :
COMPARATOR : The portion of control system that computes the magnitude and direction of mismatch between perceptual and reference signal.
Boris
On Sat, Nov 25, 2017 at 4:13 PM, Bruce Abbott bbabbott@frontier.com wrote:
Sorry, I meant to include it, and forgot. It’s
The title I gave to the site is actually “Perceptual Control System Demos.�
I’m surprised that a Google search didn’t turn it up!
Bruce A.
From: Bruce Nevin [mailto:bnhpct@gmail.com]
Sent: Saturday, November 25, 2017 11:50 AM
To: Bruce Abbott bbabbott@frontier.com
Subject: Re: source code
URL, please? A search on “Perceptual Control Demos abbott” doesn’t nail it in an obvious way. I see
http://users.ipfw.edu/abbott/pct/ (I don’t see demos there)
http://www.pct-labs.com (apparently hosted by Dag)
On Fri, Nov 24, 2017 at 2:04 PM, Bruce Abbott bbabbott@frontier.com wrote:
Hi Bruce,
I have a Google website, Perceptual Control Demos, that offers downloadable copies of all the programs in LCS III, plus a few others that I have written. Each is a zip file that includes source code and the executable. These are all Delphi (a descendent of Pascal) programs. I don’t know anything about posting to SourceForge, but I suppose that could be another repository option. The Google site is probably going to be less permanent as it’s tied to me personally.
Most of the effort (and code) that goes into writing these programs centers on creating the user interface; in fact readers of the code often are surprised at how little code involves simulating the control systems and the physics. So there’s usually not much that’s reusable
Bruce A.
From: Bruce Nevin [mailto:bnhpct@gmail.com]
Sent: Wednesday, November 22, 2017 9:07 PM
To: Adam Matic adam.matic@gmail.com
Cc: Tom Bourbon tombourbon@sbcglobal.net; Bruce Abbott bbabbott@frontier.com; j richard kenneway jrk@cmp.uea.ac.uk; Rupert Young rupert@perceptualrobots.com; Greg and Pat Williams gwill@mis.net; Gary Cziko gcziko@gmail.com; board@iapct.org
Subject: Re: source code
BN: Thanks, Rick. Adam now included.
BN: Bill’s source code was an initial concern, but it’s a much broader issue. We should have a common repository for code developed by others and as new code is developed ongoing. Many code management issues should be considered. Does it make sense to think of libraries of reusable and adaptable code? There’s an obvious modularity to any CT hierarchy. Should we think of the architecture of a control loop in modular terms? Should we organize a project or collection of related projects in sourceforge? Other questions will occur to anyone experienced in software development. These questions are outside my scope–above my pay grade, as the saying goes. But I’m confident they should be addressed.
/Bruce
On Wed, Nov 22, 2017 at 7:00 PM, Richard Marken rsmarken@gmail.com wrote:
If it’s source code for Bill’s demos you want you should also copy to Adam Matic (adam.matic@gmail.com).
On Wed, Nov 22, 2017 at 8:17 AM, Bruce Nevin bnhpct@gmail.com wrote:
Friends,
I sent the below query in August. Greg said he has no source code and referred me to Dag and Gary
My concern was archiving, but maybe better would be an active development repository in sourceforge. Should I ask more widely on CSGnet for volunteers competent to take this up?
/Bruce
On Mon, Aug 7, 2017 at 9:56 AM, Bruce Nevin bnhpct@gmail.com wrote:
Tom, Bruce, Richard, Rupert, Greg,
For years when I have told people about demos and simulations I have said that the source code is available so that they can see that we are not ‘cheating’ in any way. For example, the 2011-2012 web-published joint paper says this. This needs to be a true statement. We need to have the source code assembled in an accessible archive.
Another reason (of which I know you, Tom, have a vivid awareness) is the value of the source code for learning how make computer simulations and other PCT programs. I, personally, have felt the lack.
Would you five be willing to work together to pull together all the source code you can, and associated documentation if any, and get it to appropriate archives?
It should be archived at Northwestern and on our several websites. Replication is welcome, on the LOCKSS principle beloved of archivists (“lots of copies keep stuff safe”).
You may well determine among you that other people should be involved. Please let me, Richard Pfau, and Allie know your decisions and how it’s going.
/Bruce
–
Richard S. Marken
"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
–Antoine de Saint-Exupery