[Bruce Abbott (2018.05.21.1945 EDT)]
Bruce,
BA : Let’s see if I can clarify it for you, Boris.
HB : You are very kind Bruce, but you don’t need to clarify me anything. Beleive me that I understand Bills’ theory. And I hope you beleive me that I understand how organisms function. Including PCT view.
Clarify it to Bill. I’m wondering how many times did you try to clarify to him that he was wrong when you worked together ? You already appologized once to me for misleading CSGnet. I don’t understand why you are trying to prove the same what was once already proven to be wrong with your consensus.
Sorry, but I have no recollection of having appologized to you for misleading CSGnet. About what?
BA : I’ll start with the familiar and relatively simple example of a car’s cruise control, which as you know, controls the car’s speed.
HB : This is very “chewed examples”. I’d rather see if you explain to me sleeping, sunshining, observing, walking…etc. and add also expalinations how Amoeba, Bacteria, Plannton function etc. And then try to make general theory about how organisms function and diagram. How can Amoeba drive a car ?
Boris, Bill did not invent control theory, nor was he the first to recognize that control theory can help one to understand animal (including human) behavior. Norbert Wiener’s classic book, Cybernetics: Control and Communication in the Animal and the Machine (1948), the book that got Bill started down the road to PCT, explicitly sets out to establish a new field of theory and research that would apply equally well to control exhibited by human-engineered systems (e.g., regulators, servo-mechanisms) and to biological control systems.
What Wiener and others realized is that one could have a science of the machine that focused on the organization of the machine’s parts, independent of the specific nature of the parts themselves. For example, a certain organization of parts (forming a closed loop with negative feedback) will function as a control system whose behavior under given conditions can be predicted mathematically. The machine will behave the same (allowing for possible differences in speed) whether the machine’s internal signals are transmitted electrically, hydraulically, pheumatically, or via nerve impulses or horomonal concentrations. Thus, standard engineering control theory can be applied to electromechanical control systems (such as cruise control) and to biological control systems (such as those regulating body temperature or the joint angles of a human arm.
One must, of course, first have a correct model of the system whose behavior one is trying to understand, and part of the research now going on in neuroscience involves identifying the physiological components of a given system and their connections. A nice example is the research being conducted on so-called motor control at the spinal level. Bill Powers outlined one proposal in B:CP based on the evidence that was available to him at the time.
Any control-systems engineer would immediately recognize the standard PCT diagram of a control system as being that of a proportional control system; indeed, Bill applies standard engineering computations to show how such a system is expected to behave. In the steady state, with a constant disturbance acting on the environmental variable q.i., the control system will not quite eliminate the error between reference and perception; instead, there will be a residual error that is inversely proportional to the loop gain of the system. (See the Appendix to B:CP for Bill’s steady-state analysis.)
Bill thought that biological control systems can be analyzed successfully with the mathematical tools of classical engineering control theory – tools originally developed for analysis of nonbiological systems – and did so in his book and many other writings. Do you reject his view?
As Bill understood, human engineered and biological control systems that are organized in the same way will behave in the same way, thus allowing standard engineering control theory to be applied successfully to understand and predict how such systems will behave under given conditions.
If follows that what can be learned from studying how a system such as cruise control works will apply equally  well to understanding how the biological control system whose diagram appears in B:CP (and elsewhere) works.
This is why I chose to use a car’s cruise control to explain why q.i. (which Bill sometimes refers to as the »controlled quantity«) is being controlled when p is being controlled.  Whatever conclusion emerges logically from the analysis of this system applies with equal force to the standard PCT control system.
Cruise control provides an especially clear example because the perceptual variable, p derives from a single environmental variable q.i., the speed of the car. I demonstrated that in such a case, if p is controlled, then q.i. is also controlled, one being a simple proportion of the other. (p is just q.i. multiplied by a constant gain factor). Do you identify some flaw in my logic that leads you to reject that conclusion? If so, please tell me where my analysis has gone off the rails. Don’t just tell me it’s wrong, tell me the logical basis on which you are able to conclude that it is wrong.
Bill succeded in forming general theory and you agreed once :
Bill P. at all (50th Anniversary, 2011) :
Perceptual Control Theory (PCT) provides a general theory of functioning for organisms.
HB : If you will find common points to all behaviors I mentioned above then try to make general theory and diagram how organisms fucntion with all other examples : Ricks, yours, Martins’ and Freds’ and Bruce Nevin’s etc. I’m sorry I have enough of “one case – one theory” discussion.
<
My objective in writing the post was to explain to you why it is true that when p is controlled, then q.i. may also be controlled. You apparently reject that conclusion, so I was trying to help you understand why your opinion in this matter is incorrect. If you reject my analysis, I hope you will take the trouble to identify where my analyis (in the cruise control example) is incorrect.
BA : So why the assertion that control systems control their perceptions (p) and not q.i., the environmental variable of which p is a function? I think the statement should read AND NOT NECESSARILY q.i.
HB : What can I say Bruce ? I already wrote that PCT is general theory and that generally orgsnisms control only perception (uncluding Amoeba, Bacterias and Plankton), How can they control environmental variable (q.i.) ?
Yes, that is what you wrote, and it is wrong. If you go back and read my analysis of how cruise control works, you should be able to see why it is wrong.
I advise you to try to change PCT or make your own theory where you’ll use “cannonical principle”. Afterall I think you are the author of this principle.
Boris, I am not trying to change PCT here, I am trying to explain to you how p and q.i. can be controlled within the same control system, at the same time.
Why are you asking me this ? Find places in PCT where you don’t agree with Biil and ask Powers ladies to change PCT theory. And of course ask them if they can change the diagram. We are “chewing” all the time the same problem. I’m sorry Bruce I have enough. If you think that you are right make your own theory, but it’s not fair that you want to “push” your theory under PCT “umbrella”.
Let me see how PCT would look like with those changes you mentioned. Make example model that we can see how PCT would look like with changes you mentioned " AND NOT NECESSARILY q.i."
.
My post analyzed the operation of a standard PCT-style control system. In what way does that make changes to PCT? Please explain
Until you’ll not do so I’ll use PCT diagram (LCS III) and definitions of control loop (B:CP).
Precisely what I am using.  If you think my description departs from these, please describe those departures. Simply claiming that I am changing them is not helpful – I cannot understand from this precisely what you find objectionable.
I also assume that you proposed Bill many times these changes but he didn’t accept them. Did he ?
Again, what changes? Please be specific.
Once for all. Bills theory PCT is general theory of why and how Living beings behave. If somebody can find better diagram and explanation I’ll be the first to support it.
This is valid for all members. Specialy for Rick who is trying to change PCT into RCT from Bills’ death. Why just do empty talkings. I’d like to see how PCT would look like with your changes. Change definitions. Change diagram if you think that you have better proposal of how generally organisms
My advise is Bruce that you find other ways of proving what you are aiming at ? Sorry to say it Bruce. As I’m concerned, my oppinion is, that you are on wrong way.
I’m sorry you feel that way. I don’t believe that my description has departed in any way from PCT.  In my own opinion, I have merely described how a standard PCT-style control system works, using cruise control as a simple example. Here are the equivalences:
PCT system                        Cruise control
Reference level               Set point
Input quantity                  Input quantity (vehicle speed)
Input function                  Input function (converts vehicle speed to proportionate electrical signal)
Perceptual signal             Electrical signal proportional to vehicle speed
Comparator                       Comparator (compares set point to speed signal)
Error signal                        Error signal (difference between set point speed and speed sensor reading)
Output function               Output function (converts error signal to output physical throttle setting)
Output                                Output (engine throttle setting)
Feedback function          Feedback function (gives effect of output on q.i., vehicle speed)
Feedback                           Physical effect of system’s output on q.i., mediated by laws of physics
Disturbance                       Disturbance: effects of wind, friction, and other forces on vehicle speed
The diagram of the cruise control system is identical to that of the standard PCT-style diagram of a control system. The components form a closed loop with negative feedback. The system operates identically to a standard PCT-style control system, so what you can learn by studying how cruise control works will also apply to any biological control system organized in the same way.
If you believe that they are fundamentally different, somehow, please explain IN DETAIL what those differences are, as you see them. At present, I don’t see them.
Best wishes,
Bruce
Best,
Boris
···
From: “Bruce Abbott” (bbabbott@frontier.com via csgnet Mailing List) csgnet@lists.illinois.edu
Sent: Saturday, May 19, 2018 8:00 PM
To: CSGnet csgnet@lists.illinois.edu
Subject: RE: Lies (was On “variables” (was Re: Do we control “environmental variables”?))
[From Bruce Abbott (2018.19.1400 EDT)]
From: “Boris Hartman” (boris.hartman@masicom.net via csgnet Mailing List) csgnet@lists.illinois.edu
Sent: Friday, May 18, 2018 10:42 AM
To: csgnet@lists.illinois.edu
Subject: FW: Lies (was On “variables” (was Re: Do we control “environmental variables”?))
[From Bruce Abbott (2018.05.15.1845 EDT)]
Boris: do you think that what you wrote on Rick “rubber band” game “theme” is in accordance with Bills’ diagram (LCS III) and his definitions (B:CP) ? > BA: Yes, of course! (See below.) …
BA: Boris, I’ve now indicated how each of the components of a control system, as Bill P. Defined them above, are instantiated in the rubber band demo. I hope you agree with my analysis.
RM: Excellent post Bruce.
HB : I’ll not explain here what was and what wasn’t excelent about Bruces’ post. He is great thinker and I respect him. We agreed in some specific way about only we know how it works
. Bruce has a point but I have a feeling that he doesn’t know how to prove it in scientific way. His view upon “environmental control problem” is totaly different from Ricks.
But I know a statement about “control of perceptual signal and environment” that is perfectly PCT right :
RY earlier : Sure, a perceptual signal (q.i*g) may correspond to, or be a function of, variable aspects of the environment (q.i) but it is the perceptual signal that is controlled not the variable aspects of the environment
HB : It’s a perfect statement. When I saw it first time I wasn’t sure who wrote it. Bill or Rupert. I now know it was Rupert. It’s a PCT flower. Rupert has some really great statements on CSGnet.
I can support this statement with any means that is possible. With PCT means, biological, physiological, name it. But I can’t think of any scientific mean that could support Bruce’s way of proving how “environmental variable and it’s analog perceptual signal” are both controlled. Maybe there is one “vague” way, but it got some other name, not control.
BA : Let’s see if I can clarify it for you, Boris.
I’ll start with the familiar and relatively simple example of a car’s cruise control, which as you know, controls the car’s speed.
The driver sets the reference speed for the car by bringing the car to the desired speed (e.g., 100 kph) and then pressing the SET button of the cruise control. This SET POINT is stored as the value of an electrical reference signal. A sensor senses the car’s speed (q.i.) and converts it to an electrical signal (p) that is proportional to that speed.
The cruise control’s comparator compares the sensed value of the car’s speed (p) to the reference value and generates an error signal based on the difference between them. If the sensed speed is below the set point, then the cruise control advances the throttle of the car’s engine, thus increasing the power delivered by the car’s wheels to the road; consequently the car accelerates, reducing the error until the sensed speed matches the set point. If the sensed speed is above the set point, then the cruise control retards the throttle, which allows the forces such as wind drag and mechanical friction to slow the car, thus reducing the error. As the car’s speed approaches the set point, the throttle setting is adjusted until there is just sufficient power being developed to keep the car’s speed at the set point.
Disturbances: If the car begins to climb or descend a hill, or encounters changes in drag caused by variations in wind speed, the car’s actual speed will change, decelerating in response to headwinds or hill-climbing, or accelerating in response to tailwinds or hill-descending. The sensed speed, which is proportional to the actual speed, will change as well, and this will result in an error between set point and sensed speed. Depending on the direction of that error, the throttle will be either advanced or retarded, the power to the wheels will change, and the car’s actual speed will change so as to reduce the error and bring the sensed speed back to its set point.
For the purpose of this discussion, I am assuming that there is a real car on a real road and that the car has a real speed relative to the road surface. But to regulate the car’s speed, the cruise control must have a way to sense what that speed is, and convert the car’s actual speed to an electrical signal whose value is proportional to that speed. The car’s actual speed, out there in the environment outside the cruise control’s electronic system, is the cruise control’s input quantity, q.i. The electrical signal produced by the speed sensor is the cruise control’s internal representation of that speed, p.
If the electrical speed signal p is proportional to the car’s actual speed, q.i., then by controlling the speed signal, p (keeping it near the set point and defending against deviations from that value by adjusting the throttle setting), the car’s cruise control ALSO CONTROLS THE CAR’s ACTUAL SPEED, keeping it at or near some particular speed despite disturbances.
HB : You just made constantaion. How this exactly happens ? Show me through afferent nerve to afferent neuron and than show me hoe in the same time also “q.i.” IS “CONTROLLED!”.
We could verify this fact experimentally by using other devices to measure the car’s road-speed, such as radar, counting mile posts and dividing by the time between them, and so on, while setting cruise control to various speeds. If the various measures agree, then we can be highly confident that the speed being measured by the car’s cruise control does reflect the car’s actual speed.
HB : We can verify this only with
So why the assertion that control systems control their perceptions (p) and not q.i., the environmental variable of which p is a function? I think the statement should read AND NOT NECESSARILY q.i.
HB : What can I say Bruce. I advise you to try to change PCT. Why are you teliling me this ? Find places in PCT where you don’t agree with Biil and ask Powers ladies to change PCT theory. And of course ask them if they can change the diagram. We are “chewing” all the time the same problem. I’m sorry Bruce I have enough.
Once for all. Bills theory is general theory of why and how Living beings behave. If somebody can find better diagram and explanation I’ll be the first to support it.
Let me see how PCT would look like with the changes you mentioned. But I assume that you proposed Bill many times these changes but he didn’t accept them. Did he ?
I advise you Bruce that you find other ways of proving what you are aiming at ? Sorry to say it Bruce. As I’m concerned my oppinion is that you are on wrong way.
Best,
Boris
BA : Not necessarily, because sensors can fail. Consider the case in which the speed sensor fails and now generates an electrical signal that varies irratically relative to the car’s actual speed. Cruise control will still attempt to make the sensed speed match the set point, and to the extent that it succeeds, p will be controlled. But since the actual speed may bear little relation to actual speed at any given moment, control of q.i. will be very poor (to the extent that some correlation between q.i. and p remains) or even nonexistent ( if the correlation is zero). Thus a control system will always control p, even when q.i. is not controlled. However, in the vast majority of cases in which the correlation is strong, both p and q.i. are controlled. Here, p is the controlled perception and q.i. is the controlled environmental quantity.
In the more complex case in which p is a function of several environmental variables, we might define q.i. in terms of the formula relating these environmental variables to q.i. For example, we might have a control system that controls the perceived area of a rectangle, in which case the two environmental variables might be height and width. In this case q.i. = Rectangle Area = height X width, and p might be a signal that is proportional to q.i., e.g., p = ki * q.i., where ki is the input gain. The control system might attempt to keep the area constant at some reference value by changing the width of the rectangle as drawn on the computer screen, whereas disturbances might act to change the height and width independently. Although p is based on two sepatate input quantities, height and width, its value reflects the area of the actual rectangle on the screen, q.i. Thus if p is controlled, then q.i. is also controlled.
Bruce