General

[PHILIP_JERAIR_YERANO]

If a system uses inverse kinematics to calculate the effects of its output, is it not ultimately doing so in order for its actions to produce a desired effect on its input? Does it really matter if a system is using a simulated model of the environment to predict its input before calculating its output? Â

All control systems have reference values for their inputs. There is no system that only controls its output. Â

All control systems are doing “perceptualâ€? control. Who does not agree with this proposition?Â

[wmansell]

H Philip, I think It does matter for a few reasons:

- calculating inverse kinemetics is computational laboursome and we don’t ev n know if it is possible within a biological system in the timescale required

- the environment is changing continually and an element of this change will be either random, or not anticipated within the current model, or be deliberately evasive as in other agents in conflict, or even be an unanticipated result of one’s own actions.

- even though the mathematics of models derived from some other theories control do control their input, this is rarely, if ever, acknowledged within a diagram or within a verbal description of the theory. So, Science, and society, without PCT, continue obliviously regurgitating the view that we control our output, and fail to advance through the advantages of shifting perspective to behaviour as the control of input.

All the best

Warren

···

On 28 Jan 2018, at 05:36, PHILIP JERAIR YERANOSIAN <pyeranos@ucla.edu> wrote:

If a system uses inverse kinematics to calculate the effects of its output, is it not ultimately doing so in order for its actions to produce a desired effect on its input? Does it really matter if a system is using a simulated model of the environment to predict its input before calculating its output?
All control systems have reference values for their inputs. There is no system that only controls its output.
All control systems are doing “perceptual�? control. Who does not agree with this proposition?

[MartinT]

[Martin Taylor 2018.01.28.00.59]

If a system uses inverse kinematics to calculate the effects of its output, is it not ultimately doing so in order for its actions to produce a desired effect on its input? Does it really matter if a system is using a simulated model of the environment to predict its input before calculating its output?

Does it matter to who? In other words, just what are you asking?

To a person wanting to make a mechanical system, who is willing to program the proper kinematics and tune it before moving the device to a slightly different environment, it probably doesn't matter, because they would have the fun of fiddling with getting it right if they had sufficient computing power. To the person interested in how organisms function, it matters if they can find data that suggests this is more like what the organism in question really does. To a person who wants to make a simple device that works in a variety of environments and can quickly correct for disturbances it matters because simply making the input move toward its reference by any means is enough to produce the right result, perhaps using a hierarchy of such simple "direction of correction only" units, and does it without needing any model of the environment.

On CSGnet, more people are of the second kind than of the first or third. What matters to them is whether there are data showing that new kinematic calculations are produced from the new environmental simulations every time the organism moves to a new place with new structures and constraints in it, and that these new calculations for every new place describe the actual operation of the organism more accurately than do models that use simple "move the input in the right direction" units for the same task.

All control systems have reference values for their inputs. There is no system that only controls its output.
All control systems are doing “perceptual�? control. Who does not agree with this proposition?

I guess it depends on what the proposition actually is, and that depends on what you call a control system. Can you define a control system without including the conclusion of the proposition in the definition of control?

Martin

···

On 2018/01/28 12:36 AM, PHILIP JERAIR YERANOSIAN wrote:

[rsmarken]

[Rick Marken 2018-01-28_09:55:50]

···

On Sat, Jan 27, 2018 at 9:36 PM, PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu wrote:
Â

PY: If a system uses inverse kinematics to calculate the effects of its output, is it not ultimately doing so in order for its actions to produce a desired effect on its input? Does it really matter if a system is using a simulated model of the environment to predict its input before calculating its output? Â

All control systems have reference values for their inputs. There is no system that only controls its output. Â

All control systems are doing “perceptualâ€? control. Who does not agree with this proposition?Â

RM: RM: I think you are asking whether everyone agrees that all control systems (properly designed closed- loop negative feedback systems) even those that include model-based calculations of inverse kinematics, are doing perceptual control. I can’t speak for everyone but I certainly agree with that. And I think it’s a very good observation.Â

RM: This fact about how control systems work is not particularly important for engineers who build such systems to understand since the engineers know which variables they want their control systems to control and they develop perceptual systems (sensors) that provide analogs of these variables for the system to control. But it is very important for students of the controlling done by living systems to understand since figuring out what perceptions these systems are controlling is essential to understanding their behavior.Â

Best

Rick

–
Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery

[PHILIP_JERAIR_YERANO]

The purpose of my question is to figure out why PCT is to be developed as a scientific discipline in its own right.Â
What predictions does PCT make that are different from those of other theories?   Â

Virus-free. www.avast.com

···

On Sun, Jan 28, 2018 at 9:56 AM, Richard Marken rsmarken@gmail.com wrote:

[Rick Marken 2018-01-28_09:55:50]

On Sat, Jan 27, 2018 at 9:36 PM, PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu wrote:
Â

PY: If a system uses inverse kinematics to calculate the effects of its output, is it not ultimately doing so in order for its actions to produce a desired effect on its input? Does it really matter if a system is using a simulated model of the environment to predict its input before calculating its output? Â

All control systems have reference values for their inputs. There is no system that only controls its output. Â

All control systems are doing “perceptualâ€? control. Who does not agree with this proposition?Â

RM: RM: I think you are asking whether everyone agrees that all control systems (properly designed closed- loop negative feedback systems) even those that include model-based calculations of inverse kinematics, are doing perceptual control. I can’t speak for everyone but I certainly agree with that. And I think it’s a very good observation.Â

RM: This fact about how control systems work is not particularly important for engineers who build such systems to understand since the engineers know which variables they want their control systems to control and they develop perceptual systems (sensors) that provide analogs of these variables for the system to control. But it is very important for students of the controlling done by living systems to understand since figuring out what perceptions these systems are controlling is essential to understanding their behavior.Â

Best

Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery

Rick

[bnhpct]

[Bruce Nevin 2018-02-19_14:00:13 ET]

PHILIP JERAIR YERANOSIANÂ (Sun, Feb 18, 2018 at 3:28 PM) –

Philip, we do model-based control without calculating inverse kinematics. When you walk through a familiar room in the dark or with your eyes closed, the model is made of perceptions restored from memory and controlled in imagination. Controlling in imagination to avoid the chair that is there and the table that is over there accompanies ordinary perceptual control processes by which you walk over to the light switch by the door there and turn it on. In the final stages of controlling a perception of having light in the room by means of flipping that switch no amount of prediction is going to guarantee that your fingers find the switch reliably every time, but repeated practice will sharpen the precision of your memory, not to mention your confidence in your imagination.Â

I will leave it to others to refer you to the abundant literature laying out predictions that PCT makes that are different from those of other theories. And I have no choice but to leave it up to you to read that literature.

···

On Sun, Feb 18, 2018 at 3:28 PM, PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu wrote:

The purpose of my question is to figure out why PCT is to be developed as a scientific discipline in its own right.Â
What predictions does PCT make that are different from those of other theories?   Â

Virus-free. www.avast.com

On Sun, Jan 28, 2018 at 9:56 AM, Richard Marken rsmarken@gmail.com wrote:

[Rick Marken 2018-01-28_09:55:50]

On Sat, Jan 27, 2018 at 9:36 PM, PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu wrote:
Â

PY: If a system uses inverse kinematics to calculate the effects of its output, is it not ultimately doing so in order for its actions to produce a desired effect on its input? Does it really matter if a system is using a simulated model of the environment to predict its input before calculating its output? Â

All control systems have reference values for their inputs. There is no system that only controls its output. Â

All control systems are doing “perceptualâ€? control. Who does not agree with this proposition?Â

RM: RM: I think you are asking whether everyone agrees that all control systems (properly designed closed- loop negative feedback systems) even those that include model-based calculations of inverse kinematics, are doing perceptual control. I can’t speak for everyone but I certainly agree with that. And I think it’s a very good observation.Â

RM: This fact about how control systems work is not particularly important for engineers who build such systems to understand since the engineers know which variables they want their control systems to control and they develop perceptual systems (sensors) that provide analogs of these variables for the system to control. But it is very important for students of the controlling done by living systems to understand since figuring out what perceptions these systems are controlling is essential to understanding their behavior.Â

Best

Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery

Rick

[PHILIP_JERAIR_YERANO]

That is a very strange example. Â
I have read some PCT literature. One remark is that PCT doesn’t employ the same model as stimulus-response psychology. How do the predictions of a stimulus-response theory of behavior differ from PCT predictions? Â

···

On Mon, Feb 19, 2018 at 11:01 AM, Bruce Nevin bnhpct@gmail.com wrote:

[Bruce Nevin 2018-02-19_14:00:13 ET]

PHILIP JERAIR YERANOSIANÂ (Sun, Feb 18, 2018 at 3:28 PM) –

Philip, we do model-based control without calculating inverse kinematics. When you walk through a familiar room in the dark or with your eyes closed, the model is made of perceptions restored from memory and controlled in imagination. Controlling in imagination to avoid the chair that is there and the table that is over there accompanies ordinary perceptual control processes by which you walk over to the light switch by the door there and turn it on. In the final stages of controlling a perception of having light in the room by means of flipping that switch no amount of prediction is going to guarantee that your fingers find the switch reliably every time, but repeated practice will sharpen the precision of your memory, not to mention your confidence in your imagination.Â

I will leave it to others to refer you to the abundant literature laying out predictions that PCT makes that are different from those of other theories. And I have no choice but to leave it up to you to read that literature.

/Bruce

On Sun, Feb 18, 2018 at 3:28 PM, PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu wrote:

The purpose of my question is to figure out why PCT is to be developed as a scientific discipline in its own right.Â
What predictions does PCT make that are different from those of other theories?   Â

Virus-free. www.avast.com

On Sun, Jan 28, 2018 at 9:56 AM, Richard Marken rsmarken@gmail.com wrote:

[Rick Marken 2018-01-28_09:55:50]

On Sat, Jan 27, 2018 at 9:36 PM, PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu wrote:
Â

PY: If a system uses inverse kinematics to calculate the effects of its output, is it not ultimately doing so in order for its actions to produce a desired effect on its input? Does it really matter if a system is using a simulated model of the environment to predict its input before calculating its output? Â

All control systems have reference values for their inputs. There is no system that only controls its output. Â

All control systems are doing “perceptualâ€? control. Who does not agree with this proposition?Â

RM: RM: I think you are asking whether everyone agrees that all control systems (properly designed closed- loop negative feedback systems) even those that include model-based calculations of inverse kinematics, are doing perceptual control. I can’t speak for everyone but I certainly agree with that. And I think it’s a very good observation.Â

RM: This fact about how control systems work is not particularly important for engineers who build such systems to understand since the engineers know which variables they want their control systems to control and they develop perceptual systems (sensors) that provide analogs of these variables for the system to control. But it is very important for students of the controlling done by living systems to understand since figuring out what perceptions these systems are controlling is essential to understanding their behavior.Â

Best

Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery

Rick

[EetuP]

[Eetu Pikkarainen 2018-02-20_08:45:11 UTC]

Sorry, I am not a psychologist so I hope that someone will correct and complement my short answer.

S-R psychology makes predictions about subject’s reactions based on the current stimulus and previous knowledge. Usually the success of these predictions are quite low.

PCT explains why the correlation between stimulus and response must be slow: there is always an important intervening variable, namely the reference value, which determines how the stimulus
will affect the response. So PCT research does not even try to predict reactions from stimuluses. Instead it tries to find out what perception the subject is controlling (the Test for controlled variable). If it can be found then it is possible to predict
with high accuracy how the subject responds by her output (“reaction�) to the disturbances (“stimulus�) against the perceived value of the controlled variable.

Perhaps the example in Powers: Behavior: The Control of Perception (2005) about the analysis of a typical rat experiment in PCT terms could be lightening about the difference, pages
238-243.

An extensive description about PCT type of research in relation to more typical psychological research is in Runkel, P. (1990). Casting Nets and Testing Specimens: Two Grand Methods
of Psychology.

Eetu

···

That is a very strange example.

I have read some PCT literature. One remark is that PCT doesn’t employ the same model as stimulus-response psychology. How do the predictions of a stimulus-response theory of behavior differ from PCT predictions?

On Mon, Feb 19, 2018 at 11:01 AM, Bruce Nevin bnhpct@gmail.com wrote:

[Bruce Nevin 2018-02-19_14:00:13 ET]

PHILIP JERAIR YERANOSIAN (Sun, Feb 18, 2018 at 3:28 PM) –

Philip, we do model-based control without calculating inverse kinematics. When you walk through a familiar room in the dark or with your eyes closed, the model is made of perceptions restored from memory and controlled in imagination. Controlling
in imagination to avoid the chair that is there and the table that is over
there accompanies ordinary perceptual control processes by which you walk over to the light switch by the door
there and turn it on. In the final stages of controlling a perception of having light in the room by means of flipping that switch no amount of prediction is going to guarantee that your fingers find the switch reliably every time, but repeated practice
will sharpen the precision of your memory, not to mention your confidence in your imagination.

I will leave it to others to refer you to the abundant literature laying out predictions that PCT makes that are different from those of other theories. And I have no choice but to leave it up to you to read that literature.

/Bruce

On Sun, Feb 18, 2018 at 3:28 PM, PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu wrote:

The purpose of my question is to figure out why PCT is to be developed as a scientific discipline in its own right.

What predictions does PCT make that are different from those of other theories?

Virus-free.

www.avast.com

On Sun, Jan 28, 2018 at 9:56 AM, Richard Marken rsmarken@gmail.com wrote:

[Rick Marken 2018-01-28_09:55:50]

On Sat, Jan 27, 2018 at 9:36 PM, PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu wrote:

PY: If a system uses inverse kinematics to calculate the effects of its output, is it not ultimately doing so in order for its actions to produce a desired effect on its input? Does it really matter if a system is using a simulated model
of the environment to predict its input before calculating its output?

All control systems have reference values for their inputs. There is no system that only controls its output.

All control systems are doing “perceptual� control. Who does not agree with this proposition?

RM:
RM: I think you are asking whether everyone agrees that all control systems (properly designed closed- loop negative feedback systems) even those that include model-based calculations of inverse kinematics, are doing perceptual control. I can’t speak for everyone
but I certainly agree with that. And I think it’s a very good observation.

RM: This fact about how control systems work is not particularly important for engineers who build such systems to understand since the engineers know which variables they want their control systems to control and they develop perceptual
systems (sensors) that provide analogs of these variables for the system to control. But it is very important for students of the controlling done by living systems to understand since figuring out what perceptions these systems are controlling is essential
to understanding their behavior.

Best

Rick

–

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

[PHILIP_JERAIR_YERANO]

I understand that PCT explains how S-R psychology works. But these two theories involve what is essentially the same constitutive equation: response is proportional to stimulus. In order to constitute a different theory, this constitutive equation would need to be different. Would you disagree?   Â

···

On Tue, Feb 20, 2018 at 1:06 AM, Eetu Pikkarainen eetu.pikkarainen@oulu.fi wrote:

[Eetu Pikkarainen 2018-02-20_08:45:11 UTC]

Â

Sorry, I am not a psychologist so I hope that someone will correct and complement my short answer.

S-R psychology makes predictions about subject’s reactions based on the current stimulus and previous knowledge. Usually the success of these predictions are quite low.

PCT explains why the correlation between stimulus and response must be slow: there is always an important intervening variable, namely the reference value, which determines how the stimulus
will affect the response. So PCT research does not even try to predict reactions from stimuluses. Instead it tries to find out what perception the subject is controlling (the Test for controlled variable). If it can be found then it is possible to predict
with high accuracy how the subject responds by her output (“reaction�) to the disturbances (“stimulus�) against the perceived value of the controlled variable.

Perhaps the example in Powers: Behavior: The Control of Perception (2005) about the analysis of a typical rat experiment in PCT terms could be lightening about the difference, pages
238-243.

An extensive description about PCT type of research in relation to more typical psychological research is in Runkel, P. (1990). Casting Nets and Testing Specimens: Two Grand Methods
of Psychology.

Â

Eetu

Â

From: PHILIP JERAIR YERANOSIAN [mailto:pyeranos@ucla.edu]
Sent: Tuesday, February 20, 2018 3:43 AM
To: csgnet@lists.illinois.edu
Subject: Re: General

Â

That is a very strange example. Â

I have read some PCT literature. One remark is that PCT doesn’t employ the same model as stimulus-response psychology. How do the predictions of a stimulus-response theory of behavior differ from PCT predictions? Â

Â

On Mon, Feb 19, 2018 at 11:01 AM, Bruce Nevin bnhpct@gmail.com wrote:

[Bruce Nevin 2018-02-19_14:00:13 ET]

Â

PHILIP JERAIR YERANOSIANÂ (Sun, Feb 18, 2018 at 3:28 PM) –

Â

Philip, we do model-based control without calculating inverse kinematics. When you walk through a familiar room in the dark or with your eyes closed, the model is made of perceptions restored from memory and controlled in imagination. Controlling
in imagination to avoid the chair that is there and the table that is over
there accompanies ordinary perceptual control processes by which you walk over to the light switch by the door
there  and turn it on. In the final stages of controlling a perception of having light in the room by means of flipping that switch no amount of prediction is going to guarantee that your fingers find the switch reliably every time, but repeated practice
will sharpen the precision of your memory, not to mention your confidence in your imagination.Â

Â

I will leave it to others to refer you to the abundant literature laying out predictions that PCT makes that are different from those of other theories. And I have no choice but to leave it up to you to read that literature.

Â

/Bruce

Â

On Sun, Feb 18, 2018 at 3:28 PM, PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu wrote:

The purpose of my question is to figure out why PCT is to be developed as a scientific discipline in its own right.Â

What predictions does PCT make that are different from those of other theories?   Â

Â

Virus-free.

www.avast.com

Â

On Sun, Jan 28, 2018 at 9:56 AM, Richard Marken rsmarken@gmail.com wrote:

[Rick Marken 2018-01-28_09:55:50]

Â

On Sat, Jan 27, 2018 at 9:36 PM, PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu wrote:

Â

PY: If a system uses inverse kinematics to calculate the effects of its output, is it not ultimately doing so in order for its actions to produce a desired effect on its input? Does it really matter if a system is using a simulated model
of the environment to predict its input before calculating its output? Â

All control systems have reference values for their inputs. There is no system that only controls its output. Â

All control systems are doing “perceptualâ€? control. Who does not agree with this proposition?Â

RM:
RM: I think you are asking whether everyone agrees that all control systems (properly designed closed- loop negative feedback systems) even those that include model-based calculations of inverse kinematics, are doing perceptual control. I can’t speak for everyone
but I certainly agree with that. And I think it’s a very good observation.Â

Â

RM: This fact about how control systems work is not particularly important for engineers who build such systems to understand since the engineers know which variables they want their control systems to control and they develop perceptual
systems (sensors) that provide analogs of these variables for the system to control. But it is very important for students of the controlling done by living systems to understand since figuring out what perceptions these systems are controlling is essential
to understanding their behavior.Â

Â

Best

Â

Rick

Â

–

Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you

have nothing left to take away.�

                --Antoine de Saint-Exupery

Â

Â

Â

[rsmarken]

[Rick Marken 2018-02-20_09:52:49]

Eetu Pikkarainen (2018-02-20_08:45:11 UTC)

Â

EP: Sorry, I am not a psychologist so I hope that someone will correct and complement my short answer.

RM: I see nothing to correct and only things to compliment in this post. I compliment your excellent description of what PCT explains that S-R psychology doesn't. And I also compliment your excellent command of English as revealed in your understanding of the distinction between complementing and complimenting.Â
Best
Rick
 >

···

S-R psychology makes predictions about subject’s reactions based on the current stimulus and previous knowledge. Usually the success of these predictions are quite low.

PCT explains why the correlation between stimulus and response must be slow: there is always an important intervening variable, namely the reference value, which determines how the stimulus will affect the response. So PCT research does not even try to predict reactions from stimuluses. Instead it tries to find out what perception the subject is controlling (the Test for controlled variable). If it can be found then it is possible to predict with high accuracy how the subject responds by her output (“reaction�) to the disturbances (“stimulus�) against the perceived value of the controlled variable.

Perhaps the example in Powers: Behavior: The Control of Perception (2005) about the analysis of a typical rat experiment in PCT terms could be lightening about the difference, pages 238-243.

An extensive description about PCT type of research in relation to more typical psychological research is in Runkel, P. (1990). Casting Nets and Testing Specimens: Two Grand Methods of Psychology.

Â

Eetu

Â

From: PHILIP JERAIR YERANOSIAN [mailto:pyeranos@ucla.edu>pyeranos@ucla.edu]
Sent: Tuesday, February 20, 2018 3:43 AM
To: <mailto:csgnet@lists.illinois.edu>csgnet@lists.illinois.edu
Subject: Re: General

Â

That is a very strange example. Â

I have read some PCT literature. One remark is that PCT doesn't employ the same model as stimulus-response psychology. How do the predictions of a stimulus-response theory of behavior differ from PCT predictions? Â

Â

On Mon, Feb 19, 2018 at 11:01 AM, Bruce Nevin <<mailto:bnhpct@gmail.com>bnhpct@gmail.com> wrote:

[Bruce Nevin 2018-02-19_14:00:13 ET]

Â

PHILIP JERAIR YERANOSIANÂ (Sun, Feb 18, 2018 at 3:28 PM) --

Â

Philip, we do model-based control without calculating inverse kinematics. When you walk through a familiar room in the dark or with your eyes closed, the model is made of perceptions restored from memory and controlled in imagination. Controlling in imagination to avoid the chair that is there and the table that is over there accompanies ordinary perceptual control processes by which you walk over to the light switch by the door there and turn it on. In the final stages of controlling a perception of having light in the room by means of flipping that switch no amount of prediction is going to guarantee that your fingers find the switch reliably every time, but repeated practice will sharpen the precision of your memory, not to mention your confidence in your imagination.Â

Â

I will leave it to others to refer you to the abundant literature laying out predictions that PCT makes that are different from those of other theories. And I have no choice but to leave it up to you to read that literature.

Â

/Bruce

Â

On Sun, Feb 18, 2018 at 3:28 PM, PHILIP JERAIR YERANOSIAN <<mailto:pyeranos@ucla.edu>pyeranos@ucla.edu> wrote:

The purpose of my question is to figure out why PCT is to be developed as a scientific discipline in its own right.Â

What predictions does PCT make that are different from those of other theories?   Â

Â

<Safe emailing Mac | Avast; <https://ipmcdn.avast.com/images/icons/icon-envelope-tick-round-orange-animated-no-repeat-v1.gif&gt;&gt;&gt;&gt;

Virus-free. <Safe emailing Mac | Avast

Â

On Sun, Jan 28, 2018 at 9:56 AM, Richard Marken <<mailto:rsmarken@gmail.com>rsmarken@gmail.com> wrote:

[Rick Marken 2018-01-28_09:55:50]

Â

On Sat, Jan 27, 2018 at 9:36 PM, PHILIP JERAIR YERANOSIAN <<mailto:pyeranos@ucla.edu>pyeranos@ucla.edu> wrote:

Â

PY: If a system uses inverse kinematics to calculate the effects of its output, is it not ultimately doing so in order for its actions to produce a desired effect on its input? Does it really matter if a system is using a simulated model of the environment to predict its input before calculating its output? Â

All control systems have reference values for their inputs. There is no system that only controls its output. Â

All control systems are doing “perceptualâ€? control. Who does not agree with this proposition?Â

RM: RM: I think you are asking whether everyone agrees that all control systems (properly designed closed- loop negative feedback systems) even those that include model-based calculations of inverse kinematics, are doing perceptual control. I can't speak for everyone but I certainly agree with that. And I think it's a very good observation.Â

Â

RM: This fact about how control systems work is not particularly important for engineers who build such systems to understand since the engineers know which variables they want their control systems to control and they develop perceptual systems (sensors) that provide analogs of these variables for the system to control. But it is very important for students of the controlling done by living systems to understand since figuring out what perceptions these systems are controlling is essential to understanding their behavior.Â

Â

Best

Â

Rick

Â

--

Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery

Â

Â

Â

--
Richard S. MarkenÂ
"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery

[MartinT]

[Martin Taylor 2018.02.20.13.12]

Allow me to compliment and complement Eetu's comment (pace [Rick

Marken 2018-02-20_09:52:49][Rick Marken 2018-02-20_09:52:49].

It is true that for a single isolated control loop, if the internal

component of the loop has a perfect model of the way the
environmental part works, and has sufficient computing resources,
then no external test could tell the difference between them. The
S-R version would have been trained with all possible applied
stimuli before the experiment, so that no matter what it perceived
at any moment in the rial, it would be able to produce the same
output as would the PCT control loop at that moment. The PCT loop,
on the other hand, would have needed no training at all, contains no
model of what “stimuli” it might encounter, nor what to output on
any occasion, other than to be provided with a leaky integrator as a
computing process to produce the output given the error.

The experimenter would, of course, have to be ignorant of the

different precursor training conditions for the two systems. But
given that ignorance, the experimenter would not know that the S-R
system had built an internal model that inverted the function
relating the output to the “stimulus”, whereas the control loop did
not need one.

Eetu pointed out how the introduction of the reference variable

differentiated between an S-R view and a control-loop view, since it
means that any well defined stimulus is followed by a statistically
varying response. The PCT researcher may recognize this possibility
and seek a second controlled variable. The S-R researcher has to
find a better way to describe the stimulus, adding a variable that
can be found in the environment rather than in the organism. Both
need something that was not in the original description of the
experiment, but the S-R researcher has the harder job.

A second, related version of this is the PCT mantra: "Many means to

the same end." With more than a single level of control, a
higher-level system can often (always?) control its perceptual value
by altering the balance between the lower-level control units that
support it, possibly using one means to the exclusion of another.
“Clever Hans” is an example from a long time ago, after which
experimenters in vision tried to make sure that no sounds
accompanied their changes of “stimulus”.

An anecdote that I have used before, to illustrate the problem. A

colleague of mine described his participation in a simple tracking
study when he had been a first-year undergraduate. The experimenter
asked him to keep the point of a stylus on a small conductive disk
set into a larger rotating disk like an old-fashioned gramophone
record. My colleague perceived this to be a study of intelligence,
and dismounted the large “gramophone record” disk so that he could
simply rest the stylus on the small conductive disk. He followed the
instructions he was given, but the experimenter (who could not see
the trial in progress) was first astonished by the perfection of his
“tracking” performance, and then angry that he “cheated”.

I think that an S-R psychologist would have a problem finding the

environmental variable that was the part of the “stimulus” that
caused my colleague’s output to differ so much from that of the
other subjects. PCT has no problem in finding a perceptual variable
that might have been controlled by my colleague so that it changed
reference values down the hierarchy to cause the output that removed
the large disk from its mounting rather than causing outputs that
continued to move the stylus as the large disk rotated (as the
experimenter intended).

I guess the fundamental answer is that you are correct that an S-R

system can always be invented that will mimic the externally
detectable results of any control system, however complex. But even
the most naive version of Ockham’s Razor: “the simplest theory
wins,” the Razor comes down very heavily in favour of hierarchic
control.

Martin

···

    I understand that PCT explains how S-R psychology

works. But these two theories involve what is essentially the
same constitutive equation: response is proportional to
stimulus. In order to constitute a different theory, this
constitutive equation would need to be different. Would you
disagree?   Â

      On Tue, Feb 20, 2018 at 1:06 AM, Eetu

Pikkarainen eetu.pikkarainen@oulu.fi
wrote:

                [Eetu Pikkarainen

2018-02-20_08:45:11 UTC]

Â

                Sorry, I am not

a psychologist so I hope that someone will correct
and complement my short answer.

                S-R psychology

makes predictions about subject’s reactions based on
the current stimulus and previous knowledge. Usually
the success of these predictions are quite low.

                PCT explains why

the correlation between stimulus and response must
be slow: there is always an important intervening
variable, namely the reference value, which
determines how the stimulus will affect the
response. So PCT research does not even try to
predict reactions from stimuluses. Instead it tries
to find out what perception the subject is
controlling (the Test for controlled variable). If
it can be found then it is possible to predict with
high accuracy how the subject responds by her output
(“reaction�) to the disturbances (“stimulus�)
against the perceived value of the controlled
variable.

                Perhaps the

example in Powers: Behavior: The Control of
Perception (2005) about the analysis of a typical
rat experiment in PCT terms could be lightening
about the difference, pages 238-243.

                An  extensive

description about PCT type of research in relation
to more typical psychological research is in Runkel,
P. (1990). Casting Nets and Testing Specimens: Two
Grand Methods of Psychology.

Â

Eetu

Â

From: PHILIP JERAIR
YERANOSIAN [mailto:pyeranos@ucla.edu ]
Sent: Tuesday, February 20, 2018 3:43 AM
To: csgnet@lists.illinois.edu
Subject: Re: General

Â

                    That is a very strange

example. Â

                      I have read some PCT

literature. One remark is that PCT doesn’t
employ the same model as stimulus-response
psychology. How do the predictions of a
stimulus-response theory of behavior differ
from PCT predictions? Â

Â

                      On Mon, Feb 19, 2018 at

11:01 AM, Bruce Nevin <bnhpct@gmail.com >
wrote:

                          [Bruce Nevin

2018-02-19_14:00:13 ET]

Â

                            PHILIP JERAIR

YERANOSIANÂ (Sun, Feb 18, 2018 at 3:28
PM) –

Â

                              Philip, we do

model-based control without
calculating inverse kinematics. When
you walk through a familiar room in
the dark or with your eyes closed, the
model is made of perceptions restored
from memory and controlled in
imagination. Controlling in
imagination to avoid the chair that is
there  and the table that is
over
there  accompanies ordinary
perceptual control processes by which
you walk over to the light switch by
the door
there  and turn it on. In the
final stages of controlling a
perception of having light in the room
by means of flipping that switch no
amount of prediction is going to
guarantee that your fingers find the
switch reliably every time, but
repeated practice will sharpen the
precision of your memory, not to
mention your confidence in your
imagination.Â

Â

                              I will leave it to

others to refer you to the abundant
literature laying out predictions that
PCT makes that are different from
those of other theories. And I have no
choice but to leave it up to you to
read that literature.

Â

/Bruce

Â

                                On Sun, Feb 18,

2018 at 3:28 PM, PHILIP JERAIR
YERANOSIAN <pyeranos@ucla.edu >
wrote:

                                    The purpose

of my question is to figure out
why PCT is to be developed as a
scientific discipline in its own
right.Â

                                      What

predictions does PCT make that
are different from those of
other theories?   Â

Â

                                              Virus-free.

www.avast.com

Â

                                          On Sun,

Jan 28, 2018 at 9:56 AM,
Richard Marken <rsmarken@gmail.com >
wrote:

                                              [Rick

Marken
2018-01-28_09:55:50]

Â

                                                  On

Sat, Jan 27, 2018
at 9:36 PM, PHILIP
JERAIR YERANOSIAN
pyeranos@ucla.edu wrote:

Â

                                                      PY:

If a system
uses inverse
kinematics to
calculate the
effects of its
output, is it
not ultimately
doing so in
order for its
actions to
produce a
desired effect
on its input?
Does it really
matter if a
system is
using a
simulated
model of the
environment to
predict its
input before
calculating
its output? Â

                                                      All

control
systems have
reference
values for
their inputs.
There is no
system that
only controls
its output. Â

                                                      All

control
systems are
doing
“perceptual�
control. Who
does not agree
with this
proposition?Â

                                                RM:                                                       RM:

I think you are
asking whether
everyone agrees
that all control
systems (properly
designed closed-
loop negative
feedback systems)
even those that
include
model-based
calculations of
inverse
kinematics, are
doing perceptual
control. I can’t
speak for everyone
but I certainly
agree with that.
And I think it’s a
very good
observation.Â

Â

                                                RM:

This fact about how
control systems work
is not particularly
important for
engineers who build
such systems to
understand since the
engineers know which
variables they want
their control
systems to control
and they develop
perceptual systems
(sensors) that
provide analogs of
these variables for
the system to
control. But it is
very important for
students of the
controlling done by
living systems to
understand since
figuring out what
perceptions these
systems are
controlling is
essential to
understanding their
behavior.Â

Â

Best

Â

Rick

Â

                                                  --

Richard S. MarkenÂ

                                                      "Perfection

is achieved
not when you
have nothing
more to add,
but when you

                                                      have nothing

left to take
away.�

                                                      Â  Â  Â  Â  Â  Â  Â 

      Â
  --Antoine
de
Saint-Exupery

Â

Â

Â

[PHILIP_JERAIR_YERANO]

Yes Eetu, your response is fair. I will repost my next message for convenience.
Â

I understand that PCT explains how S-R psychology works. But these two theories involve what is essentially the same constitutive equation: response is proportional to stimulus. In order to constitute a different theory, this constitutive equation would need to be different. Would you disagree?Â

···

On Tue, Feb 20, 2018 at 9:52 AM, Richard Marken rsmarken@gmail.com wrote:

[Rick Marken 2018-02-20_09:52:49]

Eetu Pikkarainen (2018-02-20_08:45:11 UTC)

Â

EP: Sorry, I am not a psychologist so I hope that someone will correct and complement my short answer.

RM: I see nothing to correct and only things to compliment in this post. I compliment your excellent description of what PCT explains that S-R psychology doesn’t. And I also compliment your excellent command of English as revealed in your understanding of the distinction between complementing and complimenting.Â

Best

Rick

Â

S-R psychology makes predictions about subject’s reactions based on the current stimulus and previous knowledge. Usually the success of these predictions are quite low.

PCT explains why the correlation between stimulus and response must be slow: there is always an important intervening variable, namely the reference value, which determines how the stimulus
will affect the response. So PCT research does not even try to predict reactions from stimuluses. Instead it tries to find out what perception the subject is controlling (the Test for controlled variable). If it can be found then it is possible to predict
with high accuracy how the subject responds by her output (“reaction�) to the disturbances (“stimulus�) against the perceived value of the controlled variable.

Perhaps the example in Powers: Behavior: The Control of Perception (2005) about the analysis of a typical rat experiment in PCT terms could be lightening about the difference, pages
238-243.

An extensive description about PCT type of research in relation to more typical psychological research is in Runkel, P. (1990). Casting Nets and Testing Specimens: Two Grand Methods
of Psychology.

Â

Eetu

Â

From: PHILIP JERAIR YERANOSIAN [mailto:pyeranos@ucla.edu]
Sent: Tuesday, February 20, 2018 3:43 AM
To: csgnet@lists.illinois.edu
Subject: Re: General

Â

That is a very strange example. Â

I have read some PCT literature. One remark is that PCT doesn’t employ the same model as stimulus-response psychology. How do the predictions of a stimulus-response theory of behavior differ from PCT predictions? Â

Â

On Mon, Feb 19, 2018 at 11:01 AM, Bruce Nevin bnhpct@gmail.com wrote:

[Bruce Nevin 2018-02-19_14:00:13 ET]

Â

PHILIP JERAIR YERANOSIANÂ (Sun, Feb 18, 2018 at 3:28 PM) –

Â

Philip, we do model-based control without calculating inverse kinematics. When you walk through a familiar room in the dark or with your eyes closed, the model is made of perceptions restored from memory and controlled in imagination. Controlling
in imagination to avoid the chair that is there and the table that is over
there accompanies ordinary perceptual control processes by which you walk over to the light switch by the door
there  and turn it on. In the final stages of controlling a perception of having light in the room by means of flipping that switch no amount of prediction is going to guarantee that your fingers find the switch reliably every time, but repeated practice
will sharpen the precision of your memory, not to mention your confidence in your imagination.Â

Â

I will leave it to others to refer you to the abundant literature laying out predictions that PCT makes that are different from those of other theories. And I have no choice but to leave it up to you to read that literature.

Â

/Bruce

Â

On Sun, Feb 18, 2018 at 3:28 PM, PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu wrote:

The purpose of my question is to figure out why PCT is to be developed as a scientific discipline in its own right.Â

What predictions does PCT make that are different from those of other theories?   Â

Â

Virus-free.

www.avast.com

Â

On Sun, Jan 28, 2018 at 9:56 AM, Richard Marken rsmarken@gmail.com wrote:

[Rick Marken 2018-01-28_09:55:50]

Â

On Sat, Jan 27, 2018 at 9:36 PM, PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu wrote:

Â

PY: If a system uses inverse kinematics to calculate the effects of its output, is it not ultimately doing so in order for its actions to produce a desired effect on its input? Does it really matter if a system is using a simulated model
of the environment to predict its input before calculating its output? Â

All control systems have reference values for their inputs. There is no system that only controls its output. Â

All control systems are doing “perceptualâ€? control. Who does not agree with this proposition?Â

RM:
RM: I think you are asking whether everyone agrees that all control systems (properly designed closed- loop negative feedback systems) even those that include model-based calculations of inverse kinematics, are doing perceptual control. I can’t speak for everyone
but I certainly agree with that. And I think it’s a very good observation.Â

Â

RM: This fact about how control systems work is not particularly important for engineers who build such systems to understand since the engineers know which variables they want their control systems to control and they develop perceptual
systems (sensors) that provide analogs of these variables for the system to control. But it is very important for students of the controlling done by living systems to understand since figuring out what perceptions these systems are controlling is essential
to understanding their behavior.Â

Â

Best

Â

Rick

Â

–

Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you

have nothing left to take away.�

                --Antoine de Saint-Exupery

Â

Â

Â

–
Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery

[FredNickols]

[From Fred Nickols (2018.02.20.1419 ET)]

Regarding the experiment you describe below, Martin, it seems clear to me that the more intelligent of the two was the subject, not the experimenter. Do you suppose the experimenter learned anything from the experiment? I always thought experiments were intended to find out how things would turn out, not ensure that they turn out as planned.

Fred Nickols

···

From: Martin Taylor [mailto:mmt-csg@mmtaylor.net]
Sent: Tuesday, February 20, 2018 1:49 PM
To: csgnet@lists.illinois.edu
Subject: Re: General

[Martin Taylor 2018.02.20.13.12]

Allow me to compliment and complement Eetu’s comment (pace [Rick Marken 2018-02-20_09:52:49][Rick Marken 2018-02-20_09:52:49].

It is true that for a single isolated control loop, if the internal component of the loop has a perfect model of the way the environmental part works, and has sufficient computing resources, then no external test could tell the difference between them. The S-R version would have been trained with all possible applied stimuli before the experiment, so that no matter what it perceived at any moment in the rial, it would be able to produce the same output as would the PCT control loop at that moment. The PCT loop, on the other hand, would have needed no training at all, contains no model of what “stimuli” it might encounter, nor what to output on any occasion, other than to be provided with a leaky integrator as a computing process to produce the output given the error.

The experimenter would, of course, have to be ignorant of the different precursor training conditions for the two systems. But given that ignorance, the experimenter would not know that the S-R system had built an internal model that inverted the function relating the output to the “stimulus”, whereas the control loop did not need one.

Eetu pointed out how the introduction of the reference variable differentiated between an S-R view and a control-loop view, since it means that any well defined stimulus is followed by a statistically varying response. The PCT researcher may recognize this possibility and seek a second controlled variable. The S-R researcher has to find a better way to describe the stimulus, adding a variable that can be found in the environment rather than in the organism. Both need something that was not in the original description of the experiment, but the S-R researcher has the harder job.

A second, related version of this is the PCT mantra: “Many means to the same end.” With more than a single level of control, a higher-level system can often (always?) control its perceptual value by altering the balance between the lower-level control units that support it, possibly using one means to the exclusion of another. “Clever Hans” is an example from a long time ago, after which experimenters in vision tried to make sure that no sounds accompanied their changes of “stimulus”.

An anecdote that I have used before, to illustrate the problem. A colleague of mine described his participation in a simple tracking study when he had been a first-year undergraduate. The experimenter asked him to keep the point of a stylus on a small conductive disk set into a larger rotating disk like an old-fashioned gramophone record. My colleague perceived this to be a study of intelligence, and dismounted the large “gramophone record” disk so that he could simply rest the stylus on the small conductive disk. He followed the instructions he was given, but the experimenter (who could not see the trial in progress) was first astonished by the perfection of his “tracking” performance, and then angry that he “cheated”.

I think that an S-R psychologist would have a problem finding the environmental variable that was the part of the “stimulus” that caused my colleague’s output to differ so much from that of the other subjects. PCT has no problem in finding a perceptual variable that might have been controlled by my colleague so that it changed reference values down the hierarchy to cause the output that removed the large disk from its mounting rather than causing outputs that continued to move the stylus as the large disk rotated (as the experimenter intended).

I guess the fundamental answer is that you are correct that an S-R system can always be invented that will mimic the externally detectable results of any control system, however complex. But even the most naive version of Ockham’s Razor: “the simplest theory wins,” the Razor comes down very heavily in favour of hierarchic control.

Martin

I understand that PCT explains how S-R psychology works. But these two theories involve what is essentially the same constitutive equation: response is proportional to stimulus. In order to constitute a different theory, this constitutive equation would need to be different. Would you disagree?

On Tue, Feb 20, 2018 at 1:06 AM, Eetu Pikkarainen eetu.pikkarainen@oulu.fi wrote:

[Eetu Pikkarainen 2018-02-20_08:45:11 UTC]

Sorry, I am not a psychologist so I hope that someone will correct and complement my short answer.

S-R psychology makes predictions about subject’s reactions based on the current stimulus and previous knowledge. Usually the success of these predictions are quite low.

PCT explains why the correlation between stimulus and response must be slow: there is always an important intervening variable, namely the reference value, which determines how the stimulus will affect the response. So PCT research does not even try to predict reactions from stimuluses. Instead it tries to find out what perception the subject is controlling (the Test for controlled variable). If it can be found then it is possible to predict with high accuracy how the subject responds by her output (“reaction�) to the disturbances (“stimulus�) against the perceived value of the controlled variable.

Perhaps the example in Powers: Behavior: The Control of Perception (2005) about the analysis of a typical rat experiment in PCT terms could be lightening about the difference, pages 238-243.

An extensive description about PCT type of research in relation to more typical psychological research is in Runkel, P. (1990). Casting Nets and Testing Specimens: Two Grand Methods of Psychology.

Eetu

From: PHILIP JERAIR YERANOSIAN [mailto:pyeranos@ucla.edu]
Sent: Tuesday, February 20, 2018 3:43 AM
To: csgnet@lists.illinois.edu
Subject: Re: General

That is a very strange example.

I have read some PCT literature. One remark is that PCT doesn’t employ the same model as stimulus-response psychology. How do the predictions of a stimulus-response theory of behavior differ from PCT predictions?

On Mon, Feb 19, 2018 at 11:01 AM, Bruce Nevin bnhpct@gmail.com wrote:

[Bruce Nevin 2018-02-19_14:00:13 ET]

PHILIP JERAIR YERANOSIAN (Sun, Feb 18, 2018 at 3:28 PM) –

Philip, we do model-based control without calculating inverse kinematics. When you walk through a familiar room in the dark or with your eyes closed, the model is made of perceptions restored from memory and controlled in imagination. Controlling in imagination to avoid the chair that is there and the table that is over there accompanies ordinary perceptual control processes by which you walk over to the light switch by the door there and turn it on. In the final stages of controlling a perception of having light in the room by means of flipping that switch no amount of prediction is going to guarantee that your fingers find the switch reliably every time, but repeated practice will sharpen the precision of your memory, not to mention your confidence in your imagination.

I will leave it to others to refer you to the abundant literature laying out predictions that PCT makes that are different from those of other theories. And I have no choice but to leave it up to you to read that literature.

/Bruce

On Sun, Feb 18, 2018 at 3:28 PM, PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu wrote:

The purpose of my question is to figure out why PCT is to be developed as a scientific discipline in its own right.

What predictions does PCT make that are different from those of other theories?

Virus-free. www.avast.com

On Sun, Jan 28, 2018 at 9:56 AM, Richard Marken rsmarken@gmail.com wrote:

[Rick Marken 2018-01-28_09:55:50]

On Sat, Jan 27, 2018 at 9:36 PM, PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu wrote:

PY: If a system uses inverse kinematics to calculate the effects of its output, is it not ultimately doing so in order for its actions to produce a desired effect on its input? Does it really matter if a system is using a simulated model of the environment to predict its input before calculating its output?

All control systems have reference values for their inputs. There is no system that only controls its output.

All control systems are doing “perceptual� control. Who does not agree with this proposition?

RM: RM: I think you are asking whether everyone agrees that all control systems (properly designed closed- loop negative feedback systems) even those that include model-based calculations of inverse kinematics, are doing perceptual control. I can’t speak for everyone but I certainly agree with that. And I think it’s a very good observation.

RM: This fact about how control systems work is not particularly important for engineers who build such systems to understand since the engineers know which variables they want their control systems to control and they develop perceptual systems (sensors) that provide analogs of these variables for the system to control. But it is very important for students of the controlling done by living systems to understand since figuring out what perceptions these systems are controlling is essential to understanding their behavior.

Best

Rick

–

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

[MartinT]

[From Fred
Nickols (2018.02.20.1419 ET)]

        Regarding

the experiment you describe below, Martin, it seems clear to
me that the more intelligent of the two was the subject, not
the experimenter. Do you suppose the experimenter learned
anything from the experiment? I always thought experiments
were intended to find out how things would turn out, not
ensure that they turn out as planned.

Fred Nickols

···

From: Martin Taylor
[mailto:mmt-csg@mmtaylor.net]
Sent: Tuesday, February 20, 2018 1:49 PM
To: csgnet@lists.illinois.edu
Subject: Re: General

[Martin Taylor 2018.02.20.13.12]

      ...n anecdote that

I have used before, to illustrate the problem. A colleague of
mine described his participation in a simple tracking study
when he had been a first-year undergraduate. The experimenter
asked him to keep the point of a stylus on a small conductive
disk set into a larger rotating disk like an old-fashioned
gramophone record. My colleague perceived this to be a study
of intelligence, and dismounted the large “gramophone record”
disk so that he could simply rest the stylus on the small
conductive disk. He followed the instructions he was given,
but the experimenter (who could not see the trial in progress)
was first astonished by the perfection of his “tracking”
performance, and then angry that he “cheated”.

[EetuP]

[eetu.pikkarainen@oulu.fi 21.2.2018]

Philip, that constitutive equation is really different generally in these approaches. It is similar in that special case when the reference is stable (and also the environmental feedback chain is stable). But if the reference changes then the output or reaction
will change indepedently from stimulus. Also that “many ways to same end” phenomenon described by Martin can cause independent (surprising) reactions. Thirdly the changes in environmental feedback function can cause independent changes in reaction as described
in Rick’s demo in a recent message (so called “behavioral illusion”).

So I think that s-r theory is a narrow partial description or model of a certain special case of action or behavior of a living organism and thus it could sometimes predict quite correctly. It is not unnatural if an organism sometimes behaves like an automaton
or even like an inanimate object and gives reactions which are proportional to stimuluses. But that is not the general and typical case.

If your goal is to keep the cursor in a point x and a disturbance or stimulus moves it some distance away then isn’t it quite natural that you try to move it backwards just the same distance to keep it in x. Here the reaction is strictly proportional to stimulus.
But if you decide that now the cursor should not anymore stay in x but in y somewhere else then there is no proportionality any more. PCT researcher can understand that latter situation but even she cannot predict it.

(I feel this reply still misses something but this role of reference i.e. goal of action is anyway the core point of my understanding of PCT.)

Rick, for your disappointment I must tell that I searched the word complemet from dictionary and had no idea that is so dangerously near to compliment

Eetu

···

---

From: PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu
Sent: Tuesday, February 20, 2018 9:00:00 PM
To: csgnet@lists.illinois.edu
Subject: Re: General

Yes Eetu, your response is fair. I will repost my next message for convenience.

I
understand that PCT explains how S-R psychology works. But these two theories involve what is essentially the same constitutive equation: response is proportional to stimulus. In order to constitute a different theory, this constitutive equation would need
to be different. Would you disagree?

On Tue, Feb 20, 2018 at 9:52 AM, Richard Marken
rsmarken@gmail.com wrote:

[Rick Marken 2018-02-20_09:52:49]

Eetu Pikkarainen (2018-02-20_08:45:11 UTC)

EP: Sorry, I am not a psychologist so I hope that someone will correct and complement my short answer.

RM: I see nothing to correct and only things to compliment in this post. I compliment your excellent description of what PCT explains that S-R psychology doesn’t. And I also compliment your excellent command of English as revealed in your understanding
of the distinction between complementing and complimenting.

Best

Rick

S-R psychology makes predictions about subject’s reactions based on the current stimulus and previous knowledge. Usually the success of these predictions are quite low.

PCT explains why the correlation between stimulus and response must be slow: there is always an important intervening variable, namely the reference value, which determines how the stimulus will affect the response. So
PCT research does not even try to predict reactions from stimuluses. Instead it tries to find out what perception the subject is controlling (the Test for controlled variable). If it can be found then it is possible to predict with high accuracy how the subject
responds by her output (“reaction”) to the disturbances (“stimulus”) against the perceived value of the controlled variable.

Perhaps the example in Powers: Behavior: The Control of Perception (2005) about the analysis of a typical rat experiment in PCT terms could be lightening about the difference, pages 238-243.

An extensive description about PCT type of research in relation to more typical psychological research is in Runkel, P. (1990). Casting Nets and Testing Specimens: Two Grand Methods of Psychology.

Eetu

From: PHILIP JERAIR YERANOSIAN [mailto:pyeranos@ucla.edu]
Sent: Tuesday, February 20, 2018 3:43 AM
To: csgnet@lists.illinois.edu
Subject: Re: General

That is a very strange example.

I have read some PCT literature. One remark is that PCT doesn’t employ the same model as stimulus-response psychology. How do the predictions of a stimulus-response theory of behavior differ from PCT predictions?

On Mon, Feb 19, 2018 at 11:01 AM, Bruce Nevin bnhpct@gmail.com wrote:

[Bruce Nevin 2018-02-19_14:00:13 ET]

PHILIP JERAIR YERANOSIAN (Sun, Feb 18, 2018 at 3:28 PM) –

Philip, we do model-based control without calculating inverse kinematics. When you walk through a familiar room in the dark or with your eyes closed, the model is made of perceptions restored from memory and controlled in imagination. Controlling
in imagination to avoid the chair that is there and the table that is over
there accompanies ordinary perceptual control processes by which you walk over to the light switch by the door
there and turn it on. In the final stages of controlling a perception of having light in the room by means of flipping that switch no amount of prediction is going to guarantee that your fingers find the switch reliably every time, but repeated practice
will sharpen the precision of your memory, not to mention your confidence in your imagination.

I will leave it to others to refer you to the abundant literature laying out predictions that PCT makes that are different from those of other theories. And I have no choice but to leave it up to you to read that literature.

/Bruce

On Sun, Feb 18, 2018 at 3:28 PM, PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu wrote:

The purpose of my question is to figure out why PCT is to be developed as a scientific discipline in its own right.

What predictions does PCT make that are different from those of other theories?

Virus-free.

www.avast.com

On Sun, Jan 28, 2018 at 9:56 AM, Richard Marken rsmarken@gmail.com wrote:

[Rick Marken 2018-01-28_09:55:50]

On Sat, Jan 27, 2018 at 9:36 PM, PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu wrote:

PY: If a system uses inverse kinematics to calculate the effects of its output, is it not ultimately doing so in order for its actions to produce a desired effect on its input? Does it really matter if a system is using a simulated model
of the environment to predict its input before calculating its output?

All control systems have reference values for their inputs. There is no system that only controls its output.

All control systems are doing “perceptual” control. Who does not agree with this proposition?

RM:
RM: I think you are asking whether everyone agrees that all control systems (properly designed closed- loop negative feedback systems) even those that include model-based calculations of inverse kinematics, are doing perceptual control. I can’t speak for everyone
but I certainly agree with that. And I think it’s a very good observation.

RM: This fact about how control systems work is not particularly important for engineers who build such systems to understand since the engineers know which variables they want their control systems to control and they develop perceptual
systems (sensors) that provide analogs of these variables for the system to control. But it is very important for students of the controlling done by living systems to understand since figuring out what perceptions these systems are controlling is essential
to understanding their behavior.

Best

Rick

–

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

–
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

[rsmarken]

[Rick Marken 2018-02-20_17:50:31]

···

[eetu.pikkarainen@oulu.fi 21.2.2018]

Â

EP: (I feel this reply still misses something but this role of reference i.e. goal of action is anyway the core point of my understanding of PCT.)

RM: You are missing nothing. The goal of action, which is the (possibly varying) reference state of the controlled variable, is what S-R (or cognitive or equilibrium ot any cause-effect theory of behavior) cannot explain.

Â

EP: Rick, for your disappointment I must tell that I searched the word complemet from dictionary and had no idea that is so dangerously near to compliment

RM: Nope, not disappointed. Still very impressed!

BestÂ

Rick

Â

Eetu

---

From: PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu
Sent: Tuesday, February 20, 2018 9:00:00 PM

To: csgnet@lists.illinois.edu
Subject: Re: General

Â

Yes Eetu, your response is fair. I will repost my next message for convenience.

Â

I
understand that PCT explains how S-R psychology works. But these two theories involve what is essentially the same constitutive equation: response is proportional to stimulus. In order to constitute a different theory, this constitutive equation would need
to be different. Would you disagree?Â

–

On Tue, Feb 20, 2018 at 9:52 AM, Richard Marken
rsmarken@gmail.com wrote:

[Rick Marken 2018-02-20_09:52:49]

Eetu Pikkarainen (2018-02-20_08:45:11 UTC)

Â

EP: Sorry, I am not a psychologist so I hope that someone will correct and complement my short answer.

RM: I see nothing to correct and only things to compliment in this post. I compliment your excellent description of what PCT explains that S-R psychology doesn’t. And I also compliment your excellent command of English as revealed in your understanding
of the distinction between complementing and complimenting.Â

Best

Rick

Â

S-R psychology makes predictions about subject’s reactions based on the current stimulus and previous knowledge. Usually the success of these predictions are quite low.

PCT explains why the correlation between stimulus and response must be slow: there is always an important intervening variable, namely the reference value, which determines how the stimulus will affect the response. So
PCT research does not even try to predict reactions from stimuluses. Instead it tries to find out what perception the subject is controlling (the Test for controlled variable). If it can be found then it is possible to predict with high accuracy how the subject
responds by her output (“reaction�) to the disturbances (“stimulus�) against the perceived value of the controlled variable.

Perhaps the example in Powers: Behavior: The Control of Perception (2005) about the analysis of a typical rat experiment in PCT terms could be lightening about the difference, pages 238-243.

An extensive description about PCT type of research in relation to more typical psychological research is in Runkel, P. (1990). Casting Nets and Testing Specimens: Two Grand Methods of Psychology.

Â

Eetu

Â

From: PHILIP JERAIR YERANOSIAN [mailto:pyeranos@ucla.edu]
Sent: Tuesday, February 20, 2018 3:43 AM
To: csgnet@lists.illinois.edu
Subject: Re: General

Â

That is a very strange example. Â

I have read some PCT literature. One remark is that PCT doesn’t employ the same model as stimulus-response psychology. How do the predictions of a stimulus-response theory of behavior differ from PCT predictions? Â

Â

On Mon, Feb 19, 2018 at 11:01 AM, Bruce Nevin bnhpct@gmail.com wrote:

[Bruce Nevin 2018-02-19_14:00:13 ET]

Â

PHILIP JERAIR YERANOSIANÂ (Sun, Feb 18, 2018 at 3:28 PM) –

Â

Philip, we do model-based control without calculating inverse kinematics. When you walk through a familiar room in the dark or with your eyes closed, the model is made of perceptions restored from memory and controlled in imagination. Controlling
in imagination to avoid the chair that is there and the table that is over
there accompanies ordinary perceptual control processes by which you walk over to the light switch by the door
there  and turn it on. In the final stages of controlling a perception of having light in the room by means of flipping that switch no amount of prediction is going to guarantee that your fingers find the switch reliably every time, but repeated practice
will sharpen the precision of your memory, not to mention your confidence in your imagination.Â

Â

I will leave it to others to refer you to the abundant literature laying out predictions that PCT makes that are different from those of other theories. And I have no choice but to leave it up to you to read that literature.

Â

/Bruce

Â

On Sun, Feb 18, 2018 at 3:28 PM, PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu wrote:

The purpose of my question is to figure out why PCT is to be developed as a scientific discipline in its own right.Â

What predictions does PCT make that are different from those of other theories?   Â

Â

Virus-free.

www.avast.com

Â

On Sun, Jan 28, 2018 at 9:56 AM, Richard Marken rsmarken@gmail.com wrote:

[Rick Marken 2018-01-28_09:55:50]

Â

On Sat, Jan 27, 2018 at 9:36 PM, PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu wrote:

Â

PY: If a system uses inverse kinematics to calculate the effects of its output, is it not ultimately doing so in order for its actions to produce a desired effect on its input? Does it really matter if a system is using a simulated model
of the environment to predict its input before calculating its output? Â

All control systems have reference values for their inputs. There is no system that only controls its output. Â

All control systems are doing “perceptualâ€? control. Who does not agree with this proposition?Â

RM:
RM: I think you are asking whether everyone agrees that all control systems (properly designed closed- loop negative feedback systems) even those that include model-based calculations of inverse kinematics, are doing perceptual control. I can’t speak for everyone
but I certainly agree with that. And I think it’s a very good observation.Â

Â

RM: This fact about how control systems work is not particularly important for engineers who build such systems to understand since the engineers know which variables they want their control systems to control and they develop perceptual
systems (sensors) that provide analogs of these variables for the system to control. But it is very important for students of the controlling done by living systems to understand since figuring out what perceptions these systems are controlling is essential
to understanding their behavior.Â

Â

Best

Â

Rick

Â

–

Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you

have nothing left to take away.�

                --Antoine de Saint-Exupery

Â

Â

Â

–
Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery

Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery

[PHILIP_JERAIR_YERANO]

When scientists developed S-R psychology, they did not suppose that behavior did not have purpose. The premise of S-R psychology is that the purpose of behavior is not arbitrary. If this premise is false, then the theory is not valid. Do you disagree with the proposition that this same premise applies to PCT?

···

[eetu.pikkarainen@oulu.fi 21.2.2018]

Â

EP: (I feel this reply still misses something but this role of reference i.e. goal of action is anyway the core point of my understanding of PCT.)

RM: You are missing nothing. The goal of action, which is the (possibly varying) reference state of the controlled variable, is what S-R (or cognitive or equilibrium ot any cause-effect theory of behavior) cannot explain.

Â

EP: Rick, for your disappointment I must tell that I searched the word complemet from dictionary and had no idea that is so dangerously near to compliment

RM: Nope, not disappointed. Still very impressed!

BestÂ

Rick

Â

Eetu

---

From: PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu
Sent: Tuesday, February 20, 2018 9:00:00 PM

To: csgnet@lists.illinois.edu
Subject: Re: General

Â

Yes Eetu, your response is fair. I will repost my next message for convenience.

Â

I
understand that PCT explains how S-R psychology works. But these two theories involve what is essentially the same constitutive equation: response is proportional to stimulus. In order to constitute a different theory, this constitutive equation would need
to be different. Would you disagree?Â

–
Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery

On Tue, Feb 20, 2018 at 9:52 AM, Richard Marken
rsmarken@gmail.com wrote:

[Rick Marken 2018-02-20_09:52:49]

Eetu Pikkarainen (2018-02-20_08:45:11 UTC)

Â

EP: Sorry, I am not a psychologist so I hope that someone will correct and complement my short answer.

RM: I see nothing to correct and only things to compliment in this post. I compliment your excellent description of what PCT explains that S-R psychology doesn’t. And I also compliment your excellent command of English as revealed in your understanding
of the distinction between complementing and complimenting.Â

Best

Rick

Â

S-R psychology makes predictions about subject’s reactions based on the current stimulus and previous knowledge. Usually the success of these predictions are quite low.

PCT explains why the correlation between stimulus and response must be slow: there is always an important intervening variable, namely the reference value, which determines how the stimulus will affect the response. So
PCT research does not even try to predict reactions from stimuluses. Instead it tries to find out what perception the subject is controlling (the Test for controlled variable). If it can be found then it is possible to predict with high accuracy how the subject
responds by her output (“reaction�) to the disturbances (“stimulus�) against the perceived value of the controlled variable.

Perhaps the example in Powers: Behavior: The Control of Perception (2005) about the analysis of a typical rat experiment in PCT terms could be lightening about the difference, pages 238-243.

An extensive description about PCT type of research in relation to more typical psychological research is in Runkel, P. (1990). Casting Nets and Testing Specimens: Two Grand Methods of Psychology.

Â

Eetu

Â

From: PHILIP JERAIR YERANOSIAN [mailto:pyeranos@ucla.edu]
Sent: Tuesday, February 20, 2018 3:43 AM
To: csgnet@lists.illinois.edu
Subject: Re: General

Â

That is a very strange example. Â

I have read some PCT literature. One remark is that PCT doesn’t employ the same model as stimulus-response psychology. How do the predictions of a stimulus-response theory of behavior differ from PCT predictions? Â

Â

On Mon, Feb 19, 2018 at 11:01 AM, Bruce Nevin bnhpct@gmail.com wrote:

[Bruce Nevin 2018-02-19_14:00:13 ET]

Â

PHILIP JERAIR YERANOSIANÂ (Sun, Feb 18, 2018 at 3:28 PM) –

Â

Philip, we do model-based control without calculating inverse kinematics. When you walk through a familiar room in the dark or with your eyes closed, the model is made of perceptions restored from memory and controlled in imagination. Controlling
in imagination to avoid the chair that is there and the table that is over
there accompanies ordinary perceptual control processes by which you walk over to the light switch by the door
there  and turn it on. In the final stages of controlling a perception of having light in the room by means of flipping that switch no amount of prediction is going to guarantee that your fingers find the switch reliably every time, but repeated practice
will sharpen the precision of your memory, not to mention your confidence in your imagination.Â

Â

I will leave it to others to refer you to the abundant literature laying out predictions that PCT makes that are different from those of other theories. And I have no choice but to leave it up to you to read that literature.

Â

/Bruce

Â

On Sun, Feb 18, 2018 at 3:28 PM, PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu wrote:

The purpose of my question is to figure out why PCT is to be developed as a scientific discipline in its own right.Â

What predictions does PCT make that are different from those of other theories?   Â

Â

Virus-free.

www.avast.com

Â

On Sun, Jan 28, 2018 at 9:56 AM, Richard Marken rsmarken@gmail.com wrote:

[Rick Marken 2018-01-28_09:55:50]

Â

On Sat, Jan 27, 2018 at 9:36 PM, PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu wrote:

Â

PY: If a system uses inverse kinematics to calculate the effects of its output, is it not ultimately doing so in order for its actions to produce a desired effect on its input? Does it really matter if a system is using a simulated model
of the environment to predict its input before calculating its output? Â

All control systems have reference values for their inputs. There is no system that only controls its output. Â

All control systems are doing “perceptualâ€? control. Who does not agree with this proposition?Â

RM:
RM: I think you are asking whether everyone agrees that all control systems (properly designed closed- loop negative feedback systems) even those that include model-based calculations of inverse kinematics, are doing perceptual control. I can’t speak for everyone
but I certainly agree with that. And I think it’s a very good observation.Â

Â

RM: This fact about how control systems work is not particularly important for engineers who build such systems to understand since the engineers know which variables they want their control systems to control and they develop perceptual
systems (sensors) that provide analogs of these variables for the system to control. But it is very important for students of the controlling done by living systems to understand since figuring out what perceptions these systems are controlling is essential
to understanding their behavior.Â

Â

Best

Â

Rick

Â

–

Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you

have nothing left to take away.�

                --Antoine de Saint-Exupery

Â

Â

Â

–
Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery

[EetuP]

[Eetu Pikkarainen 2018-02-21_07:39:44 UTC]

Philip,

When scientists developed S-R psychology, they did not suppose that behavior did not have purpose. The premise of S-R psychology is that the purpose of behavior is not arbitrary. If this premise
is false, then the theory is not valid. Do you disagree with the proposition that this same premise applies to PCT?

···

I wonder and disagree with your first sentences and would really like to hear some evidence for them. Watson as a leader figure of behaviorist movement wanted absolutely to get rid of concepts like soul, thinking, feelings
etc. and purpose which were thought to be unmeasurable and so unscientific. S-R theory is a theory about relationship between stimulus and response and nothing else. Later there was added some intervening variables and created so called S-O-R theory but these
organism variables should also be easily measured, not inferred or assumed like purpose. I think
Tolman was an exception to try
to use the concept of purpose.

Purpose was exorcised out of the theory, but of course not from the practice. The experiments and the concepts like conditioning or reinforcement necessarily require it, thus for me it seems that S-R theory is a big cheat.

Eetu

Please, regard all my statements as questions,

no matter how they are formulated.

On Tuesday, February 20, 2018, Richard Marken rsmarken@gmail.com wrote:

[Rick Marken 2018-02-20_17:50:31]

[eetu.pikkarainen@oulu.fi
21.2.2018]

EP: (I feel this reply still misses something but this role of reference i.e. goal of action is anyway the core point of my understanding of PCT.)

RM: You are missing nothing. The goal of action, which is the (possibly varying) reference state of the controlled variable, is what S-R (or cognitive or equilibrium ot any cause-effect theory of behavior) cannot explain.

EP: Rick, for your disappointment I must tell that I searched the word complemet from dictionary and had no idea that is so dangerously near to compliment

RM: Nope, not disappointed. Still very impressed!

Best

Rick

Eetu

---

From: PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu
Sent: Tuesday, February 20, 2018 9:00:00 PM

To: csgnet@lists.illinois.edu
Subject: Re: General

Yes Eetu, your response is fair. I will repost my next message for convenience.

I understand that PCT explains how S-R psychology works. But these two theories involve what is essentially the same constitutive equation: response
is proportional to stimulus. In order to constitute a different theory, this constitutive equation would need to be different. Would you disagree?

On Tue, Feb 20, 2018 at 9:52 AM, Richard Marken rsmarken@gmail.com wrote:

[Rick Marken 2018-02-20_09:52:49]

Eetu Pikkarainen (2018-02-20_08:45:11 UTC)

EP: Sorry, I am not a psychologist so I hope that someone will correct and complement my short answer.

RM: I see nothing to correct and only things to compliment in this post. I compliment your excellent description of what PCT explains that S-R psychology doesn’t. And I also compliment your excellent command of English
as revealed in your understanding of the distinction between complementing and complimenting.

Best

Rick

S-R psychology makes predictions about subject’s reactions based on the current stimulus and previous knowledge. Usually the success of these predictions are
quite low.

PCT explains why the correlation between stimulus and response must be slow: there is always an important intervening variable, namely the reference value, which
determines how the stimulus will affect the response. So PCT research does not even try to predict reactions from stimuluses. Instead it tries to find out what perception the subject is controlling (the Test for controlled variable). If it can be found then
it is possible to predict with high accuracy how the subject responds by her output (“reaction�) to the disturbances (“stimulus�) against the perceived value of the controlled variable.

Perhaps the example in Powers: Behavior: The Control of Perception (2005) about the analysis of a typical rat experiment in PCT terms could be lightening about
the difference, pages 238-243.

An extensive description about PCT type of research in relation to more typical psychological research is in Runkel, P. (1990). Casting Nets and Testing Specimens:
Two Grand Methods of Psychology.

Eetu

From: PHILIP JERAIR YERANOSIAN [mailto:pyeranos@ucla.edu]
Sent: Tuesday, February 20, 2018 3:43 AM
To: csgnet@lists.illinois.edu
Subject: Re: General

That is a very strange example.

I have read some PCT literature. One remark is that PCT doesn’t employ the same model as stimulus-response psychology. How do the predictions of a stimulus-response
theory of behavior differ from PCT predictions?

On Mon, Feb 19, 2018 at 11:01 AM, Bruce Nevin <bnhpct@gmail.com >
wrote:

[Bruce Nevin 2018-02-19_14:00:13 ET]

PHILIP JERAIR YERANOSIAN (Sun, Feb 18, 2018 at 3:28 PM) –

Philip, we do model-based control without calculating inverse kinematics. When you walk through a familiar room in the dark or with your eyes closed, the model
is made of perceptions restored from memory and controlled in imagination. Controlling in imagination to avoid the chair that is
there and the table that is over there accompanies ordinary perceptual control processes by which you walk over to the light switch by the door
there and turn it on. In the final stages of controlling a perception of having light in the room by means of flipping that switch no amount of prediction is going to guarantee that your fingers find the switch reliably every time, but repeated practice
will sharpen the precision of your memory, not to mention your confidence in your imagination.

I will leave it to others to refer you to the abundant literature laying out predictions that PCT makes that are different from those of other theories. And I
have no choice but to leave it up to you to read that literature.

/Bruce

On Sun, Feb 18, 2018 at 3:28 PM, PHILIP JERAIR YERANOSIAN <pyeranos@ucla.edu >
wrote:

The purpose of my question is to figure out why PCT is to be developed as a scientific discipline in its own right.

What predictions does PCT make that are different from those of other theories?

Virus-free.
www.avast.com

On Sun, Jan 28, 2018 at 9:56 AM, Richard Marken <rsmarken@gmail.com >
wrote:

[Rick Marken 2018-01-28_09:55:50]

On Sat, Jan 27, 2018 at 9:36 PM, PHILIP JERAIR YERANOSIAN <pyeranos@ucla.edu >
wrote:

PY: If a system uses inverse kinematics to calculate the effects of its output, is it not ultimately doing so in order for its actions to produce a desired effect
on its input? Does it really matter if a system is using a simulated model of the environment to predict its input before calculating its output?

All control systems have reference values for their inputs. There is no system that only controls its output.

All control systems are doing “perceptual� control. Who does not agree with this proposition?

RM: RM: I think you are asking whether everyone agrees that all control systems (properly designed closed- loop negative feedback systems) even
those that include model-based calculations of inverse kinematics, are doing perceptual control. I can’t speak for everyone but I certainly agree with that. And I think it’s a very good observation.

RM: This fact about how control systems work is not particularly important for engineers who build such systems to understand since the engineers know which variables
they want their control systems to control and they develop perceptual systems (sensors) that provide analogs of these variables for the system to control. But it is very important for students of the controlling done by living systems to understand since
figuring out what perceptions these systems are controlling is essential to understanding their behavior.

Best

Rick

–

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

–

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

–

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

[PHILIP_JERAIR_YERANO]

Can the purpose of behavior be to have no purpose?
What is the difference between measuring the intervening variable in S-O-R theory and inferring the reference variable in PCT?Â

···

On Wed, Feb 21, 2018 at 12:44 AM, Eetu Pikkarainen eetu.pikkarainen@oulu.fi wrote:

[Eetu Pikkarainen 2018-02-21_07:39:44 UTC]

Â

Philip,

Â

When scientists developed S-R psychology, they did not suppose that behavior did not have purpose. The premise of S-R psychology is that the purpose of behavior is not arbitrary. If this premise
is false, then the theory is not valid. Do you disagree with the proposition that this same premise applies to PCT?

Â

I wonder and disagree with your first sentences and would really like to hear some evidence for them. Watson as a leader figure of behaviorist movement wanted absolutely to get rid of concepts like soul, thinking, feelings
etc. and purpose which were thought to be unmeasurable and so unscientific. S-R theory is a theory about relationship between stimulus and response and nothing else. Later there was added some intervening variables and created so called S-O-R theory but these
organism variables should also be easily measured, not inferred or assumed like purpose. I think
Tolman was an exception to try
to use the concept of purpose.

Purpose was exorcised out of the theory, but of course not from the practice. The experiments and the concepts like conditioning or reinforcement necessarily require it, thus for me it seems that S-R theory is a big cheat.

Â

Eetu

Â

 Please, regard all my statements as questions,

 no matter how they are formulated.

Â

Â

Â

On Tuesday, February 20, 2018, Richard Marken rsmarken@gmail.com wrote:

[Rick Marken 2018-02-20_17:50:31]

Â

[eetu.pikkarainen@oulu.fi
21.2.2018]

Â

EP: (I feel this reply still misses something but this role of reference i.e. goal of action is anyway the core point of my understanding of PCT.)

Â

RM: You are missing nothing. The goal of action, which is the (possibly varying) reference state of the controlled variable, is what S-R (or cognitive or equilibrium ot any cause-effect theory of behavior) cannot explain.

Â

EP: Rick, for your disappointment I must tell that I searched the word complemet from dictionary and had no idea that is so dangerously near to compliment

Â

RM: Nope, not disappointed. Still very impressed!

Â

BestÂ

Â

Rick

Â

Â

Â

Eetu

Â

---

From: PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu
Sent: Tuesday, February 20, 2018 9:00:00 PM

To: csgnet@lists.illinois.edu
Subject: Re: General

Â

Yes Eetu, your response is fair. I will repost my next message for convenience.

Â

I understand that PCT explains how S-R psychology works. But these two theories involve what is essentially the same constitutive equation: response
is proportional to stimulus. In order to constitute a different theory, this constitutive equation would need to be different. Would you disagree?Â

Â

On Tue, Feb 20, 2018 at 9:52 AM, Richard Marken rsmarken@gmail.com wrote:

[Rick Marken 2018-02-20_09:52:49]

Â

Eetu Pikkarainen (2018-02-20_08:45:11 UTC)

Â

EP: Sorry, I am not a psychologist so I hope that someone will correct and complement my short answer.

Â

RM: I see nothing to correct and only things to compliment in this post. I compliment your excellent description of what PCT explains that S-R psychology doesn’t. And I also compliment your excellent command of English
as revealed in your understanding of the distinction between complementing and complimenting.Â

Â

Best

Â

Rick

Â

S-R psychology makes predictions about subject’s reactions based on the current stimulus and previous knowledge. Usually the success of these predictions are
quite low.

PCT explains why the correlation between stimulus and response must be slow: there is always an important intervening variable, namely the reference value, which
determines how the stimulus will affect the response. So PCT research does not even try to predict reactions from stimuluses. Instead it tries to find out what perception the subject is controlling (the Test for controlled variable). If it can be found then
it is possible to predict with high accuracy how the subject responds by her output (“reaction�) to the disturbances (“stimulus�) against the perceived value of the controlled variable.

Perhaps the example in Powers: Behavior: The Control of Perception (2005) about the analysis of a typical rat experiment in PCT terms could be lightening about
the difference, pages 238-243.

An extensive description about PCT type of research in relation to more typical psychological research is in Runkel, P. (1990). Casting Nets and Testing Specimens:
Two Grand Methods of Psychology.

Â

Eetu

Â

From: PHILIP JERAIR YERANOSIAN [mailto:pyeranos@ucla.edu]
Sent: Tuesday, February 20, 2018 3:43 AM
To: csgnet@lists.illinois.edu
Subject: Re: General

Â

That is a very strange example. Â

I have read some PCT literature. One remark is that PCT doesn’t employ the same model as stimulus-response psychology. How do the predictions of a stimulus-response
theory of behavior differ from PCT predictions? Â

Â

On Mon, Feb 19, 2018 at 11:01 AM, Bruce Nevin <bnhpct@gmail.com >
wrote:

[Bruce Nevin 2018-02-19_14:00:13 ET]

Â

PHILIP JERAIR YERANOSIANÂ (Sun, Feb 18, 2018 at 3:28 PM) –

Â

Philip, we do model-based control without calculating inverse kinematics. When you walk through a familiar room in the dark or with your eyes closed, the model
is made of perceptions restored from memory and controlled in imagination. Controlling in imagination to avoid the chair that is
there and the table that is over there accompanies ordinary perceptual control processes by which you walk over to the light switch by the door
there  and turn it on. In the final stages of controlling a perception of having light in the room by means of flipping that switch no amount of prediction is going to guarantee that your fingers find the switch reliably every time, but repeated practice
will sharpen the precision of your memory, not to mention your confidence in your imagination.Â

Â

I will leave it to others to refer you to the abundant literature laying out predictions that PCT makes that are different from those of other theories. And I
have no choice but to leave it up to you to read that literature.

Â

/Bruce

Â

On Sun, Feb 18, 2018 at 3:28 PM, PHILIP JERAIR YERANOSIAN <pyeranos@ucla.edu >
wrote:

The purpose of my question is to figure out why PCT is to be developed as a scientific discipline in its own right.Â

What predictions does PCT make that are different from those of other theories?   Â

Â

Virus-free.
www.avast.com

Â

On Sun, Jan 28, 2018 at 9:56 AM, Richard Marken <rsmarken@gmail.com >
wrote:

[Rick Marken 2018-01-28_09:55:50]

Â

On Sat, Jan 27, 2018 at 9:36 PM, PHILIP JERAIR YERANOSIAN <pyeranos@ucla.edu >
wrote:

Â

PY: If a system uses inverse kinematics to calculate the effects of its output, is it not ultimately doing so in order for its actions to produce a desired effect
on its input? Does it really matter if a system is using a simulated model of the environment to predict its input before calculating its output? Â

All control systems have reference values for their inputs. There is no system that only controls its output. Â

All control systems are doing “perceptualâ€? control. Who does not agree with this proposition?Â

RM: RM: I think you are asking whether everyone agrees that all control systems (properly designed closed- loop negative feedback systems) even
those that include model-based calculations of inverse kinematics, are doing perceptual control. I can’t speak for everyone but I certainly agree with that. And I think it’s a very good observation.Â

Â

RM: This fact about how control systems work is not particularly important for engineers who build such systems to understand since the engineers know which variables
they want their control systems to control and they develop perceptual systems (sensors) that provide analogs of these variables for the system to control. But it is very important for students of the controlling done by living systems to understand since
figuring out what perceptions these systems are controlling is essential to understanding their behavior.Â

Â

Best

Â

Rick

Â

–

Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you

have nothing left to take away.�

                --Antoine de Saint-Exupery

Â

Â

Â

Â

–

Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you

have nothing left to take away.�

                --Antoine de Saint-Exupery

Â

Â

–

Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you

have nothing left to take away.�

                --Antoine de Saint-Exupery

[Alex_Gomez-Marin]

Mostly, it is a difference on scope about one what projects as “relevantly the matter” in studying behaviour. For many, to understand is to know that “B” will happen when I present “A”. For others, more conceptually inclined, to understand is to grasp what is going on from the point of view of the animal. This can lead to good predictions too. But usually I find the differences to emerge from a difference in taste — like in a previous post ddiscussion Rick and I were having on whether the transfer function is right or wrong; it seems it is right (it can indeed statistically predict what the animal will do given what you present to it), but one may say it is wrong because it does not give us what “we want”, and what we want is to know, not what it will do, but what it will try to perceive.

···

On Thu, Feb 22, 2018 at 12:29 AM, PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu wrote:

Can the purpose of behavior be to have no purpose?
What is the difference between measuring the intervening variable in S-O-R theory and inferring the reference variable in PCT?Â

On Wed, Feb 21, 2018 at 12:44 AM, Eetu Pikkarainen eetu.pikkarainen@oulu.fi wrote:

[Eetu Pikkarainen 2018-02-21_07:39:44 UTC]

Â

Philip,

Â

When scientists developed S-R psychology, they did not suppose that behavior did not have purpose. The premise of S-R psychology is that the purpose of behavior is not arbitrary. If this premise
is false, then the theory is not valid. Do you disagree with the proposition that this same premise applies to PCT?

Â

I wonder and disagree with your first sentences and would really like to hear some evidence for them. Watson as a leader figure of behaviorist movement wanted absolutely to get rid of concepts like soul, thinking, feelings
etc. and purpose which were thought to be unmeasurable and so unscientific. S-R theory is a theory about relationship between stimulus and response and nothing else. Later there was added some intervening variables and created so called S-O-R theory but these
organism variables should also be easily measured, not inferred or assumed like purpose. I think
Tolman was an exception to try
to use the concept of purpose.

Purpose was exorcised out of the theory, but of course not from the practice. The experiments and the concepts like conditioning or reinforcement necessarily require it, thus for me it seems that S-R theory is a big cheat.

Â

Eetu

Â

 Please, regard all my statements as questions,

 no matter how they are formulated.

Â

Â

Â

On Tuesday, February 20, 2018, Richard Marken rsmarken@gmail.com wrote:

[Rick Marken 2018-02-20_17:50:31]

Â

[eetu.pikkarainen@oulu.fi
21.2.2018]

Â

EP: (I feel this reply still misses something but this role of reference i.e. goal of action is anyway the core point of my understanding of PCT.)

Â

RM: You are missing nothing. The goal of action, which is the (possibly varying) reference state of the controlled variable, is what S-R (or cognitive or equilibrium ot any cause-effect theory of behavior) cannot explain.

Â

EP: Rick, for your disappointment I must tell that I searched the word complemet from dictionary and had no idea that is so dangerously near to compliment

Â

RM: Nope, not disappointed. Still very impressed!

Â

BestÂ

Â

Rick

Â

Â

Â

Eetu

Â

---

From: PHILIP JERAIR YERANOSIAN pyeranos@ucla.edu
Sent: Tuesday, February 20, 2018 9:00:00 PM

To: csgnet@lists.illinois.edu
Subject: Re: General

Â

Yes Eetu, your response is fair. I will repost my next message for convenience.

Â

I understand that PCT explains how S-R psychology works. But these two theories involve what is essentially the same constitutive equation: response
is proportional to stimulus. In order to constitute a different theory, this constitutive equation would need to be different. Would you disagree?Â

Â

On Tue, Feb 20, 2018 at 9:52 AM, Richard Marken rsmarken@gmail.com wrote:

[Rick Marken 2018-02-20_09:52:49]

Â

Eetu Pikkarainen (2018-02-20_08:45:11 UTC)

Â

EP: Sorry, I am not a psychologist so I hope that someone will correct and complement my short answer.

Â

RM: I see nothing to correct and only things to compliment in this post. I compliment your excellent description of what PCT explains that S-R psychology doesn’t. And I also compliment your excellent command of English
as revealed in your understanding of the distinction between complementing and complimenting.Â

Â

Best

Â

Rick

Â

S-R psychology makes predictions about subject’s reactions based on the current stimulus and previous knowledge. Usually the success of these predictions are
quite low.

PCT explains why the correlation between stimulus and response must be slow: there is always an important intervening variable, namely the reference value, which
determines how the stimulus will affect the response. So PCT research does not even try to predict reactions from stimuluses. Instead it tries to find out what perception the subject is controlling (the Test for controlled variable). If it can be found then
it is possible to predict with high accuracy how the subject responds by her output (“reactionâ€?) to the disturbances (“stimulusâ€?) against the perceived value of the controlled variable.

Perhaps the example in Powers: Behavior: The Control of Perception (2005) about the analysis of a typical rat experiment in PCT terms could be lightening about
the difference, pages 238-243.

An extensive description about PCT type of research in relation to more typical psychological research is in Runkel, P. (1990). Casting Nets and Testing Specimens:
Two Grand Methods of Psychology.

Â

Eetu

Â

From: PHILIP JERAIR YERANOSIAN [mailto:pyeranos@ucla.edu]
Sent: Tuesday, February 20, 2018 3:43 AM
To: csgnet@lists.illinois.edu
Subject: Re: General

Â

That is a very strange example. Â

I have read some PCT literature. One remark is that PCT doesn’t employ the same model as stimulus-response psychology. How do the predictions of a stimulus-response
theory of behavior differ from PCT predictions? Â

Â

On Mon, Feb 19, 2018 at 11:01 AM, Bruce Nevin <bnhpct@gmail.com >
wrote:

[Bruce Nevin 2018-02-19_14:00:13 ET]

Â

PHILIP JERAIR YERANOSIANÂ (Sun, Feb 18, 2018 at 3:28 PM) –

Â

Philip, we do model-based control without calculating inverse kinematics. When you walk through a familiar room in the dark or with your eyes closed, the model
is made of perceptions restored from memory and controlled in imagination. Controlling in imagination to avoid the chair that is
there and the table that is over there accompanies ordinary perceptual control processes by which you walk over to the light switch by the door
there  and turn it on. In the final stages of controlling a perception of having light in the room by means of flipping that switch no amount of prediction is going to guarantee that your fingers find the switch reliably every time, but repeated practice
will sharpen the precision of your memory, not to mention your confidence in your imagination.Â

Â

I will leave it to others to refer you to the abundant literature laying out predictions that PCT makes that are different from those of other theories. And I
have no choice but to leave it up to you to read that literature.

Â

/Bruce

Â

On Sun, Feb 18, 2018 at 3:28 PM, PHILIP JERAIR YERANOSIAN <pyeranos@ucla.edu >
wrote:

The purpose of my question is to figure out why PCT is to be developed as a scientific discipline in its own right.Â

What predictions does PCT make that are different from those of other theories?   Â

Â

Virus-free.
www.avast.com

Â

On Sun, Jan 28, 2018 at 9:56 AM, Richard Marken <rsmarken@gmail.com >
wrote:

[Rick Marken 2018-01-28_09:55:50]

Â

On Sat, Jan 27, 2018 at 9:36 PM, PHILIP JERAIR YERANOSIAN <pyeranos@ucla.edu >
wrote:

Â

PY: If a system uses inverse kinematics to calculate the effects of its output, is it not ultimately doing so in order for its actions to produce a desired effect
on its input? Does it really matter if a system is using a simulated model of the environment to predict its input before calculating its output? Â

All control systems have reference values for their inputs. There is no system that only controls its output. Â

All control systems are doing “perceptualâ€? control. Who does not agree with this proposition?Â

RM: RM: I think you are asking whether everyone agrees that all control systems (properly designed closed- loop negative feedback systems) even
those that include model-based calculations of inverse kinematics, are doing perceptual control. I can’t speak for everyone but I certainly agree with that. And I think it’s a very good observation.Â

Â

RM: This fact about how control systems work is not particularly important for engineers who build such systems to understand since the engineers know which variables
they want their control systems to control and they develop perceptual systems (sensors) that provide analogs of these variables for the system to control. But it is very important for students of the controlling done by living systems to understand since
figuring out what perceptions these systems are controlling is essential to understanding their behavior.Â

Â

Best

Â

Rick

Â

–

Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you

have nothing left to take away.�

                --Antoine de Saint-Exupery

Â

Â

Â

Â

–

Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you

have nothing left to take away.�

                --Antoine de Saint-Exupery

Â

Â

–

Richard S. MarkenÂ

"Perfection is achieved not when you have nothing more to add, but when you

have nothing left to take away.�

                --Antoine de Saint-Exupery