PCT an equilibrium Theory

CSGnet Archive CSG2010
1

(Gavin Ritz 2010.07.25. 00.43NZT)

Hi there Boris
and Bruce

I can’t see how
PCT can be an equilibrium theory when it shows clearly the asymmetry between input
and the output. Unless what one means by equilibrium is “stability”.
Further an organism can’t be in equilibrium in the scientific sense (thermodynamical
sense) because then it would be dead.

Any organism is always
in some state far-from-equilibrium (Prigogine) this of-course doesn’t mean
that it can’t seek some stability like a particular temperature but to
keep that temperature it requires amounts of energy to keep it there. In PCT
the reference signal at the highest level can never be zero if it was then we
would be dead. The brain actually uses more energy when in its so called stable
state. Almost like stretching rubber band further in anticipation of some input.

The difference between
any living organism and inorganic matter is information; once matter and
information are combined it seems that we have a far-from equilibrium system.
Remove the information and it’s just matter (equilibrium system- a rock).
The PCT type model is the result of information and matter combined.

Further once you add
HPCT into the mix, and I’m still not quite sure how in HPCT qualities
(sensations) and quantities (intensities) are quite reconciled, but as HPCT
stands now it does, we know that science has no place for qualities, qualities
add a sort of variety that cannot be measured so this also makes PCT irreconcilable
with physics at some level (.See Schrödinger ”What is Life”).

I must conclude from my
perspective that PCT is a far-from-equilibrium theory.

Regards

Gavin

···
2

[From Bruce Gregory (2010.0724.1254 EDT)]

(Gavin Ritz 2010.07.25. 00.43NZT)

Hi there Boris and Bruce

I can’t see how PCT can be an equilibrium theory when it shows clearly the asymmetry between input and the output. Unless what one means by equilibrium is “stability”. Further an organism can’t be in equilibrium in the scientific sense (thermodynamical sense) because then it would be dead.

You are quite correct. I was using equilibrium in the sense of stablity. But here again one can argue that living systems cannot be stable either. As for the use of “in the scientific sense.” “Energy conserving” makes no sense because energy is always conserved. Likewise, saying that organisms minimize free energy can be misread as suggesting that organisms maximize entropy.

Homeostasis may be the best term, but it too invites misuderstading.

I agree that I should not have used equilibrium, but as Martin points out, this still leaves two questions that I would have thought to merit some consideration.

Give that all studies in social psychology, sociology, and behavioral economics are deeply flawed. Would it not make sense to identify a handful of studies judged to be exemplary by their communities and to show: (1) how these studies would have been markedly improved if the researchers had applied the principles of PCT in designing and conducting the experiments, and (2) indicating what else the researchers might have learned from a proper (PCT-based) study.

Absence this guidance fro the CSGnet community, researchers might be forgiven for seeing PCT as a distinction that has yet been shown to make a difference.

.

3

[From Rick Marken (2010.07.24.1730)]

Bruce Gregory (2010.0724.1254 EDT)--

Give[n] that all studies in social psychology, sociology, and behavioral
economics are deeply flawed, [w]ould it not make sense to identify a handful
of studies judged to be exemplary by their communities and to show: (1) how
these studies would have been markedly improved if the researchers had
applied the principles of PCT in designing and conducting the experiments,
and (2) indicating what else the researchers might have learned from a
proper (PCT-based) study.

These are two good questions. Before trying to answer them I should
say that I don't think that "all studies in social psychology,
sociology, and behavioral economics are deeply flawed". Many of these
studies are not flawed at all. Some are perfectly elegant (such as my
doctoral research;-). The only problem with these studies (from a
control theory perspective) is that they are done in the context of a
causal model that assumes that sensory inputs are the cause of
behavioral outputs. Assuming that the systems under study are control
systems, PCT would see these studies as looking at "side effects" of
control; specifically, disturbance-output relationships. So if there
is a "flaw" in these studies it is just that they typically ignore, or
take for granted what is most important about behavior from a control
theory point of view: its purposes. These purposes (according to PCT)
involve keeping perceptual variables under control. Therefore, the
discovery of controlled perceptual variables and the determination of
how these variables are controlled would be a central focus of
research based on control theory (PCT).

So how would studies in social psychology, sociology, and behavioral
economics be improved by PCT? I guess my answer would be a copy of the
above: by being oriented toward discovering the purposes of behavior
and determining how these purposes are achieved. And I would add that
this research would be done _on an individual basis_: "one subject at
a time". The data from such studies would be analyze using closed loop
control models rather than open loop statistical models. The results
of such studies would not be considered "facts" unless the model
accounted for at least 99% of the variance in the data. There would be
no more need for statistical analysis of the data.

I think this also answers your question regarding what else
researchers might learn from PCT-based research. What else researchers
would learn is something they never thought of studying: the subjects'
purposes (what variables the subject control) and how they achieve
those purposes.

Most current studies in social psychology, sociology, and behavioral
economics are not oriented toward detecting purposes. I think the
people who do this research are quite satisfied with what they do find
out from these studies; they are comfortable with the statistical
relationships that they find and, when necessary, they will throw in
speculations about purposes when this seems warranted. For example,
I've seen studies of behavioral economics that talk about the findings
in terms of purposes like maximizing relative or absolute gain. But I
have seen no systematic attempt to test these hypotheses on an
individual level (the inferences about purpose are typically based on
average data) using the appropriate closed-loop models, at least not
in behavioral economics. Maybe I just haven't noticed.

I do think it would be a good idea to make this more tangible by doing
what you suggest: identify a handful of studies judged to be exemplary
by their communities and to show how they would be improved if the
researchers had applied the principles of PCT. I think it would be
best to start with one study and discuss in some detail the actual
results of the study, not just a high level description of those
results. For example, maybe we could look at a classic social
psychology experiment like Asch's "conformity" study. The high level
description of that study says that subjects' judgments were found to
be influenced by peer pressure. But I think we should learn as much as
possible about the details of what actually went on in the study and
see what we might learn from looking at it from a control theory
perspective. Or perhaps you could suggest a more recent, highly
regarded study in behavioral economics. That seems to be a trendy
area. Let's take a detailed look at one of those studies and see if
PCT has anything to contribute.

Best

Rick

···

---
Richard S. Marken PhD
rsmarken@gmail.com
www.mindreadings.com

4

[From Kenny Kitzke (2010.07.24)]

The following video is a study/experiment of behavior in chimps and children. I find it full of the complaints that Rick has been expressing. The conclusions are exasperating science in my view.

I wonder how Rick and Martin analyze this type of behavioral experiment and whether they believe the experimenter’s discoveries are valid. After all, multiple chimps and children seemed to behave the same way.

Thanks

http://www.care2.com/greenliving/whos-smarter-you-or-a-chimp-2-videos.html

In a message dated 7/24/2010 8:30:57 P.M. Eastern Daylight Time, rsmarken@GMAIL.COM writes:

···

[From Rick Marken (2010.07.24.1730)]

Bruce Gregory (2010.0724.1254 EDT)–

Give[n] that all studies in social psychology, sociology, and behavioral
economics are deeply flawed, [w]ould it not make sense to identify a handful
of studies judged to be exemplary by their communities and to show: (1) how
these studies would have been markedly improved if the researchers had
applied the principles of PCT in designing and conducting the experiments,
and (2) indicating what else the researchers might have learned from a
proper (PCT-based) study.

These are two good questions. Before trying to answer them I should
say that I don’t think that “all studies in social psychology,
sociology, and behavioral economics are deeply flawed”. Many of these
studies are not flawed at all. Some are perfectly elegant (such as my
doctoral research;-). The only problem with these studies (from a
control theory perspective) is that they are done in the context of a
causal model that assumes that sensory inputs are the cause of
behavioral outputs. Assuming that the systems under study are control
systems, PCT would see these studies as looking at “side effects” of
control; specifically, disturbance-output relationships. So if there
is a “flaw” in these studies it is just that they typically ignore, or
take for granted what is most important about behavior from a control
theory point of view: its purposes. These purposes (according to PCT)
involve keeping perceptual variables under control. Therefore, the
discovery of controlled perceptual variables and the determination of
how these variables are controlled would be a central focus of
research based on control theory (PCT).

So how would studies in social psychology, sociology, and behavioral
economics be improved by PCT? I guess my answer would be a copy of the
above: by being oriented toward discovering the purposes of behavior
and determining how these purposes are achieved. And I would add that
this research would be done on an individual basis: “one subject at
a time”. The data from such studies would be analyze using closed loop
control models rather than open loop statistical models. The results
of such studies would not be considered “facts” unless the model
accounted for at least 99% of the variance in the data. There would be
no more need for statistical analysis of the data.

I think this also answers your question regarding what else
researchers might learn from PCT-based research. What else researchers
would learn is something they never thought of studying: the subjects’
purposes (what variables the subject control) and how they achieve
those purposes.

Most current studies in social psychology, sociology, and behavioral
economics are not oriented toward detecting purposes. I think the
people who do this research are quite satisfied with what they do find
out from these studies; they are comfortable with the statistical
relationships that they find and, when necessary, they will throw in
speculations about purposes when this seems warranted. For example,
I’ve seen studies of behavioral economics that talk about the findings
in terms of purposes like maximizing relative or absolute gain. But I
have seen no systematic attempt to test these hypotheses on an
individual level (the inferences about purpose are typically based on
average data) using the appropriate closed-loop models, at least not
in behavioral economics. Maybe I just haven’t noticed.

I do think it would be a good idea to make this more tangible by doing
what you suggest: identify a handful of studies judged to be exemplary
by their communities and to show how they would be improved if the
researchers had applied the principles of PCT. I think it would be
best to start with one study and discuss in some detail the actual
results of the study, not just a high level description of those
results. For example, maybe we could look at a classic social
psychology experiment like Asch’s “conformity” study. The high level
description of that study says that subjects’ judgments were found to
be influenced by peer pressure. But I think we should learn as much as
possible about the details of what actually went on in the study and
see what we might learn from looking at it from a control theory
perspective. Or perhaps you could suggest a more recent, highly
regarded study in behavioral economics. That seems to be a trendy
area. Let’s take a detailed look at one of those studies and see if
PCT has anything to contribute.

Best

Rick

Richard S. Marken PhD
rsmarken@gmail.com
www.mindreadings.com

5

(Gavin Ritz 2010.07.25.13.28NZT)

[From Bruce
Gregory (2010.0724.1254 EDT)]

(Gavin Ritz 2010.07.25.
00.43NZT)

Hi there Boris and Bruce

I can’t see how PCT can be an equilibrium theory when it
shows clearly the asymmetry between input and the output. Unless what one means
by equilibrium is “stability”. Further an organism can’t be
in equilibrium in the scientific sense (thermodynamical sense) because then it
would be dead.

You are quite correct. I
was using equilibrium in the sense of stablity. But here again one can argue
that living systems cannot be stable either. As for the use of “in the
scientific sense.” “Energy conserving” makes no sense because
energy is always conserved. Likewise, saying that organisms minimize free
energy can be misread as suggesting that organisms maximize entropy.

Homeostasis may be the
best term, but it too invites misuderstading.

I agree that I should not
have used equilibrium, but as Martin points out, this still
leaves two questions that I would have thought to merit some consideration.

Give that all studies in
social psychology, sociology, and behavioral economics are deeply flawed. Would
it not make sense to identify a handful of studies judged to be exemplary by
their communities and to show: (1) how these studies would have been markedly
improved if the researchers had applied the principles of PCT in designing and
conducting the experiments, and (2) indicating what else the researchers might
have learned from a proper (PCT-based) study.

I couldn’t agree more I too have
been asking this question for a few years.

Absence this guidance fro
the CSGnet community, researchers might be forgiven for seeing PCT as a
distinction that has yet been shown to make a difference.

Point taken.

I spend quite a bit of time on PCT thinking about
it re-reading it. I think I probably have most of the books on this subject but
quite can’t reconcile some things.

One thing that I did reconcile some weeks
ago is the concept of PCT and energy. Which was liberating for me, but probably
known by everyone else was that all incoming input is transduced energy, electromagnetic,
sound, temperature, pressure, chemical. All this is transduced into a signal. (A
perceptual signal). So a control system is really an energy transducer. But in
the CS systems, (many millions) huge amounts of asymmetry go on all the time, these are actually little
energy systems with blockages. Which requisitely would require reorganization
otherwise the system would explode. Now this takes me closer and closer to a another
loved theory of mine Engpass Konzentrierten Strategie (EKS). Developed by Wolfgang Mewes in the early 70’s and
late 60’s. It’s a Gestalt theory of energy and cybernetics.

I think PCT needs one breakthrough
situation that makes the audience go wow that’s something we could not understand
this explains it all. I also think the focus on psychology is just a waste of
energy. PCT explains psychological phenomenon but it’s not a theory of
psychology is a theory of “wholes or Gestalt”. But that doesn’t
help anyone either.

Just frustrated sometimes with PCT not because
of PCT but because I am unable to reconcile so many things about it.

I do believe that Bill has uncovered something
but how to get this aligned with other knowledge and power systems is the real challenge.
PCTers should know that it’s useless to say another theory is wrong when
so many have invested so much in other disciplines.

Regards

Gavin

.

···
6

(Gavin Ritz 2010.07.25.13.50NZT)

[From Rick Marken
(2010.07.24.1730)]

Bruce Gregory
(2010.0724.1254 EDT)–

I do think it would be a good idea to make
this more tangible by doing

what you suggest: identify a handful of
studies judged to be exemplary

by their communities and to show how
they would be improved if the

researchers had applied the principles of
PCT.

This is now making sense.

I think it would be

best to start with one study and discuss in some
detail the actual

results of the study, not just a high level
description of those

results. For example, maybe we could look at a classic
social

psychology experiment like Asch’s “conformity” study. The
high level

description of that study says that subjects’
judgments were found to

be influenced by peer pressure. But I think we should
learn as much as

possible about the details of what actually went on in
the study and

see what we might learn from looking at it from a
control theory

perspective.

Or perhaps you could suggest a more recent, highly

regarded study in behavioral economics.

I think this is an excellent
place to start Rick, behavioural economics,
this is where I can help, and I have huge amounts of personal research and plenty
of direct experience and connected to enough people to get their input. There would
be many people that would love to get involved if there was maturity and openness,
plenty Noble Laureates like Joseph Stiglitz
would be very interest he swims against the tide all the time.
I also have very powerful models that I would happily share. I have presented
some of these models at international conferences.

That seems to be a trendy

area. Let’s take a detailed look at one of those
studies and see if

PCT has anything to contribute.

Very good idea, Rick.
Let’s choose a project and get the PCT ground rules in and see what we
can do. You are in a perfect position to take your 30 plus years of PCT knowledge
and guide us through such a process.

I will look around a see
what I can dig up. The most topical subject at the moment is the economic collapse
of the financial systems. There are many technical reasons but the human ones
are always glossed over. Let’s be targeted and choose something from there.

Regards

Gavin

···
7

[From David Goldstein (2010.07.25.04:45 EDT)]

[About Kenny Kitzke (2010.07.24)]

What a fun series of videos to watch!

I noticed that even with the opaque box, the chimp was more experimental and deviated from the observed sequence a bit.

The chimp was amazing in the touching numbers in order video. It did occur to me that a lot of training of the chimp on this task was probably needed. Did the human have equivalent practice?

David

···

----- Original Message -----

From:
Kenneth Kitzke Value Creation Systems

To: CSGNET@LISTSERV.ILLINOIS.EDU

Sent: Saturday, July 24, 2010 8:55 PM

Subject: Re: PCT an equilibrium Theory

[From Kenny Kitzke (2010.07.24)]

The following video is a study/experiment of behavior in chimps and children. I find it full of the complaints that Rick has been expressing. The conclusions are exasperating science in my view.

I wonder how Rick and Martin analyze this type of behavioral experiment and whether they believe the experimenter’s discoveries are valid. After all, multiple chimps and children seemed to behave the same way.

Thanks

http://www.care2.com/greenliving/whos-smarter-you-or-a-chimp-2-videos.html

In a message dated 7/24/2010 8:30:57 P.M. Eastern Daylight Time, rsmarken@GMAIL.COM writes:

[From Rick Marken (2010.07.24.1730)]

> Bruce Gregory (2010.0724.1254 EDT)--

> Give[n] that all studies in social psychology, sociology, and behavioral
> economics are deeply flawed, [w]ould it not make sense to identify a handful
> of studies judged to be exemplary by their communities and to show: (1) how
> these studies would have been markedly improved if the researchers had
> applied the principles of PCT in designing and conducting the experiments,
> and (2) indicating what else the researchers might have learned from a
> proper (PCT-based) study.

These are two good questions. Before trying to answer them I should
say that I don't think that "all studies in social psychology,
sociology, and behavioral economics are deeply flawed". Many of these
studies are not flawed at all. Some are perfectly elegant (such as my
doctoral research;-). The only problem with these studies (from a
control theory perspective) is that they are done in the context of a
causal model that assumes that sensory inputs are the cause of
behavioral outputs. Assuming that the systems under study are control
systems, PCT would see these studies as looking at "side effects" of
control; specifically, disturbance-output relationships.  So if there
is a "flaw" in these studies it is just that they typically ignore, or
take for granted what is most important about behavior from a control
theory point of view: its purposes. These purposes (according to PCT)
involve keeping perceptual variables under control. Therefore, the
discovery of controlled perceptual variables and the determination of
how these variables are controlled would be a central focus of

research based on control theory (PCT).

So how would studies in social psychology, sociology, and behavioral
economics be improved by PCT? I guess my answer would be a copy of the
above: by being oriented toward discovering the purposes of behavior
and determining how these purposes are achieved. And I would add that
this research would be done _on an individual basis_: "one subject at
a time". The data from such studies would be analyze using closed loop
control models rather than open loop statistical models. The results
of such studies would not be considered "facts" unless the model
accounted for at least 99% of the variance in the data. There would be
no more need for statistical analysis of the data.

I think this also answers your question regarding what else
researchers might learn from PCT-based research. What else researchers
would learn is something they never thought of studying: the subjects'
purposes (what variables the subject control) and how they achieve

those purposes.

Most current studies in social psychology, sociology, and behavioral
economics are not oriented toward detecting purposes. I think the
people who do this research are quite satisfied with what they do find
out from these studies; they are comfortable with the statistical
relationships that they find and, when necessary, they will throw in
speculations about purposes when this seems warranted. For example,
I've seen studies of behavioral economics that talk about the findings
in terms of purposes like maximizing relative or absolute gain. But I
have seen no systematic attempt to test these hypotheses on an
individual level (the inferences about purpose are typically based on
average data) using the appropriate closed-loop models, at least not

in behavioral economics. Maybe I just haven’t noticed.

I do think it would be a good idea to make this more tangible by doing
what you suggest: identify a handful of studies judged to be exemplary
by their communities and to show  how they would be improved if the
researchers had applied the principles of PCT. I think it would be
best to start with one study and discuss in some detail the actual
results of the study, not just a high level description of those
results. For example, maybe we could look at a classic social
psychology experiment like Asch's "conformity" study. The high level
description of that study says that subjects' judgments were found to
be influenced by peer pressure. But I think we should learn as much as
possible about the details of what actually went on in the study and
see what we might learn from looking at it from a control theory
perspective. Or perhaps you could suggest a more recent, highly
regarded study in behavioral economics. That seems to be a trendy
area. Let's take a detailed look at one of those studies and see if
PCT has anything to contribute.

Best

Rick

Richard S. Marken PhD

rsmarken@gmail.com
www.mindreadings.com

8

[From Rick Marken (2010.07.25.1145)]

David Goldstein (2010.07.25.04:45 EDT) re Kenny Kitzke (2010.07.24)

What a fun series of videos to watch!

I agree. Very interesting. I don't see what Kenny thinks is wrong with
the study. It looks like a pretty obvious (and rather amazing) example
of control. The chimps are clearly able to control a relationship
between the order of the numbers flashed on the screen and the order
in which they touch the locations in which these numbers had appeared
in the array. And they can do this despite disturbances like
scrambling the location of the numbers on the screen or showing only a
subset of the numbers. And they seem to do this much better than
humans.

It did occur to me that a lot of training of the chimp on this task was
probably needed. Did the human have equivalent practice?

Yes, the same thing occurred to me. But I think that even if human and
chimp had equivalent training there might be a difference still.
That's because I think it's possible that the chimps are controlling a
different variable than humans when they do this. This is where PCT
might be able to make some contribution to this research. The
researchers seem to be assuming that the chimps are controlling for a
relationship between a number sequence--1,2,3...-- and the touch
sequence. But according to my research (see
http://www.mindreadings.com/ControlDemo/HP.html), perception and
control of a sequence requires about 400msec per item in the sequence.
Notice that in this chimp research the array of numbers is flashed for
a very brief period, less than a second, which would presumably make
it impossible to perceive all nine numbers as a sequence (that would
take at least a 3.5 second presentation, of chimp perceptual systems
are anything like humans').

My hypothesis is that the chimps perceive the number array as a
configuration and then produce the sequence of finger presses by
controlling the sequence of finger presses relative to the remembered
configuration (I timed the output sequence of screen touches and it
takes about 3.5 seconds, which is about the right amount of time for
sequence control). I think of it like this: the flashed array of
numbers is like a flashed picture of a face; the face is in a
different orientation on each trial but your job is to touch the left
then right eye, nose, left ear, right ear, etc. This will require
touching different points on the screen in sequence for each picture
(since the face has a different orientation each time, like the
different shape of the array of numbers for the chimp). This would be
easy to do without having to store any sequence information about the
picture itself. I think this is what the chimps have learned to do
with the numbers.

If this hypothesis about what the chimps are controlling (a
configuration rather than a sequence) is correct then it could also
explain why humans have so much difficulty with this task. I think
that when the number array is flashed at a human, the human tries to
remember it as a sequence -- "one" upper right, "two" middle left,
etc. -- rather than as a configuration (I do anyway) and the flashes
are far too brief for that. I think humans could learn to see the
array as a configuration, but it would take some training because
humans are "too smart"; I think it's difficult for people to see 1 2 3
as a configuration -- a pattern of pixels in this case -- rather than
as the higher level sequence -- "one", "two", "three". This is not a
problem for the chimps since 1 2 3 for the chimps is just different
configurations of lines.

Now we just have to figure out a way to test my hypothesis about the
controlled variable in this situation. One possibility is this:
Instead of flashing the entire array of numbers all at once have each
number come on and off in random positions in random order for a brief
time (100 msec?). After all numbers have been presented the subject is
to tap the now blank squares in an order corresponding to the
sequential order of the numbers (not the sequential order in which
they were presented). In this case there is no array of numbers to
perceive as a configuration. The subject would have to remember which
number occurred in which position and also know the sequential value
of the number itself. This could be a much more demanding task, for
chimp or human, so maybe you could use only 4 or five numbers. I bet
the human would be better at this than the chimp.

Other suggestions?

Best

Rick

···

--
Richard S. Marken PhD
rsmarken@gmail.com
www.mindreadings.com

9

[From Rick Marken (2010.07.29.1510)

(Gavin Ritz 2010.07.25.13.50NZT)

Very good idea, Rick. Let�s choose a project and get the PCT ground rules in
and see what we can do. You are in a perfect position to take your 30 plus
years of PCT knowledge and guide us through such a process.

I will look around a see what I can dig up. The most topical subject at the
moment is the economic collapse of the financial systems. There are many
technical reasons but the human ones are always glossed over. Let's be
targeted and choose something from there.

So, nu? Have you found anything yet?

Best

Rick

···

--
Richard S. Marken PhD
rsmarken@gmail.com
www.mindreadings.com

10

(Gavin Ritz 2010.07.30.10.25NZT)

[From Rick Marken (2010.07.29.1510)

(Gavin Ritz 2010.07.25.13.50NZT)

Yes I have got something just
give me a few more days to put it together, I have just come back from a business
trip. Trolled thru Stiglitz’s latest book Freefall and think I found
something.

Very good idea, Rick. Let’s choose a
project and get the PCT ground rules in

and see what we can do. You are in a perfect
position to take your 30 plus

years of PCT knowledge and guide us through such
a process.

I will look around a see what I can dig up. The
most topical subject at the

moment is the economic collapse of the financial
systems. There are many

technical reasons but the human ones are always
glossed over. Let’s be

targeted and choose something from there.

So, nu? Have you found anything yet?

Best

Rick

···

–

Richard S. Marken PhD

rsmarken@gmail.com

www.mindreadings.com