Approaches to PCT and Friction on CSGNet

[rsmarken]

[From Rick Marken (2018.05.05.1500)]

RM: I must admit that when I posted my “Minority
Report”, announcing that I would stop unsolicited posting to CSGNet, I was
busy feeling rather sorry for myself. Since Bill passed away I have been
alternately surprised, frustrated, angered and hurt by the response to my posts
to CSGNet. But over the last few days I’ve had what I think was a nice “up
a level” experience, which basically involved realizing that one of my
highest level goals is to present an accurate account of PCT that can be the
basis of furthering PCT science; and that I realized that I can achieve this
goal by, among other things, posting to CSGNet, regardless of how well those
posts are received.Â

RM: So I will start my return to unsolicited posting by
explaining why the “fundamental misunderstandings” about PCT that I
mentioned in my “Minority Report” are, indeed, fundamental
misunderstandings. (This post is not really unsolicited; after giving me some
MOL therapy, my co-author on “Controlling People”, Tim Carey,
suggested that it would be useful if I would post an explanation of why these
misunderstandings about PCT are, indeed, misunderstandings, and maybe even post
a copy of it to the MOL list.) So here it goes:

RM: First here is the list of misunderstandings about PCT
that I posted in my “Minority Report”:

• Organisms control only perceptions, not the aspects of the
  environment that correspond to those perceptions.
• The test for the controlled variable is not an essential
  component of PCT-based research.
• PCT shows that you can’t control the behavior of another
  person.
• Social stability arises from interpersonal conflict.
• The power law of movement is not an example of a behavioral
  illusion.

RM: I believe that all of these misunderstandings are a
result from taking a “mathematical-logical” rather than a
“scientific/engineering” approach to PCT. The mathematical-logical
approach treats PCT as a set of axioms from which conclusions about behavior
are derived like mathematical theorems. The scientific/engineering approach, on
the other hand, treats PCT as a set of tentative guesses about the mechanisms
that explain the behavior that is actually observed. The correctness of PCT
explanations of behavior is, therefore, evaluated very differently by these two
approaches. The mathematical-logical
approach evaluates the correctness of these explanations in terms of whether
they are properly derived from the theoretical “axioms” .The
scientific/engineering approach, on the other hand, evaluates the correctness
of these explanations by testing to see whether they correspond to what is
actually observer in appropriate tests or demonstrations.

RM: It’s easy to see how taking the mathematical-logical
approach to PCT could lead one to conclude that the five misunderstandings
listed above are actually correct implications about behavior from a PCT
perspective. For example, PCT does say that organisms control only a perceptual
signal. This perceptual signal is not the same as the aspect of the environment
to which it corresponds. So it’s logically correct to conclude that PCT says
that organisms control only perceptions, not the aspects of the environment
that correspond to those perceptions. And the logical corollary to this is that
one cannot necessarily tell what perception an organism is controlling by
testing to see whether a particular aspect of the environment is being
controlled. So the test for the controlled variable, which involves only
testing to see whether certain aspects of the environment are under control,
can’t possibly be an essential component of PCT because there is no necessary
relationship between any aspect of the environment that might be under control
and the perceptual variables that the organism is actually controlling.

RM: Because PCT says that a person’s behavior involves
control of a perceptual signal that is accessible only to the person themselves
it is also logical to conclude that it is impossible to control that person’s
behavior. And PCT also shows that, under the appropriate conditions, when
several control systems control the same perceptual variable relative to
somewhat different reference levels, there will be a conflict but the
perceptual variable will be maintained at virtual reference level that is the
average of the reference levels of all the systems involved in the conflict.
Since the perceptual variable is being stabilized by a “society” of
control systems, it is logical to conclude that social stability arises from
interpersonal conflict.

RM: Finally, PCT says that there will be a consistent,
negative relationship between the disturbance to a controlled variable and the
actions that compensate for the effect of that disturbance. The power law of
movement seems to be an example of such a disturbance-action relationship; the
power law is the observation that when we make curved movements we slow down in
proportion to the degree of curvature through which we are moving. So it looks
like variations in the speed of movement are actions that compensate for the
disturbance of variations in the degree of curvature through which the movement
is being made. So the power law is not a behavioral illusion because that
relationship is not assumed to reflect characteristics of the organism that
transform the “stimulus” of curvature in to “response” of
movement speed.

RM: So all five misunderstandings of PCT listed above are
perfectly logical deductions about behavior when one takes the
mathematical-logical approach to PCT. They can only be seen as
misunderstandings when one takes the scientific/engineering approach to PCT.
from the scientific engineering perspective the idea that organisms control
only perceptions, not the aspects of the environment that correspond to those
perceptions, is instantly seen as having things completely backwards; the idea
that organisms control perceptions is a theory designed to account for the
observation that organisms can be seen to control aspects of their environment.
What we know for a fact is that organisms control aspects of their environment;
we guess that this happening because organisms are controlling perceptual
signals that are analogs of variations in the aspects of the environment that
we see being controlled.

RM: We use the test for the controlled variable to determine
the aspects of their environment that organisms control. So the idea that the
test for the controlled variable is not essential because the results do not
necessarily reveal the aspects of the environment that correspond to the
perceptions that are controlled is nonsense from the point of view of the
scientific/engineering approach to PCT. The test for the controlled variable is
the only basis we have for determining what perceptual variables an organism
might be controlling. And it is easy to demonstrate
(http://www.mindreadings.com/ControlDemo/Mindread.html) that the results of the
test for the controlled variable provide a very reliable indication of what
aspects of the environment are under control and, thus, what perceptions we
should imagine are being be controlled; the theoretical perceptions we imagine
to be controlled correspond exactly to the aspects of the environment that we
have found to be controlled using the test for the controlled variable.

RM: A crucial difference between the mathematical-logical
and scientific/engineering approach to PCT to in the attitude toward observation
and test. Those who take the mathematical-logical approach to PCT express a
particular disdain for observation and test in the form of what are called the
PCT “demos”, such as the computer demos at http://www.mindreadings.com/demos.htm. But
these demos have been an essential part of the scientific/engineering approach
to PCT from the very start, as can be seen in the lovely “portable
demos” described in the 1960 Powers, Clark and McFarland article “A
General Feedback Theory of Human Behavior: Part II” reprinted in Living
Control Systems, pp 25-45. Using demos
such as this it is easy to show that you can control the behavior of another
person (http://www.mindreadings.com/ControlDemo/BehavioralControl.html) and that that the power law of movement is an example of a
behavioral illusion (as shown by the demo described in the “Facts of
Logic” section of Marken and Shaffer, 2018
(https://www.dropbox.com/s/3m51ko4vs1xdult/MarkenShafferReappraisal.pdf?dl=0). Â

RM: The scientific/engineering approach to PCT doesn’t deny
that social stability can arise from interpersonal conflict. However, there is,
to my knowledge, no demonstration of that it can. Such a demonstration would
involve showing that groups of real organisms produce stable results like those
produced by the simulations of multiple control systems control the same
variable relative to different references. Until such demonstrations are
produced, the idea that social stability can arise from interpersonal conflict
must remain a conclusion that is true only from the perspective of the
mathematical-logical approach to PCT.

RM: I’ll conclude this already rather long post with a copy
of an even longer post from Bill Powers. This post was sent to me by someone
who is not on CSGNet but is one of the few people who respect my work on PCT
(and the feeling is mutual). It’s a post from 1995 where Bill provides and
analysis of why he thinks there is so much friction between himself and Martin
Taylor. I’m posting it to show that Bill’s approach to PCT was most
emphatically the scientific/engineering approach. And that he was not a big fan
of the mathematical-logical approach. I have bolded some of the sections that I
think are particularly germane to the problems I have been having in
discussions on CSGNet. Like Bill, who
says at the end of this post “I have no illusions about changing your
style to correct what I see as mistakes”, I also have no illusions about
changing the approach to PCT of those on CSGNet to correct what I see as
mistakes. But I hope this helps you see why the friction on CSGNet exists.

···

========================================

Date: Wed, 17 May 1995 05:24:50 -0600

Subject: Friction

Hello, Martin – (no CCs)

I have become increasingly frustrated with our communications and have
been

trying to figure out what is wrong. In the middle of the night a
possibility

occurred to me. A bit of browsing through the archives – not
exhaustive –

has brought up a number of topics all of which have led me to the same

frustration with your approach that I am currently experiencing. The
ones I

recall now, which are probably not all of them, are (in no particular
order)

---

Information about the disturbance flowing through the perceptual signal
to

enable control to take place.

The perceptual function composed of an S-shaped response followed by an

integrator.

A discussion on bandwidth in relation to maximum realizable gain in a
control

system.

The “bomb” effect.

Flip-flops or cross-connections as explanations of category
perceptions,

association, contrast.

Categories as existing parallel to the analogue hierarchy.

Control system organization as being a model of the environment.

---

I finally realized that there is a common element in your treatment of
all

these subjects. It is very much like the way you took off on the basis
of

assuming that my limitation of the disturbance magnitude in Hans’ set
of

disturbances was due to insufficient output strength in my model, which
in

turn was caused by too short a word length. Having assumed the truth of
your

premise without particularly checking to see if it was true, you then
built a

series of plausible deductions from the assumption, which happened to
support

a general principle you were trying to get across. Unfortunately, the
premise

was false. I would not be surprised, however, if you decided that even
if the

premise happened to be false in that case, the deductions you made from
it

were probably true.

Â

In each of the above subjects, you began with a theoretical possibility
and

developed it just far enough to see some possible implications of it.
Then you

quickly built a plausible and ever-more-detailed series of deductions
from

those implications, and arrived at what seemed to you an interesting
new

phenomenon. You could see in your mind’s eye how the Bomb would sit
there

ticking, ready to go off if the right combination of disturbances
occurred.

You could imagine information flowing from the disturbance through the

perceptual system to the output, where it got used up in producing the
effects

that would counteract the disturbance. You could see the s-shaped
curves and

integrators acting like a perceptron for the input part of a control
system.

You could see a whole hierarchy of discrete categories with hysteresis,

running in parallel to the analog hierarchy. And the fact that you
could see

in principle how certain other phenomena might flow from the initial

conceptualizations was enough to convince you that the initial

conceptualizations must be correct.

Â

So what happens is that the tail wags the dog: the attractiveness and
richness

of the conclusions drawn from the initial assumptions convinces you
that the

initial assumptions must have been right. And once that has happened,
you

forget completely that the initial assumptions were never established
as true,

and you speak of the conclusions as if they were now established facts;
you

even start using them to prove other conclusions.

Â

The name of this type of reasoning process, or one name, is of course

"mathematics." In mathematics (including logic, or is it the
other way

around), it doesn’t matter whether the initial assumptions are
factually true

or in some way supportable by evidence. The assumptions are simply the
initial

process of setting up the chessboard with a problem, so you can work
out a

solution to it. Once the field of play is established, you can then
start

working out the theorems and proofs, encountering beauty and
entertainment at

many stages along the way. You begin to get a feel for the system you
have

created, so its major conclusions become familiar parts of that
conceptual

world. These major conclusions become theorems on which to build
further; they

get names like “information about the disturbance” and
“The Bomb” andÂ

"crossconnections."Â Since they have been derived by correct reasoning from the

premises, there is no reason to doubt them any more; they become real.
The

premises drop out of sight; they were never very important anyway,
except as a

way to get the game started. The real fun is in building the structure
of

ideas on those premises.

Â

Judging from various comments you have made about your interests and

preferences, I don’t think that this is a completely inappropriate
assessment

of your modus operandi. Your approach is not the engineering approach
to a

physical system, but the mathematical-logical approach to a
hypothetico-deductive

system.

Â

This hypothesis explains to me your disdain for “mere
demonstrations.” If you

have worked out the logic correctly, what is the point in doing an
actual

demonstration of it, and doing different demonstrations to bring out
one point

or another? If you understand addition, what is the point of
demonstrating

that 9 + 1 = 10, and 8 + 2 = 10, and so forth? If you understand the
complete

structure of information theory from Shannon on up, what is the point
in

demonstrating what you already know to be true: that the signals inside
a

control system must contain or pass along information about the
disturbance,

and that it is this information that makes control (and everything
else)

possible? And most important, if you have shown that there are no
logical

errors in reaching a conclusion about real behavior, what is the point
in

going through the labor of showing by direct experiment that the
conclusion

actually fits the data? If the data do not agree with the conclusion,
there

must have been some error or something unaccounted for in the
experiment.

That last if-then is the only way I can explain your reaction to
difficulties

when we actually try out some of your proposals. In the long
information

-in-perception debacle, we tried computing the reduction in the
uncertainty in

The perception, then in its first derivative, then both again with
temporal

shifts, and in every case the results disagreed with your deductions
about

what we should find. By rights, this should have brought you up short
and

caused you to question the very basis on which you built your
deductions. But

that didn’t occur: you simply abandoned the attempt to make a correct

deduction that would fit the data and turned to other subjects.

Â

If I had been in your shoes, I would have had to backtrack through the
logic

trying to find the error, and eventually (if no logical mistake could
be found

that would fix the problem) I would have gone all the way back to the
simple

starting premises on which the whole logical structure is built: if
there are

no mistakes in the logic, yet the conclusions do not fit observation,
then the

only place left to find an error is in the premises. And for me,
however

painful the decision, the only conclusion I could then reach is that
the

entire system is built on false-to-reality premises.

Â

When I went through the process of computing reduction in uncertainty
about

the disturbance due to the perceptual signal, under your tutelage, I
noticed a

fact, and mentioned it, that seemed significant to me. In the process
of

computing the conditional probabilities, I noticed that I would get the
same

conditional probabilities no matter in what order I did the sampling of
the

disturbance waveform. So in principle there was an infinity of
different

waveforms that would allow me to compute the same quantity of
information in

the perception. This made it very hard for me to see how the outcome
could be

an output waveform based on the “information” that was
arranged in the same

sequence as the elements of the disturbance waveform, which of course
is

necessary if the effect of the disturbance is to be canceled.

Your reply was brief and dismissive: you just compute the conditional

probabilities on pairs of successive values of the waveform, and get
the

probabilities of the first derivatives. But after thinking that over, I

realized that the same problem still existed: one could rearrange the
pairs

and get the same conditional probabilities. So how could the
information

passed in the perceptual signal possibly be responsible for producing
the

RIGHT output waveform?

Â

When I mentioned this (I am pretty sure I mentioned it), there was no
reply

that I recall. The failure to get the right results when we used the
first

derivatives as elements, even time-shifted, reinforced my doubts about
the

process, but not being an expert in information theory I did not feel

competent to ferret out the cause of the problem.

I now realize that you did not search for the cause of the problem by

backtracking through information theory. You just gave up on it. This
did not

solve the problem, but it left the intellectual structure of
information

theory in your head undisturbed. If PCT is correct, we can use this
phenomenon

to guess at the nature of the variable you were – and are –
controlling.

I remember getting a frantic phone call from Chris Love shortly after
the

start of the Little Baby project. He had tried to set up a big complex

hierarchy of control systems in which, per the boss’s suggestion, the

perceptual function was an S-shaped curve followed by an integrator.
The

reason he called was that he hadn’t been able to get even a single
elementary

control system to work. I tried to explain to him that a control system

organized that way would be trying to control a variable that was the
inverse

function of the proposed form, namely a nonlinear first derivative that
went

to infinity at zero and maximum perceptual signal. He was not then

knowledgeable about control theory, so I just suggested that he move
the

integrator to the output function, and preferably make the input
function

linear. He tried that, and got a working control system for the first
time,

several months into the project. I felt very sorry for Chris, because
he had

to try to make the suggested model work, and it could not work.

On other occasions, I have pointed out to you a shortcoming of the
perceptron

approach, in that it doesn’t yield perceptual signals which are
continuous

representations of controlled variables. The nonlinearities and other

properties limit the output to a yes-no signal, which is good only for

discrete control. However, in the fairly recent past, I noticed that
you were

still referring to the S-shaped input function with an integrator as
part of

the model. Chris’ problems do not seem to have shaken your faith one
bit. Or

perhaps they have simply led you to abandon that problem, and go to
modeling

discrete systems. Obviously it has not led you to re-examine the
premises

behind the perceptron approach.

---

I think that in deciding to be an abstract theoretician, you have
simply cut

off your higher level systems from lower-level perceptions, operating
the

higher-level systems in the imagination mode. And I think that this is
a

mistake. If you don’t continually check your higher-level models
against

experiences by interacting with the outside world at the lowest levels,
you

run the risk of creating a systematic delusion about the nature of the
world;

one that is internally consistent, but which is not consistent with
what your

senses could tell you if you consulted them. Abstract thought alone is
simply

not a reliable way to learn about nature.

Â

This is why I am so adamant about demonstrations and experiments. You
have to

close the loop through the external environment if you’re to achieve
real

control. No matter how self-evident or obvious or logically necessary a

conclusion may seem, it is still necessary to find a way to test it by

interacting with the world. And when you do such tests, it is necessary
to pay

attention to the outcome, because if the outcomes don’t agree with the
logic,

it says that something is wrong with the logic or with the premises on
which

it’s founded. No matter how convinced you are that you have the right
idea,

nature is perfectly capable of contradicting you.

Â

And this says something else, too. It says that there is really very
little

point in building up big deductive structures on premises that have not
been

experimentally demonstrated. Your cross-connection ideas about category

perception may prove to be quite right, but you have no way to verify
that

such cross connections exist or work in the ways you assume they work.

Technology has simply not reached the stage where we can do this in a
living

working brain. Perhaps it would be possible to do experiments to check,
at

least, the conclusions, to see if people actually work in the way that
your

hypothetical model works. But unless you can also check the premises,
you are

on very uncertain ground. For any circuit that accomplishes a given
result,

there are a dozen different ones that would do the same thing. There
will

always be uncertainties in our models, but why deliberately make them
as large

as possible?

---

I have no illusions about changing your style to correct what I see as

mistakes. What you make of what I say is in your hands alone. But if
you want

to understand where our frictions come from, you have to know how I
perceive

the way you work, and how limited it looks to me. You have to
understand that

even where you think you see agreement, you may be considering only a
narrow

range of meanings of what I or others say, meanings that fit your
world-view

but that may only represent one point of intersection of trajectories
that are

headed in different directions. And you have to realize that you often
read

hastily, making assumptions that a more careful reading would quickly
set

straight and then leaping ahead to draw unwarranted conclusions –
largely,

seeing agreement where there is actually no agreement, or only a very
partial

agreement. This is another penalty for working in the imagination mode.
You

are far from the only person to work this way, of course.

Obviously, I have considered only YOUR problems, not my own. I am sure
that

all of this looks quite different to you. If you want to turn the
tables, you

have every right.

Bill

–
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

[DagArchive]

[From Dag Forssell (2018.05.05.1610 PDT)]
Content-Type: multipart/alternative; boundary="000000000000d71aa8056b7c90f8"
X-Spam-Details: rule=notspam policy=default score=0 suspectscore=2
malwarescore=0
  phishscore=0 bulkscore=0 spamscore=0 mlxscore=0 mlxlogscore=999
  adultscore=0 classifier=spam adjust=0 reason=mlx scancount=1
  engine=8.0.1-1711220000 definitions=main-1805050219
X-Spam-Details: rule=cautious_plus_nq_notspam policy=cautious_plus_nq score=0
  suspectscore=2 malwarescore=0 phishscore=0 bulkscore=0 spamscore=0
  mlxscore=0 mlxlogscore=999 adultscore=0 classifier=spam adjust=0
  reason=mlx scancount=1 engine=8.0.1-1711220000 definitions=main-1805050219
X-Spam-Score: 0
X-Spam-Score: 0
X-Spam-OrigSender: csgnet-owner@lists.illinois.edu
X-Spam-OrigSender: rsmarken@gmail.com
X-Spam-Bar:
X-Spam-Bar:
X-Loop: csgnet@lists.illinois.edu
X-Sequence: 8524
Errors-to: csgnet-owner@lists.illinois.edu
Precedence: list
Precedence: bulk
X-no-archive: yes
List-Id: <csgnet.lists.illinois.edu>
List-Archive: <https://lists.illinois.edu/lists/arc/csgnet>
List-Help: <mailto:lists@lists.illinois.edu?subject=help>
List-Owner: <mailto:csgnet-request@lists.illinois.edu>
List-Post: <mailto:csgnet@lists.illinois.edu>
List-Subscribe: <mailto:lists@lists.illinois.edu?subject=subscribe%20csgnet>
List-Unsubscribe:
<mailto:lists@lists.illinois.edu?subject=unsubscribe%20csgnet>

[Rick Marken (2018.05.05.1500)]

Rick, your post strikes me as important but the formatting of Bill's
long post made it difficult to handle, methinketh.

I have taken the liberty of pulling the text into Word and formatting
it. I'll attach both the Word file and a pdf print. It may help
people print, read, digest, and copy for coming discussion.

Best, Dag

FrictionCSGnet.docx (22.5 KB)

FrictionCSGnet.pdf (152 KB)

···

Reply-To: csgnet@lists.illinois.edu

[rsmarken]

[Rick Marken 2018-05-05_16:25:19]

···

[From Dag Forssell (2018.05.05.1610 PDT)]

[Rick Marken (2018.05.05.1500)]

Rick, your post strikes me as important but the formatting of Bill’s long post made it difficult to handle, methinketh.

I have taken the liberty of pulling the text into Word and formatting it. I’ll attach both the Word file and a pdf print. It may help people print, read, digest, and copy for coming discussion.

Best, Dag

RM: Super! Thanks!Â

BestÂ

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

[MartinT]

[Martin Taylor 2018.05.05.23.07]

[From Rick Marken (2018.05.05.1500)]

Welcome back!

…
RM: First here is the list of misunderstandings
about PCT
that I posted in my “Minority Report”:

•         Organisms control only perceptions, not the aspects of the
  

  environment that correspond to those perceptions.

•         The test for the controlled variable is not an essential
  

  component of PCT-based research.

•         PCT shows that you can't control the behavior of another
  

  person.
• Social stability arises from interpersonal conflict.

•         The power law of movement is not an example of a
  

  behavioral
  illusion.
  …

  I wondered about these

when you posted them, mostly about why you thought that they had
been supported by people in a from that led to long discussions. I
had off-line discussion with a couple of others about who might be
the targets. All we could come up with was that the last three had
not been espoused by anyone we could think of. Boris seems to be
the only one we could think of for the first, except for occasions
when it was noted that many perceptions include inputs from
imagination to their perceptual functions. That leaves the second
“misunderstanding”, which depends on whether there is an implicit
“all” missing after the word “component.” If there is, then it
would be true that a lot of research based on PCT is defined as
not being “PCT-based research”. But that’s a matter of definition,
not of fact. What is “misunderstood” is only the meaning of
“PCT-based”.

  Anyway, I'm glad to know that you are feeling better.

  And thanks for posting that message from Bill. I had forgotten

that he was so volatile and variable in his expressed opinions.

  Martin

[MartinT]

[Martin Taylor 2018.05.06.11.27]

[From Rick Marken (2018.05.05.1500)]

Rick's posting of Pill Powers's 1995 message about his frustration with my ideas has led me to think a bit about how science advances, not just PCT, but science in general. In particular, it got me thinking about the need for variety in the evolutionary improvement of species, treating science, or rather, scientific theories, as species. Species that do not have variant phenotypes are highly liable to extinction when exposed to changing environments. Scientific theories that do not have variant forms are highly liable to elimination from the body of Science when new discoveries are made.

What constitutes a variant form of a theory, and when does a variant form become a new species? I don't think there is a sharp boundary, but there is a kind of measure, which is the degree to which predictions from one variant contradict predictions from another. This is quite different from species of discussants of the variants. When variant A and variant B are discussed by people who assume to be true, and control for perceiving them to be true, some elements of variant A that differ from corresponding elements of variant B, then discussions of the variants is likely to cause what Rick, and Bill in the quoted message, call "friction." Let's look at Bill's list of what he says caused friction between him and me 23 years ago. I think that's long enough ago for some of the heat of friction to have dissipated. I have numbered the list, which Bill did not, because I want to distinguish among types of disagreement.

···

==============
1. Information about the disturbance flowing through the perceptual signal to enable control to take place.
2. The perceptual function composed of an S-shaped response followed by an integrator.
3. A discussion on bandwidth in relation to maximum realizable gain in a control system.
4. The "bomb" effect.
5. Flip-flops or cross-connections as explanations of category perceptions, association, contrast.
6. Categories as existing parallel to the analogue hierarchy.
7. Control system organization as being a model of the environment.

Of these, only 5 and 6 represent variant forms of HPCT. 1, 3, and 4 represent the use of knowledge imported from other (physical) sciences for analysis of perceptual hierarchies. They are purely consequences of "standard" HPCT, at least they are if HPCT is to be considered a theory within "normal science" -- a proposition sometimes seemingly refuted by Rick. Irritant number 2 is a testable proposal within any form of PCT, based on the successful use (outside PCT) of that hypothesis in the prediction of an old unexplained effect. I don't know what 7 is, but in recent CSGnet discussions the consensus seems to have been that control system organization does, through evolution and individual reorganization, tend toward becoming a model of the environment, or rather, of the way the environment works, the many perceptual signals forming the model of the way the environment is. I would therefore eliminate 7 from further consideration.

Actually I don't know where Bill got the cited form of 2 from, but I can't remember claiming it. Perception is indeed based on integrating incoming data over some period of time, longer the higher the level in the hierarchy. I don't think that is contentious in any way. The "S-form" is one way of suggesting a quasi-logarithmic form for positive values with a threshold at low levels, but the non-linearity and the integration are very unlikely to be sequential, at least not in the order Powers suggests. Powers and Marken even invoked the quasi-logarithmic form of the perceptual output to account for the "Stevens Power Law" of perceptual magnitude.

In its influence on control, perceptual integration usually happens very fast compared to the loop integration, which we usually model as occurring in the output function (for very good information-theoretic reasons), so it can usually be ignored in all but the most precise modelling with no ill effects. Likewise, the "S-shape", like any other non-linearity, doesn't much, if at all, influence the quality of control. The non-linearity must be sought in different kinds of study, such as, for example, measuring neural impulse rates or making the kinds of prediction Marken and Powers did.

As for the main body of Bill's message, I plead guilty as charged to the offence of trying to think for myself rather than following slavishly the results of Bill's thought. My reasoning is that if you want to understand a wheel, the best way is to try to reinvent it. When my thinking came up with results that Bill's independent thought had not (yet) produced, I did (and do) propose them as consequences of or as alternative possibilities to the "standard" accepted consequences of a particular form of HPCT. Bill was not and should not be the only person allowed this privilege.

I do not plead guilty to two other charges Bill levelled and Rick highlighted: that (1) when I find a possible consequence, I treat it as the only possible consequence, and (2) that I disdain experiments and demonstrations. Here are some separate paragraphs from the current draft of the preface to the book on which I am working, about PCT and its very wide ramifications. (1) is the very first paragraph of the book, and they are all in the short Preface. Make of them what you will.

====1====
It takes a certain genius for a person to create something about which other people say to themselves “How obvious. Why didn’t I think of that" Once you understand it, you cannot easily go back to your previous way of seeing the world. Perceptual Control Theory (PCT) is a creation of that kind. However, simple exposure is not by itself sufficient for one to “see it", as my own experience attests. You have to explore it for yourself, and you probably will not do that unless you have some reason to believe the exercise will be worth the effort.
====2=====
This book should be read with a critical attitude throughout. Niels Bohr said something along the lines of: “Every statement I make should be treated as a question," and his comment applies to this whole book. You should always be thinking about in what ways its deductions might be wrong, and what evidence might be sought by direct experiment or from other research domains to falsify any of the claims, hypotheses, and proposals scattered throughout. Particularly after Part 5 much of what is stated as though it is established fact is really unsupported by experiment or mathematical analysis, but seems to follow from the better supported earlier material.
======3====
It is said that when you have only a hammer, everything looks like a nail. Is PCT such a “hammer"? I think it is more like a Swiss Army knife that includes a hammer, a laser rangefinder and a smartphone along with all the other tools. In this book we consider neural synapses and Hebbian learning, how dialogues stay on track, how systems can be stable despite conflict, as well as why Euro-American economies have steadily declined over the last few decades, why Northern European politics are more gender-equal than Southern (an effect of climate change some 5 or 6 thousand years ago), why conspiracy theories seem to be believed more readily than the truth, and much else besides. Everything is derived directly from the (retrospectively) simple insight of Bill Powers, the Swiss Army knife of psychology, that everything we do intentionally is to control some perception or perceptions.
=======4====
Throughout, I have tried to claim only what seems to follow from what can be supported, but over the history of science, similar conjectures have many times been proven false. I hope this is not the case in too many places, but I do warn the reader to guard carefully against the possibility, by thinking for herself about the issues discussed.

======end sample paragraphs==========

I think that PCT also suggests why science pursued with the attitude suggested by these paragraphs will always engender conflicts with those who control for particular hypotheses (perhaps their own) to be the only correct possibilities for explaining some phenomenon. Therein lies the "friction" spoken of by both Bill in 1995 and Rick in 2018. It's a friction between what in politics might be called a "liberal" and a "conservative" approach to the world.

Martin

[_Boris_Hartman]

He, he… old Rick is back… with his RCT (Ricks’ Control Theory). I thought for a moment that “PCT Rick” will appear not “RCT Rick”. You didn’t change, so we’ll go on with our 5 years old discussion. It’s Anniversary. Obviously we’ll be doing this until “death” appart us. I hope you underdstand that you haven’t got a chance from the beggining. By my oppinion you’ll never prove that Bill was wrong. Scientifically…

···

From: Richard Marken rsmarken@gmail.com
Sent: Sunday, May 6, 2018 12:00 AM
To: csgnet@lists.illinois.edu; mol@mail-list.com
Cc: Richard Marken rsmarken@gmail.com
Subject: Approaches to PCT and Friction on CSGNet

[From Rick Marken (2018.05.05.1500)]

RM: I must admit that when I posted my “Minority Report”, announcing that I would stop unsolicited posting to CSGNet,

HB : It would be good if you keep your word. A gentleman word. Otherwise your words a worthless Rick. You don’t have any credibility because we already know that you are working just for your interest. But that’s what “LCS” are designed for. Aren’t they ? Did it ever occur to you that it would be good if you start with proffesional posting about PCT on CSGnet. CSGnet is meant for Bill and Mary memory not yours. Even Bruce Nevin noticed that.

BN earlier : We are here to investigate, test, demonstrate, and promulgate perceptual control theory.

HB : Why shouldn’t you listen to your friend.

RM : I was busy feeling rather sorry for myself.

HB: It was you who is responsable for your state of mind. Only you can make right decissions that you’ll not feel bad.

RM : Since Bill passed away I have been alternately surprised, frustrated, angered and hurt by the response to my posts to CSGNet.

HB : It’s no wonder that we criticized you. You tryed to change PCT (Perceptual Control Theoy) into RCT (Ricks’ Control Theory) and you are still trying to do so. See archives and my answers to your RCT assumptions.

RM : But over the last few days I’ve had what I think was a nice “up a level” experience,

HB : Really ? Where can we see the advancement ? You are explaining again your RCT theory for who knows which time. And I’m answering with the same Bills’ arguments. I hope you don’t think that you can “beat” Bills’ arguments. So where is a “up a level” experience ?

RM : …which basically involved realizing that one of my highest level goals is to present an accurate account of PCT that can be the basis of furthering PCT science;

HB : It can’t be seen that you could have and higher level PCT goals. But you coulèd change them from RCT to PCT.

When will you understand Rick that your RCT is confusing all arround not just here on CSGnet but also in public. You don’t care for PCT, do you ? Only your interests ? You are not presenting PCT you are presenting RCT. I proved you for who knows how many times.

RM : …and that I realized that I can achieve this goal by, among other things, posting to CSGNet, regardless of how well those posts are received.

HB : You can post whatever you want, but it would be good if you start posting in PCT manner. And it would be good if you show some evidences for your “RCT” theory.

RM: So I will start my return to unsolicited posting by explaining why the “fundamental misunderstandings” about PCT that I mentioned in my “Minority Report” are, indeed, fundamental misunderstandings.

HB : It seems that you are the one who causes misunderstanding.

RM : (This post is not really unsolicited; after giving me some MOL therapy, my co-author on “Controlling People”, Tim Carey, suggested that it would be useful if I would post an explanation of why these misunderstandings about PCT are, indeed, misunderstandings, and maybe even post a copy of it to the MOL list.) So here it goes:

HB : Well I don’t know what kind of treatment you got, but I would say this is not good solution. We went through these problems so many times that you should reorganize in other direction in PCT direction. I would advice you to start with PCT arguments. I personaly think that with explaining your RCT theory you are directly opposing Bill, your teacher. Your theoretical arguments (RCT) are just opposite to Bills’.

RM: First here is the list of misunderstandings about PCT that I posted in my “Minority Report”:

• Organisms control only perceptions, not the aspects of the environment that correspond to those perceptions.

HB : What is not clear about this one ? Bills general diagram (LCS III) shows that organisms can generally only affect environment. He proved that also with physiological means. It’s on you to prove that you can “Control behavior” and of course that there is some “Controlled Perceptual Variable” or PCV. You didn’t do it till now and you are talking about misunderstanding. You are causing misunderstanding by not providing evidences for your theory.

Organisms generally control only perceptions but effects of control in organisms can be seen in environment. How interpretations are made of those effects it’s up to individual. Diagram LCS III is more than clear about this one. No misunderstandings as PCT is concerned.

• The test for the controlled variable is not an essential component of PCT-based research.

HB : You wrote it for yourself.

RM (2013) : But the intentional behavior that occurs in real life often involves the control of variables that are impossible to represent as simple function of physical variables, e.g., the honesty of a communication or the intimacy of a realtionship. A quantitative approcah to the TCV will not work when trying to study such abstract variables….

• PCT shows that you can’t control the behavior of another person.

HB : I think Alison agrees with this. And you wrote a book about how people are in control and they can try to control people. If I understood right.

• Social stability arises from interpersonal conflict.

HB : Where did you get this one ? I think I saw in Kent literature that social stability is matter of cooperation and conflicts.

• The power law of movement is not an example of a behavioral illusion.

HB : Well I admitt I didin’t study the case but I beleived Martin and others…

RM: I believe that all of these misunderstandings are a result from taking a “mathematical-logical” rather than a “scientific/engineering” approach to PCT.

HB : You forgot biological-physiological approach about how organisms function. PCT is about real substance of life not imaginational one as your “RCT” is. It’s about how organisms really function not how in demos about behavior are produced. Whether you understand how organisms function or you don’t. It’s doesn’t depend from approach. Now again how PCT function from physiological view.

Bill P :

Our only view of the real world is our view of the neural signals that represent it inside our own brains. When we act to make a perception change to our more desireble state – when we make perception of the glass change from “on the table” to " near the mouth" – we have no direct knowledge of what we are doing to the reality that is the origin of our neural signal; we know only the final result, how the result looks, feels, smells, sounds, tastes, and so forth.

HB : Do you understand what PCT is about ?

RM : The mathematical-logical approach treats PCT as a set of axioms from which conclusions about behavior are derived like mathematical theorems. The scientific/engineering approach, on the other hand, treats PCT as a set of tentative guesses about the mechanisms that explain the behavior that is actually observed. The correctness of PCT explanations of behavior is, therefore, evaluated very differently by these two approaches. The mathematical-logical approach evaluates the correctness of these explanations in terms of whether they are properly derived from the theoretical “axioms” .The scientific/engineering approach, on the other hand, evaluates the correctness of these explanations by testing to see whether they correspond to what is actually observer in appropriate tests or demonstrations.

HB : It doesn’t matter from which approach you are observing behavior. It’s important that you understand how organisms function. And obviously you don’t. You are making also wrong tests and demos. .

RM: It’s easy to see how taking the mathematical-logical approach to PCT could lead one to conclude that the five misunderstandings listed above are actually correct implications about behavior from a PCT perspective.

HB : As I said. You are confussing things. Behavior is result of organisms functioning. And PCT approach deals with this problem with all kind of approaches. So I think PCT is the best explanation of how organisms function. You should spread your reading to many books from different sciences like Bill did.

RM : For example, PCT does say that organisms control only a perceptual signal. This perceptual signal is not the same as the aspect of the environment to which it corresponds. So it’s logically correct to conclude that PCT says that organisms control only perceptions, not the aspects of the environment that correspond to those perceptions. And the logical corollary to this is that one cannot necessarily tell what perception an organism is controlling by testing to see whether a particular aspect of the environment is being controlled. So the test for the controlled variable, which involves only testing to see whether certain aspects of the environment are under control, can’t possibly be an essential component of PCT because there is no necessary relationship between any aspect of the environment that might be under control and the perceptual variables that the organism is actually controlling.

HB : So what’s wrong with PCT logic ?

RM: Because PCT says that a person’s behavior involves control of a perceptual signal that is accessible only to the person themselves it is also logical to conclude that it is impossible to control that person’s behavior. And PCT also shows that, under the appropriate conditions, when several control systems control the same perceptual variable relative to somewhat different reference levels, there will be a conflict but the perceptual variable will be maintained at virtual reference level that is the average of the reference levels of all the systems involved in the conflict. Since the perceptual variable is being stabilized by a “society” of control systems, it is logical to conclude that social stability arises from interpersonal conflict.

HB : As much as from cooperation if I understood Kent right, that’s the main conclussions Kent showed with his experiments. And it’s stability, not controlling “controlled variables” in environment.

RM: Finally, PCT says that there will be a consistent, negative relationship between the disturbance to a controlled variable…

HB : Where does PCT says that ? Disturbance to which “controlled variable” ?

RM : …and the actions that compensate for the effect of that disturbance.

HB : Actions that compensate which effect of this disturbances ? Disturbances to “controlled variable” in outer environment ??? It seems that you have to read some things 1000 x to remember.

Bill P (B:CP):

CONTROL : Achievement and maintenance of a preselected state in the controlling system, through actions on the environment that also cancel the effects of disturbances.

HB : Do you understand now what control is in PCT and what is compensated ? It’s still old Ricks story about “controlling in environment”. Boring…

RM : The power law of movement seems to be an example of such a disturbance-action relationship; the power law is the observation that when we make curved movements we slow down in proportion to the degree of curvature through which we are moving. So it looks like variations in the speed of movement are actions that compensate for the disturbance of variations in the degree of curvature through which the movement is being made. So the power law is not a behavioral illusion because that relationship is not assumed to reflect characteristics of the organism that transform the “stimulus” of curvature in to “response” of movement speed.

RM: So all five misunderstandings of PCT listed above are perfectly logical deductions about behavior when one takes the mathematical-logical approach to PCT.

HB : There is no misunderstandings of PCT listed above. Everything is in accordance to PCT definitions and diagram :

PCT Definitions of control loop :

Bill P (B:CP):

1. CONTROL : Achievement and maintenance of a preselected state in the controlling system, through actions on the environment that also cancel the effects of disturbances.

Bill P (B:CP):

1. OUTPUT FUNCTION : The portion of a system that converts the magnitude or state of a signal inside the system into a corresponding set of effects on the immediate environment of the system

Bill P (LCS III):…the output function shown in it’s own box represents the means this system has for causing changes in it’s environment.

Bill P (LCS III):

1. FEED-BACK FUNCTION : The box represents the set of physical laws, properties, arrangements, linkages, by which the action of this system feeds-back to affect its own input, the controlled variable. That’s what feed-back means : it’s an effect of a system’s output on it’s own input.

Bill P (B:CP) :

1. INPUT FUNCTION : The portion of a system that receives signals or stimuli from outside the system, and generates a perceptual signal that is some function of the received signals or stimuli.

Bill P (B:CP) :

1. COMPARATOR : The portion of control system that computes the magnitude and direction of mismatch between perceptual and reference signal.

Bill P (B:CP)

1. ERROR : The discrepancy between a perceptual signal and a reference signal, which drives a control system’s output function. The discrepancy between a controlled quantity and it’s present reference level, which causes observable behavior.

Bill P (B:CP) :

1. ERROR SIGNAL : A signal indicating the magnitude and direction of error.

RM : They can only be seen as misunderstandings when one takes the scientific/engineering approach to PCT.

HB : Speccially they are not seen as misunderstanding from the point of scientific/enginnering approach.

RM : From the scientific engineering perspective the idea that organisms control only perceptions, not the aspects of the environment that correspond to those perceptions, is instantly seen as having things completely backwards; the idea that organisms control perceptions is a theory designed to account for the observation that organisms can be seen to control aspects of their environment.

HB : It’s a little bit unclear but do I understand right that organisms can be seen “to control aspects of environment”. Well if somebody sees it like that it can be seen. But you must understand the general logic of PCT. It accounts for all behaviors thst organism can produce not just those which have “controlled aspect of environment”. Nobody said that effects of internal control are not seen in environment. But they are just effects not “controlled effects”.

What we know for a fact is that organisms control aspects of their environment; we guess that this happening because organisms are controlling perceptual signals that are analogs of variations in the aspects of the environment that we see being controlled.

HB : And when will you prove this fact that “organisms control aspect of environment” ??? Is this general fact ? Are you talking about PCT ??? What you see being controlled is your problem. That’s why we have to cosider your private “RCT” theory.

RM: We use the test for the controlled variable to determine the aspects of their environment that organisms control. So the idea that the test for the controlled variable is not essential because the results do not necessarily reveal the aspects of the environment that correspond to the perceptions that are controlled is nonsense from the point of view of the scientific/engineering approach to PCT. The test for the controlled variable is the only basis we have for determining what perceptual variables an organism might be controlling. And it is easy to demonstrate (http://www.mindreadings.com/ControlDemo/Mindread.html) that the results of the test for the controlled variable provide a very reliable indication of what aspects of the environment are under control

HB : In PCT “aspects of environment” are not under control. Organisms inside "aspects of environment are under control. So whtever is happening in outer environment is being affected by control inside organism. It’s t which aspects of environment are being controlled but which aspects of organisms inside structure are being under control. You are violating definition of control in PCT.

Bill P (B:CP):

CONTROL : Achievement and maintenance of a preselected state in the controlling system, through actions on the environment that also cancel the effects of disturbances.

RM : ….and, thus, what perceptions we should imagine are being be controlled; the theoretical perceptions we imagine to be controlled correspond exactly to the aspects of the environment that we have found to be controlled using the test for the controlled variable.

HB : I showed you with Maturana’s experiments that “perceptions we imagine” does not “correspond” to the aspects of the environment. This fact can be aslo confirmed by Bills writing in B:CP. Show us any proof that “perceptions” exactly correspond to aspects of environment and which “Controlled behaviors” are found to be controlled when “using the test for the controlled variable” ???

And again your nonsense logic about “behavior is control”, “controlled aspect of environment” and “Controlled Perceptual Variable”. You said it for your self that TCV is not reliable method for understanding what people are controling inside.

RM (2013) : But the intentional behavior that occurs in real life often involves the control of variables that are impossible to represent as simple function of physical variables, e.g., the honesty of a communication or the intimacy of a realtionship. A quantitative approcah to the TCV will not work when trying to study such abstract variables….

HB : People can lie, manipulate, trick you…

RM: A crucial difference between the mathematical-logical and scientific/engineering approach to PCT to in the attitude toward observation and test. Those who take the mathematical-logical approach to PCT express a particular disdain for observation and test in the form of what are called the PCT “demos”, such as the computer demos at http://www.mindreadings.com/demos.htm. But these demos have been an essential part of the scientific/engineering approach to PCT from the very start,

HB : Demos were your part of PCT approach and has almost nothing to do with any scientific approcah. That was clearly seen in our discussions (between Bill, you and me) about “baseball catch” when Bill thought you applayed scietific approcah to “baseball catch”. Oh, how he was dissapointed when you told him that it was your pure imagination. You forgot on first rule in scientific approach as PCT is concerned. Final arbiter is nature. Remember ? If you don’t remember you have CSGnet archives to refresh your memory.

Most of your demos are your imaginational constructs and wouldn’t survive “nature critics”. Your RCT thinking is wrong (behavioristic). You are educated psychologist. I showed you quite some times that your RCT doesn’t hold the “final natural arbitrage” (school system, tennis game, sheep-dog…). Go through archives.

RM : …as can be seen in the lovely “portable demos” described in the 1960 Powers, Clark and McFarland article “A General Feedback Theory of Human Behavior: Part II” reprinted in Living Control Systems, pp 25-45. Using demos such as this it is easy to show that you can control the behavior of another person (http://www.mindreadings.com/ControlDemo/BehavioralControl.html) and that that the power law of movement is an example of a behavioral illusion (as shown by the demo described in the “Facts of Logic” section of Marken and Shaffer, 2018 (https://www.dropbox.com/s/3m51ko4vs1xdult/MarkenShafferReappraisal.pdf?dl=0).

HB : Again EPP (economic propaganda) for your worthless, behavioristic demos. When will you stop promoting behaviorism and “manipulations” with “controlled variables” in outer environment. When will you prove that your RCT theory holds ? I don’t know what you meant by “scientific/engineering” approach, but I do know that you don’t know anything about how orgsnism function as you prove in your worthless M/S article about “Power Law”. You are again “infecting” CSGnet with “behaviorism”. Your RCT theory is wrong :

RCT (Ricks Control Theory) definition of control loop

1. CONTROL : Keeping of some »aspect of outer environment« in reference state, protected (defended) from disturbances.

2. OUTPUT FUNCTION : controlled effects (control of behavior) to outer environment so to keep some »controlled variable« in reference state

3. FEED-BACK FUNCTION : »Control« of some »aspect of outer environment« in reference state.

4. INPUT FUNCTION : produce »Controlled Perceptual Variable« or »Controlled Perception«, the perceptual correlate of »controlled q.i.«

5. COMPARATOR : ???

6. ERROR SIGNAL : ???

RM: The scientific/engineering approach to PCT doesn’t deny that social stability can arise from interpersonal conflict. However, there is, to my knowledge, no demonstration of that it can.

HB : You are not reading enough real PCT material.

RM : Such a demonstration would involve showing that groups of real organisms produce stable results like those produced by the simulations of multiple control systems control the same variable relative to different references. Until such demonstrations are produced, the idea that social stability can arise from interpersonal conflict must remain a conclusion that is true only from the perspective of the mathematical-logical approach to PCT.

HB : Are you talking hypothetically ?

RM: I’ll conclude this already rather long post with a copy of an even longer post from Bill Powers. This post was sent to me by someone who is not on CSGNet but is one of the few people who respect my work on PCT (and the feeling is mutual). It’s a post from 1995 where Bill provides and analysis of why he thinks there is so much friction between himself and Martin Taylor. I’m posting it to show that Bill’s approach to PCT was most emphatically the scientific/engineering approach. And that he was not a big fan of the mathematical-logical approach. I have bolded some of the sections that I think are particularly germane to the problems I have been having in discussions on CSGNet. Like Bill, who says at the end of this post “I have no illusions about changing your style to correct what I see as mistakes”, I also have no illusions about changing the approach to PCT of those on CSGNet to correct what I see as mistakes.

HB : Yes Rick. Bill supported you whatever you wrote or said. But I didn’t. And if you’ll list through History of CSGnet archives you’ll find many places where I criticised your “RCT” even in time when Bill was with us.

RM : But I hope this helps you see why the friction on CSGNet exists.

HB : It will not help you see that your RCT is very wrong idea about how organisms function, specialy how “nervous system” function. You proved in your M/S article your full ignorancy about biological and physiological facts. You have no knowledge and no clue to understand how organisms function. And until you’ll not understand the nature of orgsnisms you’ll not also understand PCT.

Boris

========================================

Date: Wed, 17 May 1995 05:24:50 -0600

Subject: Friction

Hello, Martin – (no CCs)

I have become increasingly frustrated with our communications and have been

trying to figure out what is wrong. In the middle of the night a possibility

occurred to me. A bit of browsing through the archives – not exhaustive –

has brought up a number of topics all of which have led me to the same

frustration with your approach that I am currently experiencing. The ones I

recall now, which are probably not all of them, are (in no particular order)

---

Information about the disturbance flowing through the perceptual signal to

enable control to take place.

The perceptual function composed of an S-shaped response followed by an

integrator.

A discussion on bandwidth in relation to maximum realizable gain in a control

system.

The “bomb” effect.

Flip-flops or cross-connections as explanations of category perceptions,

association, contrast.

Categories as existing parallel to the analogue hierarchy.

Control system organization as being a model of the environment.

---

I finally realized that there is a common element in your treatment of all

these subjects. It is very much like the way you took off on the basis of

assuming that my limitation of the disturbance magnitude in Hans’ set of

disturbances was due to insufficient output strength in my model, which in

turn was caused by too short a word length. Having assumed the truth of your

premise without particularly checking to see if it was true, you then built a

series of plausible deductions from the assumption, which happened to support

a general principle you were trying to get across. Unfortunately, the premise

was false. I would not be surprised, however, if you decided that even if the

premise happened to be false in that case, the deductions you made from it

were probably true.

In each of the above subjects, you began with a theoretical possibility and

developed it just far enough to see some possible implications of it. Then you

quickly built a plausible and ever-more-detailed series of deductions from

those implications, and arrived at what seemed to you an interesting new

phenomenon. You could see in your mind’s eye how the Bomb would sit there

ticking, ready to go off if the right combination of disturbances occurred.

You could imagine information flowing from the disturbance through the

perceptual system to the output, where it got used up in producing the effects

that would counteract the disturbance. You could see the s-shaped curves and

integrators acting like a perceptron for the input part of a control system.

You could see a whole hierarchy of discrete categories with hysteresis,

running in parallel to the analog hierarchy. And the fact that you could see

in principle how certain other phenomena might flow from the initial

conceptualizations was enough to convince you that the initial

conceptualizations must be correct.

So what happens is that the tail wags the dog: the attractiveness and richness

of the conclusions drawn from the initial assumptions convinces you that the

initial assumptions must have been right. And once that has happened, you

forget completely that the initial assumptions were never established as true,

and you speak of the conclusions as if they were now established facts; you

even start using them to prove other conclusions.

The name of this type of reasoning process, or one name, is of course

"mathematics." In mathematics (including logic, or is it the other way

around), it doesn’t matter whether the initial assumptions are factually true

or in some way supportable by evidence. The assumptions are simply the initial

process of setting up the chessboard with a problem, so you can work out a

solution to it. Once the field of play is established, you can then start

working out the theorems and proofs, encountering beauty and entertainment at

many stages along the way. You begin to get a feel for the system you have

created, so its major conclusions become familiar parts of that conceptual

world. These major conclusions become theorems on which to build further; they

get names like “information about the disturbance” and “The Bomb” and

“crossconnections.” Since they have been derived by correct reasoning from the

premises, there is no reason to doubt them any more; they become real. The

premises drop out of sight; they were never very important anyway, except as a

way to get the game started. The real fun is in building the structure of

ideas on those premises.

Judging from various comments you have made about your interests and

preferences, I don’t think that this is a completely inappropriate assessment

of your modus operandi. Your approach is not the engineering approach to a

physical system, but the mathematical-logical approach to a hypothetico-deductive

system.

This hypothesis explains to me your disdain for “mere demonstrations.” If you

have worked out the logic correctly, what is the point in doing an actual

demonstration of it, and doing different demonstrations to bring out one point

or another? If you understand addition, what is the point of demonstrating

that 9 + 1 = 10, and 8 + 2 = 10, and so forth? If you understand the complete

structure of information theory from Shannon on up, what is the point in

demonstrating what you already know to be true: that the signals inside a

control system must contain or pass along information about the disturbance,

and that it is this information that makes control (and everything else)

possible? And most important, if you have shown that there are no logical

errors in reaching a conclusion about real behavior, what is the point in

going through the labor of showing by direct experiment that the conclusion

actually fits the data? If the data do not agree with the conclusion, there

must have been some error or something unaccounted for in the experiment.

That last if-then is the only way I can explain your reaction to difficulties

when we actually try out some of your proposals. In the long information

-in-perception debacle, we tried computing the reduction in the uncertainty in

The perception, then in its first derivative, then both again with temporal

shifts, and in every case the results disagreed with your deductions about

what we should find. By rights, this should have brought you up short and

caused you to question the very basis on which you built your deductions. But

that didn’t occur: you simply abandoned the attempt to make a correct

deduction that would fit the data and turned to other subjects.

If I had been in your shoes, I would have had to backtrack through the logic

trying to find the error, and eventually (if no logical mistake could be found

that would fix the problem) I would have gone all the way back to the simple

starting premises on which the whole logical structure is built: if there are

no mistakes in the logic, yet the conclusions do not fit observation, then the

only place left to find an error is in the premises. And for me, however

painful the decision, the only conclusion I could then reach is that the

entire system is built on false-to-reality premises.

When I went through the process of computing reduction in uncertainty about

the disturbance due to the perceptual signal, under your tutelage, I noticed a

fact, and mentioned it, that seemed significant to me. In the process of

computing the conditional probabilities, I noticed that I would get the same

conditional probabilities no matter in what order I did the sampling of the

disturbance waveform. So in principle there was an infinity of different

waveforms that would allow me to compute the same quantity of information in

the perception. This made it very hard for me to see how the outcome could be

an output waveform based on the “information” that was arranged in the same

sequence as the elements of the disturbance waveform, which of course is

necessary if the effect of the disturbance is to be canceled.

Your reply was brief and dismissive: you just compute the conditional

probabilities on pairs of successive values of the waveform, and get the

probabilities of the first derivatives. But after thinking that over, I

realized that the same problem still existed: one could rearrange the pairs

and get the same conditional probabilities. So how could the information

passed in the perceptual signal possibly be responsible for producing the

RIGHT output waveform?

When I mentioned this (I am pretty sure I mentioned it), there was no reply

that I recall. The failure to get the right results when we used the first

derivatives as elements, even time-shifted, reinforced my doubts about the

process, but not being an expert in information theory I did not feel

competent to ferret out the cause of the problem.

I now realize that you did not search for the cause of the problem by

backtracking through information theory. You just gave up on it. This did not

solve the problem, but it left the intellectual structure of information

theory in your head undisturbed. If PCT is correct, we can use this phenomenon

to guess at the nature of the variable you were – and are – controlling.

I remember getting a frantic phone call from Chris Love shortly after the

start of the Little Baby project. He had tried to set up a big complex

hierarchy of control systems in which, per the boss’s suggestion, the

perceptual function was an S-shaped curve followed by an integrator. The

reason he called was that he hadn’t been able to get even a single elementary

control system to work. I tried to explain to him that a control system

organized that way would be trying to control a variable that was the inverse

function of the proposed form, namely a nonlinear first derivative that went

to infinity at zero and maximum perceptual signal. He was not then

knowledgeable about control theory, so I just suggested that he move the

integrator to the output function, and preferably make the input function

linear. He tried that, and got a working control system for the first time,

several months into the project. I felt very sorry for Chris, because he had

to try to make the suggested model work, and it could not work.

On other occasions, I have pointed out to you a shortcoming of the perceptron

approach, in that it doesn’t yield perceptual signals which are continuous

representations of controlled variables. The nonlinearities and other

properties limit the output to a yes-no signal, which is good only for

discrete control. However, in the fairly recent past, I noticed that you were

still referring to the S-shaped input function with an integrator as part of

the model. Chris’ problems do not seem to have shaken your faith one bit. Or

perhaps they have simply led you to abandon that problem, and go to modeling

discrete systems. Obviously it has not led you to re-examine the premises

behind the perceptron approach.

---

I think that in deciding to be an abstract theoretician, you have simply cut

off your higher level systems from lower-level perceptions, operating the

higher-level systems in the imagination mode. And I think that this is a

mistake. If you don’t continually check your higher-level models against

experiences by interacting with the outside world at the lowest levels, you

run the risk of creating a systematic delusion about the nature of the world;

one that is internally consistent, but which is not consistent with what your

senses could tell you if you consulted them. Abstract thought alone is simply

not a reliable way to learn about nature.

This is why I am so adamant about demonstrations and experiments. You have to

close the loop through the external environment if you’re to achieve real

control. No matter how self-evident or obvious or logically necessary a

conclusion may seem, it is still necessary to find a way to test it by

interacting with the world. And when you do such tests, it is necessary to pay

attention to the outcome, because if the outcomes don’t agree with the logic,

it says that something is wrong with the logic or with the premises on which

it’s founded. No matter how convinced you are that you have the right idea,

nature is perfectly capable of contradicting you.

And this says something else, too. It says that there is really very little

point in building up big deductive structures on premises that have not been

experimentally demonstrated. Your cross-connection ideas about category

perception may prove to be quite right, but you have no way to verify that

such cross connections exist or work in the ways you assume they work.

Technology has simply not reached the stage where we can do this in a living

working brain. Perhaps it would be possible to do experiments to check, at

least, the conclusions, to see if people actually work in the way that your

hypothetical model works. But unless you can also check the premises, you are

on very uncertain ground. For any circuit that accomplishes a given result,

there are a dozen different ones that would do the same thing. There will

always be uncertainties in our models, but why deliberately make them as large

as possible?

---

I have no illusions about changing your style to correct what I see as

mistakes. What you make of what I say is in your hands alone. But if you want

to understand where our frictions come from, you have to know how I perceive

the way you work, and how limited it looks to me. You have to understand that

even where you think you see agreement, you may be considering only a narrow

range of meanings of what I or others say, meanings that fit your world-view

but that may only represent one point of intersection of trajectories that are

headed in different directions. And you have to realize that you often read

hastily, making assumptions that a more careful reading would quickly set

straight and then leaping ahead to draw unwarranted conclusions – largely,

seeing agreement where there is actually no agreement, or only a very partial

agreement. This is another penalty for working in the imagination mode. You

are far from the only person to work this way, of course.

Obviously, I have considered only YOUR problems, not my own. I am sure that

all of this looks quite different to you. If you want to turn the tables, you

have every right.

Bill

–

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