Say ya wanna revolution

CSGnet Archive CSG1996
1

[From Rick Marken (960624.1600)]

Me:

If we know the form of f(), we can determine the value of d(t) from our
measurements of p(t) and o(t). We can then say (as Martin says) that we are
able to do this because there is "information about d(t) in p(t)". This is
obviously just a weird way of saying that, if we know that p(t) = f(d(t)+
o(t)) and we know o(t), p(t) and f(), then we can solve for d(t).

Bruce Gregory (960624.1555 EDT) --

I'm not sure why you find this wording "wierd"

You're right. I guess it's not really weird to say it that way. If I know
that 2X = Y then when you tell me the value of X you are giving me
information about Y. I can handle that.

I don't think I would have flown off the handle about this "information"
stuff if Martin had said, right off, that the perceptual signal (analogous
to X in the above equation) contains information that a control engineer can
use to determine the effective disturbance (Y in the equation). I went
bananas because it seemed like Martin was saying that the perceptual signal
contains (or carries) information that the control system itself uses to
determine the effective disturbance acting on the controlled variable.

Me:

The only reason I can imagine why people might want to believe that control
systems work by "extracting" or "using" or "consuming" information in p(t)
is if they want to maintain the idea that input _drives_ or _guides_
behavior.

Bruce:

I don't perceive this motive in Martin's writings, but he can plead guilty
or innocent without my help.

I don't think it's a conscious motive (purpose); and that might not be
Martin's purpose at all. Another, more likely, possibility is that Martin's
purpose is to perceive the relevance of a beloved theory (information theory)
to another beloved theory (PCT). Many people do this -- try to find a place
for one or another of their favorite theories in PCT (or vise versa). I have
never seen a case where this can be done without violating some fundamental
fact about how perceptual control systems work. But I sympathize with efforts
to do this. I tried it myself for the first couple of years I was learning
PCT.

The problem with trying to reconcile PCT with theories that _seem_
consistent with PCT (from my point of view) is that it keeps one from
striking off in the new direction that PCT requires. I think it limits the
contribution one can make to developing PCT as a science. Not that Martin
hasn't made some wonderful contributions to PCT; I just have a feeling
that such well-meaning attempts at reconciling irreconcilable theories
prevents one from diving into PCT "all the way".

I think there is evidence that attempts to reconcile PCT with non-PCT
(typically input-output) appoaches to understanding behavior have stifled the
development of PCT science. For example, Jeff Vancouver's Psych Bulletin
paper describes research by a _huge_ collection of people who are presumably
studying purposive ("goal-oriented") behavior. But all of these people assume
(unconsciously) that the study of purposive behavior can be done in the
context of the beloved cause-effect model research. Of course, it can't (for
reasons Jeff [Jeff Vancouver (960624.1600)] apparently finds insufferably
boring -- and I have repeated these reasons enough that they surely are
boring by now) so tons of time and money are being wasted by people who think
that they are contributing to the PCT database when they are not. I think
this is most unfortunate.

I myself didn't start doing worthwhile PCT research until I _stopped_ trying
to reconcile PCT with all the conventional data and theories that I thought
were important. I don't know if Tom Bourbon has connected to this Net again
or not but I know that he would echo this sentiment. I knew Tom (a fellow
"ex" conventional psychologist) while he went through the same process I went
through; the painful process of realizing that PCT is a whole new ballgame
that is completely inconsistent with everything we had learned to take for
granted in psychology. But once we got thru the pain, we started doing some
pretty good work, I think;-)

Best

Rick

2

From Stefan Balke (960625.1600 CET)

Rick Marken (960624.1600) --

But all of these people assume
(unconsciously) that the study of purposive behavior can be done in the
context of the beloved cause-effect model research. Of course, it can't (for
reasons Jeff ÕJeff Vancouver (960624.1600)þ apparently finds insufferably
boring -- and I have repeated these reasons enough that they surely are
boring by now) so tons of time and money are being wasted by people who think
that they are contributing to the PCT database when they are not. I think
this is most unfortunate.

For me, it's not boring to hear this stuff again and again, as long as
different perspectives/examples are illuminated. And I love the many
different items which are discussed here in CSG. No, it's just the opposite,
it can't be enough :-). Is PCT the name of a new designer drug? Of course,
it's hard to forget everything from the pre-PCT period, but for me it's even
harder to live according to the PCT view. Every time I try to convince
others, who already have their own opinion about the item in question, I
recognize too late, that they are disturbed and caused to defend their own
opinion. I think the everyday art of PCT is, to be able, to find the next
common goal. The goal which is accepted as important from all opponents and
which is not to abstract, such as the goal to be a good guy.

Shouldn't we try to convince others from the PCT-view, knowing that it
sometimes is impossible to be successful? In the worst case, it's just a
test of the controlled variable. By the way, does anybody know an everyday
example/observation/phenomenon, which couldn't be explained/analysed with
the (H)PC-Theory. Most theories define their own limits, where are the
limits of PCT?

Best, Stefan

3

[From Bruce Gregory (960625.1020 EDT)]

(Rick Marken 960624.1600)

The problem with trying to reconcile PCT with theories that _seem_
consistent with PCT (from my point of view) is that it keeps one from
striking off in the new direction that PCT requires. I think it limits the
contribution one can make to developing PCT as a science. Not that Martin
hasn't made some wonderful contributions to PCT; I just have a feeling
that such well-meaning attempts at reconciling irreconcilable theories
prevents one from diving into PCT "all the way".

My experience is that this is true when it comes to adopting any new
"paradigm" (or of convincing yourself that it isn't so new after
all!) Several years ago I wrote a book entitled _Inventing
Reality: Physics as Language_ in which I looked at the
development of physical theory as the creation and adoption of
new "languages" to describe nature. One point was that if I want
to know what you think the world is made of, the best place to
look is at your current scientific language. To the extent that
anyone paid attention to the book at all, they invariably failed to
understand what I was saying. It was simply too difficult to
reconcile the paradigm I was using with the paradigm to which they
were committed. (They also had trouble getting beyond the
notion that physics was "nothing but" a language. Sort of like
behavior being "nothing but" the control of perception. That's a
hard one to overcome. Humanists have less trouble with the idea
than most "hard" scientists do.) Yet, as you say, it is impossible
to _really_ appreciate the strengths of new paradigm without
committing to it "all the way". When you finally appreciate the
strengths of a new paradigm, there is plenty of time to explore its
weaknesses. Of course this approach requires a significant commitment
of intellectual (and emotional) resources. In the case of PCT I need
no convincing that this commitment is worth the time and trouble.

I myself didn't start doing worthwhile PCT research until I _stopped_ trying
to reconcile PCT with all the conventional data and theories that I thought
were important. I don't know if Tom Bourbon has connected to this Net again
or not but I know that he would echo this sentiment. I knew Tom (a fellow
"ex" conventional psychologist) while he went through the same process I went
through; the painful process of realizing that PCT is a whole new ballgame
that is completely inconsistent with everything we had learned to take for
granted in psychology. But once we got thru the pain, we started doing some
pretty good work, I think;-)

Many years ago I abandoned psychology as a career path after
getting a master's degree in human learning. There simply
wasn't enough science to keep me engaged. Astrophysics had no such
shortages and involved pretty pictures to boot! PCT has turned
my thinking around 180 degrees. In astrophysics if you don't
solve a problem, somebody else will. The same can't be said for
PCT. The fact that such a potentially revolutionary theory has
attracted so little attention makes it irresistible. The work
you and Tom are doing sets a very high standard, but what the
hey, its an olympic year after all!

Regards,

Bruce

4

[From Rick Marken (960625.0830)]

Stefan Balke (960625.1600 CET) --

Shouldn't we try to convince others from the PCT-view, knowing that it
sometimes is impossible to be successful?

If you change "sometimes" to "almost always" you have just described my
life for the last 10 years:-)

Most theories define their own limits, where are the limits of PCT?

One answer to that will be given by research; I suspect that the basic PCT
model (control of perceptual variables) will hold in all cases where control
of a single variable can be identified. I think that the limitations of the
existing PCT model will be found when we start looking at structural
relationships between control systems; I'm pretty sure, for example, that
the elegant hierarchical scheme proposed in B:CP will have to be changed
significantly on the basis of data; we just haven't got much data yet.

Another answer is that PCT is limited to the explanation of purposeful
behavior (control). This means that PCT will never be needed to account for
the behavior of balls accelerating down planes, hurricances tracking toward
the Carolinas or salts going into solution. This strikes me as being a very
strong limit on the application of PCT. It would be nice if some other
theories of behavior, theories that do _not_ apply to purposful behavior
(like chaotic attractor theories) understood their limits as well.

Bruce Gregory (960625.1020 EDT) --

In astrophysics if you don't solve a problem, somebody else will. The same
can't be said for PCT. The fact that such a potentially revolutionary theory
has attracted so little attention makes it irresistible. The work you and
Tom are doing sets a very high standard, but what the hey, its an olympic
year after all!

Thanks for the compliment but, don't worry, I won't get a swelled head
because I know I don't deserve it. The only reason my PCT research is
"pretty good" is becuase, as you suggested above, almost nobody else is
doing it. There are tons of problems to be solved in PCT but for the last 15
years or so Tom and I could rest assured that there would be nobody else
trying to solve them. It's easy to win a beauty contest when we're the only
contestents. But it looks like things are going to get tougher. For
example, Bruce Abbott (a _real_ researcher) has recently joined in the study
of PCT problems and he is currently doing a study of weight control that
would bring a tear to your eye. I have been getting progress reports on this
study and it is really a work of art (well, modern art -- Bruce has actually
gone so far as to measure the fecal output of rats in order to track _all_
the contributions to the variance in body weight as precisely as possible).
This study goes well beyond anything I have (or could have) ever done -- yet
it is elegant, precise, and clear. I hope Bruce will be able to report on
some of the preliminary results in a few months.

But, thanks to the fact that 99.99% of all behavioral scientists have no
interest in PCT, it is still possible for mediocre researchers like me (the
Salieri's of science;-)) to be leaders in the study of PCT. (Now, why is it
that I'm trying to get behavioral scientists interested in PCT??)

Best

Rick