[Martin Taylor 2017.06.23.12.34]
[From Rick Marken (2017.06.22.2230)]
If you observe that feedback loop, have your methods been as
successful as they might have been? If you go up a level, why do you
want to “teach it to people who want to learn it”?
If a person wants to learn more about PCT and you use your methods
of teaching, but the person decides to leave the lessons, do you
perceive their leaving as success for your methods because they no
longer belong to the class of “people who want to learn it”, or is
the “teaching” perception still in error because they left without
learning PCT “properly” as you see PCT?
That is a very suggestive comment, is it not?
PCT has been studied for 50 years, but you say that the implications
of a new idea must immediately prove that the new idea is an
improvement, before any studies can be done to test whether the idea
makes things worse, better, or no change. Even then, if the new idea
and the original PCT structure seem to be equally viable on the
basis of further study, the new idea must be discarded if it is as
good at explaining Nature as is the original version of PCT. Well,
that’s your perception. But it is not mine.
Not at all. I think Alex would not have been driven away, and I
would not have been so angry (I can’t speak for the others) if you
had provided even the slightest hint of an argument for why you were
right, rather than calling us enemies of PCT and moles for the
propagation of S-R theory.
Never yet have you explained how the curvature at a point on a curve
painted by Michelanglo is a function of time, or how the time
derivatives of the x and y positions of that point are calculated.
Anything, any hint of how time gets into the specification of shape,
would have helped. And that is something you have still not done,
not on CSGnet and not in the published paper.
If you thought we were all completely wrong, why did you not provide
evidence for your position instead of simply reiterating time after
time that your were correct that any shape in space must be
described
only as a function of time?
Here's what Alex wrote 70 minutes after your first posting about the
so-called behavioural illusion:
[AGM]
I am sorry, Rick, but the general equation for curvature, when
combined with the equation for speed, does not give rise
to the power law. Please revise your math.
[AGM]
A second reason (for those not versed in math) is a physical
one: curvature is a geometric quantity (thus, only in space),
whereas speed is a kinematic one (in space and time). The shape
of a scribble in space does not say anything in principle about
how it should be drawn in time.
That's hardly "dismissing the PCT explanation of the Power Law", is
it?
And what do you mean by "THE" PCT explanation??? I guess that when
you perceive PCT to be your private preserve, you can say that THE
PCT explanation of anything is exactly what you say it is, and need
no argument to defend it against suggestions that it might not be.
FYI, there are those among us who think of PCT as a science, and who
also think that decisions on matters of science are best left to
Mother Nature. But as a second best, adherence to logic and
generally accepted mathematics can often allow pretty good
decisions.
Is it even now too late to ask for just the teeniest hint of why you
still think (as you emphasised a month or so ago) that every shape
in space should be defined as a function of time?
Yes, that certainly would have helped. I did find a condition that
would lead to a 1/4 power law, and told Alex about it. I also
suggested experiments that might lead toward finding the controlled
variable(s) in the different kinds of task that produce the
different power laws. But neither I, nor, so far as I know, anyone
else has yet solved just what controlled perceptions would lead to
1/3 power or to the intermediate powers that are usually found.
Phew!!! What a naive misreading of not just Alex's comments both
immediately after your first offering your proposal, but also later,
and not only of his but also mine and others. Since these trivially
simple points got re-interpreted in such a nonsense way, it’s not
surprising you never attempted an argument in favour of your
proposal. I don’t know what Alex and you communicated off-line, but
I would have presumed either that Alex intended to say you were
dividing by dt when you should have been dividing by ds, or simply
that your “were not” above was a misreading of “should not be”.
I don't think there were many different ones, though they may have
looked different to you. I remember three classes that differed on
the surface: (1) In practice, if one chooses, one can go slow on the
straight and fast around curves; (2) when the power law is measured
under different conditions, such as surface friction, different
values are obtained for the power in the power law; and (3) in the
“dotty” expression for curvature, the derivatives and accelerations
of x and y are with respect to distance along the curve (ds), not
with respect to time (dt). Can you think of any more different kinds
of criticism? No more are really needed, since any one of these is a
definitive objection to your conclusion.
Did you ask any of them specifically whether the dots and double
dots in the expression for curvature represented differentiation
with respect to time or with respect to arc length, or did you just
ask them to look at the paper and see if it looked OK? I asked you
that several times, but you never said that you had specifically
done asked about the dot notations in the expressions for velocity
and for curvature.
But you shouldn't even need to, if you could see in a way you could
express to a reader how a point on a curve in a painting or drawing
moves over time differently when it is on a sharply curved part of
the shape or on a relatively straight part. Just where does the
point go and how fast does it go there while you are looking at the
picture?
Even now, some hint as to why you thought that curvature at a point
on a shape must be (or even could be) a function of time, might be
helpful. An explanation in full would be even more helpful.
Martin
···
Martin Taylor (2017.06.21.07.56)–
[MT] I have no doubt that this is your
perception. The question isn’t whether what you say is
as others perceive the history, but whether you were
and are controlling for bringing people with new ideas
and different tools to a state where they want to
learn more about PCT.
RM: No, that is not what I am controlling for at all.
This is what salesmen control for. I don’t want to sell
PCT; I want to teach it to people who want to learn it.
RM: Looking back over the history of
CSGNet, in memory as well as in the archive,
I can recall having occasionally made angry
outbursts that I regret and I wish I had
more consistently adopted the “cordial
conflict” approach to argumentation of Bill
and Mary. But I think that in general my
arguments were always substantive and
reflected my understanding of PCT. I have
tried to avoid ad hominum argumentation and
personal insults and I believe I have
largely succeeded.
[MT] FWIW, my perception, using
the “random disturbance” observational version of the
TCV, is that you control for NOT having new people
introducing new possibilities and new views that might
augment PCT.
RM: If by "new possibilities" you mean new phenomena
that might be explained by PCT then I am certainly NOT
controlling for NOT having new people introducing new
possibilities, as long as they are testable
possibilities. And if by “new views” you mean revisions
of the exiting model then I am certainly NOT controlling
for NOT having new people introducing new views, as long
as there is empirical evidence that these new views are
necessary.
[MT] Not by doing that, but by ignoring both
common sense and mathematical logic.
RM: So I am driving people away when you think my
arguments are wrong. Apparently the way for me to have
kept Alex from being “driven away” was to simply agree
with you and Bruce and Alex and everyone else on the
net?
RM: A relatively recent example that
comes to mind was Alex’s repeatedly saying
that my PCT interpretation of the power law
was “bullshit”. …So was I driving people
away by doing that [presenting my PCT
interpretation of he power law]?
But that would have been impossible for me because I
thought (and still think) that you were all completely
wrong.
I was amazed that everyone who participated in the
debate, even those who have been at this for years,
dismissed the PCT explanation of the power law as Alex
did. That’s why I decided to publish a paper on it; I
wanted to get some confirmation that my explanation was
as reasonable as I thought it was, and I did.
RM: I'm really sorry that this power law thing drove
Alex away from CSGNet. He’s obviously a very bright guy
with excellent research and entrepreneurial skills (he
runs a rather large neuropsychology lab). But I have a
very different idea of how Alex could have been
encouraged to both stay on CSGNet and move his “power
law” research in a direction consistent with an
understanding of behavior as a process of control. I
think it could have happened if all the PCT “experts” on
CSGNet had 1) contributed to an explanation of what’s
wrong with the idea that the power law tells us
something important about how behavior is produced and
2) suggested directions for research that would properly
test whether there is a relationship between curvature
and velocity of movement and why any such relationship
might exist.
[MT] I believe both. But I also believe that
“people’s misconceptions” are as you perceive them. I
ask you to think about your own behaviour when someone
tries to correct your own “misconceptions” as they are
perceived by the one trying to correct you. Since you
use Alex as an example (I agree that I used his name
first), let’s just ask about one misconception among
many in your Power Law paper – the idea that the
curvature at a point on a curve is a function of time.
Alex tried to point out to you within hours of your
first mentioning it on CSGnet that this cannot be
true. Your only “argument” for months, culminating in
a publication, was a flat statement that it is true.
Never, ever, have you actually made an argument to
support the idea.
RM: I don't believe Alex ever disagreed with my way
of computing the variables. He did mention that my
computation of the derivatives (actually differentials)
were not divided by dt so I divided them by dt and,
since this was a constant, it made no difference in the
results.
RM: So you think I am driving people away
and I think I am trying to teach PCT and
correct people’s misconceptions about it.
[RM] I tried to take everyone's criticisms into
account, but there were so many different ones it was
hard to deal with it all,
which was another motivation for writing the paper
(of course the main motivation for writing the paper was
that I thought it was a great way to get PCT in front of
a wider audience). But Alex has apparently written a
rebuttal to the my (and Dennis Shaffer’s) power law
paper. So we shall see what he comes up with. But
however it goes, I now know that at least 7 other
people besides myself, all of whom are at least as smart
as you, thought the analysis in the paper was correct.
