[Martin Taylor 2017.10.31.13.35]
Yes, but that includes your team, too. A long time ago I suggested a
start to seeking possible controlled variables in different
conditions. You said you would try some. Have you?
True.
So far as I am aware, only Rick’s bogus mathematics has led anyone
on CSGnet even close to saying such a thing.
Why, if none of the obvious tests have been done?
First, let’s define the challenge. You have a considerable body of
observations that relate some environmental states to a power-law
relationship between local curvature of a track followed by
(including produced by) organisms in motion and the tangential
velocity of movement along the track. You also have conditions under
which the relationship is not followed. What are the essential
differences among these conditions? Is the challenge for PCT to
predict/explain that? Or is it in general to explain why power laws
are observed sometimes and not other times? PCT does offer a generic
explanation for the observation of power laws, which Bill and Rick
used to explain the Stevens power law for perceived (reported)
sensory magnitudes. It is that a power law will result when a
perception is controlled of a relationship between two perceptions
that have a near-logarithmic relation to their relevant magnitudes.
In this case, maybe the challenge resolves into discovering the two
perceptions in question. Might that be true? Or is this particular
power law not an example of the generic case? Maybe that’s the
challenge.
If you are going to assert that PCT is “unable” to explain any data,
whatever the experiment, you first have to seek a controlled
variable. It is highly unlikely in any of the conditions of any of
the experiments that the mover controlled a perception either of
following a power law relation between speed and curvature, and even
if they did, that they were controlling for making the exponent take
on some particular value. So the first step is to propose some
plausible perception(s) that might be being controlled, and then try
to disturb them to include or exclude them from the plausible list.
If a perception of the power-law relationship and the particular
exponent is implausible, then the power-law itself is a side-effect
of control, and the problem becomes one of explaining what
environmental conditions affect the perception and the action
influence on whatever is being controlled.
A perceptual variable that is available to larvae and to humans
would naturally be more plausible than a requirement that different
species control different variables but wind up with the same
results. But it may prove that several different perceptual controls
working through similar environmental constraints would all give the
same power law. A power law is, after all, the result of subtracting
one logarithm from another, and since most perceptions do seem to
have a near logarithmic relation to the relevant environmental
variable (Weber-Fechner Law), such a result is not too far-fetched
to contemplate.
Whatever the controlled variables might be, difficulty in finding
them is not the same as proof that they do not exist, or that PCT is
a crumbling edifice. It helps if the search has been at least begun
without successful conclusion, but such searches sometimes take
thousands of years. Ptolemy’s laws explained the planetary motions
relative to the starts well enough not to be discarded for a
thousand years, within one conceptual frame – Earth-centred
Universe. Kepler and Newton provided a better explanation, and that
was good enough in a different conceptual frame – a Universe
conceptually known synoptically by an omniscient being. But it had
little niggles in it such as the advance of the perihelion of
Mercury. Einstein came up with a better explanation in a third
conceptual frame – “what you see is what you get” (relativity), but
it also has niggles in its incompatibility with quantum
chromodynamics, so I guess the next advance is a fourth conceptual
frame yet to be found.
PCT is, I think, in the relativistic conceptual frame – “what you
perceive is what you get”. Earlier theories that base behaviour
strictly on the environment are in the Newtonian conceptual frame
“You would know it all if you were God”. As with Ptolemy-Newton or
Newton-Einstein, the earlier and older theories explain the data
pretty well, because that’s what they are designed to do. In each
case, the newer theory explained a wider range of data and explained
it more accurately because it started from “why”, not “what”.
Newton’s “why” for the planetary motions was the gravitational law
that produced Kepler’s ellipses, which were a descriptive
improvement on Ptolemy’s epicycles. Einstein’s “why” for Newton’s
gravitational law was the distortion of space-time by mass-energy. I
suppose the next conceptual revolution in that area of physics will
be the finding of a “why” for space-time deformation that resolves
the problems with quantum theory. One “why” is simpler than a whole
lot of “whats” in the Occam’s Razor sense. PCT offers a relativity-level “why” for a whole mass of data that is
observed and for which many people have provided Ptolemy or
Newton-level “what” explanations. Maybe optimal control theory does,
too, but PCT has the advantage of not having to do complex
computations of such things as joint angles and of not having to
make special provision for the effects of unexpected external events
and forces, since dealing with them is built into the structure. I
am not aware of optimal control theory having been applied to
sociology, but I suppose it must have been. However that may be, I
suspect that another change of conceptual framing will someday
provide a “why” for the “whats” of PCT, further simplifying science
as a whole.
Enough philosophy of science. It would be nice to be able to figure
out what controlled perceptions in what environmental circumstances
lead to power laws, and with what exponents when power laws are
found. Incidentally, has anyone ever analyzed movies of a skater
doing school figures? It is my impression that their skates move
faster along the ice when the curvature is high (low radius of
curvature), but that may be an illusion.
Martin
···
On 2017/10/31 12:01 PM, Alex
Gomez-Marin wrote:
regardless of your helicopter data and RCT
mantras, it would be good if someone from CSGnet took
seriously the challenge to PCT that the speed-curvature power
law entails.Â
any figure panel of our paper proves rick's
mathematical claims wrong: the PL is not a must and when it
takes place it is not trivial and can have different
exponents.
now, how can "control of perception" explain
that phenomenon? claiming it is an illusion because it does
not fit in the dogma is like creationists insisting that dino
fossils are bogus.
so, as adam and myself take this job seriously,
and given how many optimal control and nonPCT theories explain
the data, I think Bill would really find his edifice
crumbling, or at least unable.
so, take your best shot at it and really
challenge your “revolutionary paradigm changing” theory of
behavior.
On Sun, 29 Oct 2017 at 02:48, Richard Marken <rsmarken@gmail.com >
wrote:
[From Rick Marken (2017.10.28.1745)
On Sat, Oct 28, 2017 at 3:34
PM, Alex Gomez-Marin agomezmarin@gmail.com
wrote:attachedÂ
RM: Finally! Thank you, Alex. I hope the journal gives
us an opportunity to respond. But for now I have only
one word for you: helicopter movements. Oh, that’s two
words But you know how bad I am at math;-)
Best
Rick
–
Richard S. MarkenÂ
"Perfection
is achieved not when you
have nothing more to add,
but when you
have
nothing left to take away.�
Â
             Â
 --Antoine de Saint-Exupery
–
Alex
Gomez-Marin
[behavior-of-organisms.org](http://behavior-of-organisms.org)