Test for feedforward behavior, Extremism

[From Rick Marken (931110.0900)]

CHUCK TUCKER (931109) --

It seems to me that we continue to have the same conversation
about PCT (or HPCT) because people don't understand how the
model can account for (deal with, make sense of, understand,
comprehend, explain, and the like) such experiences that we

I agree. But I also think there has not been a clear description
of what behaviors would count as involving feedforward processes.
PCT focuses on purposive behavior (control) and we know how to
determine when purpose IS involved in behavior (using "the test").
I think the problem with feedforward is the we don't have a way of
testing for feedforward behavior.

My impression is that people think of feedforward behavior as
involving something more than just UNCONTROLLED, automatically
varied variables. The behaviors you describe -- particularly habitual
and planned behavior -- suggest REPETITION and CONSISTENCY as
important aspects of what is considered a feedforward behavior. Hans
talks about REPEATELY and CONSISTENTLY walking right to his bed in
the dark and reaching to the exact location of the bed covers.
The idea seems to be that these CONSISTENT results are produced by
an internal plan or program and that this consistency is achieved
by REPETITION of the program (in a disturbance free environment)
NOT by control of perception (becuase there is, presumably, no
perception to control).

This leads me to a proposal for a "test for feedforward behavior".
The first part of the test is to identify some result that is
produced repeatedly and consistently when there are no disturbances.
The second part of the test is to apply disturbances. There should
be NO RESISTENCE to these disturbances. If there is resistence then
the result might be controlled and the test can proceed to identify
the control system involved ( the sensor and output components).
So "the test for feedforward behavior" assumes that a feedforward
behavior is a REPEATABLE, CONSISTENT result that is NOT under control.
I think it must also be established that this feedforward behavior
is directly the result of neural outputs -- and NOT a side effect of
control of other variables.

A simple example of a feedforward behavior that would pass this test
is pointing at a particular location on the computer screen with the
eyes closed. After some practice, a person could probably point
FAIRLY consistently at the same point on the screen. If the point
or the starting position of the hand is moved (disturbance) there
will be no compensation for the disdurbance -- and the result (finger
near target) will NOT repeat. Clearly, the relationship between finger
and target is a feedforward behavior. This result is produced by arm
movements, which might also be feedforward behaviors (they are repeated
and fairly consistent-- so now we just have to disturb them). If we push
on the arm (gently) while the pointing is going on there will be
resistance -- arm trajectory will be defended against disturbance so
there is probably perceptual control involved. So arm movement is
NOT a feedforward behavior.

Many examples of REPEATED/CONSISTENT behaviors -- that seem to be the
result of plans, habits, predictions, etc -- will (using this test)
probably be found to be consistent side effects of purposeful
behaviors. The apparent "prevalence" of feedforward behaviors that
is seen by some people will (once the research is done) turn out to
be those "irrelevant side effects" of control you hear tell of.

It seems to me that the ball is now in the court of the advocates
of feedforward behavior; show me an example of a feedforward behavior
(as established by "the test for feedforward behavior") that can only
be explained as a computed output of the human nervous system. That is,
show me an example of a consistently produced behavioral result (such
as reaching for the bed sheets in the dark) that is NOT controlled
OF THE NERVOUS SYSTEM (ie. is not a side effect of control of
other perceptual variables).

Martin Taylor (931109 16:40) --

I see all-or-none thinking in numerous postings to this net. "There is
NO information about the fluctuation [replacing "disturbance" with the
word agreed by Bill and me] in perception" is claimed as the ONLY possible
alternative to "Perception has perfect information about the fluctuation."

I think it's important to get your "alls" and "nones" right here. The
"none" position is (and, I predict, will remain through all eternity,
along with 2+2=4) "In a control system, there is NO information about
which component of perceptual fluctuations is caused by disturbances
to the controlled variable; not even a teensey, weensey little bit."
The "all" position is "In a control system, there is PERFECT information
about which component of perceptual fluctuations is caused by disturbances
to the controlled variable". The intermediate position is "In a control
system there is PARTIAL information about which component of perceptual
fluctuations is caused by disturbances to the controlled variable".
Since we have provided perceptual fluctuations to those who maintain
the intermediate position and have never received even a PARTIAL answer
regarding the components of those fluctuations that were caused by
disturbances to the controlled variable, I have to assume that the
"none" position still prevails. If moderation, per se, is your goal,
then perhaps we should get used to thinking of facts in a spirit of
moderation: so 2+2 is "around" 4; not ALL 4, not NOT 4, just "sort
of" 4. Is that better?