[From Rick Marken (930329.0900)]
CHUCK TUCKER (930329) --
Gee Chuck. I feel like I just got read the riot act by Innocent XII.
So, the subjects, contrary to your claim that they didn't, DID have
information that a disturbance was going to influence the cursor
Bill P. suggested that this might be the misconception under which
Martin and Allan were laboring. I never said that subjects had
no information THAT a disturbance was going to influence the
cursor; the debate is about whether there is information in the
movements of the cursor itself that can be used to determine how to
move the handle so that handle movements compensate for the disturbance.
(BTW, your description of the subjects leaves open
the possibility that one is you - a real problem if so)
Why?
BUT the point
is that you previous stated that the subject had NO INFORMATION about
a disturbance;
The ambiguity here can be helped by the mathematical notation that
you seem to dislike. When I say "disturbance" I mean the time series,
d(t), of values that is added to the cursor. I am claiming that there
is no information about the values of this time series in the time
series of cursor movements to which the values of the disturbance
time series contribute. This is amazing because the time series of
cursor movements is all that the subject can see.
that simply is not true if your own description (as poor
as it is) is an accurate representation of what was done in the study.
I think the meaning of my claim was clear from the context of the
paper; but apparently you think the fact that the subject's knew
THAT there was a disturbance is enough to disprove my claim
that there is no information in perception that will allow the
subject to make the appropriate outputs. I hope you will excuse me
for not being convinced by your disproof.
YES, this study (and all others that I have seen you do) does bother
me. It does so because it is so poorly done that I would not consider
it worthy of inclusion in a list of scientific work.
So you are offended by the poor quality of the study, not the results.
Perhaps you could do a high quality version of this experiment (or
describe what an acceptable version of the experiment might be)
so that I can see the right way to do it.
The claims that
you make for such work are usually not supported for me by your
study.
Well, could you suggest a study that might convince you, one way
or the other, on this "information in perception" question? Of
course, if you already know the answer then I don't need to waste
any more time on these sloppy, biased experiments.
You never indicate to the reader WHAT you have told the subjects
to do; this for me is basic in all research - what procedures were used
in the study!!! (e.g., I just found out in your answer to my post that
you did not TELL the subjects that there was a disturbance - did you
use deception in this study? - if you did this creates not only empirical
problems for me it creates some serious ethical ones - I hope you aren't
lying to your subjects in your research !?!). I suggest that you begin to
consider language and words more seriously in your research.
I don't know how to "consider" it? Would you be happier if I did the
same study with a white rat -- then language would not be an issue,
would it?
My evaluation of the results of this (and most other studies) is
ultimately based on matching the behavior of the subjects to that of a
control model in the same situation. As I mention at the end of the paper,
the result I consistently obtained with people (perfect correlation
between outputs on two occasions, very low correlations between the
inputs on the same occasions) was also found for a working control model
(which didn't talk at all and which had no knowledge "that" a disturbance
was influencing its input). Given the fact that people and model behaved
in exactly the same way, I conclude that the characteristics of the study
that you mention as being its shortcomings (how I instructed the subjects,
their probable knowledge that there was a distrubance, the fact that I
was one of the subjects) are irrelevant to the results. If you think they
are relevant, I think it would be helpful if you would explain why they
are and how the results would be different if these "shortcomings" were
changed.
Markenisms:
I have a sense that you and I don't use a similar criterion to make a
judgment on whether a statement is categorized as "hyperbolic" but I
do judge the first two statements that way;
Here they are again:
It's why we say PCT is revolutionary. I know that it seems
impossible that all the old revered theories in psychology are
invalidated by the work of a noce engineer from Chicago who
doesn't even have a PhD in psychology - but that's the fact
Jack."
"There is no information about the disturbance in the stumulus.
Goodby behavioal science as usual, hello looney bin."
I agree that the first one could be considered hyperbolic; I stand by
it, though; at least when "old revered theories" is taken as I intended
it -- as working model implementations of these theories. The second
statement is not hyperbolic at all -- it's just true. Perhaps you
misread the "looney bin" comment. What I meant was that people (like
me) who claim that there is no information in perception (which is
what the quoted comment implies) would be considered (by conventional
behavioral scientists) as candidates for the looney bin. I know that
the claim that there is no information in perception is a pretty strong
one; it goes against the most basic assumption of conventional behavioral
science. I'm sure it must seems like mysticism. Why do you think I'm
leaning on it so strongly. As Bill P. has said, it's a pretty obvious
point (because p = o + d -- sorry about the symbols); but, as you can see,
obvious or not, there is VERY strong resistance to this fundemental
fact about the operation of a control system -- as evidenced so
marvelously by your posts.
Greg Williams (930329)--
But then
you call for a complete rebuilding of psychology based on PCT; I guess
the lint-catchers are out of luck. I haven't heard CBSers saying
(arrogantly) that there is NO place in psychology for explaining
purposeful behavior.
Whenever I have suggested that conventional behavioral science (CBS)
will have to be rebuilt, I always meant to imply that this is only
true if behavioral scientists want to understand purposeful behavior
(catch mice). No rebuilding is necessary if CBS is only interested in
estimating population parameters or input -- output relationships from
sample data (catch lint).
Best
Rick