c = h + d? Don't bet on it!

[From Bruce Abbott (950122.1950 EST)]

Rick Marken (950121.1800y) --

In other words, it still doesn't matter, unless c=d+h+t, which it does,
so it does matter. Rick, when I'm right, I'm right. (;->

But if c = d+h+t then t IS KNOWN; you added it!! There is nothing to
"estimate" whether you call t a reference value, a target or a toosie roll.
The value of t is right there in your source code. Tom (like me) asked
you about your strange comment (about having to estimate the reference
value in order to compute d) because it suggested a bit of confusion on
your part about the difference between the reference value (which exists in
a person's brain) and the target (which exists on the screen). The
person's reference for the position of the cursor is quite independent
of (and may be quite different than) the target position on the screen.

So it seems that, sometimes, when you're right, you're wrong;-)

O.K., so YOU compute it. Here's a line from one of my data files:

325 323 302 17

The first three numbers are the three cursor positions; the last is the
mouse position. What are the three disturbance values?

The answer is -11, -13, and -34. Does c = h + d? Answer: no. Does the
analyst need to know t in order to calculate d? Answer: yes.

In my first attempt to get this across, I called t the reference, r, because
that is where we ask the participant to keep the cursor chosen for control.
Of course, Rick had to point out that the participant's reference could be
other than t, which is true, so to avoid further confusion between the
experimenter's intended reference and the participant's actual reference, I
renamed the former "t" and reserved "r" for the latter. Because I was in
the beginning using r for what I am now calling t, I was not wrong then, and
I was not wrong after changing the label to t, either. It was just a matter
of the our defining r differently.

What we have been bickering about are the values generated and saved in
arrays by Bill P.'s THREECV1 program. It seems to me that the easy way for
Rick and Tom to determine whether my statements regarding the program's data
were correct or not would have been to look at the program, rather than
relying on abstract principles. The program, not verbal argument, is the
final arbiter of what the program actually does, guys. I originally saved
disturbances to disk along with cursor and mouse positions; the problem
arose when I deleted the disturbances and discovered that I could not
reconstruct them from the cursor and mouse positions without knowing the
target positions. Saving the target position with the rest of the data was
necessary for a precise reconstruction of the disturbances.

So it seems that, sometimes, when you're right, you're wrong;-)

So it seems that, sometimes, when you THINK I'm wrong, I'm right anyway.
Kind of reminds me of the following statement my old friend John Burger once
told me:

"I never make a mistake. Once I THOUGHT I made a mistake, but I was wrong!"

John had a wry sense of humor. Have you gotten those disturbance values
without knowing t yet?

But if c = d+h+t then t IS KNOWN; you added it!! There is nothing to
"estimate" whether you call t a reference value, a target or a toosie roll.
The value of t is right there in your source code.

This is an example of forgetting the context in which a dispute arose.
Remember, I was saving the data to disk so my colleagues could apply their
own analyses to them. The analysis program knows only what is provided to
it in the data file. As I demonstrated with one of my Minitab analyses, it
is possible to ESTIMATE the reference value (the participant's r, not t)
from the data; the analysis showed that the estimated r was very close to t,
indicating that the subject (me) was using the target as reference, as
instructed. This has also been true so far of all my
colleague-participants. What it is NOT possible to do is calculate t; it
must be provided in the file. How are my colleagues going to deduce that r
is close to t if they are not told the value of t? To steal a quote from
Rick (and the National Enquirer), "inquiring minds want to know."

Have you come up with those correct disturbances yet? Funny how, for
example, 325 - 17 just won't lie down and equal -11, isn't it? (;->

Rightly yours,

Bruce