Thanks to Rich, Uncertain about uncertainty

[From Rick Marken (940604.1115)]

Richard Thurman (940603.1000) --

I really did mean "no offense."

None taken -- at all. I really thought you made an EXCELLENT
point and one worth discussion. I think the net is a great place to
consider questions like "what is the most effective way to get the
point of PCT across? " and "what are the most useful directions
for individual research?"

For you the results might seem negligible. But not for me! I
study PCT as a direct result of your article in Psychological
Science -- "Degrees of Freedom in Behavior"

This is really wonderful to hear. If the result of my work was to
get just one person of your caliber (ie. you) interested in PCT
then it was all worth it. Thank you. Thank you.

Bill Powers (940604.0820 MDT)] re: Martin Taylor (940603.1400) --

The subject is seeing a periodic repetition of a1/a2 events
following presentation of a symbol S, and per instructions is
perceiving the fraction of the time that S is followed by a1. That
fraction is the CEV with respect to this control system.

The fact that the display is a function of an invisible cursor and
target is unknown to the subject. So this is equivalent to what
Rick proposed, the situation in which the handle position plus
a disturbance affects the probability that a1 or a2 will appear
after S.

Thank you for clearing this up for Martin.

This experiment is worth doing, simply to show that the
relative requency of occurrance of two mutually-exclusive
events can become a controlled variable.

I've already run a preliminary test. I simplified it further by just
having a square appear to the left or right of a reference line
with some probability determined at each instant by the joint value
of mouse and sinusoidal distrubance. The square appeared about 2
times/sec. It's possible to keep the probability at near about .5
but control does not seem impressive. It is easier, of course, to
keep the probablity at 1 (square always bleeps on on left) or 0
(square always bleeps on on right). What is being controlled
probably depends on the rate of display. If the probability
that results in the observed display of the square is considered
the controlled variable then control is not very impressive. At
the rates that I'm using now it seems to me that I am
controlling something more like the average interval between
square appearances on each side of the reference line. In other
words, even though there is "uncertainty" about which side a
square will appear on at any instance, I am not actually
controlling a perception of uncertainty (probability) but of
something else, like temporal density of square appearances on
each side of the reference line.

People may sometimes control a perception of uncertainty but
it seems like it's not exactly one of the main things we control.
At least, I haven't found evidence of it yet.