[From Bruce Nevin (980407.1928 EDT)] <Yes, this is really me this time.>
Bill Powers (980405.0859 MST)--
Bruce Gregory (980405.0544 EDT)
Attention sometimes seems to alter the gain on a loop. Is it possible that
attention represents (our only?) access to loop gain?
Good possibility. Now all we have to do is figure out how to see if it's
right.
If it is not true, having the participant in an experiment pay closer
attention to the controlled perception will have no effect on gain. If
paying closer attention does correlate with increased gain, that is
unfortunately not conclusive because we don't have a good understanding of
what it is to pay more attention or less.
How do you pay more attention? A perception of the writing or the seams on
a fast-approaching, spinning tennis ball is difficult to control. Effort is
expended focusing the eyes, probably widening the pupils, widening the eyes
or squinting.
How do you pay less attention? by attending to other things at the same
time. By generally paying less attention to anything i.e. being sleepy. If
one is expending a lot of effort as above, then reducing the effort amounts
to paying less attention. I know of no other way. Do you?
Studying the seams or reading the label on that approaching tennis ball is
attending to perceptions that are available only when you are relatively
close to the ball leading up to the perception at close hand of contacting
the ball with the raquet. Is this what we mean by paying closer attention?
When we concentrate, less attention is given to other perceptions. Perhaps
it's just as we turn off the radio when we're lost in a car and finding our
way, we "turn off" awareness of perceptions that would be distractions.
Presumably, that means lowering the gain on control of the other
perceptions. But if we're just monitoring a perception that we might
control if error becomes large enough, then presumably gain is already at
or near zero. Metaphorically, it's as though gain could become less than
zero, perhaps it's control of excluding (not perceiving) anything but the
perception we're concentrating on.
You can pay attention to a perception without controlling it, that is, with
zero gain. Or is paying attention to the perception a form of controlling
the perception? Or, as above, is it controlling a perception of not
perceiving other things?
When I close my eyes and pay attention to the sound of the fan in that
external disk drive, I also find myself hearing other environmental sounds
that I had hitherto not noticed: the conversation upstairs, the hum of the
overhead lights. I could ignore those and just listen to the fan, but that
would take some effort. It is unclear to me what the effort is. What it
accomplishes is that I appear not to notice the other sounds that I have
chosen to ignore.
So maybe the real question is not how do we attend to perceptions, but how
do we ignore perceptions. We're already using gain=0 for monitoring without
control. And anyway, there seems to be some effort involved. So, while
attention might correlate with an increase in loop gain (TBD), it does not
follow that that is all it does, or that that is how it works.
Bruce Nevin