[From Rick Marken (930405.1800)]
Bill Powers (930405.1330 MDT) --
The "information" contained in a perceptual signal
could not even be used to reconstruct that same perceptual
signal.
Excellent observation. "Information" theory was irrelevant all
along. What we were actually doing in this long debate was
exploring, in analog terms, the details of how control systems
work. It was quite a worthwhile experience for me. I think
I came to understand, more deeply than I ever did before, why
PCT is truly revolutionary. The perceptual input really doesn't
guide the ultimate output, g(o), of a control system (where g()
is the feedback function). It really is the other way around --
output guides perceptual input: behavior is the control of
perception. Now I know what Bill Powers meant when he said
(p. 157 in LCS):
q.o = g-1[-h(q.d)]
Now THAT'S revolutionary.
Best
Rick