[from Mary Powers (991123)]
Fred Nickols 991121 21:06:12
About PCT vs. the cognitive branch of conventional psychology: there has
been plenty of tilting, but not on the net. It has taken place on the
doorsteps of the journals, as PCT articles by a number of people have been
barred from entry, for reasons ranging from the ludicrous to the outrageous.
BCP, being written in the 60's and early 70's, is probably more about
behaviorism than it would be if written today. Cog psych was in
development then. What has emerged is focussed on action. The trouble with
planned action is not in the planning, it is in the notion that what is
planned _is_ action. Because as the command to act proceeds down the
hierarchy to muscular outputs, the detailed computation required to carry
out the action grows more and more complex and more and more vulnerable to
the slightest variation in the environment in which the action is to occur.
Planned perception, a la PCT, permits any variation in action required to
bring about the desired result.
Cog psychologists do talk, some of them, about control, and self-regulation
and so on. Some are even taking control engineering courses. But they are
learning from engineers who have learned what to WTP's eyes is a pretty
bizarre form of control engineering - as far as I understand it (not a lot)
the concept of feedback is not as central to their thinking as it is to
Bill's. The whole business of model-based control seems to be very clumsy.
An example would be the Presser article in Psychological Review 199 v. 106
no. 4. I admit to not having read it, but simply looking at fig. 1,
"control via feedback" is to look at something that, to paraphrase Bruce
Gregory, is incoherent.
Mary P.