[From Rick Marken (950503.2240)]
Bill Powers (950503.1130 MDT) to Hans Blom (950503) --
This control system doesn't seem to be able to resist disturbances of
its controlled variable, xt.
Gee, and I was hoping that predictive control really could worked. Curse
you, real world!
While we're on the subject of control models that don't control, I
received the Spring, 1995 MIT Press catalog of books on Cognition,
Brain & Behavior. In there we find books that present the State of the
Art on, what else-- cognition, brain and behavior. The MIT catalog gives
us the books that represent the state of the art in modelling living
systems. And here they are: A new book (1995) called "Cognitive
Science" that "surveys the theories and empirical results of cognitive
science within a consistent computational perspective" (computed
output, anyone?). Or how about three new books on the "dynamical
approach to cognition". One, by Kelso, "extends the physical concepts
of self-organization and the mathematical tools of non-linear dynamics
to understand how human beings perceive, intend, learn, control and
coordinate complex behaviors" (trendy attractors, anyone?).
It's a pretty grim catalog, overall. Very little on behavior as the control
of perception. I rank it well below the New View catalog in scientific
quality.
But wait. Nestled near the back, in the section on "Complex adaptive
systems" we find "A comparative approach to cognitive science" edited
by H. Roitblat and J-A Meyer. Among the contributors is one W. T
Bourbon; PCT noses it's way into the mainstream! And right next to
that listing is one for "Intelligent behavior in animals and robots" by
McFarland and Bosser. This book explores the "novel insight" that
animals and robots "can be analyzed as multi-task autonomous control
systems". If that gets your heart pounding, wait 'til you read the next
part of the description: "In defining intelligent behavior, what matters
is the behavioral outcome, not the nature of the mechanism by which
the outsome is acheived". Is this a book about behavior as the control
of perception? Inquiring minds want to know -- but, with a list price of
$39.95, inquiring minds won't know until their lazy bodies manage to
get them to the library. This book was apparently reviewed favorably
by Randall Beer. Could Dr. Beer have become a control theorist since
"abandoning the conversation" on csg-l a couple years ago?
Has anyone read "Intelligent behavior in animals and robots"? Are McFarland,
Borrer and Beer PCTers without portfolio?
Best
Rick