Behavior: The Control Of Perception

by William T. Powers

6 X 9, 332 pages, 42 illustrations
select bibliography
historic and contemporary commentary (End Papers)
layflat laminated color cover
endorsements on back cover from leading scholars
individually shrinkwrapped
includes previously omitted chapter on Emotion

Library of Congress Cataloging-in-Publication Data

Powers, William T. (William Treval), 1926- Behavior : the control of perception / William T. Powers. 2nd ed., rev. and expanded.
p. cm.
Includes bibliographical references and index.
ISBN 0-9647121-7-2 (alk. paper)
 1. Human information processing.
2. Perception. 3. Human behavior.
  4. Control theory. 5. Psychology Philosophy

I. Title.   BF455.P65 2005 150.19'8--dc22

Genres: Psychology, Nonfiction, Science

318 pages, Paperback

First published June 30, 1973

About the author

William T. Powers (August 29, 1926 – May 24, 2013) was a medical physicist, science fiction author, and an independent scholar of experimental and theoretical psychology.

Powers developed the perceptual control theory (PCT) model of behavior as the control of perception. PCT demonstrates and explains how rather than controlling their behavioral outputs, living things vary their behavior as the means of controlling their sensory inputs (perceptions). Living control systems differ from those specified by Engineering control theory (a thermostat is a simple example), for which the reference value (setpoint) for control is specified outside the system by what is called the controller, whereas in living systems the reference variable for each feedback control loop in a control hierarchy is generated within the system, usually as a function of error output from a higher-level system or systems. Powers and his students and colleagues in diverse fields have developed many demonstrations of negative feedback control, and computer models or simulations that replicate observed and measured behavior of living systems (human and animal, individuals and groups of individuals) with a very high degree of fidelity (0.95 or better), and corresponding control structures have been demonstrated neurophysiologically.

Powers also designed the board game Trippples, produced by Aladdin Industries and granted US Patent 3,820,791 in 1974.

Reviews

[Steve Battle · Rating 5 out of 5]

Part of the cybernetics canon or not, and if not why not? Despite 'Perceptual Control Theory' (PCT) sometimes being presented as a separate field, it tackles topics that are central to (first order) cybernetics; feedback control and analogue neural circuits. Powers is interested in *how* the brain works rather than abstract symbolic descriptions *what* the brain does. The book contains one of the clearest accounts I've read of how 'neural currents' carry perceptual signals. Signals not symbols, because these currents only make sense in the context of a neural circuit. I'd never really grokked that because neural currents are continually replenished along their length, the signal can be split without loss of signal strength. This, combined with the fact that neural currents can't be negative makes them quite counter-intuitive for anyone more used to electronics. This lays the ground-work for looking at how neural circuits form living control systems.

Powers describes how feedback control manifests itself in organisms, which don't respond directly to error, but to their *perceptions*. The central message of the book being that behaviour is the control of perception - not of action - action being the means by which an organism influences the world and, in turn, its own perceptions. The perceptual signal is compared to an *internal* reference signal, so the goals of an organism emerge from within rather than being externally imposed. To make complex behaviour possible these control systems must be stacked *hierarchically*, with high-level control systems setting the goals for lower-level systems. The evidence stacks up too, the longer cycles of the higher levels correspond to lower frequencies. Where an over-flexed muscle oscillates around 3Hz, Parkinsonian tremors oscillate around 10Hz so must occur within a higher-level control system (though still within the brain stem). Like Merleau-Ponty before him, Powers claims that *all behaviour* is goal-directed, and that what we call reflexes have been misidentified. For example, the so-called pupillary reflex, where the pupil contracts when exposed to a bright light, is a goal-directed activity with the purpose of controlling the light level at the retina. Even the prototypical knee-jerk reflex involves a neural circuit for maintaining balance; the sharp tap on the patellar tendon fooling the body into trying to correct what it perceives as falling backwards.

Ross Ashby observed that psychology experiments are normally designed to eliminate feedback, looking at its subjects through behaviourist, stimulus-response spectacles. Powers challenges us to design our experiments accordingly. The Test for the Controlled Variable (TCV) emphasises modelling individuals over population based statistics. It's a powerful tool indeed, if a perceptual variable remains stable in the face of experimental disturbances then we can say that the subject controls that perception. It empowers us to once again, think about the *purpose* of organisms as being internally generated, without getting tied up in fruitless teleological arguments.

[Mr R · Rating 5 out of 5]

The chief work of reference on the subject of Perceptual Control Theory, by the theory's original formulator. Falling within the domain of cognitive psychology, PCT applies the framework of Cybernetics to explain how in principle the brain generates purpose-governed actions, guided by perceptual feedback from our environment, so as to control their perceivable consequences and progressively match these to those specifically intended. PCT, then, is essentially a scientific account of the cultivating of skilled actions, of invaluable insight to anyone seeking to improve their level of performance in a skill, and equally for understanding the workings of the brain on a global, system-level scale. The book's five-star rating is warranted not only for its authoritative and comprehensive treatment of the subject, but equally for Powers' ever-lucid style of writing, immediately intelligible even to readers with little or no background knowledge.

[James]

Recommended by Malcolm Ocean, on perceptual control theory