This is the first textbook based on the control-theory approach to the science of psychology. It reviews, and makes theoretical reinterpretations of, many facts found by researchers working within the framework of older traditions in psychology. The basic control-theory model for organismic behavior as the control of perception via hierarchically arranged negative feedback loops is detailed early on, providing a foundation for later chapters on learning, development, perception, personality, social psychology, psychotherapy, and other topics.

Robertson, Powers, and their colleagues provide what is lacking in other general psychology texts: a unified approach to the entire field, from laboratory studies of animal behavior, through ethology and studies of human social behavior, to clinical work. This volume is suitable as the primary text for introductory college-level psychology courses, as supplementary reading for advanced courses in the behavioral sciences, and for independent study. Its treatment of control-theory ideas is completely self-contained, but ample references are provided for those who want to learn more.

Richard J. Robertson is a professor of psychology at Northeastern Illinois University, Chicago. William T. Powers, the originator of the control-theory model, is an independent investigator in Northbrook, Illinois. Other contributors are David M. Goldstein, a clinical psychologist and consultant to various hospitals in New Jersey; Richard S. Marken, a human-factors engineer with The Aerospace Corporation, Los Angeles; and Frans X. Plooij, a developmental psychobiologist at the Pedological Institute of the City of Amsterdam, The Netherlands.

It is certainly true that on one particular Thursday afternoon in 1957 (I forget the month, probably October), I had an experience which, as far as I am concerned, really did change the course of my life. In it, I found an approach to psychology which has felt worth pursuing ever since. Three fellows came to present a lecture at one of the open seminars held on Thursday afternoons at the University of Chicago Counseling Center, where I interned. I had been a somewhat indifferent graduate student in my basic studies up to this point. Not that I was uninterested in them. I have always especially revered the core topics in psychology learning theory and developmental theory—in the opinion that there is nothing as practical as good theory. However, I had not been satisfied with the views of these subjects which I had been offered up to that point, and consequently I had been unable to invest much energy in mastering the explanations of behavior put forth in them.

That 1957 lecture “opened my eyes” and filled me with an excitement about the underlying nature of human behavior which has never left. The material presented was subsequently published as Powers, Clark, and McFarland (1960a and 1960b), and, of course, was the initial statement of the views presented in this volume.

<snip>

About the Authors

David M. Goldstein is a clinical psychologist, a consultant to various hospitals and training programs, and a pioneer in the development of a truly workable research- and theory-based approach to clinical psychology.

Richard S. Marken was for 12 years a professor of psychology at Augsburg College. He is now a human-factors engineer with The Aerospace Corporation.

Frans X. Plooij is a developmental psychobiologist heading the Department of R & D, Pedological Institute of the City of Amsterdam (Institute of Child Studies). With his wife, anthropologist Hefty van de Rijt-Plooij, he studied infant chimpanzee development at the Gombe Research Center, Africa, and described the evidence of step-by-step development through successive control-system levels consonant with Powers’ model, later paralleling these observations on human infants.

William T. Powers is associate editor of this volume and originator of the model presented here. He has authored many articles (see Powers, 1989) and has invented many demonstrations furthering the development of that model.

Richard J. Robertson is editor of this book and professor of psychology at Northeastern Illinois University.

Each author’s contributions are his own views. While the editors have exercised a certain amount of stylistic control to slant the exposition toward the intelligent layman and the university undergraduate, no attempt was made to force the various chapters into total harmony or consistency, other than that which has come about naturally from the common position provided by the basic model.

Richard J. Robertson
Northeastern Illinois University
Chicago, Illinois
January 1990

This book is a textbook on the new, control-theory approach to the science of psychology and also a general psychology text based upon the control-system model. In its function as an introductory general psychology text it reviews, and frequently makes theoretical reinterpretations of, many of the facts found by researchers working within the framework of the older traditions in psychology.

In keeping with the usual mold of general psychology texts, we begin with a discussion of the nature of science and where psychology fits into that picture. Our view of the history of science has been greatly influenced by Thomas Kuhn’s (1962, 1970) theory of how science has evolved through history, especially his conception that scientific progress consists of periodic revolutions in thought called “paradigm shifts.”

This introductory text was begun owing partly to Kuhn’s suggestion that new paradigms in science tend to be accepted first among young scholars who have a sense (however dim) that current theories are no longer adequate to extend the frontiers of their science. The present situation in psychology appears to be just such a situation. It hinges on a problem with the meaning of control in behavior.

The traditional Behaviorist position is that behavior is controlled by the environment. For example, Bandura (1978) quotes B. F. Skinner (1971), the leading Behaviorist, “A person does not act upon the world, the world acts upon him.” Laymen typically use the term in just the opposite way. For example, a book titled The Writer’s Control of Tone (White, 1970), explains how an author can foster the mood he or she desires in a reader by choice of writing style, implying that “control” means bringing about the condition you desire. In other words, we try to make our environment match our specifications; when we do, that is controlling it.

Laymen link control with one’s intentions, or purposes, as when a reporter writes, “The driver lost control of his car,” meaning that the driver failed to steer it where he intended. This view does, indeed, represent a paradigm shift in the basic assumptions underlying traditional psychology. The philosophy of science which underlies Behaviorism ruled out the concept of intention as being unscientific, because it seems to imply that behavior could be “stimulated” from the future. That would be an impossible scientific concept.

This is a mistaken conclusion. It is based upon the assumption that our actions are “stimulated” by forces in our environment. We shall show below that a per-

Psychology has been defined in different ways. Early definitions tended to view psychology as the study of mind. It came to be defined as the study of behavior with the ascendency of Behaviorism as the dominant approach. This definition was thought superior because it satisfied the requirements for science laid down early in the present century by the Logical Positivist school of philosophy, holding that the only genuine knowledge is that obtained by objective techniques or procedures. “Objective” procedures mean that I can duplicate what you did to achieve the same experience. Another way of saying it would be that the only facts on which we can definitely agree are those which can be obtained by the senses of touching, seeing, hearing, tasting or smelling.

John B. Watson, the father of American Behaviorism, adopted this view of science in attempting to develop a more scientific psychology. He selected behavior-external, observable activity—as the only fit subject for scientific study by psychologists.

It might have been coincidental that Pavlov’s (1937) studies of “conditioning” were the most notable new facts about behavior at the time Watson was looking for a way to make psychology more “scientific.” (It probably was not wholly coincidental, because the same Zeitgeist was behind both developments.) Although Watson’s desire to objectify psychology might have focused upon some other phenomenon, since Pavlov’s work happened to dominate the attention of those interested in scientific psychology just then, Watson took “conditioning” as the basic concept in Behaviorism. “Conditioning” was based upon the Cartesian paradigm: Stimuli from the environment cause, or trigger, reflexive responses within the organism.

The already existing work of the Structuralist (Wundtian) school of psychology was equally qualified to meet Watson’s requirement of “objectivity,” in that the Structuralists also studied externally measurable capacities, such as how small a difference a human can detect between the weights of two objects (psychophysics), and how many times one must rehearse a poem (or set of nonsense syllables) in order to recite it from memory (early learning theory).

But Pavlov’s work went beyond the work of the Structuralist school in one important sense. It investigated the formation of associations between seemingly unrelated actions and /or events. That went beyond simply reporting new facts, as the Structuralist studies did. (The American, Twitmeyer, previously had dis-

What we mean by working with the model is that it is necessary to examine some quantitative examples to understand it fully. That is probably the only way to get free of the cause and effect way of thinking—in which things necessarily happen one after the other—and to learn to think in terms of several things all happening at once, with each affecting the others simultaneously.

You will be given illustrations in the form of actual behavioral performances allowing you to experience control directly, as did the exercises in Chapter 2, but going a step further in being analyzable quantitatively. You will be surprised at how your understanding of the way behavior really works will be enhanced as you seriously work through the demonstrations.

All of the computations necessary will be provided, with no steps skipped, so you should be able to follow them without getting lost if you do your part, which is to do the calculations for yourself. It is easy to assume that, since the numbers are provided in tables, you don’t have to work them out. This is a common mistake of people who think that they have no aptitude for mathematics, and it can become a self-fulfilling prophecy. One can no more build in the habits of quantitative thinking by simply looking at examples of it than one can learn to talk by simply listening to other people.

It might seem that after a history of over 120 years, the science of psychology would have a crisp answer for the question heading this chapter. It does not, and that fact is behind most of the theoretical difficulties which have held psychology back. Behavior is a catchall word meaning almost anything you want it to mean. It can mean anything from tensing a muscle to solving a differential equation, from jumping after a pinprick to despairing after getting your bank statement. Behavior (in the life sciences) is simply everything that any organism can be seen doing. This definition is so nebulous, philosophers of behavioral science can’t even decide whether falling down an elevator shaft is a behavior. This is why theory in psychology is in such poor shape: how can we devise a theory of everything which happens?

In the attempt to bring order into this enormous variety of phenomena, psychologists have tried to simplify it by using generalizations—observations which, while somewhat vague, can be applied to many instances of behavior differing in detail. But generalizing, while it can give us an illusion of understanding, becomes less useful as it becomes more successful. In order to generalize, we have to ignore observations, to say that things are the same which are not the same, to proclaim that all men are mortal and ignore the fact that they can be mortal in immensely significant yet different and even contradictory ways. We are forced to treat classes of human beings as if all individuals classified in the same way are in fact the same. We are led to say that human behavior has certain properties even when the experiments behind these statements actually show that 30, 40, 50, or even 60 per cent of the subjects in those experiments did not have those properties. The more we generalize about human characteristics, the more falsehoods we utter.

When a physicist proposes a “general law” for the behavior of matter, what is meant is something quite different from what psychologists have meant by a generalization. In physics, the term “general law” means a law which applies generally: all of the time, in every case, with no exceptions whatsoever. If this criterion were applied to the laws of behavior which have been found in the life sciences, we would have to conclude that no general laws of behavior have been found. To arrive at laws of human or animal behavior which are general in the sense meant by a physicist, it is clear that we will have to go about the whole project in a new way.

In the previous chapter, we began by asking about an exceptionally simple kind of behavior moving a finger. In this chapter, we’re going to analyze an even simpler behavior: holding something still. We’ll be examining “the rubber-band experiment” from Powers (1973), a demonstration of control theory which makes clear all of the visible relationships of control. One great advantage of using this demonstration is that anyone can afford the equipment needed to set up the experiment and try it out. One small warning you will find this chapter tough going if you read it without a couple of rubber bands at hand to help make the relationships clear.

4.2 The Rubber-Band Experiment

To do the rubber-band experiment, you need two rubber bands. Pull each one’s end through the other so they become tied together end-to-end. In a two-person experiment, one person is the subject and the other is the experimenter. The subject takes hold of one end of the rubber bands, the experimenter takes hold of the other, and the two people stretch the rubber bands slightly, just above the surface of a table. The task which the subject tries to carry out is to hold the knot exactly over a mark on the table while the experimenter moves the other end of the rubber bands. If the table top is featureless, make a spot on a piece of paper. For our purposes here, let’s suppose that the people agree to keep the rubber bands lying in one line running from left to right between them, so rotations about the spot don’t come into the picture. You can do this experiment two-handed, letting one hand be the experimenter and the other the subject, just to see the relationships.

The rubber-band experiment illustrates some basic feedback relationships which can be found in many examples of ordinary behavior. The position of the knot represents something in the environment which you can perceive; the rubber band between your finger and the knot represents the effects your behavior has on that perceived thing. The other rubber band represents a connection

Let’s go on just a little further with the rubber bands, before getting into more interesting subjects. We analyzed a situation in which the subject was controlling the separation of the knot from the spot, keeping it at 3 inches to the right. To model this action, we had to introduce a reference signal against which the perception of separation was compared. If you tried the reasoning-backward method using a high loop gain, such as 200 or 2000 (the range of actual loop gains you could measure), you found that the knot was kept at this reference separation from the spot—quite accurately, regardless of the amount of disturbance. Starting here, we can bring in a new aspect of the experiment which leads to a new kind of control system.

5.2 Motion Control

If the reference signal gradually changed, at a given moment the knot still would be held at a distance from the spot which corresponds to the setting of the reference signal. You can observe this effect if you do the experiment again, playing the part of the subject while someone else acts as experimenter. Just start the knot about 3 inches to the left, and make it move slowly and smoothly to a position 3 inches to the right. While this is going on, the experimenter should vary the pull on the rubber bands, increasing and decreasing it. There will be some effect on the smooth motion of the knot, but not much—if the experimenter doesn’t take this as a challenge and move so much and so fast that you have difficulty keeping the knot where you want it. If the experimenter seems determined to have an effect on the knot, explain what you’re trying to do.

As you move the knot slowly to the right, you will notice two things going on at the same time. First, you’re moving your end of the rubber band so the knot is where it should be all during its movement. This requires that you sometimes pull more and sometimes pull less—your hand doesn’t move the way the knot does, because of the disturbance. You can’t make the knot stay still or move smoothly by holding your hand still or moving it smoothly. If you tried that, the changing disturbance would make the knot move very irregularly. The second thing you’re doing is making the position of the knot change in a specific way: a smooth slow motion toward the right. You know how the motion does look and

In the remaining three chapters of Part 3, we shall expand our speculations about the development of control systems in the growing person, and then discuss how these views apply to the traditional fields of developmental psychology, learning, motivation, and perception. Thus, Chapter 7 corresponds to an overlapping of physiological and developmental psychology in the traditional approach, Chapter 8 corresponds to the traditional chapter on the basics of learning theory, and Chapter 9 examines the subjects of motivation and perception as aspects of control, rather than as autonomous functions within the organism.

The model of the organism which we present here—as a hierarchically organized environment control system—contains several assumptions which need to be made explicit.

(1) It is the organism, not its nervous system, which should be perceived as an environment control system. The nervous system comprises a major component of that function, but as part of the organism as a whole. It, along with the glandular systems, functions first of all to regulate the internal environment of the body —the tissue-fluid medium in which the cells of the body live. This medium must be kept at the proper temperature and chemical composition essential for the life of the cells.

All of the activities of the body (including behavior) ultimately serve this purpose—preserving the life of the organism by maintaining the internal conditions needed for the life of the cells. The body can control its internal environment, in the long run, only through its ability to control its external environment. This is accomplished mainly with action by its muscles—by moving the body or other objects around—although many organisms also exert control over their environments by means of chemical control systems.

Some glandular secretions, such as pheromones, act directly upon the environment, creating “disturbances” in the variables which other organisms control. Certain genetically structured (instinctual) systems apparently are set at “0” until a given pattern of sense receptor signals (chemical, visual, or auditory) is received; then the system turns “on”—to perform genetically programmed interactions between members of a species. Such signal patterns have been called “releasers” by the ethologists who have described these phenomena. When glandular outputs, rather than actions, supply the signal patterns functioning as releasers, chemical reactions play a part in controlling the external environment; however, the major aspect of chemical control is found in the internal environment.

Such biochemical regulation seems involved especially in aspects of sexual behavior. In some species, pheromones act like the scents of flowers, being detected by special sense receptors possessed by members of the opposite sex. These sense receptors initiate feedback signals in genetically fixed control systems regulating programs for reproductive behavior.

Genetic factors apparently are mediated through the regulatory function of

Chapter 5 presented Powers’ most recent tentative model of the control-system hierarchy of a fully developed human being. How does that system develop? The field of developmental psychology is concerned with the description of the increasing complexity of overt behavior as the human organism grows into maturity. But it relates the development of behavior to its “psychic” mechanisms, leaving the discussion of the development of neuron circuitry mainly to the field of biopsychology. This latter field, on the other hand, investigates brain development with only the barest beginnings of relating it to overt action. Such connections began to be explored in “evoked potential” studies. Evoked potential studies attempt to isolate specific patterns in EEG tracings associated with single overt “behaviors,” using special techniques to tease out the pattern of interest from all of the other “squiggles” in an EEG tracing.

For example, in one such study (Shucard, Shucard, and Thomas, 1977) the experimenters used “evoked auditory potential” measurements to test the hypothesis that random clicks superimposed on either a verbal task—listening for “key” words in a speech—or a musical task—listening for a melody in a musical piece—would show a larger brain wave in the right hemisphere for the musical task and a larger wave in the left hemisphere for the verbal task. They found evidence for their hypothesis, thus adding support to the view that the left hemisphere of the brain controls sequential activities like the train of words in speaking, while the right hemisphere controls global activities, like the melodic “figure” of music.

Considering the above example in the light of Chapters 4 and 5, you can imagine how some particular behavior is produced by a complex control-circuit of nerves. You can speculate that it would change the shape of a brain wave tracing when any particular circuit is activated. Then imagine how the EEG waveforms might be blends of many such circuits’ waves, all acting at the same time, as EEG researchers have long suspected. How do such circuits come to be organized in the first place? Powers (1973) proposed a general model for the development of the kind of control hierarchy we discussed in Chapter 5. The newborn, lying in his or her bassinet, has, as far as we know now, all the brain cells he or she ever will have. But obviously, he or she does not have control circuits for even such a seemingly elementary task as tracking a moving object with his or her eyes beyond a narrow span, as Piaget (1965) noted.

8.1 Learning as Development of Control Systems

Chapter 7 explained the theory of how new control systems develop. Powers’ model postulates that error conditions within the intrinsic-system of an organism must result in increased neural firing throughout the organism’s nerve network. Eventually, when intrinsic-system errors decrease, the circuit (or path of the signals) active at that time becomes a new control system to counteract such errors in the future. The new circuit is a control system for the disturbance which initiated the error condition. When an intrinsic-system error condition (such as need for water, food oxygen, or other genetically referenced life-support variables) leads to widespread signalling in the brain, it results in movements which are random until a reduction of intrinsic error occurs. Then the Organizing/ Reorganizing System shuts down, the current configuration of neural circuitry is preserved, and we call the recorded action pattern “behavior.”

Recall the example in Chapter 7, in which “frantic flailing about” became ‘swimming” at the moment when oxygen-level error signals began to decrease. In that example, intrinsic-system errors began to decrease when the disorganized collection of arm and leg movements coalesced into the proper, rhythmically alternating pattern we recognize as swimming. While all this was happening, it also would have been necessary for analogues of the oxygen feedback signals and of the arm and leg movement sequence signals to be stored together as a unit in memory. Thus, on a future occasion (of plunging into water, for example), oxygen-level error signals could not mount up to the extreme condition of the earlier case, because before they could trigger that extreme condition, corrective action would have begun in the new control system which was built during the prior occasion.

The field of developmental psychology grew partly in reaction to the fact that learning theorists in traditional psychology took almost no account of the influence of previous experience upon learning during childhood. Based as it was in the Pavlovian conditioning paradigm, learning theory was filled with research studies about variables relating to “reinforcement schedules,” motive strengths and the like—all treating each instance of learning as if it began on a blank slate, independent of what the individual could and could not do before the current learning episode began. Educators and clinicians working with children found much lacking in this approach of learning theory. They knew that education is in many ways a sequential process, in which new tasks are begun upon the level of mastery acquired in previous tasks. Hence, they became interested in the psychoanalytic concept of “stages” of development, as formulated by Freud and greatly elaborated by Erik Erikson (1950).

9.11 Freud

Sigmund Freud (1905/1953) postulated three stages through which he claimed all children progress during normal development (and a fourth, inserted between childhood and adult phases of the third), as discussed in chapter 11, below. His basic theory was built upon a model of energy investment, rather than being directly concerned with learning, but it contained a clear implication that the child acquires different abilities during the oral, anal, genital, and latent stages or developmental periods. During the oral stage, the child’s attention and awareness is focused mainly on all the activities surrounding intake of nourishment—the mouth zone—and the major energy of growth is invested there. Then, sometime around the age of 2, the major focus of attention shifts to matters surrounding excretory control, in potty training. Between 3 and 5 it shifts again, to interest in genital development. Finally, in the elementary school years, in what Freud called the “latent period” in genital maturation, it shifts away from the individual’s body to the acquisition of skills for dealing with the outside world.

Chapter 12 analyzes several studies from the overlapping traditional areas of learning, perception, and social psychology. The control-theory perspective on these studies should further enhance viewing human behavior from the standpoint of the individual outward toward the environment, as opposed to the traditional view of the person as object of environmental forces, rather than as controller of them.

In traditional psychology, perception has been viewed as an independent function within the organism. It has been distinguished from sensation, the process by which an organism takes in “raw” information about its environment. Perception has been regarded as the process by which sensory data are converted into the more complex patterns which most behavior has been thought to involve.

The traditional approach was to separate behavior into three independent processes: perception, motivation, and action. Perception was not organized in a hierarchy of increasingly complex variables, as in the control-system scheme of Part 2. In this chapter, we compare the traditional view of perception, and some of the research studies derived from it, with the control-theory view. In the control-theory view, the perceptual signal is the input aspect of the control process at each level in the hierarchy; it is not independent of what the organism is doing. Perception and motivation, as aspects of the individual’s control over his or her environment, are separate processes only in a very artificial sense.

We shall approach the subject matter according to the traditional organization in the literature, examining it from a control-theory point of view. Researchers using the traditional approach have gathered many interesting observations, and they have helped to define many of the issues concerning how physical variables in the environment become information for the person. We shall take a look at some of their findings, accepting the “phenomena” as defined by the psychologists specializing in the traditional field of perception, keeping in mind that in the traditional approach, perception has been considered to happen before the organism acts, rather than as an inseparable part of action.

Consider the research studies on the following topics, and look particularly for what the subjects in each experiment had to do as the experimenter gathered his or her data. This will provide a bridge for moving from the traditional to the control-theory orientation in understanding the subjects’ actions in the experiments described.

The topic of personality grew up in psychology as the study of the relatively stable aspects of a person’s behavior. The stable aspects of behavior show up as habitual attitudes, moods, styles, or patterns of action displayed by the same individual over extended time periods and different situations. The terms used by observers who know the person well (which may or may not include himself or herself), for these consistent patterns, comprise the description of personality. Such descriptions often employ adjectives like “friendly,” ” “generous,” “taciturn,” “stingy,” and their opposites. These terms express a general tendency; for example, we say, “X is a friendly person,” thus categorizing something we think we notice running through many of X’s actions.

However, in recent decades this simple, common-sense notion has become more complicated by the work of some professional psychologists. As stated in a recent article in American Psychologist, “For the past two decades the person situation debate has dominated personality psychology and had important repercussions in clinical, social, and organizational psychology.” (Kenrick and Funder, 1988, p. 23)

What is the person-situation debate? It is, briefly, a controversy over whether personality is a true aspect of human nature, or a misperception resulting from a peculiarity in the way people tend to explain experiences to themselves. Do people generally act in individually characteristic ways, even in the same environment, because of their separate needs and motives? Or do we act predictably in many situations because the environment in each situation tends to require the same action from anyone in those circumstances?

Considerable evidence has been gathered in seeming support of each of these opposing views. For personality, for example, a research measure called the “guess who technique”—used extensively to study social relations among grade school children, with questions such as “who is the most popular,” “who is the quietest,” and “who is the friendliest”—has found regularly that children show considerable agreement in making such identifications (Hartshorne, May, and Shuttleworth, 1930; MacFarlane, Honzik, and Davis, 1937). This suggests that grade-school children already have enough consistency in some characteristics to enable peers to recognize each other by them.

On the other hand, attempts to use established personality measures for various kinds of practical applications have frequently proved disappointing. An

The concept of reorganization was described theoretically in Chapter 7. There we speculated about how a person’s learned hierarchy gradually gets reorganized. This happens when existing systems cannot bring some perceptual variable under control. Error signals increase, intrinsic-system values begin to be violated, and the reorganizing system becomes “turned up.” The reorganizing system then injects random signals into the hierarchy until the affected life-maintenance (intrinsic-system) values return to normal.

The current organization of the hierarchy remains relatively stable until the next reorganization is forced by a new loss of control. The individual becomes more complex, or more sophisticated, when reorganization results in the ability to control situations which he or she previously could not. Not every reorganization necessarily has this result, since any reorganization which is followed by reduction in intrinsic error signals tends to persist. Thus, reorganization theoretically can result in circuitry which might prove ineffective on some future occasions, if intrinsic errors happened to be reduced coincidentally to the ongoing reorganization.

The above view of reorganization contains complications which are not initially apparent. Different persons can come to control different variables in controlling what would seem to be the same environment. For example, among a group of people playing golf, one could be mainly controlling a desire to see a better score than his last one, another could be mainly controlling a desire to make a good impression on another player, who happens to be his boss. Still another could be controlling a personal value, such as demonstrating that he can outdo any of his peers, and so on. How each person performs does not just depend on his or her technical skill. The way one uses one’s technical skill must satisfy all of the principles which that person’s self system has called up in that situation at that time. The final result might be any combination of the lower-order repertoires producing the perceptual condition which satisfies the highest-level reference signal.

The traditional field of social psychology is something of a hodgepodge. Different psychologists have seen it variously as the study of interactions between individuals, the influence of groups upon individual behavior and development, the processes of group formation and action, or the principles of communal/ organizational behavior. Others organize the subject in terms of supposedly different, basic processes of social behavior, such as attitudes, social motives, social perceptions, social norms, interpersonal attraction, and affiliation. However, some anthropologists, and notably the American investigator of social class, Lloyd Warner (1952), view all social behavior—all forms of group action, as well as cooperation and competition between individuals—as expression of one basic principle: control over the environment.

Warners’s view is congenial with the control-theory view that, in the long run, all action ultimately aims to maintain the necessary conditions for life of the organism and the species. In this sense, there is an unbroken chain of cause and effect from the most fundamental life-sustaining actions, such as breathing, to the most complex human social actions, such as fulfilling one’s role in society. Complex control functions, such as cooperation, directly or indirectly improve the conditions of survival. In the course of evolution, those humans who cooperated in communities survived more often than those who lived alone.

This level of generality says nothing, of course, about any specific kinds of social behavior, attitudes, or values. But the postulate that social behavior, like all other behavior, serves the survival of the organism (and hence the species) helps to organize our thinking about social behavior. It suggests two specific questions about any social behavior of interest: What condition is kept under control by the action? Where does that fit in the control hierarchy of higher level systems?

To illustrate the above, let us refer to one of Lloyd Warner’s analyses of large-scale social behavior—his hypothesis about the underlying purpose of Memorial Day parades, ending at the cemetery on the Memorial Day holiday. Warner (1953) offered extensive evidence for his view that the rituals of the celebration serve the purpose of bringing together community members of all social classes and walks of life in commemorating loved ones who died in war! He claimed that the particular Memorial Day rituals carry largely unconscious, symbolic messages to the different classes of society that they, and their dead, had done the right thing in obeying the leaders of society by going to war. The main sym-

Clinical psychology is a subfield of applied psychology in which the psychologist applies the facts, theories, and methods of psychology to the goal of helping people solve personal problems. Clinical psychologists engage in two major kinds of activities, psychological testing and psychotherapy. Clinical psychologists currently do not prescribe drugs. For those instances where drugs might be helpful, clinical psychologists work with consulting psychiatrists or other physicians.

In this chapter I (DMG) shall describe applications derived from control theory as I use them in my work as a clinical psychologist. They help to show how this new model is beginning to contribute to the clinical field of applied psychology.

The first such contribution is the possibility of bringing about a unified approach to assessment and treatment. This means that the same set of concepts forms the basis of assessment and treatment methods. This makes it easier for the clinician to translate the results of psychological testing into diagnostic and treatment suggestions. A second benefit is that the clinician can measure therapy progress in the same terms as the initial assessment.

14.1.1 Brief Review of Control Theory Concepts

The terms printed in italics are the basic concepts of control theory which I shall use in this presentation. A person checks perceptions (perceptual signals, p) against his or her reference value for the perception in question (reference signal, r). As an example, imagine a person driving a car. He or she compares (automatically, unconsciously) what he or she is sensing against what he or she should be sensing. One typical driving goal might be: Am I safe? The goodness of fit between the current perception, p, and the reference perception, r, determines the size and direction of the momentary error condition (error signal, r-p). Error signals can vary from zero (perfect match) to some large (positive or negative) number. The person’s brain functions to keep error signals as small as possible.

Continuing with the above example, if the driver perceives that the car ahead has stopped suddenly, this creates an error signal with respect to the goal of driving safely. When there are no error signals related to a given perception, actions related to this perception do not change. When there is a non-zero error signal, actions do change to bring about a correction. Error signals in the control system

When scientists introduce a new theory—a new way of understanding reality—they typically devote considerable energy initially to offering evidence that it “works.” If it is a theory about how to do something, the attempt is to prove that it has the looked-for results. If it is a theory for explaining something, the attempt is to demonstrate that it explains things which its predecessor (and rival) cannot explain, or explains inconsistently.

I (RJR) have observed both of these processes in the course of my career in psychology. First was the then-new “Client-Centered Therapy” of Carl Rogers (1951). He pioneered in applying psychological research to evaluate his new method of doing psychotherapy. The initial studies on this approach were devoted to determining whether his method worked—that is, whether individuals treated with the client-centered approach gained improved mental health as compared with similar people who remained untreated. Later on, after establishing that client-centered procedures did work in many cases, further studies went into the more specific questions of how well it worked, with whom, under what conditions, what the effective ingredients were, and many other questions concerning details. Eventually, these early studies grew into the major specialty of psychotherapy research, involving many of the different therapeutic schools, and still continuing to develop at present (see Garfield and Bergin, 1978/1986; Howard, Kopta, Kraus, and Orlinsky, 1986; Kiesler, 1973; Parloff, London, and Wolf, 1986; Rice and Greenberg, 1984).

The other process of developing and evaluating a new approach which I have had the pleasure of observing has been that of the Control Systems Group, with initial research studies aimed at testing Powers’ hypothesis that behavior is the control of perceptions, rather than actions. Group members pioneered new research methods in the course of promoting the paradigm revolution in basic psychological theory represented by Powers’ postulate, as shown in this volume. The fundamental nature of the new direction perhaps has been expressed best by Marken (1988) in a paper distinguishing the fact that humans control their perceptions from the research methods needed to analyze how this is done, and by Runkel (in press), who discriminated two fundamentally different methods of behavioral research: the method of relative frequencies and the method of specimens.

Runkel describes the method of relative frequencies as a method useful for

A paradigm shift in science, Thomas Kuhn (1962/1970) maintained, does not result simply in new facts coming to light, but in new kinds of facts becoming of interest, and in new explanations for many old issues. The control-theory paradigm in psychology inspires just such reexaminations, as shown in Part 4 of this volume, and in Hershberger’s (1989) volume on the problem of volition.

Several other issues of perennial interest in psychology take on a new look from the standpoint of control theory. They invite theoretical speculations, even though critical tests of such speculations may lie uncertainly in the future. These are such issues as the nature of consciousness, determinism versus free will, attention and control, the nature of projection and prejudice, and the role that randomness in reorganization plays in the development of individual differences.

16.2 Speculations on the Nature of Consciousness and Will

When the foundations of psychology as a separate discipline were being laid down in the mid-19th century, the first two schools, the Structuralists and the Functionalists, tended to equate mind with consciousness, without really offering any hypothesis as to how consciousness is produced. This problem was left unresolved as Behaviorism simply steered around it, instead of confronting it.

In fact, as recently as 1977, we find the neurophysiologist Sir John Eccles saying, “I have the indubitable experience that by thinking and willing I can control my actions…. I am not able to give a scientific explanation of how thought can lead to action…. When thought leads to action, I am constrained, as a neuroscientist, to conjecture that in some way my thinking changes [my] neuronal activities…. Thinking thus eventually comes to control the discharges of impulses… [to] my muscles.” (Popper and Eccles, 1977, pp. 282-283; italics mine, RJR)

If we understand Eccles correctly in the above quote, then he is saying that a conscious thought leads to an action in the sense of producing it. But looking at the problem in terms of Powers’s conception of how the the control hierarchy functions, we could restate the description to say that the thought is the perception—in the higher-order control system—of the goal which that level issues (as a reference signal) to the order below. The perception of the body’s movement in the environment is being brought to match the thought of it, perceived a moment

Allport, G. W. (1960). Personality and social encounter. Berkeley, CA: Beacon Press.

Anand, B. K., Chinna, G. S., & Singh, D. (1962). Effect of glucose on the activity of hypothalamic “feeding centers.” Science, 138, 597-598, 1962. Cited in McFarland, 1981.

Andrews, J. D. (1989). Integrating visions of reality. American Psychologist, 44, 803-817.

Asch, S. E. (1940). Studies in the principles of judgments and attitudes: II. Determination of judgments by group and by ego-standards. Journal of Social Psychology, 12, 433-465.

Asch, S. E. (1953). Effects of group pressure upon the modification and distortion of judgments. In D. Cartwright & A. Zander (Eds.), Group dynamics: Research and theory. Evanston, IL: Row, Peterson.

Asch, S. E, Block, H., & Hertzman, M. (1938). Studies in the principles of judgments and attitudes: I. Two basic principles of judgment. Journal of Psychology, 5, 219-251.

Avia, M. D., & Kanfer, F. H. (1980). Coping with aversive stimulation: The effects of training in a self-management context. Cognitive Therapy and Research, 4, 73-81.

Bandura, A. S. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84,191-215.

Bandura, A. S. (1978). The self system in reciprocal determinism. American Psychologist, 33, 344-358.

Bates, E (1976). Language and context: Studies in the acquisition of pragmatics. New York: Academic Press.

Bayliss, L. E (1966). Living control systems. San Francisco: W. H. Freeman.

Beitman, B. D. (1987). The structure of individual psychotherapy. New York: Guilford Press.

Bennett, S. (1979). A history of control engineering. New York Peter Peregrinus.

Berkowitz, L. (n. d.). Quoted in Elms, 1975.

Berscheid, E (1985). Interpersonal attraction. In G. Lindzey & E. Aronson (Eds.), Handbook of social psychology (3rd ed.). New York Random House.

Berscheid, E. (1988). Some comments on love’s anatomy: Or, whatever happened to old fashioned lust? In R. J. Sternberg & M. L. Barnes (Eds.), The anatomy of love. New Haven: Yale University Press.

Berscheid, E, & Walster, E (1978). Interpersonal attraction (2nd ed.). Reading, MA: Addison-Wesley.

Bierhoff, H. W., Cohen, R. L, & Greenberg, J. (1986). Justice in social relations. New York: Plenum Press.

Blass, E M., & Epstein, A. N. (1971). A lateral preoptic osmosensitive zone for thirst. Journal of Comparative and Physiological Psychology, 76, 378-394.

Bloom, L. (1973). One word at a time: The use of single word utterances before syntax. The Hague: Mouton.

Boneau, C. A. (1974). Paradigm regained? Cognitive behaviorism reinstated. American Psychologist, 29, 297-309.

Boring, E G. (1950). A History of experimental psychology (2nd ed.). New York Appleton-Century-Crofts.

Boulant, J. A. (1980). Hypothalamic control of thermoregulation-neurophysiological basis. In P. J. Morgane & J. Panksepp (Eds.), Handbook of the hypothalamus (Vol. 3, Part A). New York Marcel Dekker.

Bourbon, W. T. (1988). Modelling control by dyads. Paper presented at the meeting of the Control Systems Group, Kenosha.

Cannon, W. B., 37
Cantor, N., 155
Carismith, J. M., 157
Cartwright, D. S., 180
Cartwright, R. D., 219
Carver, C. S., 20, 154, 156, 189
Castellucci, V., 115

Accomodation, as used by Piaget 125

Action, as creation of behavior 62,198, 213;
as primary fact of behavior. 34, 137, 209 if.

Adaptation: 113, 115, 118

Advertising industry: 138

Affiliation: 171, 173

Aggression: 172

Albert, in the ‘little Albert experiment” of J. B. Watson: 16

Anorexia: 193

Anti-psychotic agents: 195

Anxiety: 131, 138, 217

Aplysia:113,116,118,119

Approach-approach,
approach-avoidance,
avoidance-avoidance (types of learning behavior): 168

Approach response 208

Aptitude tests: 12

Artificial Intelligence: 78

Assimilation, as used by Piaget: 125

Association, free, Freud’s method of: Ill

Assodationism: 111

Associations, in the sense of learned connections: 6,16, 109, 110, 115, 117

Attention, problems of: 198, 209

Attitudes: 147, 171, 173, 177, 213, 214

Attribute-signals: 66

Attribution theory: 173; see also Chapter 11

Autonomic nervous system: 107

Autoshaping: 19