CYBERNETIC APPLICATIONS
The development of a more scientific psychology has increased the need for mathematical models or symbolic representations of the most recent learning theories. These are mechanistic in nature and are classified under the general concept of cybernetics, which attempts to describe control processes in precise mathematical terms. The term cybernetics comes from a Greek word, Kybernetes, which means pilot or governor. Much has been written on this discipline in relation to the study of control processes in machines, organisms, and social groups.
On a broader basis, cybernetics combines the views from different but related fields of knowledge, including engineering, mathematics, physiology, biology, and psychology. Cyberneticians contend also that principles of learning and purposeful behavior characteristic of machines apply to human functioning. The theory does not imply that electronic or mechanical analogues can adequately represent the functioning of the central nervous system, but, rather, that living organisms parallel the over-all stimulus-response behavior patterns of automatically controlled machines (servomechanisms).
The central principal, as defined in the last chapter, is that goal-directed organisms as well as machines utilize error-correcting information to achieve purposeful behavior or equilibrium. This indicates that the system is in negative feedback. If the information fed back in an automatically controlled machine or organism causes the error to increase, with a resultant instability or breakdown, the system is said to be in positive feedback.
These principles are similar to Bernard's and Cannon's views on homeostasis. However, what is new is that dynamic equilibrium involves a perpetual exchange of energy with the environment. What leaves the organism is called “output,” and what goes into it, “input.” Man, as an open control system, therefore, receives from his environment and makes his contribution to it—but input and output do not interact.
With this model in mind, it appears that the hypnotic interpersonal relationship, as well as the resultant hypnotic conditioning procedure, depends largely on the manner in which the subject is willing to have his perceptual mechanisms restructured. If the subject incorporates the therapist into this system, the type of feedback is altered in the control process. This concept allows us to understand some of the fundamentals of behavior. It is hoped that the life sciences will join forces with the behavioral sciences to reveal other hidden factors in the tangled areas of human behavior. Hypnosis, because of its discriminative ability, affords an experimental device for penetrating this bewildering maze.
From the rapid strides of the engineering and mathematical sciences, cybernetics is providing newer applications for the older concepts of self-maintenance or equilibrium. These applications should lead to a better understanding of higher nervous system functioning, especially in reference to hypnotic behavior. Also, it is more apparent that research possibilities and new insights into the psychology of learning and the processing of information (thinking) will ultimately be developed to understand the complex neuropsychologic mechanisms of human relationships, responses and behavior.
Critics of cybernetics believe that it, like behaviorism, is an excellent theory from the viewpoint of scientific methodology, but that, like most of the behavioristic theories, it is inadequate since it neglects the role of man's creativeness (something no machine yet possesses). Overlooked, too, is the role of the essential meaningfulness which underlies man's experiences, past and present.
CYBERNETIC MODELS OF LEARNING
There are several models for understanding the mechanisms involved in learning, especially as it applies to the communication of information in man and the machine. Outstanding among these are the feedback, information theory and probability theory models. The first two are related to the design of telecommunications equipment—all are concerned with goal-directed behavior, probability and decision making. The probability theory developed from the strategy of games and has been used in the prediction of behavior in several other fields. Already it is being suggested that, if human specifications can be recast into machine-compatible specifications, this “could give clarity and rigor to the language and concepts of psychology, and open the possibility of man-machine comparisons, cross-simulation studies, and substitution experiments.”32 Also, these experiments would make possible the scientific validation of hypotheses, and would be especially valuable if one wishes to use the machine to study any factor which is not accessible in people.32
The following hypotheses are presented with due regard to the dangers involved in contending that there might be a comparative relationship between the machine and man. Nevertheless, there seem to be no objections when engineers attempt to design better machines by studying the behavior of living organisms. This new science is known as bionics. The physical and behavioral sciences are now revising our notions of communication processes as they relate to signal detection theory, neural control, and self-regulating features of brain functioning. These disciplines are destined to play an ever-increasing role in medical education. It is for this reason that cybernetic principles are presented in this chapter.
The Feedback Model
The feedback principle, though not new, is a unique method for viewing things: it introduces a new model for thinking about thinking.30 A good example of negative feedback is its use in walking. Kinesthetic and postural feedbacks from muscles, joints and tendons automatically make the corrective movements necessary for effective locomotion. The gait of an alcoholic or a tabetic is illustrative of disequilibrium or positive feedback. Here, some of the necessary feedbacks are missing, and this results in incoordinated muscular activities of an oscillatory rather than a purposeful character. The placing of the feet has to be controlled visually—a less satisfactory substitute feedback.
In learning and adaptive responses to everyday life situations, the feedback principal constantly “monitors” behavioral response; the success or the failure of the results modifies future behavior. In other words, learning is by trial and error—adaptive. For instance, we have emphasized that, during hypnotic induction, the motivated subject automatically makes full use of his own internal feedback mechanisms (the ideosensory and ideomotor activities) for achieving a goal (hypnotic relaxation). Also, in the hypnotic interrelationship, patient and therapist use feedback mutually to improve their respective reactions to one another's communication processes. As these aspects of feedback mechanisms and cybernetics become more applicable to behavior disorders, the fundamental role and technics of hypnotherapy may greatly expand in this respect.
The Information Theory Model
It is not possible in a book of this type to treat the mathematics of information theory and its quantitative applications to the problems of information transmission, storage or its retrieval. For a deeper understanding of the scope and the application of information theory, the classical work of Shannon and Weaver is recommended for those with mathematical training.28 However, an understanding of the principles of information theory is valuable even though one lacks a knowledge of higher mathematics. The following discussion will, therefore, avoid abstract mathematics, and will be directed to those with little or no previous acquaintance with probability or information theory.
The relevance of information theory to psychiatry has been described.6 In psychotherapy, particularly in hypnosis, we are interested in meaningful measures of the subjective or semantic value of the information conveyed to a patient or to a therapist. If such measures are available, there is a possibility that scientific methods can be applied to a field which must otherwise remain an art. Therefore, it might be instructive to compare the technics employed by physical scientists in measuring and studying information, in the hope that information theory, allied with other theoretic and experimental work, will in the near future help to explain human communication processes on a more scientific basis.
In the physical sciences, the information in a message is defined in a purely statistical way without any reference to the importance of the message. The amount of information gained from receipt of a message is measured in terms of the amount of uncertainty removed by the message.* The resultant information, which leads to a reduction of uncertainty, enables decision making based on knowledge rather than on guesses. These ideas are intuitively obvious, but, until they are translated into the exact language of mathematics, it is not possible to formulate the concepts in other than loose verbal terms.
In engineering design, information theory answers some very basic questions as to the ability of a communication system to transfer information from sender to receiver in the presence of “noise”—defined as any type of interference. Even the clearest message can lose some of its meaningfulness before its reception. This is known as entropy, and some entropy occurs at all levels of human communication. However, the theory allows one to state explicitly how information is lost due to noise in the communication channel. Also it allows one to determine the extent to which the signal must be strengthened in order to transmit the desired amount of information in the presence of noise. This is a fundamental consideration in design of all communication or telemetry systems.
At the risk of oversimplification, these concepts seem particularly germane to the objectives of psychotherapeutic communication, especially during hypnosis. The hypnotherapist, acting as a transmitter, wishes to communicate or encode information to the patient as accurately and reliably as possible in the presence of noise. The noise may take the form of disturbing sounds in the environment or internal noise generated in the “receiver” (the patient) by virtue of his unreceptive attitudes or preoccupation with irrelevant thoughts. Understanding the technics used by the physical scientist to cope with the problems of noise may offer interesting possibilities for improving the two-way communication† during any interactional relationship.
The following simple illustrations show how the amount of information possessed (the ability to select from a large number of alternatives) allows correct predictions or decisions to be made. The unit of information is the binary digit or “bit.” It represents the amount of information necessary to resolve two equally likely alternatives. Symbolically, these two alternatives may be represented as “yes” or “no,” or “1” and “0.” Using such a pair of symbols, it is possible to encode any message—a printed page, a symphony, a picture—with any desired degree of detail. For instance, a message which gives a person's sex contains one bit of information, since there are only two equally likely alternatives. To resolve 8 equally probable alternatives, a message containing 3 bits is required. The first bit reduces the alternatives from 8 to 4, the second bit from 4 to 2, and the third from 2 to 1. Thus, in general, each bit reduces the number of alternatives by one half.
It is believed that all of man's experiences, memory and thought are based on such simple particles of information. Every perception is a pattern of impulses—unique only in that certain nerve fibers “fire” digitally (“all,” “yes,” or “1”), while others do not (“none,” “no,” or “0”).
McKay believes that general information theory may provide a mathematical description of the nature of human behavior, that is, a reduction of all communication processes to statistical data.20 He further contends that the thought processes of living organisms may soon be imitated by mechanical means. Already it has been applied usefully to learning problems which involve discrimination, judgment and decision making.19
Our interest in this fascinating area is particularly relevant during hypnosis communication, in which the prime objective is to convey meaningful verbalizations in order to narrow the attention span to a given idea and, as a result, eliminate “noise” in the form of semantic confusion. This reduces entropy or the number of possible meanings or alternatives.
For example, messages which have a high specificity contain more bits of information than do generalities. Suggesting to a subject that he is not “asleep,” “unconscious,” or in a “trance” does not effectively convey information. On the other hand, telling the subject that he is in a state of relaxed attention identifies hypnosis as a positive state, and eliminates all alternatives. Hence the therapist must be specific in saying what he means as well as meaning what he says! One might say that the heightened perception or acuity characteristic of hypnosis acts as a filtering “device” similar to that used in machines for screening out irrelevant signals.
Communications systems which stress accuracy in the receipt of a message utilize a high level of redundancy (repetition of the same information). This applies particularly to hypnotic induction, in which, for instance, a phrase such as, “Your legs are getting heavier and heavier,” is used again and again to obtain the correct ideosensory and ideomotor responses. Once these are identified as correct, the chances that the subject will make the same responses again will be greater than ever. With sufficient repetition, the correct responses will become a virtual certainty, that is, they will be learned automatically and become a “habit.” It is by such “ideoid” phenomena that one's beliefs are processed into convictions.
This section has dealt primarily with some of the basic aspects of information theory. Very few comments have been made regarding semantic information, not because this subject is unimportant, but rather because there is at present no sound quantitative theory for treating semantic information. Statistical information theory, however, has relevance to semantics insofar as it tells us what confidence we can place in the information received as truly representing the information sent.
One might conclude that information theory provides insight for analyzing and improving storage and communication processes, but does not unravel the bewildering complexities associated with significance, meaning, or value judgments.
The Theory of Games Model
Von Neumann's mathematical theory not only estimates the probabilities of outcome but makes decisions (best “bets”) based upon a course of action which has the greatest value or utility.33 Since both of these are involved in motivation, the theory may have relevance to the psychology of individual learning. It may help to treat, in meaningful quantitative fashion, the outcome of an action that is not completely determined.
This theory closely parallels the field approach used in cooperative engineering. The old concept of cause and effect has been abandoned. A phenomenon is seen not as an effect but as an event taking place in a field, and every force in it, whether active or not, has some relationship to it. In order to produce a change, it is desirable to ascertain which of the forces can be altered or eliminated to bring about the desired effect. The goal is not to find the “cause” but rather to discover a means of intervention.7 The physical scientist considers that we do not know the meaning of a concept unless we can detail the specific operations used in applying the concept in a concrete situation. Any abstraction which cannot be duplicated in terms of what the scientist does is considered unscientific. Thus such abstractions as “the underlying psychodynamics” or “hypnosis is nothing but suggestion” would be ruled out as meaningless constructs.
This approach is the essence of a relatively new concept called “operational analysis.” It has much in common with the probabilistic game theory. Although developed independently, both are applicable to all situations in which a large number of variables have to be considered for increasing efficiency. This is particularly significant for medicine, especially psychotherapy, which is essentially a two-person interactional “game.” Since game theory has apparently developed an approach to give the results of outcomes with certain theoretic assumptions for an unlimited number of multivariant processes, it may be that the “probability theory” can help us to evaluate how imaginative processes build up a notion of probability, calculate the odds and learn which decisions are most favorable based on these odds.
If higher nervous activities are to perform the task of making continuous predictions under the affective influence of a comparison of an imagined future with an experienced past, the brain must have available “counters” or images as the data (“bits” in the computer) for its computations.17 These represent the elements of behavior patterns.8 The reader interested in a new era of brain research should read a stimulating article on the subject of brain-computer analogies.15 Black and Walter have provided the first objective evidence, in the form of EEG patterns, on how the brain responds to hypnotic suggestions.3 They postulated that the anterior cortex—the silent area of the brain—acts as a “contingency computer” to extract information from the environment by the assessment of probability.
Modern neurophysiologists ultimately must deal with brain function in terms of more sophisticated analogue models. However, the staggering complexities in a system comprised of 10 to 15 billion cells, each of which might be regarded as a hybrid microcomputer, make it difficult to prove that the brain is a computer, but nevertheless the concept at this point in time is a useful one. For instance, the brain seems to function like a highly sophisticated model of a digital-analogue computer. By way of explanation, the digital machine performs numerical computations with incredible speed when the problems can be reduced to conventional arithmetic operations. The digital computer loosely approximates those of brain processes concerned with awareness and those which involve autonomic or reflex activities. The analogue machine attempts to solve a problem by recreating within the machine the physical circumstances which give rise to the problem and thus determine its outcome. It is useful for handling relatively complex situations. The cerebral mechanisms for those psychological functions which we now call, for want of a better name, unconscious and certain preconscious functions, approximate more closely the digital type of computer in their functioning.
There are other similarities of the brain to the digital machine. Control of an organism by thought processes is largely mediated by discrete or different distinct levels of neural functioning (digital). The execution of the digital “commands” are carried out in analogue fashion. For instance, humoral and endocrine functioning resembles the continuous levels of activity characteristic of analogue computers. It has been postulated that Pavlov's distinction between a primary signal system concerned with directly perceived stimuli and a secondary signal system devoted to verbal elaborations seems to parallel the above distinction between digital and analogue computers.29
Because cybernetics has disregarded nonoperational and useless constructs, it is making rapid strides in explaining, not only how information is processed in the machine, but also how perception, learning and concept formation are processed in the nervous system. Therefore, brain-computer operational analogies will be discussed more fully in an attempt to explain the phenomenology of hypnosis—its evolution and function as an adaptive response mechanism not as a singular thing, but rather as a process basic and fundamental to the organism, which, like behavior, is multifaceted as well as fluctuating. The author fully realizes the speculative nature of these assumptions, but believes that they constitute a rational hypothesis that will help place the understanding of hypnosis on a more scientific basis.
Analogy Between Computer and Brain Function
As of now, machines are not capable of thinking. However, from a purely mechanistic standpoint, devices capable of a wide range of selective behavior based on evaluation of a large number of variables are being developed, but they are incapable, even remotely, of equaling the tremendous capacity of human recall, learning, and perception.
Those who embrace the present highly formalized schools of psychology may object to a mechanistic approach since humanistic elements are ignored. But to conclude that research on the simulation of human behavior with a machine is wrong is somewhat analogous to saying that research on the simulation of the human heart with an artificial one is wrong because the latter organ is not a living one.
Let us examine some of the properties of living organisms which machines are capable of simulating. One of the key features of the behavior of living organisms is adaptability. This property can be simulated on the machine. Such a device automatically changes its internal structure in accordance with the environmental stimuli (input information signals) to function in a purposeful manner. This same adaptive property enables the mechanism to change from a positive to a negative feedback system by sensing and correcting its own performance. In the human, this trial and error process of learning causes physiochemical changes in the structure of the feedback networks to enable the organism to respond normally to the class of stimuli to which it has become adapted.
Theoretic Evaluation of Hypnotic Responses and Controlled Adaptive Behavior Based on Computer Analogies
In the past, vague and nontestable formulations have been advanced to explain the nature of hypnosis and hypnotic responses. The reasons are obvious—these are built-in mechanisms—the result of responses developed during our genetic endowment and continually refined to give the organism greater adaptability.
My hypothesis is that hypnotic response was at one time a primitive adaptive mechanism which was necessary for survival. Its evolutionary development can be descriptively equated with that of the origin and behavior of modern computers.
Modern electronic “thinking” machines were originally developed as special purpose computers (S.P.C.) for solving relatively simple problems. As the physical sciences developed, it became necessary to perfect a machine that would solve a large variety of complex problems. Since the S.P.C. was inadequate, it inevitably evolved into the present large and complex general purpose computer (G.P.C.). However, in achieving this flexibility, the G.P.C.'s capability far exceeded the demands of limited problems. Nevertheless, when the G.P.C. is committed to limited problems it can solve them with amazing speed—but at a high cost for this increased celerity.
Although one cannot as yet demonstrate that analogies exist between computers and cerebral neurophysiologic systems, the evolutionary development of this model of automatic control closely parallels the evolutionary development of brain function before it was capable of analytic thinking, a comparatively recently acquired function. Early man had a primitive mechanism—the “nose-brain”—for sensing the world around him. Its function was specialized to receive nonverbal signals or impressions only through olfactory sensations. This was the only sense which provided information for coping with his environmental problems. In this respect, the simple behavior of the primitive “nose-brain” mechanism might be compared with that of the S.P.C.
As man's brain continued to evolve, other sensory stimuli, in the form of subverbal or preverbal suggestions, helped to shape his mental processes before he had the ability to think analytically and to adapt with a greater degree of affective feeling to environmental changes. A stage of development comparable with this archaic level of functioning is the behavior of anencephalic monsters and decorticated humans and animals, who apparently, in a primitive way, see, hear, taste, smell, utter crude sounds, cry and smile, and react with pleasure or displeasure to pleasant and unpleasant stimuli.5
In this evolutionary process, as the cortex expanded from the ancient smell centers, the simple adaptive responses were integrated into the lower or the subcortical centers to provide an automatic system for maintaining vital functioning of the organism—homeostasis. One of the adaptive physiologic response mechanisms manifesting this “mechanical calming” of the organism was hypnosis, which has been known under various appellations from “nirvana” to “suggested sleep.” The fact that spontaneous quasi-hypnotic behavior is noted, to a degree, in animals and humans strongly indicates that it is still largely dependent on autonomic functioning, and as such, therefore, is an inherited behavioral response mechanism in the human. It is also known that neural control of behavior, when it becomes more complex in the process of evolution, retains simpler mechanisms as higher centers are added.
In primitive man, before the development of analytic processes, simple ideas must have been accepted by primitive mechanisms. Suggestions must have been the process which fulfilled this function.22 It is also at this psychophysiologically regressed level of mental functioning that suggestions are uncritically accepted and acted upon with precision by the human. Here, hypnotic response is strikingly similar to the limited-goal behavior of the S.P.C., that is, when arousal or perceptivity is high, and when the cognitive processes are directed toward a special purpose, a hypnotic subject behaves like an efficient S.P.C. This regression is in rather sharp contrast with the logical and highly analytic but generalized mental functioning characteristic of nonhypnotic states.
Therefore, it is plausible to conclude that when an organism can have its sensing apparatus respond selectively to specific inputs, with its fullest cognitive capacities, as during hypnosis, such functioning is a reversion to a more primitive but more adaptive level. The evidence cited below points to hypnosis as being an atavistic state or psychophysiologic regression serving as a substratum for the latter development of more complex life experiences. The author was among the first of modern writers to postulate the atavism or regression hypothesis as an explanation for hypnotic behavior.13 He stated:
The hypnotic state at one time may have been necessary in humans as a protective defense mechanism … the hypnotic state may be an atavistic reversion analogous to the inanimate state of catalepsy so commonly observed in frightened animals when they “freeze to the landscape” in order to escape detection, the difference being that the presence of fully developed cortex in the human now makes unnecessary various instinctive defense mechanisms.
Later, several theoretic concepts based on a phylogenetic core were proposed. The hypnotic state was visualized as a condition which represented the most primitive form of psychophysiologic awareness of individual environment differentiation attainable among living organisms; this capacity was to some degree retained in all biologic systems.27 Guze states that hypnosis may be defined as “a state of readiness for emotional action increasingly subordinated to cortical influence as one ascends phylogeny but nonetheless consistently present in animal organisms in a variety of forms.”9
The concept that suggestibility is an archaic mental function thus can be used to explain the nature of hypnosis. According to Meares, the regression is not at the behavioral level, but rather at the perceptual or mental functioning level.22 It is not implied that primitive man lived in a constant state of hypnosis; rather, that in the phylogenetic development of the nervous system, higher functions retained the ability to control the more primitive functions to a greater or a lesser degree. Hypnosis was one of these autonomic primitive functions to maintain homeostasis or a “steady state” in the organism.
The ability of man to survive is due largely to these autonomic functions built into the lower brain centers for selectively handling incoming information. This frees the cortex for the more specialized complex problems of adaptation. Similarly, when hypnosis is used to increase adaptive cortical responses, a comparison can be made with the G.P.C. operating with its total capacity directed toward a specific problem. This, too, represents an operational alteration or purposeful reversal in computer operation, that is, a highly developed device (G.P.C.) being used instead of an S.P.C. to solve an elementary or primitive problem.
Relationship of Neurophysiology to Psychic Processes and Hypnosis
Neurophysiologic data16 which tend to confirm our hypothesis are as follows: The reticular activating system (R.A.S.), phylogenetically speaking, is an ancient brain structure. Before the full development of cortical structures, the R.A.S. played an even more important role in regulating behavior, probably that of maintaining greater arousal. However, in the modern brain, the ascending reticular activating system (A.R.A.S.) can now selectively filter incoming sensory stimuli not only for maintaining selective arousal but for integrating incoming sensory information with awareness. This is significant in regard to autonomic responses, movements and sensations.
With reference to adaptive ability, higher nervous activity is apparently Pavlovian in type. As proof, Anokhin showed that the A.R.A.S. specifically and selectively involved only some of the synaptic endings in the brain stem.1 He demonstrated this by involving biologically opposite activities, as eating and defense, which could occur only through different functional systems. The importance of this observation is that all biologic activities consist of continuous formation of newly established conditional reflexes on the basis of unconditioned stimuli of different quality.
This implies that the A.R.A.S. and the limbic lobes, to some degree, in the brain's hierarchy of other structures, govern discriminatory functioning during hypnosis. This is obtained by maintaining arousal of the cortex (selective attention or excitation), while simultaneously excluding irrelevant stimuli from awareness (selective inattention or active, concentrated inhibition) (see Chap. 28).
The arousal results either when narrowing of the attention span occurs in response to monotonic stimuli, or when there is an input-overload. In the latter instance, the high degree of arousal induced by strong emotions or vigorous stimulation tends to prevent extraneous sensory stimuli from reaching cortical awareness. Here the law of dominant effect is followed; a strong stimulus displaces a weaker one.
It seems also that whenever the integrity of the organism is threatened by imminent danger, the A.R.A.S. allows such vital and important information to be forwarded to the cortex for discrimination and instantaneous arousal. It has been noted, for instance, that a sleeping person generally awakens in response to a strange sound such as a footstep, but is able to sleep through much louder noises such as routine traffic.
Likewise, in hypnosis, arousal is maintained by limiting the patient's attention-span to specific input information from the operator. The limiting process may be due to a summation effect reaching threshold levels or saturation of the A.R.A.S. Here there is full utilization of its pathways. West contends that feedback mechanisms limit “nearly all additional information regardless of its significance under ordinary circumstances of adaptation.”34
However, it appears that a feedback process is not necessary to explain the functioning of the A.R.A.S. under such conditions. It may be that the saturation is analogous to what happens under similar circumstances in electronic systems which filter information at their input to exclude less important or unnecessary information. In short, the A.R.A.S. reduces the saturation threshold to zero for all sensory inputs except those selectively permitted to get through to higher centers. As a result, selective attention exists, as mentioned above, to the exclusion of reality (i.e., internal inhibition).
In Chapter 3, it also was pointed out that Pavlov was the first to note this neural mechanism—internal inhibition—as it related to the neurophysiology of hypnosis. He observed that hypnosis had an inhibitory character; that is, the cortical neurons became, as it were, weaker and less efficient, the maximum limit of their possible excitability diminished. This hypothesis, too, fits in with the saturation threshold hypothesis.
The inhibitory character of internal inhibition during hypnosis also has a protective feature similar to the nonspecific therapeutic effect of sleep and tranquilizers in emotionally disturbed individuals. This has been borne out by Russian experiments involving toposcopic examination.31 In this procedure, oscilloscopic representations of the brain's electrical potentials make a bioelectric mosaic or pattern of different cortical areas. In well-adjusted persons, the resultant bioelectric mosaic shows continual and rapid changes in potential distributed at random over the cortex. In severely disturbed individuals, such as psychotics, the changes in the mosaic are greatly reduced. Tranquilizers, sleep (generalized inhibition), and hypnosis (partial inhibition) increased the activity of the bioelectric mosaic (converted it to a normal pattern).
Particularly interesting in this respect is the use of hypnotic suggestion to inhibit specific or nonspecific stressors, such as harmful words, thoughts, and memories. Hypnotic suggestion directed to elimination of conditioned and unconditioned stimuli results in their inhibition.12 The stimulation excluded by the suggestion acquires the characteristics of conditioned inhibition (neutralization of a harmful conditioned or unconditioned stimulus).
The neurophysiologic data supports Pavlov's thesis that emotional disorders are brought about by increased excitation of neurons, and that hypnosis (protective sleep inhibition) or even actual sleep prevents exhaustion or destruction of neurons, with consequent improvement.
Leading neurophysiologists are now urging reconsideration of internal inhibition as the neural mechanism which can be utilized in psychotherapy. Magoun points out that “If the inferences drawn from these many contributions [Pavlovian concepts] are correct, this is a brain mechanism whose function psychiatry must ultimately incorporate into its conceptions of inhibitions in mental activity and, I urge it, in understanding of the wellness and illness of the mind.”21
COMMENTS ON ADAPTIVE CONTROL SYSTEMS AS THEY MAY RELATE TO PSYCHOTHERAPY
In order to gain some insight into the complex processes which take place in psychotherapy, it is instructive to compare the subject's response in psychotherapy with the response characteristics of adaptive servomechanisms. Since some modern electronic systems have the ability to adapt to their environment, these comparisons are becoming more meaningful.
In comparing electronic and human systems, the environment consists of the signals (stimuli) as well as the electrical noise and interference (specific or nonspecific stress), which appear as inputs (afferent stimuli).
The internal structures of these systems are allowed to vary so that the systems can learn from previous experience how to process the input information (“think”) in an optimal way (successful adaptation). For example, if the positive feedback or the noise input to the system in a given frequency range is excessive, the system will reject this noise by means of a rejection filter centered at the noise frequency (scanning mechanisms). The system does this at the risk of rejecting useful information which may be centered at the same frequency. However, the system design essentially is based on the decision that it is preferable to run the risk of losing useful signals in a given frequency range rather than to allow the system to be swamped by noise which would prevent it from accepting useful information at other frequencies. This is rather similar to the physiologic functioning of the A.R.A.S.
Thus these adaptive systems react to interfering signals (stress) in a manner much like that of physiologic systems. For example, if an interfering signal causes instability, the system detects its own unstable behavior and causes the adjustable components in the system to change values so that the instability is decreased. In other words, after the system has been exposed to the signal environment for a period of time, it has learned where to place its rejection filters and how to adjust its internal structure to prevent unstable modes of behavior.
Automatic control systems used in engineering, like many analogous physiologic control systems, are stable in behavior when the input signal or stimuli are of one type and unstable when these signals or stimuli are of another type. In the adaptive control system, the system is required to adapt so that its mode of behavior will be stable when subjected to either type of input signal. The significance for psychophysiology, in studying the engineering uses of adaptive systems, is that it now appears possible to attempt a quantitative study of adaptive psychophysiologic behavior with simulated physiologic control systems.
The communications which take place between the psychotherapist and the subject, irrespective of whether the approach is permissive or directive, may be looked upon as a rather complex form of directed adaptive behavior. The psychotherapist may be thought of as providing the input signal environment, while the subject may be considered as the adaptive system which adjusts its behavior parameters (psychophysiologic variables) to conform with the environmental stimuli (inputs). As in engineering systems, the ability of the subject to adapt is a strong function of his present state. Also, if the range of adjustment required to cause stable behavior is very large, the parameters of the system may be unable to change their values enough to achieve stability. In engineering systems, such a situation would necessitate revision of the adaptive system to allow its adjustment parameters to be varied over wider ranges. The speculative implications to be drawn from this in the case of psychotherapy suggest that those subjects who do not respond require another approach, perhaps a revision or a restructuring of the therapeutic design.
NEUROPHYSIOLOGIC THEORIES OF MEMORY
Older theories of memory and learning maintained that experiences left “etchings” on the brain as traces. Still others contend that memory traces depended upon decreased synaptic resistances, with the resultant establishment of well-grooved pathways. These have been invalidated by pavlovian “learning,” which is not confined to the cortex. Nor is memory limited to the midbrain or the brain stem, although some storage of information takes place in these areas. Localized memory traces apparently have been demonstrated also as a function of the temporal lobes.24 However, the data are inconclusive as yet. Other theories are that “experiences establish perceptual patterns of potential gradients in cortical electrical fields” or “resonance patterns occur in neural loops to produce altered physiochemical changes.”14
The recently developed complex general computer stores “bits” of information as electrical pulses, which continually revolve until needed for computation. Since these pulses are not specifically located, they are referred to as “functional” or random memory; changes are not stored in a definite manner. Recent neurophysiologic data also indicate that human memory is random in nature, because no special part of the brain stores it—a wise provision in case of accident.30
As further evidence that memory is random, it has been hypothesized that the two-way feedback or reverberating neuronal chains are capable of manipulating thought according to symbolic or mathematical logic.18 Lashley's alternative theories are that memory is due to “potential gradients in electrical fields” or that “resonance patterns in neural loops” account for it.14 Pringle postulates a model of closed chains of neurons which act as “loosely coupled oscillators,” similar to those occurring in the brain.25 Irrespective of the validity of the above theories involving reverberating neural loops, it is certain that physiochemical alteration takes place in the circulating neuron chains to preserve “memories.” This is more in accord with the most recent theory that every incoming percept leaves its trace by alterations in the arrangement of the large protein molecules of the neuron.
Hyden has demonstrated that some stimuli alter the ribonucleic acid (RNA) molecules of the neurons which cause the synthesis of altered protein molecules that are stored as “bits” of information or memory traces.10 It is believed that the frequency modulation set up in a neuron by a specific stimulus may prescribe the arrangement of the RNA components (and thus of proteins) which acts as a code to pass on information to other neurons. In this way, whole chains of neurons can be molecularly conditioned to react to the repetition of a stimulus. From a statistical viewpoint, the molecules furnish the required permutation possibilities for the storage of all the bits of information received in a lifetime. Thus the recall of past experiences as memories is made possible.
Other scientists favor a molecular “switch” theory, which suggests that specific synaptic proteins subserve selective interaction between pre and postsynaptic elements, thereby serving as an engram. The structure of such proteins would be genetically predetermined.
The best clinical data on the recovery of memory traces is Penfield's24 work, cited earlier. It is interesting that a single recollection was recalled, not a mixture of memories or a generalization, as in ordinary memory. The evoked reaction was an exact reproduction of what the patient saw, heard, felt, and understood.
It seems that the memory records of all experiences are recorded by patterns of previous passage of nerve impulses. The patterns of neuronal memories are duplicated in both hemispheres since the removal of most of one lobe does not interfere with recollection. It is believed that these records are located in the centrencephalic circuits in the higher brain stem. Every experience seems to have access to both temporal lobes and evidently remains unchanged with the passage of time.
As discussed in Chapters 29 and 30, this may be the explanation for age-regression, revivification, sensory-imagery conditioning and hypnotic self-exploration. Scanning mechanisms unite stored memories with selected ideas, former experiences, and relate them to the incoming sensory percepts. Since peculiar disorders of memory also occur with lesions around the third ventricle, there are two other areas involved in various types of memory in addition to the temporal lobes, namely, the upper brain stem and the periaqueductal region.
These memory mechanisms also operate during dream states, reverie, hypnosis, and other dissociated states, the degree depending on selective “filtering,” mediated chiefly by the reticular activating system. Since hypnosis is a state of hyperacuity, one would infer that greater arousal is being maintained to a selected input than during ordinary attention. The only difference is that, in hypnosis, the discriminatory ability of the cortex is held in selective inhibition. Normally, the cortex does not accept incoming information without prior computation. If stored data are not available for computation, unreality is accepted as reality—hypnosis.
Raginsky described how syncope and temporary cardiac arrest were induced under hypnosis in a patient who had Stokes-Adams syndrome and who had, until the time of the experiment, remained free of such symptoms.26 Since the memory was sequential (always moved forward) in this patient as well as in several others, it was felt that hypnotic recall closely parallels electrical stimulation. Hypnotic recall is purer or more accurate than the recalls elicited by psychotherapy, in which patients bring up generalized memories or those that have been modified by subsequent thinking or experiences (screen memories).
Blum and co-workers offer the hypothesis that when the chances for interference are effectively minimized, well-entrenched memories do remain virtually intact over very long time spans, even in the absence of rehearsal.4 They showed that under hypnosis the spontaneous emergence of distant memories differed markedly from the extraordinary feats performed by memory “experts.” The latter rely upon all kinds of cues, including several forms of synesthesia, in combinations with intense eidetic imagery, to achieve success.
IMPORTANCE OF PSYCHOCYBERNETICS TO THERAPY
It has been pointed out that what the learner does in successive trials is regulated by the results of his performance—feedback. Incorrect responses usually are replaced by successful responses and, when these are remembered and automatically reinforced, they become adaptive or maladaptive habit patterns. However, incorrect responses or failures usually are forgotten and replaced by successful ones. There are numerous examples of how feedback modifies faulty behavioral responses. For instance, constructive self-criticism can correct maladaptive behavior to bring about a desirable state or goal. However, too much criticism is disastrous and is synonymous with psychological inhibition.
Kline has demonstrated that a continual delayedspeech (positive) feedback at nonhypnotic levels, with the inability to defend oneself against this feedback, produces acute emotional disorganization and signs of stress and psychopathologic behavior.11 These reactions to positive or excessive physical feedback can be inhibited by hypnosis.
Extreme carefulness or fear of error is a form of positive feedback. This dynamism is noted in the stutterer who, in the presence of increased fear, develops bad motor response patterns because of inhibition. By having the stutterer listen to his own voice, negative auditory feedback automatically monitors correction of the speech. In the section on stuttering in Chapter 43, it is suggested that the patient listen under hypnosis to the playback of a tape recording of his own voice, speaking normally. Faulty enunciation, tone and such other impediments as blocking are quickly discerned and therefore readily corrected. Optimal functioning is established more effectively under hypnosis because the learned responses make use of the built-in reflexes, and, as a result, are eventually utilized automatically in a more spontaneous manner. The only danger is that overcorrection makes the stutterer too self-conscious, and, as a result, he worsens (inhibition).
Autohypnosis and sensory-imagery conditioning can alter behavioral responses either positively or negatively. If positively, purposeful behavior is brought about by healthy autosuggestions (input), which result in proper physiologic responses (output), and then a part of these regulate further “input” to control the behavior of the system to achieve equilibrium.
In a previous chapter it was noted that the brain can only process sensory percepts and correlate them with stored impressions. If the resultant computations are perceived as harmful, then a negative self-image is produced. Everyone has special images of himself and the environment, and behaves as though the images were real rather than imagined. At the risk of oversimplification, if an individual can imagine himself sick, he also can imagine himself well. Under hypnotic sensory-imagery conditioning, many “dry runs” can be processed to implant healthy convictions based on the stored data. A new image of the self is achieved by replacing negatively stored images by positive ones. As a result, new reaction patterns are formed which become available for involuntary functioning to maintain healthy adjustments. This is, in part, the very basis for behavior modification therapy. The well-adapted individual no longer has to check his mental (“How am I doing?”) feedbacks, but, rather, can be more concerned with goal-directed activities. In general, it is the purpose of positive hypnotherapeutic suggestions to make available the healthy stored data in order to inhibit harmful impressions (disinhibition). This allows the subject to perceive and cope with reality in a more effectual manner.
With the advent of recent cybernetic concepts, the capabilities ascribed solely to humans are being chipped away slowly. Computers designed to model human mental processes can now simulate problem-solving, learning and decision making. Computers are making possible a much more intensive search for factual observations of multivariant functional relationships. It is hoped that these developments not only will give us a theoretic explanation of corresponding human behavior, but will help also to explain memory mechanisms, the bridge between nerve impulses and thought, and the continuum of awareness ranging from hypnosis to sleep.
REFERENCES
1. Anokhin, P.K.: Paper delivered at the First Pavlovian Conference on Higher Nervous Activity, Med. News, November 9, 1960.
2. Ashby, W.R.: Design for a Brain. New York, Wiley, 1952.
3. Black, S., and Walter, W.G.: Effects on anterior brain responses of an expected association between stimuli. J. Psychosomat. Res., 9:33, 1965.
4. Blum, G.S.: Distinctive mental contexts in long-term memory. Int. J. Clin. Exp. Hypn., 19:117, 1971.
5. Cairns, H.: Disturbances of consciousness with lesions of the brain stem and diencephalon. Brain 75:109, 1952.
6. Crider, D.B.: Cybernetics: a review of what it means and some of its applications to psychiatry. Neuropsychiatry, 4:35, 1956-57.
7. Dunbar, H.F.: Anxiety, stress and tuberculosis. In Sparer, P. J. (ed.): Personality, Stress and Tuberculosis. New York, International Universities Press, 1956, p. 211.
8. Glaser, G.H.: Panel Discussion: Recent concepts of central neurophysiology; their bearing on psychosomatic phenomena. Psychosom. Med., September-October, 1955.
9. Guze, H.: Hypnosis as emotional response. J. Psychol. 35:313, 1953.
10. Hyden, H.: Paper read at symposium, Control of the Mind, University of California, Feb. 27, 1961.
11. Kline, M.V.: An experimental study of the nature of hypnotic deafness: effects of delayed speech feedback. J. Clin. Exp. Hypn., 2:145, 1954.
12. Korofkin, I.I., and Suslova, M.M.: On the neural mechanisms of hypnosis. In Winn, R.B. (ed.): Psychotherapy in the Soviet Union. New York, Philosophical Library, 1961.
13. Kroger, W.S., and Freed, S.C.: Psychosomatic Gynecology. Philadelphia, W. B. Saunders, 1951; reprinted, Los Angeles, Wilshire Book Company, 1962.
14. Lashley, K.S.: In search of the engram. Society for Experimental Biology Symposia 4: Physiological Mechanisms in Animal Behavior. Cambridge, Cambridge University Press, 1950.
15. Lindgren, N.: To understand brain. I.E.E.E. Spectrum, 52-58, 1969.
16. Livingston, R.B.: Some brain stem mechanisms relating to psychosomatic functions. Psychosomatic Med., 17:351, 1955.
17. Luria, A.R.: The Mind of Mnemonist. A Little Book About a Vast Memory. New York, Basic Books, 1968.
18. McCulloch, W.S., and Pitts, W.: A logical calculus of the ideas imminent in nervous activity. Bull. Math. Biophysics, vol. 5, 1953.
19. McGill, W.J.: Applications of information theory in experimental psychology. Bull. N. Y. Acad. Sci., 19:343, 1957.
20. McKay, D.M.: In Search of Basic Symbols, Cybernetics, New York, J. Macy Foundation, 1951.
21. Magoun, H.W.: Discussion of Anokhin, P.K.: Paper delivered at the First Pavlovian Conference on Higher Nervous Activity. Med. News, November 9, 1960.
22. Meares, A.: A System of Medical Hypnosis. Philadelphia, W. B. Saunders, 1961.
23. Ostrow, M.: Psychic contents and processes of the brain, Psychosom. Med, 17:396, 1955.
24. Penfield, W.: The role of the temporal cortex in certain psychical phenomena. J. Ment. Sci., 101:451, 1955.
25. Pringle, J.W.S.: On the parallel between learning and evolution. Behavior, vol. 3, 1951.
26. Raginsky, B.B.: Temporary cardiac arrest induced under hypnosis. Int.J. Clin. Exp. Hypn., 7:53, 1959.
27. Schneck, J.M.: A theory of hypnosis. J. Clin. Exp. Hypn., 1:16, 1953.
28. Shannon, C.E., and Weaver, W.: The Mathematical Theory of Communication. Urbana, University of Illinois Press, 1949.
29. Simon, B. (ed.): Psychology in the Soviet Union. Stanford, Cal., Stanford University Press, 1957.
30. Sluckin, W.: Minds and Machines. Harmondsworth, Middlesex, Pelican Books, 1954.
31. Snechnevsky, A.V.: Paper delivered at the First Conference on Higher Nervous Activity. Med. News, November 9, 1960.
32. Uhr, L., and Vossler, C.: Suggestions for self-adapting computer models of brain functions. Behav. Sci., 6:91, 1961.
33. Von Neumann, J., and Morgenstern, O.: Theory of Games and Economic Behavior. Princeton, N.J., Princeton University Press, 1944.
34. West, L.J.: Psychophysiology of hypnosis. J.A.M.A., 172:673, 1960.
35. Wiener, N.: Cybernetics, New York, Wiley, 1948.
ADDITIONAL READINGS
Blum, G.S.: A Model of the Mind. New York, Wiley, 1961.
Coburn, H.E.: The brain analogy. Psychol. Rev., vol. 58, 1951.
Craik, K.J.W.: The Nature of Explanation. Cambridge, Cambridge University Press, 1943.
Hebb, D.O.: The Organization of Behavior: A Neuropsychological Theory. New York, Wiley, 1949.
Hilgard, E.R.: Theories of Learning. New York, Appleton-Century-Crofts, 1948.
McCulloch, W.S.: The brain as a computing machine. Electrical Engineering, vol. 68, 1949.
MacKay, D.M.: Mentality in Machines, Proceedings of the Aristotelian Society, 1952.
Rashevsky, N.: The neural mechanism of logical thinking. Bull. Math. Biophysics, 8, 1946.
Reiff, R., and Scheerer, M.: Memory and Hypnotic Age Regression: Developmental Aspects of Cognitive Function Explored Through Hypnosis. New York, International Universities Press, 1959.
Rosenblatt, F.: Principles of Neurodynamics. Perceptions of Neurodynamics. Perceptions and the Theory of Brain Mechanisms. Washington, D. C., Spartan Books, 1962.
Sheer, D.E. (ed.): Electrical Stimulation of the Brain: An Interdisciplinary Survey of Neurobehavioral Integrative Systems. Austin, University of Texas Press, 1961.
Thomson, R., and Sluckin, W.: Cybernetics and mental functioning. Br. J. Philos. Sci., vol. 3, 1953.
Walter, W.G.: Possible features of brain function and their imitation, Symposium on Information Theory, London, Ministry of Supply, 1950 (reprinted 1953).
Wisdom, J.O.: The hypothesis of cybernetics. Br. J. Philos. Sci., vol. 2, 1951.