Orne, M. T. Hypnosis. Naval Research Reviews, 1970, 23, 1-10.
Institute of the Pennsylvania Hospital and University of Pennsylvania, Philadelphia, Pennsylvania
The Navy and Marine Corps are constantly looking for ways to increase the ability of their personnel to withstand the wide variety stresses to which they are subjected. Fatigue, extremes of heat and cold, motion sickness, isolation, hostile interrogation, and pain are examples. Hypnosis has often been suggested as a useful means for increasing resistance to such stressors. Dr. Martin Orne, author of this article, is recognized internationally as an authority in the experimental study of hypnosis. The article reports his progress under Office Naval Research sponsorship towards achieving an understanding the basic nature of hypnotic phenomena, a necessary first step towards the rigorous study of the utility of hypnosis in enhancing performance in Navy and Marine Corps settings.
The mystic, the magical, the supernatural have in the past been closely associated with the hypnotic phenomenon. Even to this day there is a tendency for individuals to be for or against it, to believe that it is a panacea or to reject it as something akin to fraud. While the phenomenon was already known to the Egyptians and was discussed in the Elba Papyrus, it was in large part rediscovered and attracted wide attention with Mesmer, after which it was described almost completely by one of his students, the Marquis de Puysegur.
The problems which have made an objective understanding of the phenomenon difficult were already clearly evident when Benjamin Franklin, as chairman of a commission established by Louis XVI, wrote a report investigating Mesmer. While acknowledging that dramatic events could and would occur with mesmerized patients, the commission brough back the finding that no magnetic fluid was required to account for these changes -- indeed, that they were merely the product of imagination. This view is remarkably close to that based on modern investigations. The major disagreement would be about the word "merely"; in other words, the modern investigator has developed a healthy respect for and some understanding of the power of human imagination and its variants.
In view of the many implications that hypnosis and related phenomena may have for the mental capabilities of an individual, an understanding of hypnosis is relevant to any systematic effort to comprehend what we call personality, motivation, memory, and thinking. Furthermore, hypnosis has been proposed as a technique to modify human capabilities,
2 MARTIN T. ORNE
and some of these changes can have special relevance for the Navy. Not only has hypnosis been suggested as a means of increasing an individual's capability of adequately functioning in the face of acute or chronic stress, of increasing his memory, visual and auditory acuity, vigilance, learning ability, and motivation, but also its use has been suggested for certain special purposes. The Manchurian Candidate fictionally dramatized the possibility that posthypnotic suggestion might be employed to control an individual's behavior without his own awareness and to compel him to carry out desired actions, potentially making him an unwitting tool of a foreign power. Similarly it has been proposed that posthypnotic amnesia might be induced for sensitive information in individuals subject to capture, such as downed Naval aviators, a technique that, superficially at least, seems interesting. Though the information would not be available to the individual consciously, it would be readily available when a pre-arranged signal was given to lift amnesia. In a similar way, hypnotic techniques have been advocated as ways of controlling and modifying an opponent's thoughts and actions and have been implicated in brain washing.
It is not necessary to belabor the offtimes ingenious and thought-provoking suggested applications of hypnosis. It will be immediately clear to the reader why a phenomenon which makes it possible to alter an individual's memory for events of the past, to induce new attitudes about the past, present, and future, and to compel him to carry out a wide variety of behaviors at some future time in response to a predetermined cue, should act as a fertile stimulus to the imagination. The basic research carried out in our laboratory, first at Harvard and more recently at the University of Pennsylvania, has been directed towards clarifying a variety of basic questions about how, when, and why these events can occur. It was first necessary to develop techniques for doing systematic research on this problem before even beginning to answer questions about the feasibility of the many concrete proposals that had been put forth for the utilization of hypnosis.
What Characterizes Hypnosis?
While Mesmer is appropriately credited with reviving the interest in hypnosis in modern times, his patients were mesmerized without the kind of verbal suggestions one usually associates with hypnosis today. Instead they invariably entered the state by what he termed a "crisis," which can best be described in modern terms as a "hysteric seizure." The patients then went to sleep and awoke minutes to hours later, usually with relief of their symptoms. Certainly this sequence of events is vastly different from hypnosis as it is known today; similarly hypnosis as we know it differs markedly from hypnosis as described by
Charcot -- who claimed that when the top of a subject's head was rubbed he would pass from sleep to somnambulism -- and from the phenomenon evoked by Emile Coue under the heading of self-hypnosis.
The behavior characteristic of the hypnotized individual seemed to vary so widely from place to place and from time to time that some explanation was necessary. To account for such differences we hypothesized that the hypnotized subject behaves as he does either because it is suggested to him that he do so or, in the absence of any suggestion, because he believes certain behaviors to be typical of the hypnotized individual.
In order to test the hypothesis experimentally, we had to invent as a variable as plausible "characteristic" of hypnosis that had, however, never been observed previously -- we chose catalepsy of the dominant hand. Catalepsy -- waxen rigidity of the limbs -- is a common occurrence during hypnosis, but when it occurs it happens in all extremities, the trunk, the head, and so forth. It had never been known to occur in one part of the body while the rest remained flaccid. Catalepsy of the dominant hand was chosen because it sounded "scientific" and would thereby be plausible to students, reminding the college undergraduates of lectures about stuttering due to suppressed left-handedness, etc. A lecture about hypnosis was then arranged for a large class of college students. This included a demonstration with three subjects chosen from the class who, unbeknownst to the group, had previously been hypnotized and given the suggestion that they would subsequently develop catalepsy of the dominant hand. The classic hypnotic phenomena were then demonstrated, including amnesia, age regression, and posthypnotic behavior. During the induction process catalepsy was carefully tested, and the hypnotist casually commented upon how typical the behavior of the subjects was concerning their catalepsy of the dominant hand (two subjects were right-handed and one was left-handed). In all other respects the lecture was accurate. A matched class of students received the identical lecture and identical demonstration but catalepsy was not tested.
A month later subjects from both classes were asked to volunteer for some experiments. They were brought to the laboratory and hypnotized. We observed the spontaneous development of catalepsy of the dominant hand in most of the subjects who had been at the appropriate lecture. No such phenomenon was observed, of course, among the control subjects, who had not seen the demonstration of dominant hand catalepsy.
The rather simple experiment characterizes the problems encountered in investigating hypnosis. It would seem that we are dealing with something that has the qualities of a chameleon. The hypnotized subject will manifest whatever behavior he believes characteristic of hypnosis; furthermore, he may be taught to do this without either his own or the
4 MARTIN T. ORNE
hypnotist's awareness. As a result, two hypnotists with varying convictions can go forth (attempting to test their beliefs) and will observe diametrically opposite findings, though both are applying apparently objective procedures. It is not surprising therefore that the history of hypnosis is characterized by acrimonious disputes and that many distinguished investigators became disappointed and left this field of inquiry.
The Development of Special Research Techniques
It became crucial to develop ways of recognizing the difference between those aspects which are essential characteristics of hypnosis and those which might just as easily be epiphenomenal artifacts subtly suggested to the individual. The problem here was analogous to that encountered in the evaluation of drugs with psychic effects. It is not enough to know that aspirin relieves a headache since an inert pill, or placebo as it is called, can under appropriate circumstances do equally well. Even a placebo, however, is not adequate to evaluate a new medication, since the doctor who is aware of what he is administering would be enthusiastic, interested, concerned, and solicitous of side effects when administering a new medication but rather disinterested or even bored when giving a placebo. It is for these reasons that psychopharmacologists evolved the double-blind technique as a partial solution to some of these difficulties.
Studies utilizing hypnosis suffered from similar problems. Not only would hypnotized subjects manifest the behavior congruent with the hypnotist's expectations but there was an additional difficulty. When the hypnotist gave suggestions to the hypnotized individual, they were characteristically delivered with enthusiasm and conviction; on the other hand, during the control trials with the same subject in the waking state, suggestions were presented in a more routine, matter-of-fact fashion. Thus, even though a control procedure was employed, the results were not necessarily conclusive, for the subject could easily realize that he was expected to do better in hypnosis. Furthermore, the instructions were inadvertently being given in a manner to insure that the subject would try harder while hypnotized.
To circumvent these difficulties, it was essential to develop a technique which would guarantee that the hypnotist would treat the unhypnotized subject in the fashion as the hypnotized. He would have to provide the same kinds of information about what he expected the subject to do and do it in the manner usually employed for the hypnotized person, but without the subject's actually entering hypnosis.
The ability to enter hypnosis varies with individuals and, while most, cooperative subjects are able to enter hypnosis, five to ten percent find it impossible to do so despite repeated attempts with competent hypnotists. If one eliminates those who appear negativistic and who are
not really cooperating, one has a group who are exceedingly unlikely to enter hypnosis, a counterpart to the ten to twenty percent on the other end of the continuum who are capable of entering very deep hypnosis quickly and easily. These latter individuals are the ones who readily show all the classic phenomena of deep hypnosis, while the former do not seem to be able to do so. This appears to be an individual difference in subjects, not easily related to intelligence, independence, thought, leadership potential, and so forth. Nor does there seem to be a simple relationship to the usual kind of personality dimensions studied by psychological tests among the normal population. Yet within each individual, the ability to respond to hypnosis is relatively stable over time and largely independent of the hypnotist or the situation, so that it can be considered as a constant for experimental study. Furthermore, given repeated exposure to hypnotic induction, both highly hypnotizable and low hypnotizable subjects develop the same general knowledge about the phenomenon and about the laboratory's purposes and aims: since they are drawn from the same populations, they have similar background information available to them.
The simplest approach, then, to the experimental design problem in hypnosis research, seemed to be to ask subjects to simulate -- to make believe that they were actually hypnotized. There were many statements in the literature indicating that subjects were unable to carry out such instructions, that simulation was easily recognized by competent hypnotists. Our early efforts in this regard bore out this assertion. Subjects typically tried to simulate but performed extremely awkwardly. Their behavior was transparent and usually interspersed with sly smiles, surreptitious glances at the hypnotist and similar cues which permitted even the most unskilled hypnotist to recognize a fraudulent performance.
In discussing the experience with subjects it became clear why they found the task so difficult. They were aware that the hypnotist knew they were simulating and they felt they had been presented with a task they viewed as impossible. Regardless of their performance, there would be no way in which they could "win." To cope with this problem, the situation was altered slightly but highly significantly. Subjects unable to enter hypnosis were instructed by a colleague as follows:
"In the past you have tried to enter hypnosis and I realize that you have been unable to do so. Today we want you to take part in a particularly important experiment. You will be working with Dr. X, who has never met you and your task will be to go in and fool him, to convince him that you are a good hypnotic subject and able to enter deep hypnosis. Dr. X will know that some people will be trying to fool him by simulating hypnosis but he will not know who. If he catches on to the fact that you are simulating he will stop the experiment immediately; so, even if you don't know how you are doing, don't think you have given yourself away. You haven't because if you do Dr. X will stop the experiment. I
6 MARTIN T. ORNE
can't tell you what to do or how to behave. Do the best you can. I know it is a difficult task but we have found in the past that intelligent subjects have been able to do it successfully. Good luck!"
Subjects instructed in this fashion were randomly mixed with subjects who were in fact extremely good hypnotic subjects and both were seen by a hypnotist who was in fact blind as to the subjects' true status.
When this line of inquiry was first undertaken, we felt it might provide a useful control technique but still expected that a highly trained hypnotist would be able to distinguish readily between these two groups. To our amazement we found that this was not the case. A number of simple tests which we and our colleagues traditionally employed in evaluating hypnosis proved to be inadequate to the task of discriminating between hypnotized and simulating individuals. These included the use of moderately painful stimuli such as pinching, pinpricks, or faradic shock to the fingers. In careful studies using motion pictures and videotape, observation of the eyes, face, posture, general coordination, speech, tests of endurance, and so forth, not only were colleagues within the laboratory unsuccessful in discriminating between these groups, but investigators from elsewhere who had worked with hypnosis extensively found that the diagnostic criteria they had believed to be infallible failed to discriminate.
This did not mean that we found no differences. The experiences of these two groups were radically different, the attitudes toward the experiment once it was over were radically different, etc. It did mean, however, that many of the behaviors commonly attributed to hypnosis could as readily be explained as a function of subtle cues and expectations on the part of the hypnotist, communicated without his awareness, rather than as due to the presence of the phenomenon itself.
We had developed a technique by which we could indeed evaluate the extent to which a particular phenomenon could be ascribed to the chameleon-like quality of hypnosis and the extent to which it could be viewed as an essential characteristic. The fact that hypnosis could be simulated successfully did not, of course, imply that there is no such phenomenon, any more than the fact that a psychologist had successfully simulated schizophrenia and fooled the personnel in a leading psychiatric hospital would imply that schizophrenia is therefore nonexistent. Far from questioning the reality of hypnosis we had established a rigorous criterion by which to appraise its intrinsic properties. The simulating control group allowed us to begin the difficult task of evaluating the many claims concerning the unique properties of hypnosis and the powers it confers upon the hypnotized subject. It would now be possible to discriminate between those aspects of the behavior of the hypnotized individual that are of necessity a function of being hypnotized and which might as easily be explained in other ways.
Examples of Research Problems
A number of problems could now be examined. For example, it was possible to establish that the physical capacity of the deeply hypnotized could be matched by that of simulating subjects. These subjects not only were able to maintain their limbs in uncomfortable positions for long periods, support heavy weights, pull on a dynograph, etc., at levels of performance greater than we had anticipated, but they could also duplicate the "experiment" commonly demonstrated by stage magicians -- staying in a totally rigid position supported only by their neck and heels between two chairs while carrying the weight of another person on their abdomen. It became clear that this and similar feats of strength are within the repertoire of the normal waking individual provided he is appropriately motivated; it merely had not occurred to anyone that this might be a possibility.
Another set of studies (1) considered the question of whether deeply hypnotized subjects could be compelled to carry out antisocial, destructive actions. Rowland (2) and subsequently Young (3) had both shown that deeply hypnotized individuals could be compelled to pick up a poisonous snake with their bare hands, to remove a penny with their bare fingers while it was dissolving in a dish of concentrated nitric acid, and to take this acid and throw it at a research assistant. These actions are clearly self-destructive and antisocial. When these same subjects were asked in the waking state whether they would carry out these actions, they appeared horrified at the mere thought. The Young study was then carefully replicated with the addition of the simulating control procedure. It was indeed true, as Young had reported, that five out of six deeply hypnotized subjects could be complled to carry out the items of obviously self-destructive and antisocial behavior and would, in the waking state, deny such a possibility. However, six out of six simulating subjects did likewise.
This study did not, of course, resolve the question of whether subjects would or would not carry out antisocial behavior under hypnosis but rather clarified some of the limitations of experimental work in this field. It showed than an independent control group is essential and that one cannot rely upon a subject's report about what he might do when asked in a different setting. Furthermore, the study helped establish the unique characteristics of the experimental situation and emphasized the extent to which nonhypnotized subjects would comply with the demands of the experimenter simply because he is an experimenter, a finding that had implications for many other research problems.
The same methodological approach was used in our studies of age regression, where we found that simulating subjects were able to reproduce the kind of findings that had previously been assumed to require deep hypnosis. Again, in a different context, we found it was possible
8 MARTIN T. ORNE
for waking and simulating subjects to produce a wide range of physiological alterations at will -- a capability that in the past was felt to require the presence of hypnosis.
The Meaning of Our Findings
Our studies using the real-simulating technique to study hypnosis clarified the many previous observations where hypnosis seemed to allow the subject to transcend his own capabilities. For the most part, they demonstrated that miraculous performances are within the repertoire of normal waking individuals. Since neither the subject nor the experimenter would have believed this to be the case, the subject's waking capabilities would have been grossly underestimated, had the experimenter not been truly blind as to the status of the subject and the subject not been truly motivated to maintain his role. Some of the most important observations to come from this technique, therefore, have dealt with previously unrecognized human capacities.
To date we have been unable to demonstrate clear evidence that the hypnotized individual is able to perform behaviors outside of the waking individual's capabilities. This in no way denies the dramatic reality of the hypnotic phenomenon. When the deeply hypnotized individual reports an hallucination, it is likely that he tells of what is subjectively real to him. When he eats an hallucinated meal, he may enjoy his repast and respond physiologically by a change in gastric and urinary pH as though he had indeed eaten the hallucinated meal. The unhypnotized subject, knowingly imagining such a meal, does not tend to feel satisfied by it nor does he seem to derive the same enjoyment from it; however, his imagining will also evoke changes in urinary and gastric pH (4). The difference between the hallucinating and imagining appears to reside more in the subjective experience than in objective performance.
Subjective reports, however, have not been the only source of demonstrated differences between simulating and hypnotized individuals. The first such difference we were able to objectively show, involved suggesting an hallucination for a person actually in the room. Let's assume that Dr. Jones is actually sitting to the left of the hypnotist and observing the experiment. The hypnotist then suggests to the subject that Dr. Jones is sitting in a chair (actually empty) on his right. Care is taken to induce this hallucination effectively and the subject is encouraged to interact with the hallucinated Dr. Jones. Provided he is deeply hypnotized, he will talk to him, shake hands with him, and so forth. After a while the hypnotist points to his left where Dr. Jones is actually sitting and asks, "Who is this?" Simulating subjects characteristically report, "I don't know," "Mr. Smith," or they may deny that they see anyone. When, at the conclusion of the experiment, they are asked why they had reported in this fashion to the real Dr. Jones, they very logically point out that
since the hypnotist had told them to see Dr. Jones on his right and they had done so, he certainly could not be sitting on the hypnotist's left and they reasoned the correct response would be to deny his existence. In contrast, the deeply hypnotized individual will look at Dr. Jones, register surprise and do a "double-take," look back at his hallucination and then back to Dr. Jones. With some surprise he will recognize Dr. Jones where he is actually sitting and then say, "There are two of them!" The fact that an hallucination has been induced for Dr. Jones does not prevent the veridical perception of Dr. Jones, as no suggestion to do so has been given. Furthermore, the deeply hypnotized individual appears able to tolerate the incongruity of this situation and manifests what I have called "trance logic," ignoring the fact that one person can be in only one place at one time (5). That this phenomenon, which can, of course, be seen only in deeply hypnotized individuals, is not a function of previous knowledge or subtle cueing by the hypnotist is demonstrated by the fact that it is not manifested by simulators.
We have observed other differences between these two types of subjects that help clarify the nature of hypnosis. In a series of studies on posthypnotic behavior (6,7) we explored the occurrence of the posthypnotic response in the absence of the hypnotist outside of the experimental setting. We found that simulating subjects would respond more reliably and for a longer period to a posthypnotic cue in the presence of the hypnotist but would stop responding in a situation which they perceived to be outside of the experimental context. Deeply hypnotized subjects, however, were as likely to respond in the hypnotist's absence as in his presence.
In another study (8,9) we asked the question, "What would happen if the hypnotist disappeared while the subject was in deep hypnosis?" Using a tape-recorded hypnotic induction procedure with highly trained subjects, we arranged a "power failure." The tape-recorder came to a grinding halt, all lights extinguished, and the polygraph which was used to record subjects' physiological responses stopped. After a very brief attempt to restart the tape-recorder by flickering the switch, the technician muttered, "The fuse is out again," adding an appropriate pithy comment, and left the room. Observing the subject without his awareness once he was alone, we noted that deeply hypnotized subjects continued in the state for some period of time, apparently pulling themselves out of hypnosis after a period of some minutes had elapsed. Usually they would awaken and appear disoriented. On the other hand, simulating subjects characteristically opened their eyes and began to look as soon as they believed themselves to be alone.
It is clear from these and similar studies that, despite the superficial similarities which make it difficult to distinguish between the hypnotized subject and the simulator by simple observations, there is a vast difference between subjects who are hypnotized and those who are simulating.
10 MARTIN T. ORNE
With appropriate techniques, these differences become evident and help clarify the nature of the changes, other than those of subjective experience, that are associated with entering hypnosis.
When a series of methodological tools which made possible the scientific study of hypnosis had been developed, it became clear that few of the previously reported findings could be accepted at face value. The motivated experimental subject has a repertoire of responses and capabilities far greater than heretofore recognized, and the methodological problems raised by this discovery have had considerable impact on much psychological research. At the same time, it became clear that hypnosis does produce profound alterations in the subjective experience of individuals and in some of their behavioral responses as well. With improved methods it has become imperative to reexplore many of the psychophysiological consequences of hypnosis. We have been able to show that there are differences in the hypnotized individual's response to pain (10) and it would seem surprising indeed if these differences were not accompanied by physiological differences -- though they yet remain to be established. Again, while changes in the electroencephlograph, for example, have not been observed to date, newer techniques both in the analysis of EEG and in methodology of hypnotic research make further work essential.
Meaningful application of hypnosis for any purpose is not possible without first recognizing its nature. Our studies have not only allowed us to dismiss some of the quasi-magical ideas about hypnosis but have also evolved some hard criteria for the phenomenon. In the process of answering substative questions about human capability under hypnosis, we also inevitably are required to study more about human capabilities in the normal waking state. While we expect a great deal from the former, some of the most startling observations to date have been of the latter. Nonetheless, hypnosis remains a powerful, dramatic way of altering the subjective experience of the individual. An understanding of this process must ultimately increase our ability to predict and control human behavior.