Self-hypnosis or autohypnosis is a procedure in which the subject both induces the hypnotic state and makes suggestions to himself. When self-hypnosis is to be employed as part of a therapeutic regimen, it is necessary for the therapist to train the patient in its use. Often training is done under heterohypnosis, and the initial self-inductions are aided by a posthypnotic signal to into the hypnotic state. There is some experiemental evidence that inexperienced subjects can hypnotize themselves about as well as they can by hypnotized by another person.

There is ambiguity, however, concerning the nature of the self-hypnosis procedures typoically employed in such studies, involving, as they do, an experimenter giving a subject either initial verbal instructions or a booklet of directions on hypnotizing one's self, as well as a set of suggestions. There may be elements of both self-hypnosis anad heterohypnosis present in such a procedure. The main value of heterohypnosis in training a subject to induce autohypnosis is not providing him with a posthypnotic signal for induction but in letting him experience the subjective feelings of the hypnotic state that he must seek to attain self-induction.

The view is often expressed that all hypnosis is self-hypnosis because it is the subject's imagination that produces all of the effects in heterohypnosis. On the other hand, it could be argued that all hypnosis is basically heterohypnosis, and self-hypnotic effects resulty from posthypnotic suggestions given while training subjects in self-hypnosis. As early as 1928, Young researched this issue. He had hypnotic subhjects instruct themselves prior to hypnosis to modify specific aspects of rapport behavior and posthypnotic amnesia. He found that subjects could do this successfully and concluded that ther was no sine qua non of hypnosis. Posthypnotic amnesia was dependant on the subjects set of expectancy, and hypnotic behavior could be modified in many ways without affecting its depth. He concluded that the essential element in heterohypnosis was the autosuggestion of the subject.

Ruch (1975) also supports the notion that active self-hypnosis is the primary phenomenon and that heterohypnosis is, in effect, a case of guided self-hypnosis. He found that initial self-hypnosis facilitated subsequent heterohypnosis but that conventional heterohypnosis (of a passive subject by an active hypnotist) inhibited later attempts at self-hypnosis. This inhibitory effect was eliminated when "first-person instructions" were used in heterohypnosis. That is, instead of saying to the subject, "I am going to give you suggestions to help you to relax," the experimenter would say, "I am going to give myself instructions to help me relax." Thus the subject was able to regard the hypnotist's voice as his own, making suggestions to himself.

Ruch's view of the primacy of self-hypnosis is contrary to the conventional idea that heterohypnosis is an aid in training a person in self-hypnosis. It is premature to say whether the foregoing results are generalizable or are limited to the particular induction procedures tested. However, it seems questionable to label the procedure used as self-hypnosis, since, in the initial instructions, the experimenter made suggestions concerning the sequence of events that were to occur and then left the subject to count to himself and experience hem. This is similar to the Flower method of heterohypnosis in which all instructions to the subject are massed at the beginning of the induction. For an induction to qualify as an example of true autohypnosis, the subject should be responsible for all elements of the induction, and the hypnotist should make no suggestions of any kind beyond requesting the subject to commence the procedure.

Johnson (1981) notes that any study of self-hypnosis must be contaminated to some degree by heterohypnotic influence unless the study is limited to spontaneously developed trance states. Gardner (1981) proposed making a distinction between self-hypnosis (which she used to indicate self-hypnosis preceded or aided by heterohypnosis) and autohypnosis (which referred to spontaneous autohypnosis with no prior heterohypnosis). However, since these two terms are generally used interchangeably, such a distinction will probably prove as futile as the distinction between susceptibility and hypnotizability made in this book. If such distinctions are to be made (and they probably should be), then perhaps it will be necessary to coin new terms.

Most researchers have found few, if any, differences in success in inducing self-hypnosis as a function of previous heterohypnotic experience (Johnson, 1979; Kroger, 1977a; LeCron, 1964; Sacerdote, 1981; Shor and Easton, 1973). Sacerdote (1981) points out that with the modern trend toward more permissive inductions, the distinction between heterohypnosis and self-hypnosis is becoming vaguer.

Fromm (1975) notes that until recently, most of the serious research in hypnosis was in heterohypnosis; the literature of self-hypnosis was often the product of "quacks and laymen." She questioned on theoretical grounds the common assumption that heterohypnosis and self-hypnosis are basically the same and undertook to investigate the similarities and differences between the two. She conceptualizes hypnosis as an ego-splitting process. In heterohypnosis, the ego splits into two parts: the experimenter (participating ego) and the observer (observing ego). In autohypnosis, the ego splits into three parts: the experimenter and observer plus a director who gives the hypnotic instructions and suggestions. She found that in some subjects, a third or fourth aspect of the ego, a skeptic, was also present.

In a preliminary study, Fromm gave 18 males and 18 females one session each of heterohypnotic and autohypnotic experiences using a counterbal-anced order of presentation. The 12 least susceptible subjects described both experiences as essentially the same, but the 24 most-susceptible subjects de-scribed subjective differences between the experiences. Idiosyncratic fantasy and visual imagery arose spontaneously with a much higher frequency In autohypnosis. There was also more rational, cognitive activity going on In thes condition, and subjects were unanimous in reporting the greater number of ego splits predicted. Autohypnosis was found, as predicted, to require more effort on the part of the subjects.

Some subjects were able to reach a deeper state under heterohypnosis, while others went deeper under autohypnosis. Fromm accounts for this difference in terms of differences between subjects with respect to their need for surrender versus their need for autonomy and control. In a second study, three males and three females were instructed to practice self-hypnosis once a day for a month. Subjects were required to keep a daily diary of their experiences and were interviewed by telephone every few days. They were also subjected to two interviews plus a follow-up group discussion one month after the study.

Fromm found that with practice, self-inductions were easier to achieve. Eventually subjects began to employ methods of induction exclusively of their own design. Some used dissociative methods, such as producing an arm levitation by forgetting about the arm or commanding it to rise. Others simply "let go" and developed a passive-receptive ego state. After 2 or 3 weeks, most subjects who did not incorporate self-hypnosis into their life-style became bored with the procedure and had to be coaxed to continue the experiment. One of the causes of this problem was thought to be the tendency of prolonged self-hypnosis to reduce the transference with the experimenter. It was found that imagery was stimulated to a much greater extent in self-hypnosis but that some effects, such as positive hallucinations, profound ego regression, and role playing, were easier to produce in heterohypnosis.

The major advantage claimed for self-hypnosis in this study was that the subject was always attuned to his own responses during induction, and hence suggestions could be optimally timed. An outside hypnotist can at best make an educated guess as to the subject's subjective state. Johnson and Weight (1976), using factor analysis, found that behavioral and subjective experiences of subjects under heterohypnosis and self-hypnosis were generally similar. However, heterohypnosis invoked more feelings of unawareness of the environment, passivity, and loss of control, while autohypnosis was associated with more feelings of time distortion, disorientation, active control, and variations of trance depth.

In a later longitudinal study, Fromm and her associates (1981) essentially confirmed her earlier results. They concluded that "expansive free-floating attention" and ego receptivity to internal stimuli were state-specific for self-hypnosis, while concentrated attention and receptivity to a single external source of stimuli were state-specific for heterohypnosis. Again imagery was found to be much richer in self-hypnosis, while suggestions of age regression or positive and negative hallucinations were markedly more effective in heterohypnosis.