Essentials of Complementary and Alternative Medicine (June 1999)

History and Development
Hypnotic phenomena have been reported in all cultures across the world, across all periods of recorded human history, and in various culturally conditioned forms. 
Manifestations of hypnotic behavior have typically occurred in either a religious or a healing (medical) context. Although hypnosis did not originate over 200 years ago 
in France with the physician Anton Mesmer (1734–1815), Mesmer deserves recognition as the first person known to propose a naturalistic rather than a demonic 
explanation of hypnotic phenomena. Borrowing from contemporary physics, Mesmer formulated a theory of hypnotic phenomena based on (animal) magnetism 
radiating from himself.
The last 75 years of experimental research has clearly established that the bulk of hypnotic phenomena resides in the hypnotized person's natural hypnotic ability, not 
in any projection from the hypnotist (
8
). In controlled experiments conducted more than 200 years ago by the French Academy of Sciences, no evidence of projected 
animal or other magnetism was found. However, the Academy did not deny the empiric efficacy of Mesmer's cures. They concluded that Mesmer was wrong with 
respect to the mechanism of hypnotic behavior but correct about its clinical efficacy. The French Academy of Sciences alternatively proposed that the mechanism of 
Mesmer's cures was mere imagination. The Academy was partly right with respect to the hypothesized mechanism of imagination in that it proposed to account for the 
clinical efficacy of hypnosis. However, it was erroneous with respect to the implication that the potency of imaginative and cognitive effects is trivial. Recent work in 
cognitive neuroscience has made it clear that the effects of cognitions on biology are very significant (
9
).
The French Academy's report temporarily hurt the scientific study of hypnosis. In coining the term  hypnotism (or nervous sleep), English physician Braid proposed 
another naturalistic mechanism of explanation for hypnosis. Braid went on to demonstrate its clinical efficacy in medical practice. This second hypothesized 
mechanism of hypnotic behavior has also been proved false by controlled electroencephalographic (EEG) studies of hypnosis in the last 50 years. Hypnosis is 
definitely not sleep stages 4, 3, or 2 (
10
).
The third naturalistic, but pathologic, explanation of the mechanism of hypnotic clinical efficacy was proposed by the eminent French neurologist Charcot (1835–1893) 
in Paris. Charcot's towering scientific reputation and clinical genius resuscitated the scientific and clinical status of hypnosis. According to Charcot, hypnosis was 
associated with a psychopathologic phenomenon (hysteria) and abnormal central nervous system function. Bernheim, Professor of Medicine at Nancy, France, 
challenged Charcot's theory and proposed a fourth theory—that hypnosis was caused by the normal behavioral phenomenon of increased suggestibility.
The modern systematic study of hypnosis in the experimental psychology laboratory can be dated to the psychologist Clark Hull, whose research at Yale University 

led to the 1933 publication of  Hypnosis and Suggestibility, a classic text. More recently, the eminent experimental psychologist E.R. Hilgard (
8
) of Stanford University 
has made major contributions to establishing the scientific credibility of hypnosis with quantitative empirical studies of individual differences in hypnotic ability as well 
as a theory of hypnosis as “divided consciousness” (
11
).
Many physicians and psychologists are unaware that in 1955 the British Medical Association cautiously recommended the teaching of hypnosis in medical schools 
and its use in clinical practice. In 1958, the American Medical Association made a similar recommendation. In 1960, the American Psychological Association officially 
recognized the American Board of Psychological Hypnosis (ABPH) and its authority to examine and certify diplomates with advanced competence in either 
experimental or clinical hypnosis. There are now similar national boards identifying advanced competence in medical and dental hypnosis.
Model: Mind-Body Interaction
In 1979, a model of mind-body interaction, in which hypnotic ability was a central component and beliefs were hypothesized to have biologic consequences, was 
proposed (
1
, 
3
, 
12
, 
13
 and 
14
). There is now evidence from cognitive neuroscience that causality operates in both a top-down and a bottom-up sense (
9
). Recent 
empiric neuroscience work has been specifying the neuroendocrine and immune links between cognitions, emotions, and biology (
2
). There is growing evidence that 
beliefs may have biologic consequences through neuroendocrine and immune mechanisms (
1
, 
3
, 
12
, 
13
 and 
14
). In fact, in highly hypnotizable persons and under 
certain conditions (e.g., during stress, highly supportive environments), beliefs can have potent, specific, and reliable biologic consequences that range from allergic 
reactions, warts, or congenital skin diseases to changes in mammary glands, proneness of the skin to burn or be burnt , and the inhibition of bleeding (
1
, 
12
, 
15
). 
Converging evidence shows that if threatening beliefs are blocked from consciousness, under some circumstances they may drive somatic symptoms (
1
, 
3
, 
12
, 
14
, 
16
, 
17
 and 
18
) like chronic pain, insomnia, and irritable bowel syndrome; these circumstances may include major life changes, times of low social support, or the 
individual having a high Marlowe-Crowne score (an operational definition of repression [
19
]).
These physiologic effects in highly hypnotizable subjects can be more specific and rapid than the effects of drugs. For example, one study (
20
) demonstrated that it is 
possible under hypnotic control for an individual to demonstrate significant increases in peripheral skin temperature in one hand and significant decreases 
concurrently in the other hand. There is no drug that can have such concurrent and specific selective effects on the human body.
Hypnosis and the High-Risk Model of Threat Perception
In highly hypnotizable and motivated persons, hypnosis results in apparently involuntary changes in perception, memory, and mood; these changes can have 
profound behavioral and biologic consequences (
1
, 
3
, 
14
). This perception of changes occurring without effort or involuntarily is the litmus test of true hypnosis (
21
, 
22
). At a baseline level, there are large individual differences in people's response to hypnotic suggestions. There appears to be an association between a patient's 
measured hypnotic ability and the clinical efficacy of hypnotherapy for several somatic and psychologic symptoms (
5
, 
23
, 
24
). There is evidence from the high-risk 
model of threat perception (HRMTP) (
1
, 
3
, 
12
, 
13
 and 
14
, 
16
, 
25
) that high and also low hypnotic abilities, in interaction with conscious or unconscious negative 
affect, may be risk factors for several psychophysiologic disorders and some psychological disorders (
1
, 
12
, 
14
, 
25
). “Highs” will develop a mix of psychological (e.g., 
anxiety, depression) and somatic (e.g., pain, sleep) symptoms, but “lows” will develop mainly somatic symptoms during trauma or stress (
1
, 
12
, 
14
).
The HRMTP predicts that persons who are high in hypnotic ability will be at risk for stress-related disorders because of their hypersensitivity to the perception of 
threat; a superior capacity for operant conditioning; an ability to keep secrets from self (e.g., the ability to block memory and perception, as in posthypnotic amnesia 
and surgical hypnoanalgesia); a propensity to surplus pattern recognition (their tendency to see meaning in apparently randomly distributed events); and a tendency 
to surplus empathy (their tendency to absorb the emotions of others, with indistinct interpersonal boundaries). The following symptoms have been found to be related 
to a patient's measured baseline hypnotic ability: migraine pain intensity (
26
); experimental pain intensity (
27
); facial pain intensity (
28
); intensity of chronic urticaria 
(
29
); atopic eczema (
30
); severity of clinical and dental phobias (
31
, 
32
); negative moods (
33
, 
34
); major depression (
35
); posttraumatic stress disorder (
36
, 
37
); 
dissociative disorders (
38
, 
39
); predisposition to nightmares (
40
); EEG–defined insomnia (
3
); substance abuse and bulimia (
35
); moderate and morbid obesity (
3
, 
41
); 
and nausea and vomiting during pregnancy (
42
).
This vulnerability of highly hypnotizable persons is probably related to peculiarities in their perception, memory, and mood that amplify negative affect (
14
, 
16
). A 
subset of patients who have chronic somatic complaints that are unresponsive to standard medical or surgical therapy are high or low in hypnotic ability (
14
). Hypnotic 
ability is related, in a dose-response manner, to sympathetic electrodermal reactivity (EDR) during experimentally-induced cognitive stress, and hypnotizability 
interacts with experimentally-induced cognitive stress (
16
) to drive up EDR in chronic pain patients. High hypnotizability also appears related to alterations in immune 
function (
43
). There is evidence that low hypnotic ability during stress may be a risk factor for somatic symptoms associated with pathophysiology (
14
, 
44
, 
45
 and 
46
). 
These symptoms include chronic pain, morbid obesity, response to cardiac surgery, chest pain, and insomnia. It appears that these low hypnotizable patients, 
because of their rigidly skeptical cognitive style and denial of psychological causation, have limited psychological coping skills, are hyposensitive to psychosocial 
threat, and delay seeking diagnostic investigation of their symptoms (
1
, 
14
, 
41
). High hypnotic ability is related to sympathetic hyperactivity, and low hypnotic ability is 
related to dysregulation of the parasympathetic nervous system during stress or trauma (
19
, 
41
).
PRINCIPAL CONCEPTS
Despite prior controversy, sophisticated signal processing techniques have shown that, under baseline conditions and during and after hypnotic induction, there are 
systematic EEG differences between carefully selected high and low hypnotizable subjects (
47
). It is worth noting that these electrophysiologic differences exist in the 
frontal and temporal cortex, and that they distinguish between high and low hypnotizable subjects at baseline before any hypnotic induction. These converging data 
are consistent with the hypothesis that high and low hypnotizables have different cognitive styles and process information very differently both outside of and within 
hypnosis (
1
, 
3
, 
12
, 
14
), and that using biofeedback to increase theta EEG waves may at least temporarily increase hypnotic ability (
48
, 
49
). Recent work suggests that 
measuring baseline frontal and temporal theta waves may provide an electrophysiologic test of hypnotic ability.
Hypnotizability and hypnosis have also been shown to be related to the ability to verbally alter a variety of basic physiologic, electrophysiologic, and conditioning 
phenomena (operant and respondent) during hypnosis. For example, Klein and Spiegel (
50
) and Whorwell et al. (
51
) showed that hypnotizability or hypnosis could 
stimulate and inhibit both gastric acid secretion and a colonic motility index. In a controlled study, Ruzyla-Smith et al. (
43
) showed that hypnotizability was related to 
alterations in B cells and helper T cells of the immune system. Black (
52
) and Zachariae et al. (
53
) showed that high hypnotic ability was related to the inhibition of the 
Mantoux reaction to tuberculin, and that during hypnosis the Mantoux reaction could be selectively increased in one arm and decreased in the other. For persons who 
are highly hypnotizable, hypnotic analgesia is as effective as morphine and more effective than acupuncture (
54
, 
55
). Also, naloxone does not block the mechanism of 
hypnotic analgesia (
56
, 
57
). It has been found that hypnotically suggested visual hallucinations alter cortical EEG event-related potentials and not simply verbal 
reports (
58
, 
59
 and 
60
). High hypnotic ability can increase the rate of acquisition of learning in both operant (
61
, 
62
, 
63
, 
64
 and 
65
) and Pavlovian conditioning (
2
) 
situations. Hence, it is not surprising to find that high hypnotic subjects respond more rapidly to various types of short-term psychotherapy (
66
, 
67
) and appear to learn 
both adaptive and maladaptive responses rapidly and unconsciously (
1
, 
3
, 
12
, 
13
, 
19
).
PROVIDER-PATIENT INTERACTIONS
Hypnotizability
Hypnotherapy is the use of hypnosis for treatment and is usually added to some established form of psychosocial diagnosis and therapy (
65
), such as psychodynamic 
psychotherapy (
7
), behavior therapy (
65
, 
85
), or biofeedback therapy (
65
). It appears that a hypnotic induction can potentiate verbal instructions in any social 
influence situation (
65
, 
86
, 
87
 and 
88
). After a hypnotic induction, a variety of psychological techniques, such as age regression (psychodynamic therapy), guided 
imagery, systematic desensitization (behavior therapy), or the delayed or immediate feedback of biologic information (biofeedback), may reduce specific clinical 
symptoms (
89
). The clinical efficacy of blocking or recovering painful, fantasized, or real memories or altering experimental pain perception is related to hypnotic 
ability (
11
). The hypnotic induction ritual increases, at least moderately, the following factors:
1. Suggestibility
2. Imagery and fantasy ability
3. Access to primitive modes of information processing
4. Access to early childhood memories and fantasies
5. Tolerance for logical incongruities (e.g., “trance logic,” which is the acceptance of the suggestion that a person is in two different places at the same time)
6. Alteration or inhibition of cognition; selective amnesia

7. Creativity
8. Alterations in the perception of sensory events and muscular response
Empirical evidence exists showing that most of these eight factors can be increased to some degree by a hypnotic induction ritual particularly for persons who have 
high hypnotic ability (
90
, 
91
). Hypnotic ability, like absorption, appears to be a normally distributed stable trait (
8
, 
92
), with a .71 test-retest correlation after 25 years. 
It also appears to be partly genetically based (
93
, 
94
).
Hypnotizability associated with a “nonvolitional” response to hypnotic suggestions (
22
) is not compliance, conformity, gullibility, or social desirability (
8
, 
14
). Nor is 
hypnotizability correlated significantly with any other known personality variable measured on any standard personality tests, such as the Minnesota Multi-phasic 
Personality Inventory (MMPI), California Personality Inventory (CPI), Neuroticism Extroversion Openness (NEO) Personality Inventory, Eysenck Personality Inventory, 
or the Meyers Briggs Inventory (
8
, 
14
). Hypnotic ability is an essential, but not a sufficient, condition to demonstrate the aforementioned eight alterations in 
perception, memory, and mood. The highly hypnotizable person must also be motivated to participate in the hypnotic induction and to respond to the verbal 
instructions in hypnosis (
22
, 
65
).
Tests of Hypnotizability
There are a number of tests of hypnotizability, each of which has strengths and weaknesses. Hypnotic depth measurements made by clinicians 150 years ago are 
generally in agreement with psychometrically more reliable and valid procedures developed in the last 50 years. For example, the estimates of the percent of people 
in the general population with high hypnotic ability and those with low hypnotic ability are similar (
68
, 
69
).
S
TANFORD
 H
YPNOTIC
 S
USCEPTIBILITY
 S
CALE
, F
ORM
 C
Currently, the gold standard in the measurement of hypnotic ability is the Stanford Hypnotic Susceptibility Scale, Form C (
69
, 
70
). However, the Stanford Form C has 
many difficult cognitive items, and its distribution of scores is not normal, but it taps a broader range of hypnotic abilities than do other tests. Its use in routine clinical 
practice is not practical for several reasons. First, this test can be given to only one patient at a time, and it takes nearly one hour to administer. Second, its use in a 
clinical practice requires a skilled clinician, and it can generate a high rate (29–31%) of negative side effects (e.g., temporary headache or disorientation, nausea), 
even with otherwise healthy college students (
71
, 
72
).
H
ARVARD
 G
ROUP
 S
CALE OF
 H
YPNOTIC
 S
USCEPTIBILITY
, F
ORM
 A
For clinical research, there are several reasons to use the Harvard Group Scale of Hypnotic Susceptibility, Form A (
73
) (HGSHS:A) with congruent subjective 
validation (
21
). The Harvard scale was found to correlate .68 (p <.0001) with the gold standard in a recent study (
70
), and other studies have also found correlations 
as high as .84 (
74
). It was found that the HGSHS:A correctly classified more than 80% of highly hypnotizable persons (
74
). Group testing with the addition of 
subjective validation criteria (
21
) permits a skilled technician working under the supervision of a clinician to test 5 to 10 patients in 1 hour with a minimal rate (less 
than 3%) of negative side effects (e.g., temporary headaches and disorientation, nausea). Using the Harvard scale, Crawford et al. reported a 5% rate of similar 
negative side effects in normal college students (
71
).
After being tested on the Harvard scale, the bulk of patients (85%) in a behavioral medicine clinic reported temporarily increased comfort and relaxation. 
Approximately 5% of these patients even reported immediate temporary relief of their presenting clinical symptom or symptoms. This nonspecific therapy response is 
brief, but it grabs the patient's attention and improves general compliance. The Harvard scale correlates (r = .74) with the Stanford, Form C (
75
), which has been used 
extensively in large-scale longitudinal and cross-sectional studies of the stability and genetics of hypnotic ability. The Harvard scale has norms on large cross-cultural 
non-patient samples (
69
) and even on patients (
14
, 
35
, 
76
, 
77
). Also, this scale has high reliability, and its validity can be increased by using congruent, subjective 
scoring procedures (
21
, 
78
). Using a technician to administer the Harvard scale avoids contaminating the clinician–patient relationship with “failed” test suggestions. 
Failed test suggestions are hypnotic suggestions the patient could not experience. In clinical research today, the Harvard scale, along with subjective scoring, has 
many merits as a first test of hypnotizability. The patient's score can tell what disorders and types of symptoms (somatic or psychological) this patient is at risk for and 
what the mean rate of therapeutic response will be to learning-based treatments (e.g., biofeedback, cognitive behavior therapy).
H
YPNOTIC
 I
NDUCTION
 P
ROFILE
For purposes of rapid routine clinical screening, the Hypnotic Induction Profile (HIP) (
79
) has many merits. It is a brief test (10 minutes) that is minimally challenging to 
the patient and presents hypnosis as a subtle perceptual alteration involving a capacity for attention, responsiveness, and concentration that is inherent in the person 
and can be tapped by the examiner (
79
). The HIP has two components: the eye roll sign and the induction (IND). The eye roll sign appears to correlate .34 (p < .001) 
with the Stanford, Scale C (
69
). The eye roll is a biologic marker of hypnotic potential but not necessarily of typical hypnotic performance. The IND score on the HIP 
correlated at .63 with the Stanford, Form C, and when depth ratings from the two scales were compared, the correlation was .78 (
80
). The aforementioned 
considerations suggest that the HIP may be a useful, brief measure of hypnotic ability in clinical demonstrations and in routine clinical practice. Its limited acceptance 
by the research community is a constraint on its use in clinical research.
A
BSORPTION
 T
EST
All the previously mentioned tests of hypnotizability involve a standardized hypnotic induction and one or more behavioral or verbal report measures of response to 
standardized hypnotic suggestions. The absorption test involves neither a hypnotic induction nor an actual measure of response to standardized suggestions (
81
). 
The absorption test is a short (10 minute) paper-and-pencil test that correlates moderately with the Harvard and Stanford tests. It inquires about the frequency of 
naturally occurring hypnotic-like experiences (e.g., the ability to use fantasy and to ignore distractions in everyday life). On some theoretical and empirical grounds, it 
can be considered a good measure of true hypnotic ability and one of the most difficult cognitive tests of hypnotic capacity (
22
).
Absorption is a personality trait that is normally distributed, is stable (30-day retest, r = .91), and appears—like hypnotic ability—to be partly genetically based in 
monozygotic twins reared apart (
82
) and independent of context effects (
83
). Like hypnotic ability (
1
, 
12
, 
14
, 
16
, 
25
), absorption has also been shown to be a risk 
factor for several stress-related disorders, such as nonorganic chest pain (
45
) and somatic complaints in family medicine, as well as morbid obesity (
41
), nightmares 
(
40
), and anticipatory nausea and vomiting secondary to chemotherapy (
84
).
Absorption appears to provide a nonintrusive measure of hypnotic ability, particularly at the high (above 75%) and low (below 25%) ends of the scale. This test can be 
given to a patient before a clinical session to obtain a primitive estimate of the patient's hypnotic ability. I suspect that absorption will predict, for example, response to 
several clinical interventions like acupuncture, guided imagery, noncontact therapeutic touch, herbal therapy, massage, and so forth (
1
).