THE WORKS OF CLARK HULL

TABLE OF CONTENTS
1. CLARK HULL: PSYCHOLOGY HISTORY (1884 - 1952): JANA SCHROCK 1
2. "LOST IN THE T-MAZE" CLARK HULL AND KENNETH SPENCE

2
A. OVERVIEW

3
B. INVESTIGATIVE STRATEGY

4
C. HABIT AND BEHAVIOR

5
3. CLARK HULL: (1884 - 1952): SUNNY COOPER, M.S., M.ED.

6
4. ENCYCLOPEDIA OF WORLD BIOGRAPHY ON CLARK LEONARD HULL 7
5. CLARK L. HULL - WIKIPEDIA

8
6. CLARK LEONARD HULL: (1884-1952) ENCYCLOPEDIA OF PSYCHOLOGY 9
7. CLARK HULL: HYPNOSIS AND SUGGESTIBILITY:
JAMES HORTON, HELEN CRAWFORD

10
(CLARK HULL: HYPNOSIS AND SUGGESTIBILITY: APPLETON-CENTURY-CROFTS:
NY: COPYRIGHT 1933: 1961 #8-14 )
8. PREFACE, ACKNOWLEDGMENTS, FOREWORD: CLARK HULL

12
A. PREFACE

12
B. ACKNOWLEDGMENTS

13
C. FOREWORD BY ERNEST R. HILGARD

14
9. CH 1: HYPNOTISM IN SCIENTIFIC PERSPECTIVE: CLARK HULL

16
A. MESMER AND THE BEGINNINGS OF HYPNOTISM
 16
B. BRAID AND THE REVOLT FROM ANIMAL MAGNETISM

19
C. LIEBEAULT AND THE FRENCH REVOLT FROM ANIMAL MAGNETISM 20
D. CHARCOT AND THE REVIVAL OF ANIMAL MAGNETISM

21
E. THE PARADOX OF ALFRED BINET

22
F. BERNHEIM AND THE ECLIPSE OF ANIMAL MAGNETISM

23
G. COUE AND AUTOSUGGESTION

24
H. HYPNOSIS AND THE CONTROL EXPERIMENT

25
I. THE SCIENTIFIC OUTLOOK FOR HYPNOSIS

26
10. CH 2: ELEMENTARY PHENOMENA OF HYPNOSIS AND SUGGESTIBILITY:
CLARK HULL
 27
A. WAKING SUGGESTION, CATALEPSY

28
B. LIGHT HYPNOSIS: CATALEPSY OF THE EYES

29
C. PROFOUND HYPNOSIS

30
D. POST HYPNOSIS SUGGESTIONS

32
E. RAPPORT

34
F. THE TRANCE EXPERIENCE
 35
11. CH 3: EXPERIMENTAL PHENOMENA OF DIRECT WAKING SUGGESTION:
CLARK HULL

37
A. UNCONSCIOUS MOVEMENT AND THE IMAGINATION

38
B. INDIRECT HETEROSUGGESTION

40
C. SUMMARY AND CONCLUSIONS

43
12. CH 4: SOME RELATIONSHIPS OF EXPERIMENTAL HYPNOSIS
AND SUGGESTIBILITY: CLARK HULL

44
A. HYPNOTIC SUSCEPTIBILITY SCORING SYSTEM: DAVIS & HUSBAND 46
B. CLINICAL OBSERVATIONS OF HYPNOTIC SUSCEPTIBILITY

48
C. SEX DIFFERENCES IN HYPNOTIC SUSCEPTIBILITY
 51
D. HYPNOTIC SUSCEPTIBILITY AND CHARACTER

57
13. CH 5: THE RECOVERY OF LOST MEMORIES IN THE HYPNOTIC TRANCE:
CLARK HULL

60
A. AN EXAMPLE OF TRAUMATIC AMNESIA

61
B. SOME TYPICAL OPINIONS REGARDING HYPNOTIC HYPERMNESIA 63
C. HYPNOTIC RECALL OF EARLY MEMORIES

64
14. CH 6: EXPERIMENTAL ASPECTS OF POST-HYPNOTIC PHENOMENA:
CLARK HULL

71
15. CH 7: HYPNOSIS AND THE DISSOCIATION HYPOTHESIS: CLARK HULL 88
16. CH 8: HYPNOSIS CONCEIVED AS SLEEP: CLARK HULL

100
17. CH 9: HYPNOTIC SUGGESTIBILITY AND THE TRANSCENDENCE OF
VOLUNTARY CAPACITY: CLARK HULL

113
18. CH 10: HYPNOTIC SUGGESTIBILITY AND THE TRANSCENDENCE OF
VOLUNTARY CAPACITY (Continued): CLARK HULL

121
19. CH 11: HYPNOSIS AS A STATE OF HEIGHTENED SUGGESTIBILITY:
CLARK HULL

139
20. CH 12: HYPNOSIS REGARDED AS HABIT: CLARK HULL

159
21. CH 13: HYPNOSIS AND NON-PRESTIGE SUGGESTION: CLARK HULL

164
22. CH 14: INTERPRETATIONS: CLARK HULL

178
 THE WORKS OF CLARK HULL
1. CLARK HULL: PSYCHOLOGY HISTORY (1884 - 1952): JANA SCHROCK (MAY 1999)
Clark Hull grew up handicapped and contracted polio at the age of 24, yet he became
one of the great contributors to psychology. His family was not well off so his education had to
be stopped at times. Clark earned extra money through teaching. Originally Clark aspired to be
a great engineer, but that was before he fell in love with the field of Psychology. By the age of
29 he graduated from Michigan University. When Clark was 34 when he received his Ph.D. in
Psychology at the University of Wisconsin in 1918. Soon after graduation he became a member
of the faculty at the University of Wisconsin, where he served for 10 years. Although one of his
first experiments was an analytical study of the effects of tobacco on behavioral efficiency, his
life long emphasis was on the development of objective methods for psychological studies
designed to determine the underlying principles of behavior.
Hull devoted the next 10 years to the study of hypnosis and suggestibility, and in 1933
he published Hypnosis and Suggestibility, while employed as a research professor at Yale
University. This is where he developed his major contribution, an elaborate theory of behavior
based on Pavlov's laws of conditioning. Pavlov provoked Hull to become greatly interested in
the problem of conditioned reflexes and learning. In 1943 Hull published, Principles of
Behavior, which presented a number of constructs in a detailed Theory of Behavior. He soon he
became the most cited psychologist.
THEORY: Hull believed that human behavior is a result of the constant interaction
between the organism and its environment. The environment provides the stimuli and the
organism responds, all of which is observable. Yet there is a component that is not observable,
the change or adaptation that the organism needs to make in order to survive within it's
environment. Hull explains, "when survival is in jeopardy, the organism is in a state of need
(when the biological requirements for survival are not being met) so the organism behaves in a
fashion to reduce that need" ( Schultz & Schultz, 1987, p 238). Simply, the organism behaves in
such a way that reinforces the optimal biological conditions that are required for survival.
Hull was an objective behaviorist. He never considered the conscious, or any mentalistic
notion. He tried to reduce every concept to physical terms. He viewed human behavior as
mechanical, automatic and cyclical, which could be reduced to the terms of physics. Obviously,
he thought in terms of mathematics, and felt that behavior should be expressed according to
these terms. "Psychologist must not only develop a thorough understanding of mathematics,
they must think in mathematics" (Schultz & Schultz, 1987, p 239). In Hull's time three specific
methods were commonly used by researchers; observation, systematic controlled observation,
and experimental testing of the hypothesis. Hull believed that an additional method was needed,
- The Hypothetico Deductive method. This involves deriving postulates from which
experimentally testable conclusions could be deduced. These conclusions would then be
experimentally tested.
Hull viewed the drive as a stimulus, arising from a tissue need, which in turn stimulates
behavior. The strength of the drive is determined upon the length of the deprivation, or the
intensity / strength of the resulting behavior. He believed the drive to be non-specific, which
means that the drive does not direct behavior rather it functions to energize it. In addition this
drive reduction is the reinforcement. Hull recognized that organisms were motivated by other
forces, secondary reinforcements. " This means that previously neutral stimuli may assume
drive characteristics because they are capable of eliciting responses that are similar to those
aroused by the original need state or primary drive" (Schultz & Schultz, 1987, p 240). So
learning must be taking place within the organism.
Hull's learning theory focuses mainly on the principle of reinforcement; when a S-R
relationship is followed by a reduction of the need, the probability increases that in future
similar situations the same stimulus will create the same prior response. Reinforcement can be
defined in terms of reduction of a primary need. Just as Hull believed that there were secondary
drives, he also felt that there were secondary reinforcements - " If the intensity of the stimulus
is reduced as the result of a secondary or learned drive, it will act as a secondary reinforcement"
( Schultz & Schultz, 1987, p 241). The way to strengthen the S-R response is to increase the
number of reinforcements, habit strength.
Clark Hull's Mathematico Deductive Theory of Behavior relied on the belief that the
link between the S-R relationship could be anything that might effect how an organism
responds; learning, fatigue, disease, injury, motivation, etc. He labeled this relationship as "E", a
reaction potential, or as sEr. Clark goal was to make a science out of all of these intervening
factors. He classified his formula sEr = (sHr x D x K x V) - (sIr + Ir) +/- sOr as the Global
Theory of Behavior. Habit strength, sHr, is determined by the number of reinforces. Drive
strength, D, is measured by the hours of deprivation of a need. K, is the incentive value of a
stimulus, and V is a measure of the connectiveness. Inhibitory strength, sIr, is the number of
non reinforces. Reactive inhibition, Ir, is when the organism has to work hard for a reward and
becomes fatigued. The last variable in his formula is sOr, which accounts for random error. Hull
believed that this formula could account for all behavior, and that it would generate more
accurate empirical data, which would eliminate all ineffective introspective methods within the
laboratory (Thomson, 1968). Although Hull was a great contributor to psychology, his theory
was criticized for the lack of generalizability due to the way he defined his variables in such
precise quantitative terms. "Thus, Hull's adherence to a mathematical and formal system of
theory building is open to both praise and criticism" (Schultz & Schultz, 1987, p 242).
Time Line
1884 Hull was born
1918 Received Ph.D. from University of Wisconsin
19 publication of the literature on tests and measurements 19 Becomes research Professor at
Yale
193 Hypnosis and Suggestibility , published
1940 Mathematico - Deductive Theory of Rote Learning: A study in Scientific Methodology
was published.
19 Principles of Behavior was published
1951 The Essentials of Behavior was published
1952 A Behavior System was published
1952 Hull died

2. "LOST IN THE T-MAZE" CLARK HULL AND KENNETH SPENCE

BIOGRAPHY: As a child and young adult Hull was plagued by poor health, and he had
poor eyesight all his life. A case of polio at age 24 left him disabled in one leg and forced him
to wear a heavy iron brace and to always walk with a cane. Ernest Hilgard once tossed
horseshoes with him behind his home and he said rather plaintively that horshoes with his son
was the first athletic event he had ever participated in. His family had little money and he had to
interrupt his education several times to earn money. He studied mining engineering at the
University of Wisconsin, then switched to psychology and got his Ph.D. in 1918. He was proud
of his dissertation because "it moved experimental psychology into the area of thought
processes by investigating the learning of concepts" (Hilgard, 1987), and became very downcast
when year after year no one paid attention to it. Other early work involved the effects of
tobacco on behavior and a survey of the literature on tests and measurements. He developed
methods of statistic analysis and invented a machine to calculate correlations. Intriguingly
enough he investigated concept formation and even hypnosis and suggestibility. In 1927 he read
Pavlov and turned to problems of conditioned reflexes and learning. His major books were
Principles of Behavior (1943) and A Behavior System (1952). He was ill for a number of years
while writing the latter book, and died before reading the galley proofs.
After his contact with Pavlov, Hull's position became one of an uncompromising,
radically behavioristic analysis. He read Isaac Newton's Principia (1697) to find out where his
early efforts at theory had faltered. Any difference between psychology and physics, held Hull,
is in degree and not in kind. His mature system of ideas left no room for consciousness,
purpose, or any other mentalistic notion. When he spoke of intervening variables they closely
and specifically tied to objective stimuli and responses. He viewed human behavior as
automatic and reducable to the language of physics. He warned against giving subjective
meanings to any behavior being observed. We mustn't anthropomorphistically think, for
example, "If I were a rat, what would I do in this situation?" In 1943 he suggested that we
should consider "the behaving organism as a completely self-maintaining robot, constructed of
materials as unlike ourselves as may be."
I have seen no biography of Hull and am moved to speculate that the childhood of a
sickly child with poor eyesight from a family with very little money could have been filled with
taunts and abuse from other children, if not from parents. Such a condition could provoke
someone to clamp an iron lid on the painful memories of his inner life, and could predispose
him to an outlook that rules out all interest in such things. Please note that this is pure
speculation, best formulated as a hypothesis. I look forward to finding some data that might
support or disprove it.
OVERVIEW: For about 20 years Clark Hull held a dominant position in the history of
American academic psychology. Vast numbers of experiments by experimental psychologists
all across the country were carried out to confirm or challenge various details of his theoretical
formulations. Then during the 1960s, almost as quickly as his ideas had come to dominate
learning theory, they largely vanished from it. The conclusion became first inescapable and then
widespread that the basic tack he was taking, that of trying to formulate a precise mathematical
model which would identify just how much of behavior is due to what, was an impossible task
even for the Norway rat in a T-maze, which was the dominant animal and environment in this
research. For a human being it was not even conceivable.
In retrospect it is incredible how much time and energy went into this endeavor and how
little came out of it. I will not bother here with the details of the various mathematical models
and the controversies about whether certain independent variables were related to others in an
additive or multiplicative relationship. Hull perhaps added somewhat more to our knowledge of
the behavior of the rat than Titchener did to our understanding of human consciousness, but not
much. His basic approach turned out to be, to use a precisely appropriate metaphor in his world
of rats and mazes, a blind alley.
One of Hull's starting points was in noting that conditioning theory failed to deal
convincingly with motivation. He was astute enough to recognize that motivation may be
viewed as either a learned aspect of behavior (as Guthrie viewed it) or as a behavioral
determinant independent of learning (as Tolman viewed it). Either way, it needed to be given
greater importance. Hull drew on Freud's "instincts" as motivating forces, but changed the word
to "drives" in his own formulations. Late in his life and work, in 1952, even before the futility
of his modeling endeavors became evident, Hull admitted that his system probably applied only
to hungry rats.
INVESTIGATIVE STRATEGY
Hull held that:
1. We should begin with specific testable postulates, even if basd on minimal evidence. Then
we derive concrete, empirically verifiable deductions from these and test them.
2. The task of a theorist is to formulate postulates so they will lead to unequivocal deductions.
3. The worth of a theory resides in how much research it generates and how consistent with its
theoretical deductions the findings are.
4. He was willing to put himself on the line with his predictions. His willingness to be wrong
was a remarkable virtue. He was constantly revising his theories in light of empirical results.
These first four points represent perhaps Hull's most lasting contribution to experimental
psychology. No one before Tolman and Hull was as careful, sophisticated, or precise in
experimental design. Their research models of compared groups were later supplanted by other
models, like Skinner's single-subject designs, but the sophistication in experimental design that
grew out of their work outlived their research programs and is still a characteristic feature of
American academic psychology.
Also part of Hull's approach were:
5. In theorizing, a heavy emphasis on intervening variables, cast in mathematical form.
6. Kenneth Spence was intimately associated with Hull through most of Hull's career. It was
Spence who urged Hull to adopt Tolman's "intervening variable" concept and approach.
We turn now to a brief characterization of some of the main features of Hull's ideas, leaving out
the mathematical models which went nowhere.
DRIVE AND REINFORCEMENT
1. Drive is based on animal's need-state--hunger, thirst, sexual arousal, pain, or whatever. Drive
activates behavior--any behavior.
2. Reinforcement occurs whenever drive is reduced, leading to learning of whatever response
solves the animal's problem. Thus the reduction in need serves as reinforcement and produces
reinforcement of the response that leads to it. 3. Basic approach: need-related motivation, drive,
and S-R learning are produced by (and only by) reinforcement. The S-R connection is called
"Habit.": 4. Hull held that drives are substitutable in motivating behavior. If a hungry animal
has learned a given response to get food, it should be easy to transfer the same response to get
water. Early studies tended to confirm this motivation transfer, but more recent experiments
have failed to find such motivation transfer when proper care is taken to use sources of drive
that can be independently manipulated. It Now appears that what happens with difference
sources of drive is very unpredictable.
AN ALTERNATIVE FORMULATION was proposed by Miller & Dollard. Using a similar
habit construct, they proposed that any strong stimulus can have motivating or drive properties
without being tied to the needs of the organism.
HABIT AND BEHAVIOR
1. Drive and habit act together to determine the strength of behavior. Thus the strength of
behavior depends on both:
a. animal's motivation at time of testing
b. amount of prior learning
Neither motivation nor prior learning alone will tell us what animal do.
2. Habit is built up as result of drive reduction
3. Habit strength depends on four different classes of independent variables:
a. Number of reinforced trials
b. Magnitude of reward
c. Immediacy or delay of reinforcement
d. interval between CS onset and US.
4. Behavior can be characterized by both frequency and magnitude, and the two measures need
not be correlated. We might have a veRy frequent response of low amplitude or a rare response
of high amplitude.
EVIDENCE REGARDING DRIVE REDUCTION
1. Habit is indeed built up as result of drive reduction.
2. Experimental work has shown that there are other variables. For example, Neal Miller carried
out a series of studies showing that placing food directly in te stomach is reinforcing but food in
the mouth is much more reinforcing.
3. Sheffield proposed that it is not drive or need reduction that constitutes reinforcement; but
simply the occurrence of a consummatory response. A rat learns a response when this response
lets it eat, mate, explore, etc. 4. Ultimately, the pure drive-reduction view of reinforcement was
abandoned. PURPOSE AND INCENTIVE:
1. Hull's first papers on learning theory in 1929 and 1930 were attempts to show that the
purposiveness in behavior which Tolman had demonstrated could be explained with Pavlovian
S-R associations. Hull sought to extend the S-R framework from Pavlov's original conditioning
situation to the kind of situation in which behavior apears highly flexible, adaptive, and
intelligent (like rats in T-mazes, we might say with tongue in cheek).
2. "Fractional anticipatory goal responses."
This is the idea that we anticipate the reinforcing event at the end of the chain and it motivates
the rat back at the beginning of the chain. We make fractional anticipatory goal responses" (rG
--r for the anticipatory response and G for the goal) that are themselves somewhat rewarding to
us. For instance, someone's fantasies of sexual intercourse yet to come, and the movements and
secretions associated with those fantasies, are themselves somewhat rewarding. This led to the
very productive insight that it works best to work backward in teaching chains of behavior.
3. Hull recognized that variables can affect performance without materially affecting learning.
Crespi (1942) ran a study in which different amounts of food did not affect learning but did
affect performance. Hull concluded that a fractional ancicipatory goal reaction becomes a
conditioned reinforcing stimulus, with reinforcing consequences. His final explanation of
Crespi's results was that animals getting more food have a more vigorous rG conditioned to the
goal box.
4. Hull's explanation for the latent learning demonstrated by Tolman was that "habit strenth"
had built up on early trials but performance was poor because incentive motivation was low.
When the rat found food, considerable incentive motivation was suddenly present.
HABIT FAMILY HIERARCHIES
1. This somewhat intimidating phrase refers to alternative behavior sequences that lead to the
same goal. Component acts of these sequences become conditioned to the same fractional
anciciatory goal reactions and in this sense constitute a "family."
2. The weaker responses in this hierarchy--farther to reinforcement, or more difficult to make,
etc., are less likely behavior sequences (the law of less work), and thus lower in the habit-family
hierarchy. That is, they are less likely to occur, even though they're there and available. 3. It
appears to me (Victor) that this concept might be applied to complexes, with a variety of
different defense mechanisms serving the repressive, discomfort-reducing function.
OSCILLATION AND THRESHOLDS
1. Following Guthrie's lead, Hull was one of the earlier psychologists to recognize that behavior
is essentially probabilistic. There is considerable variability in what a rat or person does.
2. He proposed that in order for behavior to occur, the factors that produce it must create a
tendency to respond that is greater than a given threshold. He also assumed that the threshold
varies randomly in time according to an oscillating function. This is surely wrong; it is not too
difficult to identify factors inside the person or environment that could affect this threshold. But
with his rats, Hull was essentially saying that behavior can be predicted only on the average,
over a period of time, or over a group of animals.
KENNETH SPENCE
1. Kenneth Spence hypothesized that discrimination occurs through reinforcement combined
with frustration or inhibition. The excitatory tendencies of the reinforced stimulus increase;
those of the unreinforced one weaken through an active inhibitory process. This is perfectly
consistent with Pavlov. 2. Spence identified Tolman's approach as SS, or Sign-significant
learning: "...organization into some kind of functional whole of the perceptual systems of the
subject." A tendency to explain learning in terms of circumstances in te perceptual field. The
S-R model, by contrast, accounted for learning by stimulus-response conections.
3. For Spence, reinforcement, rather than affecting habit strength, affected "incentive
motivation." In regard to inhibition, for example, "early (chimpanzee) experiments suggested
that the subject had to learn to expect or anticipate a reward upon responding to a stimulus
before there would be any inhibition or frustration effect. This was a shift in a direction
reminiscent of Tolman's formulations.
4. Hull defined learning (habit) broadly. He thought many different independent variables could
affect it, but that there were only a few things which resulted in motivation (drive). Spence
reversed this relationship. For him, motivation was broadly determined, including as it did both
drive and incentive motivation, whereas habit strength was just a function of the number of
stimulus-response pairings (this is a sort of contiguity position).

3. CLARK HULL: (1884 - 1952): SUNNY COOPER, M.S., M.ED.

DRIVE REDUCTION THEORY: BIOGRAPHY: Clark Hull did his undergraduate
work at University of Michigan (1913) and later earned his Ph.D. from the University of
Wisconsin in 1918. He spent most of his professional career at the Institute of Psychology at
Yale University (1929 until his death in 1952). At the age of 24 he contracted polio and suffered
disabilities as a result of it.
Theory: Major influences on Hull's thinking included the work of Pavlov, Darwin,
Thorndike and Tolman. His theory was an attempt to create a synthesis of the theories of these
researchers plus Newtonian physics. He became interested in hypnosis, and wrote a book
entitled Hypnosis and Suggestibility in 1933. During the 1940's and 1950's Hull's work was
much-cited in the psychological literature.
He is best known for his Drive Reduction Theory which postulated that behavior occurs
in response to "drives" such as hunger, thirst, sexual interest, feeling cold, etc. When the goal of
the drive is attained (food, water, mating, warmth) the drive is reduced, at least temporarily.
This reduction of drive serves as a reinforcer for learning. Thus learning involves a dynamic
interplay between survival drives and their attainment. The bonding of the drive with the goal of
the drive was a type of reinforcement, and his theory was a reinforcement theory of learning.
Hull believed that these drives and behaviors to fulfill the drives were influential in the
evolutionary process as described by Darwin. Movement sequences lead to need reduction as
survival adaptations. He assumed that learning could only occur with reinforcement of the
responses that lead to meeting of survival needs, and that the mechanism of this reinforcement
was the reduction of a biological drive.
Hull was only interested in operational descriptions of what was observable. He did not
deny cognitive aspects such as purpose, ideas, intelligence, insight, values, or knowledge, but
since these characteristics could not be directly observed, he did not include them as part of his
theoretical constructs. He devised a complex calculus to quantify behavior.
Hull's work has been criticized because he assumed that his laws of behavior, which
were derived from experiments with rats, would account for all human behavior, including
social behavior. While his theory is not considered a major contributor to current understanding
of learning, nonetheless he contributed to the methodology for experimentation in learning
theory.
Hull had several graduate students who went on to make contributions to the body of
knowledge concerning learning. Neal Miller developed a social learning theory that was
successfully applied to psychotherapy and understanding neurosis. Carl I. Hovland studied and
wrote about the psychology of war. O.H. Mowrer developed a two-factor learning model and
demonstrated that fear is an acquirable drive. His most famous student was Kenneth W. Spence,
who developed and extended Hull's theory and developed a discrimination learning theory.

4. ENCYCLOPEDIA OF WORLD BIOGRAPHY ON CLARK LEONARD HULL

The American psychologist Clark Leonard Hull (1884-1952) was a primary
representative of the neobehaviorist school. He was also the first known psychologist to apply
quantitative experimental methods to the phenomena of hypnosis.
Clark L. Hull was born in a country farmhouse near Akron, N.Y., on May 24, 1884. He
attended high school for a year in West Saginaw, Mich., and the academy of Alma College. His
education was interrupted by bouts of typhoid fever and poliomyelitis, giving him pause to
consider possible vocational choices; he decided upon psychology. He then matriculated at the
University of Michigan, took his bachelor's degree, and went on to the University of Wisconsin,
receiving his doctorate in 1918. Staying on at Wisconsin to teach, Hull was at first torn between
two schools of psychological thought which prevailed at the time: early behaviorism and Gestalt
psychology. He was not long in deciding in favor of the former.
After an experimental project on the influence of tobacco smoking on mental and motor
efficiency, Hull was offered the opportunity to teach a course in psychological tests and
measurements. Gladly accepting it, he changed the name to "aptitude testing" and worked hard at
developing it as a sound basis for vocational guidance. The material which he collected in this
course was gathered into a book, Aptitude Testing (1928). Next, with the help of a grant from
the National Research Council, he built a machine that automatically prepared the correlations
he needed in his test-construction work.
In 1929 Hull became a research professor of psychology at the Institute of Psychology at
Yale University, later incorporated into the Institute of Human Relations. He came to certain
definite conclusions about psychology, and in 1930 he stated that psychology is a true natural
science, that its primary laws are expressible quantitatively by means of ordinary equations, and
that quantitative laws even for the behavior of groups as a whole could be derived from the same
primary equations.
The next 10 years were filled with projects dealing not only with aptitude testing but with
learning experiments, behavior theory, and hypnosis. As a representative of behaviorism, Hull
fell into that school's neobehaviorist period of the 1930s and early 1940s. His basic motivational
concept was the "drive." His quantitative system, based on stimulus-response reinforcement
theory and using the concepts "drive reduction" and "intervening variables," was highly
esteemed by psychologists during the 1940s for its objectivity.
Hull was probably the first psychologist to approach hypnosis with the quantitative
methodology customarily used in experimental psychology. This combination of experimental
methods and the phenomena provided by hypnosis yielded many appropriate topics for
experimental problems by his students. Hypnosis and Suggestibility, the first extensive
systematic investigation of hypnosis with experimental methods, was published in 1933,
incorporating the earlier, and better, part of the hypnosis program that Hull had carried out at the
University of Wisconsin.
In 1940 Hull published, jointly with C. I. Hovland, R. T. Ross, M. Hall, D. T. Perkins,
and F. B. Fitch, Mathematico-Deductive Theory of Rote Learning. Three years later his
Principles of Behavior was published, followed by a revision of his theories in Essentials of
Behavior (1951). Hull expressed learning theory in terms of quantification, by means of
equations which he had derived from a method of scaling originally devised by L. L. Thurstone.
In his last book, A Behavior System (1952), Hull applied his principles to the behavior of single
organisms. His system stands as an important landmark in the history of theoretical psychology.
He died in New Haven, Conn., on May 10, 1952.

5. CLARK L. HULL - WIKIPEDIA

This article needs additional references or sources for verification. Please help to
improve this article by adding reliable references. Unverifiable material may be challenged and
removed.
This article has been tagged since June 2007.
Clark Leonard Hull (1884-1952) was an influential American psychologist and
behaviorist who sought to explain learning and motivation by scientific laws of behavior. Born
in Akron, New York, Hull obtained bachelor's and master's degrees from the University of
Michigan, and in 1918 a PhD in from the University of Wisconsin. His doctoral research on
"Quantitative Aspects of the Evolution of Concepts" was published in Psychological
Monographs.
Hull conducted research demonstrating that his theories could predict and control
behavior. His most significant works were the Mathematico-Deductive Theory of Rote Learning
(1940), and Principles of Behavior (1943), which established his analysis of animal learning and
conditioning as the dominant learning theory of its time. Hull is known for his debates with
Edward C. Tolman.
In experimental psychology, he created the "hypothetic-deductive" systematic method, after the
observation and elaboration of hypotheses. This method brought him precise definitions and
conceptualised axioms which helped him develop his theories. He believed that behavior was a
set of interactions between an individual and their environment. He analyzed behavior from a
perspect of biological adaptation, which is an optimization of living conditions through need
reduction.
Hull is often credited with having begun the modern study of hypnosis. His work Hypnosis and
Suggestibility (1933) was a rigorous study of the phenomenon, using statistical and
experimental analysis. Hull's studies emphatically demonstrated once and for all that hypnosis
had no connection with sleep ("hypnosis is not sleep, â| it has no special relationship to sleep,
and the whole concept of sleep when applied to hypnosis obscures the situation"). The main
result of Hull's study was to rein in the extravagant claims of hypnotists, especially regarding
extraordinary improvements in cognition or the senses under hypnosis. Hull's experiments did
show the reality of some classical phenomena such as hypnotic anaesthesia and post-hypnotic
amnesia. Hypnosis could also induce moderate increases in certain physical capacities and
change the threshold of sensory stimulation; attenuation effects could be especially dramatic.
Hull is famous for his signature hypnotic induction in which he would look at someone straight
in the eyes until they were induced.

6. CLARK LEONARD HULL: (1884-1952) ENCYCLOPEDIA OF PSYCHOLOGY

Clark L. Hull was born in a country farmhouse near Akron, New York, on May 24,
1884. He attended high school for a year in West Saginaw, Michigan, and the academy of Alma
College. His education was interrupted by bouts of typhoid fever and poliomyelitis, giving him
pause to consider possible vocational choices; he decided upon psychology. He then matriculated
at the University of Michigan, took his bachelor's degree, and went on to the University of
Wisconsin, receiving his doctorate in 1918. Staying on at Wisconsin to teach, Hull was at first
torn between two schools of psychological thought which prevailed at the time: early
behaviorism and Gestalt psychology. He was not long in deciding in favor of the former. After
an experimental project on the influence of tobacco smoking on mental and motor efficiency,
Hull was offered the opportunity to teach a course in psychological tests and measurements.
Gladly accepting it, he changed the name to "aptitude testing" and worked hard at developing it
as a sound basis for vocational guidance. The material which he collected in this course was
gathered into a book, Aptitude Testing (1928). Next, with the help of a grant from the National
Research Council, he built a machine that automatically prepared the correlations he needed in
his test-construction work.
In 1929 Hull became a research professor of psychology at the Institute of Psychology at
Yale University, later incorporated into the Institute of Human Relations. He came to certain
definite conclusions about psychology, and in 1930 he stated that psychology is a true natural
science, that its primary laws are expressible quantitatively by means of ordinary equations, and
that quantitative laws even for the behavior of groups as a whole could be derived from the
same primary equations.
The next 10 years were filled with projects dealing not only with aptitude testing but with
learning experiments, behavior theory, and hypnosis. As a representative of behaviorism, Hull
fell into that school's neobehaviorist period of the 1930s and early 1940s. His basic motivational
concept was the "drive." His quantitative system, based on stimulus-response reinforcement
theory and using the concepts "drive reduction" and "intervening variables," was highly
esteemed by psychologists during the 1940s for its objectivity. Hull was probably the first
psychologist to approach hypnosis with the quantitative methodology customarily used in
experimental psychology. This combination of experimental methods and the phenomena
provided by hypnosis yielded many appropriate topics for experimental problems by his
students. Hypnosis and Suggestibility, the first extensive systematic investigation of hypnosis
with experimental methods, was published in 1933, incorporating the earlier, and better, part of
the hypnosis program that Hull had carried out at the University of Wisconsin.
In 1940 Hull published, jointly with C. I. Hovland, R. T. Ross, M. Hall, D. T. Perkins,
and F. B. Fitch, Mathematico-Deductive Theory of Rote Learning. Three years later his
Principles of Behavior was published, followed by a revision of his theories in Essentials of
Behavior (1951). Hull expressed learning theory in terms of quantification, by means of
equations which he had derived from a method of scaling originally devised by L. L. Thurstone.
In his last book, A Behavior System (1952), Hull applied his principles to the behavior of single
organisms. His system stands as an important landmark in the history of theoretical psychology.
He died in New Haven, Connecticut, on May 10, 1952. Gale Encyclopedia of Psychology, 2nd
ed. Gale Group, 2001.

7. CLARK HULL: HYPNOSIS AND SUGGESTIBILITY: JAMES HORTON, HELEN

CRAWFORD
"As observations have accumulated, it has become more and more probable that
hypnosis is not a single, unitary thing. It appears, rather, to be a more or less loosely related
group of phenomena (p. 23)." Over 70 years ago, the experimental psychologist Clark L. Hull
(1884-1952), known perhaps best for his learning theories and research, wrote his seminal
work on Hypnosis and Suggestibility: An Experimental Approach (1933). It is one of the true
classics in experimental hypnosis research that remains quite relevant today and can easily be
savored by those individuals interested in clinical or experimental hypnosis work. Thus, it is a
joy to find this classic republished now for the second time with a new introduction by Michael
Yapko. Even though some of Hull's findings may now be controversial or even incorrect, as
Dr. Yapko states in his introduction to this reprint, one of Hull's major goals and contributions
was to stimulate high quality hypnosis research by subsequent researchers.
Drawing on his many studies, first done at the University of Wisconsin and later at Yale
University, Hull's purpose in writing this book was to present a body of experimental material,
involving normal rather than pathological subjects, to the public. Hull was assisted by 20
research associates to whom his book was dedicated "in remembrance of our united efforts to
establish hypnotism on a secure experimental basis." Guided by John Stuart Mill's (1919)
"method of difference," he presents many controlled experiments with sound methodology
designed to investigate what effects hypnosis produced, if any, beyond that which could be
produced in non-hypnotic conditions. Hull managed to give the field a place of respectability in
the scientific community and literally set the stage for future researchers to investigate hypnosis.
he further stated that the applied use of hypnosis would benefit from scientific experimentation.
Due to conflict with Yale University's psychiatry department and administration over his
research program, upon the completion of his book, Hull sadly left further hypnosis work to be
done by others (Hilgard, 1987).
For the readership of the American Journal of Clinical Hypnosis, there are two chapters of
particular interest that are intriguing yet controversial even by contemporary standards: "The
Recovery of Lost Memories in the Hypnotic Trance" and "Hypnosis and the Dissociation
Hypothesis." Even though Hull reports a case of seemingly accurate hypnotic recall of an
accident for which the subject had almost complete amnesia, he also reports a case of memory
fabrication in recalling an early memory and cautions that subjects may be fabricating recalled
memories. In addition to prior experimental research, he reviewed his associate Huse's
paired-associate recall paradigm in which she found that recall was slightly better in the normal
state than in the hypnotized state. Hull concludes that hypnosis does not usually aid in the
recovery of recently acquired material. he goes on to conclude "There is some striking
experimental evidence which, while not absolutely convincing, tends strongly to confirm the
clinical observations that hypnosis facilitates the recall of childhood and perhaps other remote
memories" (p. 127).
The chapter on hypnosis and the dissociation hypothesis reviews Pierre Janet's
influential conception of dissociation, Morton Prince's attempt to test the dissociation
hypothesis, and Burnett's studies of carrying out two simultaneous tasks when both tasks were
conscious or one was "subconscious." This set the stage for the wellknown experimental studies
of the dissociation hypothesis by his laboratory associates Messerschmidt and Mitchell. he
concluded that their results suggest rather strongly that the whole concept of dissociation as
functional independence is an error. It is to be hoped that the situation is now sufficiently
clarified that the near future will see a series of well controlled, large-scale investigations which
will completely remove the uncertainties which at present becloud this extremely important
problem (p. 191).
Due to World War II and the lack of interest in scientific hypnosis research, we waited
much longer than Hull would have liked! Ernest Hilgard at Stanford University picked up his
interest in dissociation, and he and his colleague's work is detailed in Divided Consciousness:
Multiple Controls in Human Thought and Action (Hilgard, 1977). And, still we have not
removed all of the uncertainties that Hull referred to, and more research is still needed on this
extremely important problem.
Hull begins his book by briefly reviewing the history of hypnosis from its unscientific
beginnings with Mesmer, through its progression to the current status existing in the early
1930s. It was in 1923 that Hull became interested in hypnosis. According to Milton H. Erickson
(1961), he had an impact on Hull when Erickson, still an undergraduate at the University of
Wisconsin, was invited by Hull to provide demonstrations to a seminar of graduate students. As
an historian of American psychology, Hilgard (1987) explored this apparent association
between Erickson and Hull, and was perhaps surprised to find that Hull attributed his beginning
interests not to Erickson but rather to his hypnotizing a student with a phobia. "The only
mention of Erickson is of a tooth extraction reported to Hull by Erickson after he already had
his medical degree, a case in which Erickson had served as the hypnotist" (Hilgard, 1987, p.
824). Who was correct we will never know, but these little stories humanize our hypnosis
ancestors.
Hull devotes one chapter to the broad phenomena of waking and hypnotic suggestibility
and provides wonderfully rich and detailed demonstrations. he spends another chapter on direct
waking suggestions; which is reminiscent of work by Gheorghiu (Gheorghiu & Reyher, 1982)
and others. Hull shows experimentally that suggestion can influence differentially the behavior
of individuals in both waking and hypnotic states. he highlights but does not investigate
systematically individual differences to the degree that the field does today. Hull investigated
suggestibility as correlated to other fundamental variables such as sex, age, intelligence,
character, psychoneurotic tendencies, psychoses, delinquency, and drug influence. As with all
his conclusions, Hull cautiously reports his findings with the full knowledge that he is relying
on statistical probability, and that future research could support or contradict his findings.
In the chapters on hypnotic suggestibility and the transcendence of voluntary capacity,
Hull emphasized that hypnotized subjects were more responsive to suggestion and that they
experienced analgesia and were less susceptible to pain during experimental procedures, two
main findings that are still supported in hypnosis research today. he also found little if any
enhancement of motoric strengths or resistance during hypnosis.
Influenced by his interest in learning and the history of facilitation of hypnosis by
successive repetitions of the hypnotic state, Hull designed a series of experiments to determine
if hypnosis was a habit. he identified six characteristics of habituation, detailed in Chapter
Twelve, and proceeded to investigate the extent to which hypnosis conformed to those
characteristics. Additionally, Hull tested the effects of repetition on habituation in the waking
state and on tonic immobility in a fowl. The experimental results revealed that repeated
hypnosis inductions resulted in high conformity to the characteristics he detailed for habituation.
Hull concluded that "such a remarkable and detailed conformity of the phenomena of hypnosis
to the known experimental characteristics of ordinary habituation can hardly be accidental and
without significance. The indication would seem to be that, whatever else hypnosis may be, it
is-to a considerable extent, at least -a habit phenomenon..." (p. 347). Hull found, to a lesser
degree, parallel results for the waking condition. In the last chapter Hull goes into some detail
describing what hypnosis is not. Hypnosis is not inherently rapport or essential catalepsy, is not
a form of true sleep, does not inherently involve heightened sensitivity, is not pathological, and
does not necessarily involve a state of dissociation. he finally concluded that:
the only thing which seems to characterize hypnosis as such and which gives any justification
for the practice of calling it a "state" is its generalized hypersuggestibility. The difference
between the hypnotic state and the normal is, therefore, a quantitative rather than a qualitative
one (p. 391).
We anticipate that all clinicians and researchers interested in hypnosis will place this book on
their bookshelf, and that the pages will show evidence of their being read and enjoyed over the
years. We agree with Hull's reflection on his own book: "I believe...that the book itself has been
worth doing from the point of view of the advancement of science. I believe that it is an
important contribution, that it may mark a new epoch in that form of experimentation, and that
it will be read and quoted for a long time, possibly a hundred years (Hull, 1962, p. 852, cited by
Hilgard, 1968 p. xv). The year 2033 is not far off and we hope Hull's worthwhile book will still
be in print.
 (CLARK HULL: HYPNOSIS AND SUGGESTIBILITY: APPLETON-CENTURY-CROFTS:
NY: COPYRIGHT 1933: 1961 #8- 14)
8. PREFACE, ACKNOWLEDGMENTS, FOREWORD: CLARK HULL: HYPNOSIS AND
SUGGESTIBILITY: APPLETON-CENTURY-CROFTS: NY: COPYRIGHT 1933: 1961
PREFACE: [The study of hypnotism bristles with difficulties, although this has not
occurred to the numerous persons who have expected to find in these questions the occasion of a
brilliant and easy success. Although nothing is more simple than the invention of dramatic
experiments, which strike the vulgar with fear and astonishment, it is on the other hand very
difficult, in many cases, to find the true formula of the experiment which will give its result with
convincing accuracy.--Alfred Binet and Charles F ere, Animal Magnetism (1888).] I first
became interested in the subject of hypnotism some ten years ago, while giving a course of
lectures to a class of medical students. The innumerable opportunities which the dilapidated state
of the subject offered for the application of modern experimental procedures proved a temptation
too great to be resisted. From this interest developed a program of research which has yielded
thirty-two published papers, together with a number of others which have not appeared in print.
While the results of other workers have been utilized in the preparation of the present work
wherever possible, my chief object has been to make this body of experimental material
available to the general public.
Throughout the history of hypnotism the clinical approach, with its preoccupations with
remedial exigencies, has greatly predominated. In contrast to this tendency the approach of the
present work is experimental rather than clinical; the persons employed as subjects in the
program of research were normal rather than pathological; the ends sought were principles and
relationships rather than treatments and cures. This choice was not made through any lack of
appreciation of applied science, but rather in the belief that in the long run application itself will
be furthered by a proportionate development of pure science. There has accordingly been special
emphasis upon the methodology of control experimentation, in which sphere hypnotic
investigations of the past have been so largely defective. A considerable number of the more
important techniques which have proved useful are therefore described in sufficient detail to be
immediately available to persons contemplating research in this field.
Distinctly the most satisfying aspect of the years of effort which have produced the series
of experiments here recounted has been the association in scientific endeavor with a score of
enthusiastic young collaborators. The major portion of (ix) whatever may prove of value in the
work clearly must be credited to them; my own relationship to it has been essentially that of a
catalyzing agent. As such 1 have been permitted to share more or less vicariously in the joys of
pursuit and in the occasional thrills of success. The dedication of this monograph to my
associates in research is a feeble expression of my appreciation of this privilege.
To certain persons, however, my debt is so great that special mention must be made of
the fact. Miss Messerschmidt at considerable personal sacrifrce carried out, for insertion in this
work, a special study on the relation of prestige suggestion to age in children. Professor Patten
carried out two last-moment studies in a frnal attempt to clear up the question of
hypersuggestibility. Dr. Williams and Mr. Krueger were my intimate associates in New Haven
during a difficult year of concentrated investigation of these questions. Mr. Walter C. Shipley
computed the statistical reliabilities of nearly all of the experimental data cited. Dr. Sears read
and criticized the greater part of the preliminary draft of the manuscript; Mr. Switzer later gave a
similar criticism to all of it, and prepared the indexes. Miss Ruth Hays gave invaluable aid in the
preparation of the manuscript from the beginning. Finally, many valuable suggestions have been
received from professor Elliott in his capacity as editor. C.L.R. New Haven January, I933 (x)
ACKNOWLEDGMENTS: I take pleasure in making the following acknowledgments
of courtesies extended to me in the form of permission to use quotations and figures.
To The Macmillan Company for a quotation and for my Figure 34 from Janet's The Major
Symptoms of Hysteria, second edition, copyright 1920; for a quotation from Janet's
Psychological Healing, Volume I, translated by Eden and Ccdar Paul, copyright 1925; and for a
quotation from Prince's The Unconscious, copyright 1914.
To G. P. Putnam's Sons for many quotations from Bernheim's Suggestive Therapeutics,
translated by C. A. Herter.
To Dr. A. A. Brill and the Nervous and Mental Disease Publishing Company for quotations from
Freud's Selected Papers on Hysteria and Other Psychoneuroses, translated by A. A. Brill.
To the Nervous and Mental Disease Publishing Company for a quotation from Schilder and
Kauders's Hypnosis, translated by S. Rotherberg.
To The Oxford University Press for quotations from Pavlov's Conditioned Reflexes, translated
by G. V. Amep.
To the D. Appleton-Century Company, Inc., for many quotations from Binet and Fenes Animal
Magnetism; and for one quotation from Sidis's The Psychology of Suggestion.
To Charles Scribner's Sons for quotations from McDougall's Outline of Abnormal Psychology,'
for many quotations from Moll's Hypnotism, translated from the fourth enlarged edition by A. F.
Hopkirk; and for one quotation from Jacoby's Suggestion and Psychotherapy.
To Bailliere, Tindall, and Cox for quotations from Wingfreld's An Introduction to the Study of
Hypnotism.
To the Allied Book Company for several quotations from Forel's Hypnotism and Psychotherapy,
translated from the frfth German edition by H. W. Armit.(xi)
To the American Library Service for a quotation from Coue’s How io Practise Suggestion and
Autosuggestion.
To Wm. Rider and Son for several quotations from Bramwell's Hypnotism.
Numerous quotations and figures have been borrowed from the following journals: Allgem. antI.
Zeitschrift f. Psychotherapie u. psychische Hygiene; American Journal of Physiology; American
Journal of Psychology; .4.rchives de Psychologie; Archiv f. experimentelle Pathologie u.
Pharmakologie; Archives of lrzternal Medicine; Archives of Neurology and Psychiatry;
Archives of Psychology; British Journal of Psychology; Bulletin of the Johns H opkins Hospital;
Deutsches Archiv f. klinische M edizin; Deutsche medizinische W ochenschrift; Human
Biology; Journal of Abnormal and Social Psychology; Journal of Applied Psychology; Journal
of Experimental Psychology; Journal of General Psychology; Journal of General Physiology;
Journal of Genetic Psychology; Journal of Physiology; Klinische Wochenschrift; Medizinische
Klinik; Miinchener medizinische Wochenschrift; Psychological Bulletin; Psychological
Monographs; Psychological Review; Scientific Monthly; Studies from the Yale Psychological
Laboratory; University of California Publications in Psychology. (xii)
FOREWORD BY ERNEST R. HILGARD : Professor Clark L. Hull (1884-1952) was
an ardent behaviorist, but an unusual one. Instead of closing his eyes to phenomena that might
seem to others to be the very epitome of the subjective -whether the trance state within hypnosis
or matters such as foresight and purpose-he set out to show that he could treat them by his
objective methods. This book was one of the diverse products of this ingenious experimenter and
theoretician, working first at the University of Wisconsin, and later at Yale University. His other
books, on aptitude testing (1928) and on learning (1940, 1943, 1951, 1952), were all of quality
and significance, and this book on hypnosis stands up well alongside them.
It was an important event in the checkered story of hypnosis when this book, so
methodologically sound, so full of experimental data, made its appearance. Hypnosis had always
been a bit on the fringe-known too often as entertainment, or exploited by those who wished to
demonstrate some mysterious or uncanny functioning of the human mind-and here at last was a
matter-of-fact laboratory investigation. After more than 30 years Hull's findings hold up very
well indeed, and it is good to have his book widely available in accessible form.
Although devoted largely to the experiments of his own and his students and
collaborators, these investigations are placed in the context of history, and the historical material
is painstaking and accurate. It is interesting, for example, to see Hull's concern over Alfred
Binet, a distinguished investigator (whose name we know in connection with intelligence tests)
because Binet permitted the authority of Charcot to make him accept the bizarre results of some
experiments, results which could only have been obtained due to experimenter bias.
Hull's own biases did not lead him to mistake his own fancies for data, and all
experiments are beautiful illustrations of experimental design and objectivity of interpretation.
Despite his courage in attacking a range of problems not always congenial to the objectivist,
Hull's behaviorism led him to an overemphasis (xiii) upon phenomena of movement, learning,
and retention, for which methods of measurement were available, and to underplay such
phenomena as positive and negative hallucinations. To his credit it must be said that he
recognized the broad range of the standard phenomena of hypnosis as genuine, and neglect does
not mean denial. Despite an interest in individual differences that had led to an earlier book on
aptitude testing, the material on differences in susceptibility is fragmentary, and because of his
lack of interest in personality dynamics there is almost none of the clinician's "feel" for the
interpersonal aspects of hypnosis. But there is so much here that a few omissions can be excused.
Ever since the days of Mesmer, hypnosis has had a rough and undulating history of
acceptance and rejection. There are signs now of greater acceptance, such as recommendations
by both the American and British Medical Associations that hypnosis be taught in medical
schools, and the establishment with the blessing of the American Psychological Association of
an American Board of Examiners in Psychological Hypnosis, issuing certificates in both
experimental and clinical hypnosis. Those who are today finding it easier to do investigatory
work in the field without being considered "far out," and to use hypnosis in their therapeutic
work without being ostracized owe some of this acceptance to Clark L. Hull and his
collaborators for having carried out the work reported in this book. Hull, in the conclusion to the
book, noted some of the difficulties in working in this area:
Too many of the works on the subject in the past have fallen short of the scientific ideal.
Doubtless many things have contributed to this weakness, but surely a major factor must be the
inherent difficulty of the problems involved, the fundamental elusiveness of the phenomena, and
the consequent subtlety necessary in the experimental controls. These difficulties are so great
that to enter seriously on a program of investigation in this field is a little like tempting fate; it is
almost certain to court disaster. Small wonder that orthodox scientists have usually avoided the
subject! Yet each generation may be expected in the f~ture, as in the past, to produce a few rash
souls who will not only risk the danger of making scientific errors but will also have the courage
to brave the semi-superstitious fears of the general public and the uneasy suspicions of their
orthodox scientific brethren. It is to them that the present work is really addressed (page 403).
Because the problems of hypnosis lie at the very heart of human planfulness and
self-control, as well as the relinquishing of control to another, because its methods permit
explorations of significant reality distortions, hypnotic experimentation holds the promise of
broadening our understanding of personality functioning generally, apart from the intrinsic
interest in hypnotic phenomena themselves.
Hull, in the privacy of his notebook diaries, felt that this book was a good one. After
expressing some concern that he might become stigmatized for working on hypnosis, he goes on
to say:
I believe, however, that the book itself has been worth doing from the point of view of the
advancement of science. I believe that it is an important contribution, that it may mark a new
epoch in that form of experimentation, and that it will be read and quoted for a long time,
possibly a hundred years.1
The book is well along on its first hundred years without losing any of its freshness. It
still stands as a model of clarity and objectivity in the approach to what remain even today
puzzling and unsolved problems. ERNEST R. HILGARD May 1961

9. CH 1: HYPNOTISM IN SCIENTIFIC PERSPECTIVE: CLARK HULL: HYPNOSIS AND

SUGGESTIBILITY: APPLETON-CENTURY-CROFTS: NY: COPYRIGHT 1933: 1961
Phenomena more or less resembling those of the hypnotic trance appear to have been
known from very early times, especially among Oriental peoples. For the most part, these states
were associated with religious and mystical practices. While of interest to the curious, this phase
of the history of hypnosis has left no tangible contribution to science. It is sufficient to observe
that hypnotism originated in magic in much the same way that chemistry arose from alchemy
and astronomy from astrology.
MESMER AND THE BEGINNINGS OF HYPNOTISM: About the time of the
American Revolution, Franz Anton Mesmer (1733-1815), a Viennese physician, put forward to
the scientific world the theory and practice of what he called animal magnetism. Mesmer's
medical training naturally made his interests clinical. No doubt the spirit of the times tended to
tinge his practice with something of the mystical. In any case, having failed of success in
Germany, he went in 1778' to Paris where he soon had a tremendous vogue. There he opened a
remarkable clinic in which he treated all kinds of diseases.
The clinic was held in a large hall which was darkened by covering the windows. In the
center of this room was a large oaken tub, the famous baquet. The tub was about a foot high and
large enough to permit thirty patients to stand around it. It was filled with water, in which had
been placed iron filings, ground glass, and a number of bottles arranged in a symmetrical
manner. Over the tub was a wooden cover provided with openings through which projected
jointed iron rods. (3)
These rods were applied by the patients themselves to their various ailing parts. While at
the tub the subjects were commanded to maintain absolute silence, possibly to render them more
susceptible to the plaintive music that was provided. At the psychological moment, Mesmer
would appear on the scene garbed in a brilliant silk robe. He would pass among the patients,
fixing his eyes upon them, passing his hands over their bodies, and touching them with a long
iron wand. Individuals apparently suffering from the most varied disorders declared themselves
cured after two or three such treatments.
Mesmer's Character: It is difficult at this day to form a just opinion of Mesmer's
character. Widely different views have been held. He has been charged, probably not entirely
without truth, with being an avaricious charlatan. Bernheim, for example, says "In spite of the
discredit which the interested charlatanism of Mesmer threw upon his practices, magnetism
continued to have its followers ... "
Somewhat more charitable views have been held by others. Moll, perhaps the most
learned and judicious writer on the subject of hypnosis, remarks: "I do not wish to join the
contemptible group of Mesmer's professional slanderers. He is dead, and can no longer defend
himself from those who disparage him without taking into consideration the circumstances or the
time in which he lived. Against the universal opinion that he was avaricious, I remark that in
Vienna, as well as later in Morsburg and Paris, he always helped the poor without reward. I
believe that he erred in his teaching, but think it is just to attack this only, and not his personal
character. Let us consider, however-for I deem it right to uphold the honor of one who is
dead-more closely in what his alleged great crime consisted. He believed in the beginning that he
could heal by means of a magnet, and later that he could do so by a personal indwelling force
that (4) obvious relation to hypnosis as it is now known.
A number of the more interesting propositions follow:
"1. A responsive influence exists between the heavenly bodies, the earth, and animated bodies.
"2. A fluid universally diffused, so continuous as not to admit of a vacuum, incomparably subtle,
and naturally susceptible of receiving, propagating, and communicating all motor disturbances,
is the means of this influence.
"3. This reciprocal action is subject to mechanical laws, with which we are not as yet acquainted.
"4 Alternative effects result from this action, which may be considered to be a flux and reflux.
"5. This reflux is more or less general, more or less special, more or less compound, according to
the nature of the causes which determine it ....
"9. Properties are displayed, analogous to those of the magnet, particularly in the human body, in
which diverse and opposite poles are likewise to be distinguished, and these may be
communicated, changed, destroyed, and reinforced. Even the phenomenon of declination may be
observed .... (There was no numbers 10. 11, or 12 listed.)
"13. Experiments show that there is a diffusion of matter, subtle enough to penetrate all bodies
without any considerable loss of energy.
"I4. Its action takes place at a remote distance, without the
aid of any intermediary substance.
"I5. It is, like light, increased and reflected by mirrors.
"16. It is communicated, propagated, and increased by sound.
"I7. This magnetic virtue may be accumulated, concentrated,
and transported .... (There was no numbers 18, 19, 20, 21, or 22 listed.)
"23. These facts show, in accordance with the practical rules I am about to establish, that this
principle will cure nervous diseases directly, and other diseases indirectly.
"24. By its aid the physician is enlightened as to the use of medicine, and may render its action
more perfect, and he can provoke and direct salutary crises, so as completely to control them ....
"27. This doctrine will, finally, enable the physician to decide upon the health of every
individual, and of the presence of the diseases to which he may be exposed. In this way the art of
healing may be brought to absolute perfection."
Animal magnetism was held to be a kind of impalpable gas or fluid as distinguished from
the magnetism of minerals. Its distribution and action were supposed to be under control of the
human will. Mesmer's followers not only believed that this strange fluid could be reflected by
mirrors but that it could (6) be seen. Trained somnambulists were supposed to behold it
streaming forth from the eyes and hands of the magnetizer, though they disagreed as to whether
the color was white, red, yellow, or blue! It was agreed, however, that the fluid could be
confined in a bottle, and transported thus to exert its marvelous power in distant places.
Mesmer and the Verdict of the Commissioners: In 1784, owing to a dispute which
had arisen between Mesmer and some of his disciples over the right to give public lectures
revealing his supposed secrets, the French government intervened by appointing a commission to
investigate the truth of his claims. Benjamin Franklin ,vas one of the members of the
commission. These hard-headed individuals proceeded to run a series of control experiments,
which, if performed by Mesmer himself, would have saved the world an immense amount of
confusion.
"The commissioners were particularly struck by the fact that the crises did not occur
unless the subjects were aware that they were being magnetized. For instance, in the experiments
performed by J umelin, they observed the following fact. A woman who appeared to be a very
sensitive subject, was sensible of heat as soon as Tumdin's hand approached her body. Her eyes
were bandaged, she was informed that she was being magnetized, and she experienced the same
sensation, but when she was magnetized without being informed of it, she experienced nothing.
Several other patients were likewise strongly affected when no operation was taking place, and
experienced nothing when the operation was going on. But the most curious experience of this
kind was made in Deslon's presence, much to his confusion. According to the theory, when a
tree was magnetized, every person who approached it was affected by its influence. The
experiment was made at Passy when Franklin was present. Deslon magnetized one tree in an
orchard, and a boy of twelve years old, very sensitive to magnetism, was brought towards it with
his eyes bandaged. At the first, second, and third tree he turned giddy; at the fourth, when he was
still at a distance of twenty-four feet from the magnetized tree, the crisis occurred, his limbs
became rigid, and it was necessary to carry him to an adjoining grass-plat before Deslon could
recall him to consciousness." (7) Under such conditions the commission had no alternative but to
decide against the existence of animal magnetism as a physical force. Moreover, even at that
early day they were dimly aware of the psychological nature of the phenomena in question, as is
shown by the following section taken from their official report:
"Finally, they have demonstrated by decisive experiments that imagination apart from
magnetism, produces convulsions, and that magnetism without imagination produces nothing.
They have come to the unanimous conclusion with respect to the existence and utility of
magnetism, that there is nothing to prove the existence of the animal magnetic fluid; that this
fluid, since it is non-existent, has no beneficial effect; that the violent effects observed in patients
under public treatment are due to contact, to the excitement of the imagination, and to the
mechanical imitation which involuntarily impels us to repeat that which strikes our senses ... "
The Royal Society of Medicine made a very similar report at about the same time: "From
a curative point of view animal magnetism is nothing but the art of making sensitive people fall
into convulsions. From a curative point of view animal magnetism is useless and dangerous ... "
These reports marked the end of Mesmer's popular favor m Paris, and he returned to Germany.
"Artificial Somnambulism" and Related Trance Phenomena: Mesmer did not
hypnotize his subjects, although some of them appear to have had spontaneous hysterical
convulsions and to have shown other related behavior while at the tub. The sleeping trance,
which is a familiar part of hypnotism today, seems to have been discovered accidentally in 1784
by a follower of Mesmer, the Marquis de Puysegur. One day he attempted to apply Mesmer's
magnetizing methods to a young shepherd, Victor, who, instead of showing the usual hysterical
convulsions, fell into a quiet sleeping trance. From this condition he appeared not fully to awake
for some time, but went about his duties like a sleep-walker. When he finally awoke (8) from his
somnambulistic state, he was unable to recall anything that had happened during the period of its
existence. The sleeping or trance condition, with its subsequent amnesia, was quite naturally
regarded as an artificially induced somnambulism, and it at once attracted a great deal of
attention, partly, no doubt, because of the supposed clairvoyant powers of subjects while in that
state. About the same time Petetin, a physician at Lyons, described the phenomenon of hypnotic
catalepsy or muscular immobility. The discovery of the remaining major hypnotic phenomena
followed rapidly, and by 1825 hypnotically induced positive hallucinations (seeing things which
are not present), negative hallucinations (being functionally blind to things really present),
hypnotic anxsthesias, hypnotic analgesias (insensibility to pain), and the action of post-hypnotic
suggestion had all been clearly described.
The century which has elapsed since 1825 has been much less fertile in the discovery of
hypnotic phenomena than the preceding half-century. Indeed, almost the only noteworthy
tendency during this period has been the gradual though still incomplete correction of errors
which had accumulated around the pseudo-science of hypnotism previous to that date. One
development is the gradual, but only partial, escape of hypnotism from its age-long
entanglements with mysticism and magic. A second and more dramatic episode is the struggle
centering around the rivalry between the physical theories of animal magnetism and the
psychological theory of suggestion as alternative explanations of hypnotic phenomena. It is with
events of this latter conflict that we shall new concern ourselves.
BRAID AND THE REVOLT FROM ANIMAL MAGNETISM: The subjective or
psychological nature of hypnotic phenomena seems to have been discovered and eXploited quite
independently by James Braid in England (1843) and by a group of French investigators whose
work began with the Abbe Faria (1819) and culminated with Liebeault (1866) and Bernheim
(1886). In contrast with the French movement, which seems to have been a somewhat gradual
development with several persons contributing, Braid's stroke of insight appears to have (9) been
a relatively independent and isolated event. In 1841 he witnessed a mesmeric seance conducted
by a French magnetizer named Lafontaine. Braid first went to the demonstration suspecting
fraud, but upon witnessing it a second time, and after making certain tests on the magnetized
subjects himself, he became convinced that the phenomena were genuine. He was aroused to
enthusiasm and later experimented extensively on his own account. These researches very soon
led him to the view that the cause of the various phenomena was not a fluid which passed from
the body of the mesmerist into that of the subject.
Further, Braid developed a special technique for inducing the trance, a method still
extensively used. Originally he caused his patients to look at a cork bound to the forehead. His
later procedure was to have the subject look fixedly at some bright object, such as his lancet
case, which was held near, and slightly above, the eyes in such a way that the eye muscles were
under a certain amount of strain. This technique was usually combined with verbal suggestion,
though he seems not to have appreciated, at least during his early years of experimentation, the
importance of the role played by suggestion in the process. Braid coined the word hypnotism,
now in general use, and utilized the trance mainly for painless surgical operations, which he
performed in large numbers.
Braid's book, Neurypnology, in which he gives the chief exposition of his views, waS
writt.en only two years after his first contact with animal magnetism. It contains many long
notes in which he replies often with considerable heat to the numerous unjust criticisms made by
persons now otherwise forgotten. Though the book is very clumsily written, the author's
sincerity gleams forth from every page. His rejection of the theory of a magnetic fluid marks a
great advance over the magnetizers. He also saw with perfect clearness that hypnosis is not a
natural sleep. These things are lastingly to his credit, and give Braid a secure place in the history
of science. Even so, if judged by modern standards, Neurypnology must be considered as
psychologically naIve, and the experiments reported in it as decidedly amateurish. Of the nine
points which he regarded as significant and well established by his (10) researches, probably no
more than two or three would survive properly controlled repetition.
Perhaps Braid's most remarkable example of bad experimentation was the long series of
studies by which he supposed he had confirmed the then current theory of ph reno-magnetism.
He believed that pressure, when applied to certain points on the heads of hypnotized subjects,
would evoke from them behavior characteristic of a corresponding phrenological faculty. Thus,
when he pressed the "organ" of veneration, "an altered expression of countenance took place,
and a movement of arms and hands, which later became clasped in addition, and the patient ...
arose from the seat and knelt down as if engaged in prayer."
But this experimental naivete must have been largely the result of inexperience, for in
later years his methods greatly improved. The results obtained from subsequent and better
controlled experiments finally led him to abandon his belief in phrenology ami its hypnotic
correlates. He appears also in his later years to have come gradually to a very adequate
realization of the role played by suggestion in producing the phenomena, both of the hypnotic
and the waking or "vigilant" condition.
LIEBEAULT AND THE FRENCH REVOLT FROM ANIMAL MAGNETISM:
The parallel movement in France which opposed the theory of a magnetic fluid was much more
complicated. It began in 1814-1818, when the Abbe Faria showed by experiments that no special
force was necessary for the production of mesmeric phenomena such as the trance, but that the
determining cause lay within the subject himself. One of Faria's subjects was a general named
Noizet, who became converted to the Abbes views. He, in turn, passed the teachings on to a
physician, Alexander Bertrand, who elaborated them. Both Noizet and Bertrand wrote books
upon the subject.
Basing his opinion largely on the striking similarities between the systems of Noizet and
Bertrand on the one hand, and that of Liebeault on the other, Pierre Janet has advanced the view
that Noizet's book may have fallen into the hands of Liebeault. Bramwell, on the other hand,
calls attention to the fact that Braid's anti-magnetic views were being exploited in France (11)
through the influence of Azam and others in 1859-1860. At all events, we find Liebeault
seriously beginning the study of mesmerism in 1860, but entirely rejecting the theory of a
magnetic fluid.
Liebeault was a humble physician who began a country practice in 1850. In 1864 he
settled at Nancy and practised hypnotism among the poor peasants who came to his clinic. The
temper of the man is indicated by his refusal to accept fees for these services. Bramwell, who
visited Liebeault's clinic, draws such an inimitable picture of it that it must be quoted:
"His clinique, invariably thronged, was held in two rooms in the corner of this garden ...
The patients told to go to sleep apparently fell at once into a quiet slumber, then received their
dose of curative suggestions, and when told to awake, either walked quietly away or sat for a
little to chat with their friends, the whole process rarely lasting longer than ten minutes ...
No drugs were given, and Liebeault took special pains to ex- . plain to his patients that he
neither exercised nor possessedfl any mysterious powers, and that all he did was simple and .
capable of scientific explanation ... A little girl, about five years old, dressed shabbily, but
evidently in her best, with a crown of paper laurel leaves on her head and carrying a little book
in her hand, toddled into the sanctum, fearlessly interrupted the doctor in the midst of his work
by pulling his coat, and said, 'You promised me a penny if I get a prize.' This accompanied by
kindly words, was smilingly given, incitement to work having been evoked in a pleasing, if not
scientific way. Two little girls, about six or seven years of age, no doubt brought in the first
instance by friends, walked in and sat down on a sofa behind the doctor. He, stopping for a
moment in his work, made a pass in the direction of one of them, and said, 'Sleep, my little
kitten,' repeated the same for the other, and in an instant they were both asleep. He rapidly gave
them their dose of suggestion and then evidently forgot all about them. In about twenty minutes
one awoke, and wishing to go, essayed by shaking and pulling to awaken her companion -her
amused expression of face, when she failed to do so, being very comic. In about five minutes
more the second one awoke, and, hand in hand, they trotted laughingly away."
After coming to Nancy, Liebeault began writing a book on hypnotism which was
finished after two years of hard work. When it was published, however, only one copy was sold!
But Liebeault patiently pursued his gratuitous labors among (12) the poor for twenty years,
when, by a kind of accident, his remarkable work Was finally recognized. It seems that
Bernheim, a professor in the medical school at Nancy, treated without success for six months a
case of sciatica which had lasted for six years. The patient was quickly cured through hypnotic
suggestion admonished by Liebeault. This stking cure led Bernheim to investigate the novel
method of treatment. His initial incredulity Soon changed to enthusiastic admiration, and in
1884-1886 Bernheim published an attractively written book in which be directed the attention of
the World to Liebeault's work. Then tardily, twenty years after it had been written, the remaining
copies of Liebeault's book were finally sold, and the modest physician at last received
recognition. Doctors from all countries now flocked to Nancy to study his methods.
CHARCOT AND THE REVIVAL OF ANIMAL MAGNETISM: But suggestion as
an explanation of hypnotic phenomena was yet to encounter a severe struggle. Quite
independently of Liebeault, Charcot, an anatomist and neurologist of Paris, had around 1880
attracted considerable attention by his courageous experiments and lectures on the subject of
hypnosis. Warned by the unscientific extravagances which had very properly brought the
magnetizers into disrepute, Charcot resolved that his experiments, at least, should be
ultra-scientific and technically above reproach. It is largely because of this that the controversy
which eventually grew up between the Paris and Nancy schools merit our attention. Nevertheless
the despite Charcot's scientific intentions, no one has ever fallen into more scientific
interventions, no one has ever fallen into more grievous experimental errors or gone astray in
experimental method than he.
Charcot seems to have been especially fearful of being deceived by his subjects. He
therefore sought in their behavior for signs of magnetic influence which could not be simulated.
Apparently be never hypnotized anyone himself but depended upon his assistants, who brought
the subjects to him. These subjects were mainly three hysterical young Women. With these three
mentally pathological subjects, he sought diligently for objective signs characteristic of the
hypnotic sleep. Quite naturally he employed the same general metbods that he had (13) recently
applied with success to the study of locomotor ataxia and lateral sclerosis. When his subjects
were stimulated, their muscles seemed to show characteristic reactions following definite laws.
"All of these phenomena could be successfully linked to Charcot's earlier studies. They
could be examined with the guidance of the same anatomical ideas. The same method and the
same instruments could be used. The same little hammer could be used for testing the reflexes.
As of old, demonstrations could be made by the chief to an admiring circle of pupils. It was still
possible to seek upon the bared limb of the subject a place where a blow with the hammer would
readily induce a well-marked contracture, and one plainly visible to all beholders. To Charcot
this was irresistible. He declared that the study of such phenomena could be conducted by a
perfectly sound method; that the method sufficed to exclude the possibility of fraud, which had
invalidated the old experiments upon somnambulists; and that it was in the light of the data
acquired by this method that a critical review of all the recorded phenomena of animal
magnetism must be undertaken." (8, 168) 4
THE PARADOX OF ALFRED BINET: Binet and Fere, loyal and admiring pupils of
Charcot, elaborated his views in their book Animal Magnetism. In regard to the phenomenon of
transference, they remarked:
"In subjects sensitive to the magnet, the transfer of unilateral contractures may be
effected by means of this agent; thus, when the ulnar attitude has been produced in the right
hand, and a magnet is brought close to the subject's fore-arm when he is asleep, and even when
he is awake, both his hands become agitated with slight, jerking movements; then the contracture
of the right hand ceases, and is transferred to the left hand, without losing any of its
characteristics or of its precise localization."
Pursuing the methods just described, Charcot reported a number of supposed discoveries.
Major hypnotism, as it was now called, was said to show three sharply marked stages: lethargy,
catalepsy, and somnambulism. In the lethargic stage, induced by closing the subject's eyes, the
subject could hear nothing and could not speak; but, when certain nerves were pressed,
remarkable and uniform contractures resulted. If, while in the lethargic state, the subject's eyes
were opened, she at once passed into the cataleptic stage, in which the limbs remained in any
position they were placed by the experimenter, though she was still unable to hear or speak.
Lastly, if friction were applied to the top of the head, the subject passed into the somnambulistic
condition, which was substantially that of the ordinary trance. Sometimes these contractures,
catalepsies, and other hypnotic manifestations appeared only on one side of the body. In such
cases, if a large magnet were brought close to the limbs in question, the particular symptoms
would be displaced at once to the other side of the body. This phenomenon was called
transference. (14) Thus we find magnetism reappearing in the history of hypnotism, this time in
respectable, scientific garb, though quite as fallacious as when the existence of a magnetic fluid
was advanced by the old magnetizers.
Animal Magnetism was published in 1888, two years after the appearance of the first
edition of Bernheim's Suggestive Therapeutics. Possibly motivated by the opposing tendencies
of Bernheim's book, Binet and Fere performed a series of control experiments which, if properly
executed, would have removed the element of suggestion as a constant error from their
technique and have corrected their mistake. At the present time it is ironical to read their account
of these experiments: "We need not in this place prove the reality of aesthesiogenic influence, in
order to reply to those who only see in these agents the effects of suggestion and of expectant
attention, since we have already had occasion to explain this point. It only remains to show that
in the following experiments we took sufficient care to eliminate suggestion and expectant
attention. These were the points on which we insisted: Since these researches were new to us,
we were in many cases unable to foresee what would occur, and especially with respect to the
polarization of emotions, so that suggestion on our part was impossible. We repeated the
experiments on absolutely fresh subjects, and obtained the same results. The same effect was
produced when the magnet was concealed under a cloth. This was also the case when the magnet
was made invisible by suggestion. (15) although if there had been any results they could not
have afforded a counterproof, since they might have been explained by the recollection of a
previous peripheral excitement ... "
This insistence of Binet and Fere upon such a gross error in the face of Bernheim's
experiments and criticisms is especially surprising when one recalls the well-deserved scientific
fame to which Binet later attained. Only two years after the publication of Animal Magnetism,
Binet published La Suggestibilite, an extremely original and thoroughly scientific work on an
intimately related subject. And all the world knows that still later, in collaboration with Simon,
he devised the intelligence tests which bear his name and thus made to the science of psychology
one of the greatest contributions of its entire history. Even so, the fact remains that there has
rarely been written a book containing a greater aggregation of results from wretched
experiments, all put forward with loud protestations of impeccable scientific procedure and
buttressed by the most transparent sophistries, than this work by Binet and Fere.
BERNHEIM AND THE ECLIPSE OF ANIMAL MAGNETISM: To these claims of
Charcot and his followers Bernheim replied in the second edition of his book: "If, in our
researches, we failed to take as our starting-point the three phases of hysterical hypnotism
described by Charcot, this was because we were unable by our observations to confirm their
existence. We were unable to ascertain that the action of opening or closing the subject's eyes, or
friction of the vertex, modified the phenomena in any way; or that in the subjects who were not
disposed to manifest certain phenomena under the sole influence of suggestion, such phenomena
could be induced by any of the physical stimuli just mentioned.
" ... Conversely, all the phenomena can be readily obtained when they are described in
the subject's presence, and when the idea of them is allowed to permeate his mind. Not only can
all the classical effects of the magnet be induced in this way, but the same thing applies to all the
varieties of transference. I say, 'I am going to move the magnet, and when I do so there will be a
transference from the arm to the leg.' A minute later, the arm falls and the leg rises. Without
saying any more to the subject I next move the magnet back to the leg; thereupon there is a fresh
transference from the leg to the (16) arm. If, without disclosing the fact to the subject, I
substitute for the magnet a knife, a pencil, a bottle, a piece of paper, or nothing at all-still the
phenomena are witnessed."
The salutary manner in which Bernheim thus exposed the basic error in the experimental
technique of the Paris school, by means of an adequate control experiment, should be pondered
long and well by all who essay experimentation in the field of hypnosis.
COUE AND AUTOSUGGESTION: The conflict with Paris having been won, there
was yet another chapter in the history of hypnosis and suggestion to emanate from Nancy. In
1885 the good Liebeault met at Troyes a young druggist named Emile Coue. The two men at
once found much in common. For a time Coue studied and practised hypnotic suggestion
according to Liebeault's technique. Meanwhile, in his profession, he observed the influence of
waking suggestion in effecting cures when associated with the use of drugs, the latter often quite
ineffective in themselves.
Coue studied and brooded over the matter for a period of twenty-five years. In 1910, at
the age of fifty-three, he established what has sometimes been called the "neo-Nancy" school.
Following the example of his predecessor, Dr. Liebeault, Coue held his clinique in his own
home and gave gratuitously his healing suggestions to the many who flocked to receive them.
But his technique was different. Coue abandoned the trance entirely and depended wholly upon
waking suggestion. This he called autosuggestion, insisting that all suggestion is in reality
nothing but autosuggestion.
What Coue meant by the term autosuggestion may best be understood from his quaint
directions to a person suffering from pain: "Therefore every time you have a pain, physical or
otherwise, you will go quietly to your room ... sit down and shut your eyes, pass your hand
lightly across your forehead if it is mental distress, or upon the part that hurts, if it is pain in any
part of the body, and repeat the words: 'It is going, it is going,' etc., very rapidly, even at the risk
of gabbling ... The (17) essential idea is to say: 'It is going, it is going,' so quickly, that it is
impossible for a thought of contrary nature to force itself between the words. We thus actually
think it is going, and as all ideas that we fix upon the mind become a reality to us, the pain,
physical or mental, vanishes. And should the pain return repeat the process 10, 20, 50, 100, 200
times if necessary, for it is better to pass the entire day saying: 'It is going!' than to suffer pain
and complain about it."
The Tardy Development of Hypnotism as a Science: Such, in brief, is the history of
hypnotism. All sciences alike have descended from magic and superstition, but none has been so
slow as hypnosis in shaking off the evil associations of its origin. None has been so slow in
taking on a truly experimental and genuinely scientific character. Practically all of the actual
phenomena were discovered and described during the first fifty years, from 1775 to 1825. But
the century since 1825 has shown a remarkable sterility. Almost nothing of significance has been
accomplished during this period except the very gradual correction of errors which originally
flowed directly from bad experimental procedures. We have already had occasion to note a
classical case of this in the controversy between Bernheim and Charcot. The tardy development
of the science of hypnotism, moreover, is especially striking when it is recalled that practically
from the beginning hypnosis has been definitely an experimental phenomenon. Not only this, but
experimentation has been continuous and widespread during a period in which other fields of
science have made the greatest advances ever known.
The paradox in this case, as in all others, disappears with full knowledge of the attendant
circumstances. In the first place, the non-physical notions of the nature of mind fostered by
metaphysical idealism probably favored hypnotism's mystical affinities, and mysticism is
notoriously incompatible with controlled experiment. In the second place, as we have already
seen, the dominant motive throughout the entire history of hypnotism has been clinical, that of
curing human ills. A worse method for the establishment of scientific principles among highly
elusive phenomena could hardly have been devised. As (18) we shall have occasion to observe
frequently, one indispensable feature of satisfactory hypnotic investigation is the control
experiment. Charcot's magnetic experiment was utterly misleading and scientifically pernicious
until Bernheim completed it by substituting for the magnet "a knife, a pencil, a bottle, a piece of
paper, or nothing at alL" But deliberately to run a control experiment in genuine clinical practice
involves withholding from a considerable number of patients (the control group) a mode of
treatment possessing a certain presumption of curative value. This deliberate withholding of the
means of life and health from certain individuals, even though in the long run it might greatly
profit a much greater number of other individuals, is revolting to ordinary human nature. And,
when individual patients are paying individual doctors for treatment, it is quite out of the
question. The physician's task is to effect a cure in the quickest manner possible, using more or
less simultaneously any and all means at his disposal. General laws which call for the varying of
a single factor at a time do not readily emerge from such situations. Worse still (despite notable
exceptions, as we have observed in the work of Braid and Bernheim), the limitations of clinical
practice often influence the methods of experimenters accustomed to them, even when the
conditions surrounding the particular experimental situations are such as really to permit control
experiments to be carried out.
HYPNOSIS AND THE CONTROL EXPERIMENT: The control experiment has
long been known and employed by scientific investigators. It is an integral part of that most
potent of all scientific methods, the "method of difference." According to John Stuart Mill, it is
"by the method of difference alone that we can ever, in the way of direct experience, arrive with
certainty at causes." The principle upon which the method is based is stated by Mill as follows:
"If an instance in which the phenomenon under investigation occurs, and an instance in which it
does not occur, have every circumstance in common save one, that one occurring only in the
former; the circumstance in which alone the two instances differ is the effect, or the cause, or an
indispensable part of the cause, of the phenomenon." (19)
At bottom the method is very simple. Its procedure falls naturally into two parts. The
first part is usually thought of as the main or basic experiment. The second part is what we have
called the control experiment. It is the almost universal failure of the experimenters adequately
to perform part two as required by the method of difference that has proved so disastrous in the
history of hypnotism.
In the main or basic experiment by the method of difference there is set up an
experimental situation containing a factor A, which is presumed to be causally active, along with
attendant factors, B, C, and D, all of which are presumed in this particular situation to be
non-active. What follows from the joining of these factors is then noted. The result X, whatever
it chances to be, is likely by the unsophisticated to be taken forthwith as the effect of the
antecedent A. Thus when Charcot brought the magnet close to the contracted leg of the
hypnotized subject (A) and the contraction thereupon was transferred to the arm (X), he naIvely
concluded that the specific magnetic property of the magnet was the active agent. As a matter of
fact, no general conclusion whatever as to causation may safely be drawn at this stage of the
experiment. It is always possible that the observed consequent X may have been caused by the
supposedly neutral attendant circumstances B, C, or D, or some combination of them, and not by
A at all. It is still only an
hypothesis that A produced X.
The second or control part of the method of difference comes in at this point to clear up
the experimental ambiguity. A new experiment is set up which is in all respects exactly like the
first except that the antecedent factor A is absent. If, now, the consequent X is also found to be
absent, then the conclusion may be drawn that A is in truth the cause of X. But if, on the other
hand, X should really be found among the consequents, it will be quite as clear that A is not the
cause of X. This last is what happened when Bernheim carried out the control to Charcot's
experiment. With everything else the same, a bottle or a pencil, or simply the actions of the
experimenter (B), proved quite as efficacious in changing contractures as a true magnet. In short,
B-C-D-E and B-C-D-F led to X quite as readily as did A-B-C-D. If control experiments had
been general in hypnotic experimentation during the last hundred years, (20) the history of the
subject might have been shorter, but it would have been far more edifying.
No doubt an important factor contributing to the almost universal failure to perform
satisfactory control experiments in the history of hypnotism has been the grossly inadequate
training of the investigators. For the most part, their training, if any, has been in the non-mental
sciences of physics, chemistry, physiology, and anatomy. Almost without exception they have
known little or nothing of the technique and peculiar pitfalls of psychological experimentation.
The ordinary student of physics, chemistry, or physiology never encounters the phenomenon of
substitution of stimulus (or cause) which is characteristic of habit formation. Thus a magnet may
be active in two radically different ways. In the fIrst place, it may influence iron filings through
its specific magnetic properties. Training in the non-mental sciences equips the investigator to
cope with such situations. In the second place, it may acquire through habit formation the
capacity as a stimulus to evoke almost any conceivable motor response in a sensitive organism.
Such stimulus-response connections may be acquired with remarkable facility without even the
knowledge of the subject. The control experiment cited by Binet and Fed failed because, despite
their good intentions, they did not effectively prevent their subjects from receiving visual or
other sensory cues which, through previous associations, had acquired the capacity of evoking
the responses falsely believed to be due to the specific magnetic properties of the magnet. The
causal effects in this latter case are, of course, quite as physical as the action of the magnet on
iron filings, though the mechanisms involved in the two situations are radically different.
Thorough training in experimental human behavior is needed to cope successfully with this latter
type of situation.6
THE SCIENTIFIC OUTLOOK FOR HYPNOSIS: Despite the very devious and
unscientific history of hypnotism, there is excellent reason to expect a decided change for the
better. The excessive preoccupation with the clinical and 6 In some ways the psychological
methods of introspective analysis formerly in vogue in psychological laboratories were even less
adapted to hypnosis than those of the chemist or physiologist. (21) other practical applications of
hypnotism so characteristic of its history has now subsided to moderate proportions. This can
hardly be regarded as anything but a fortunate circumstance for the development of hypnotism as
a true experimental science and, ultimately, for its effective application as well. Moreover, the
rapid development of psychology as an experimental science within recent years has given to the
hypnotic investigator a large number of experimental methods and devices particularly of a
quantitative nature, which were not available to earlier workers. These may be utilized at once
with little or no modification. Indeed, the following pages will be occupied mainly with
elaborating in the concrete a variety of examples of the application of the methods of modern
experimental psychology to the solution of the elusive problems of hypnosis and suggestibility.
(22)

10. CH 2: ELEMENTARY PHENOMENA OF HYPNOSIS AND SUGGESTIBILITY:

CLARK HULL: HYPNOSIS AND SUGGESTIBILITY: APPLETON-CENTURY-CROFTS:
NY: COPYRIGHT 1933: 1961
DESPITE the very unscientific nature of most of the hypnotic investigations of the past,
a considerable number of significant manifestations have been discovered. As observations have
accumulated, it has become more and more probable that hypnosis is not a single, unitary thing.
It appears, rather, to be a more or less loosely related group of phenomena. Under present
conditions at least, the formulation of a really illuminating definition of hypnosis is hardly
possible. The exhibition of the concrete phenomena themselves is far more informing in any
case. Our immediate task, then, will be to give an acquaintance with the typical forms of
behavior ordinarily classed as hypnotic. In order to assist the reader in the attainment of an early
perspective, there will be presented at the outset a number of examples of transitional forms
involving 'waking suggestibility. Presentation will be by means of a series of descriptive
demonstrations.
Muscle Reading: Demonstration I. A number of young people have assembled for a
demonstration of thought reading. I inquire if there are any persons present who have been
successful in operating the "Ouija" board. Several have tried it and Miss X seems to have been
quite adept. She even intimates that "Ouija" has made some interesting revelations. I decide that
I shall try to "read her thought." I go out of the room and am thoroughly blindfolded. Meanwhile
Miss X carefully secretes a finger-ring. I am led back to the room. I know nothing about where
the object is hidden, or even what the object is. Miss X meets me at the door. I direct her to
grasp my right wrist firmly with her thumb on the lower side and the tips of her four fingers on
the back. I tell her to think constantly and intently of where the object lies hidden, and that I will
try to read her mind and locate the object with my right hand. I (23) hold my hand about ten
inches from my body and slightly below my chest, forearm horizontal. I keep the arm and hand
flexed just as much as possible so as to attain the maximum of sensitivity to slight unconscious
movements of the person whose mind I am trying to read. I wait for a time for some leading
movements but feel none. I begin to move slowly away from the door, at the same time
reminding the subject to think intently of where the object is. Presently I pause and try again for
signs. I make gentle movements in various directions, but find that the subject's hand makes a
slight resistance in all directions except towards the right. I therefore begin moving slowly
towards the right, watching intently for signs of resistance or of leading. After going some feet I
pause again for further signs, but am unable to distinguish anything. I make tentative movements
in several directions, but to no avail. Once more I admonish the subject to think intently. I hold
my arm perfectly still for some minutes, waiting for an indication. At length I detect a faint
pulling of my hand forward and somewhat to the left. I follow the lead slowly and tentatively.
By moving carefully I can feel that slight pulling nearly all of the time now. It leads me around
the corner of a table. At length I feel my hand touching the shoulder of a man's coat. I start to
move my hand upward, thinking the object may be lying on top of the man's head, but encounter
marked resistance from my subject's hand, which presses downward. I completely relax my arm
and the subject quite definitely moves my hand down the front of the man's coat and pauses at a
pocket. I reach into the pocket, feel the ring, and pull it out. Here the audience applauds and I
know I have been successful. I remove the blindfold and ask Miss X if she intentionally led me
to the ring. She vigorously denies leading me either intentionally or otherwise.
The phenomenon shown in Demonstration I has been variously known as mind-reading,
thought-reading, and muscle reading. The unconscious movements made by the subject, by
which I was able to locate the hidden object, have frequently been called ideomotor action.
Aside from being the naturalistic explanation of many phenomena popularly regarded as more or
less occult, among which may be mentioned the planchette phenomena and those of its modern
descendant, the "Ouija" board, ideomotor action is of basic importance in understanding
hypnotic phenomena. (24)
WAKING SUGGESTION, CATALEPSY: Demonstration II. A demonstration of
hypnotic phenomena has been promised a class in medical psychology. When the class has
assembled, I call for six young men to volunteer as subjects. After a little hesitation the required
number have come to the front of the room. They appear a trifle self-conscious. I announce that
the first experiment will not involve the trance at all, but will be concerned only with normal
waking suggestion. The men are arranged in a semicircle, standing close together. They are
instructed to remove their coats and bare their right arms to above the elbow. I stand before them
with a grave, somewhat tense expression. I extend my right arm straight and rigid, volar surface
upward, fist clenched, and instruct them to do the same. I say, "Make your arms just as straight
and rigid as you possibly can. Just as rigid as you possibly can!" After they have thus exerted
themselves to the utmost for about thirty seconds, I say, "Just as rigid as you can! Now I'll touch
a center in your arm." I now pass rapidly from one to another, giving each a sharp stab with my
index finger on the exposed upper surface of the elbow joint. Quickly stepping back to my
former position before them, I again extend my arm with an exaggerated display of muscular
effort and command them firmly, "Look at me. Look at my arm! Now I'm going to hold your
arms and I don't believe you can bend them. Try it. You can't do it ! You can't do it! YOU
CAN'T DO IT!" Two of the men bend their arms easily and are apparently quite unaffected. One
other bends his arm at once, but reports that it felt a little stiff. Two more of the men succeed in
bending their arms but, judging by their expression, it requires considerable muscular effort.
rv1oreover, the action takes place very slowly, one of the two requiring more than ten seconds.
Near the end of his struggle, I step up to him and say, UN ow you can bend it when I snap my
finger 1" As I snap my finger the tension in the arm disappears and he bends it freely. But one of
the six cannot bend his arm at all. He appears to be making a great effort and displays marked
embarrassment at his failure, but succeeds only in moving it awkwardly at the shoulder, the
elbow remaining rigid. After about half a minute of this I say, "Now you can bend it!" and snap
my finger. The arm now begins to bend, but only a little, and apparently requires great effort. I
have to repeat the command emphatically several times before the arm can be moved at the
elbow in a normal manner.
Here we observe the quite remarkable control that one person can exercise over the
behavior of another, even though (25) the latter is evidently quite wide awake and not in any
sense in a trance. The phenomenon is surprising to the average person merely because it is rarely
encountered. This, in turn, is because the particular combination of causal factors necessary to
produce it rarely occurs. Among the factors which were presumably operative to a significant
degree in the above experiment may be mentioned the prolonged voluntary contraction of the
muscles, the fixation of the eyes on the rigid arm of the experimenter, the experimenter's display
of muscular effort, and the vigorous verbal suggestions that the arms should not bend. Lastly,
there is the knowledge and belief on the part of the subjects, obtained from members of previous
classes, that the experimenter could actually produce such effects. This last is the substance of
what is called prestige and is a fairly potent factor, though almost anyone can secure results
measurably similar to those described above by a little aggressive persistence. In this connection
it may be noted that in the demonstration only one of the six showed the effect in a perfect
manner, while two of the subjects showed no discoverable
effect at all.
The same general effect as that shown in Demonstration II may be secured by an
analogous procedure, except that the subjects lock the fingers of the two hands together in front
of the chest. Some subjects cannot get their hands apart even after struggling several minutes.
The experimenter commands one of these, saying, "Now as I make passes over your hands they
will very slowly come apart until the last joints of your middle fingers meet. There they will
stick and you can't pull them any further." This last is especially convincing because, with
merely the ends of the fingers in contact, there is no longer any question of a mechanical
resistance, as there is when the fingers are tightly locked near the knuckle.
LIGHT HYPNOSIS: CATALEPSY OF THE EYELIDS: Demonstration III. We have
here a young man, aged twenty-three years, who is a student of philosophy. Though somewhat
skeptical, he has asked to be hypnotized for the sake of the experience. He is seated in an
ordinary chair near the wall. I display a globular hypnotic crystal (Fig. 3) With my left hand I
hold this about four inches from his eyes and a (26) little above, in such a manner as to put the
eye muscles under a strain during fixation. He is told to find a bright reflection in the crystal, to
look at this steadily, and to think of nothing but sleep. Presently his eyelids begin to droop
slightly. With the fingers of my right hand spread apart, I make downward passes between the
crystal and his eyes, in such a way that his view of the bright spot is interrupted by each finger in
turn. Gradually his lids close, and he appears to be asleep. After two or three minutes I say,
"No.w I shall count three, and when I snap my finger you will wake up." When I snap my finger
he opens his eyes and immediately tells me that he was not asleep at all, but only closed his eyes
because they felt a little tired, and that he heard everything that happened. To prove this he
proceeds to relate accurately everything that took place from the beginning, and assures me that
he could have opened his eyes at any time if he had wished. I suspect, on the basis of previous
observations, that he has been influenced more than he realizes. I therefore repeat the
experiment. When his eyes have closed a second time I ask him if he thinks he could open them
if he wished. He replies in the affirmative. I say, "I don't think so. Try it." He makes the attempt,
raises his eyebrows very high and moves strongly various other muscles of his face, but his lids
remain closed. He admits, with a little chagrin, that he cannot open them. I say, "Now you can,"
and snap my finger. The lids open at once. The subject is again able to give a clear account of
everything that took place while his lids were closed.
Demonstration III shows a subject in a very light stage of hypnosis. It also illustrates the
fact that subjects frequently are (27) not able to judge accurately the degree to which they have
been influenced. They almost invariably underestimate it. The technique employed by the
experimenter for inducing the trance is one of the more common ones in use at the present time.
It will be recalled (p. 10) that the practice of using a bright fixation object for inducing hypnosis
was originated by James Braid.
The particular fixation object utilized in the above demonstration (Fig. 3) is a device
especially constructed for hypnotic purposes. It has a wooden base with a short cylindrical upper
portion. Over this cylinder is pressed, rather tightly, a glass tube about a half-inch in diameter,
the upper end of which has been blown out into a spherical bulb. Before the glass is pressed on
the base there is driven into the center of the cylinder a metal pin to the end of which has been
attached a brightly polished nickel-plated ball. The pin is driven into the wood to such a distance
that the ball will be in the center of the glass sphere when the latter is in position. Lastly there is
placed inside the glass a small quantity of coarse sand, dyed blue. It is not to be supposed that
any parts or characteristics of this crystal possess, in themselves, any hypnotic potency whatever.
It is perhaps possible that the novelty of its appearance may impress na'ive subjects and in that
way -facilitate the hypnotic process. It is difficult to determine any other function of the blue
sand, unless it might be to keep the metal ball polished. The device has one very obvious
advantage, however. Its numerous bright spherical surfaces serve to reflect to the subject's eye
almost any source of strong light in the neighborhood of the experiment, without much regard to
the particular position in which the fixation object may be held. The spherical shape of the
reflecting surfaces allows the reflections, as seen by the subject, to remain steady and relatively
undisturbed, even though the involuntary movements of the experimenter's hand may keep the
crystal in rather constant motion.
PHENOMENA CHARACTERISTIC OF PROFOUND HYPNOSIS:
DEMONSTRATION IV. I am anxious to secure several good subjects for a hypnotic research
project, and call for volunteers from a class of medical students. I direct them to report to the
psychological laboratory at 7:00 P.M. When I arrive I find ten (28) men awaiting me. They are
conversing in subdued undertones. I deliberately keep them waiting for about twenty minutes;
they are then ushered into a room dimly lighted by a single bulb placed in a socket on the wall. I
choose six men from the group and perform an experiment in waking suggestion as described
above under Demonstration II. I carry out the experiment with considerable vigor, and several of
the men are strongly influenced. All are greatly impressed by the result. I then choose as a
subject the man who seemed most responsive to the waking suggestion and place him in a
high-backed Morris chair, facing away from the light. I tell him to loosen his collar and tie, to
lean back in the chair, and to relax all of his muscles. He is evidc::ntly under some emotional
strain, as is indicated clearly by the violent beating of the carotid artery above his collar-bone.
Turning to the others I say, "I shall first describe to you the various stages of going into the
trance so that you may know what to look for. I hold in my hand a powerful hypnotic crystal."
Holding the crystal before my own eyes and simulating the behavior of a subject as realistically
as possible, I continue in a tone which gradually becomes more and more slow, dull, and
monotonous as I go along, "The crystal is held before the subject's eyes like this. He looks at the
bright spot in it steadily and thinks of nothing but sleep. Presently his eyes begin to water and
the lids begin to blink. Soon the lids become very heavy; they begin to droop; and when they
close it is difficult to open them. The lids droop lower and lower, the world seems farther and
farther away, the breath comes slow and deep. Finally the eyeballs roll upward, the lids close
permanently, and the subject is fast asleep." I let my eyes remain closed for a few seconds and
then open them. I shake my head slightly as if I had nearly gone to sleep while giving the
demonstration and were under the necessity of rousing myself. Several of the men show
evidence of drowsiness, including the one I have chosen for a subject. I now step quietly to the
side of the latter and silently hold the crystal before his eyes. I ask him if he sees the bright spot
and he nods his head. He brightens up for a short time, but after about a half-minute his lids
begin to droop slightly, his eyes get watery, and the lids begin to blink. I make downward passes
with my right hand between the crystal and the subject's face. As the eyes begin to close I say in
soothing, barely audible tones, "They are closing now, there they go, there they go, there they
go. Now they are closed and you are falling fast asleep. Your arms are tired, your legs are tired,
your body is tired, your head is falling forward farther, farther, until your chin is resting on your
chest. Now you are sound asleep." As I say the words his head gradually falls forward (29) until
his chin rests upon his chest, his breathing becomes heavy, his eyes are closed, his muscles
relaxed, his body slumped in the chair. Presently I say to him gently, "Are you sound asleep I"
and he nods affirmatively with his head.
I step back from the subject for a moment. As I do so the men call my attention to an
individual sitting over at one side, slumped in his chair with his eyes closed, very much as the
man with whom I have been working. Presently I go to him and wake him in the usual manner.
When I snap my finger he wakes with a little start, rubs his eyes, and cannot tell anything that
has happened for some minutes past. He appears somewhat embarrassed at having fallen into the
trance in this way and explains that he has felt sleepy all evening.
I return to my first subject, who still rests quietly in his chair as I left him. I say to him,
"The library clock is about to strike nine. Listen and when it begins to strike, count the strokes
out loud." I wait some time but the subject says nothing. I say, "Can't you hear it?" He shakes his
head. I say, "Listen closely now. It is about to strike. There it goes. Now count! One ... " He now
begins counting in a slow, droning manner at about the speed at which the clock really strikes.
This is a hypnotically induced hallucination.
I say to him, "You can now open your eyes, though without waking. Mr. X has gone
home. Look around and tell me whether you can see him." The subject gazes around sleepily,
apparently with some care, and replies that he cannot. I te1lhim to count the men present aloud,
pointing to each with his finger. He does so. When he comes to Mr. X he passes him by without
notice, and ends with a sum one less than the number of persons really present. This
phenomenon is sometimes called a negative hallucination.
The subject is told to close his eyes and to go on sleeping. I then take his left hand and
say, "Your arm and hand are becoming rigid and stiff like a piece of wood. They no longer feel
pain or touch. If I prick your hand with a needle you won't feel it at all. Now I am going to prick
you, but you won't feel it." I then prick his hand near the thumb with a sterilized needle. He says
he felt the touch but no pain. I prick the other hand in exactly the same way, and the subject
winces and mutters, "Ouch!" I prick the first hand, this time until a drop of blood appears, but
again no shrinking response can be detected and the subject once more firmly denies feeling any
pain, though he admits feeling a light touch. This phenomenon is known as hypnotic analgesia.
I take two matches, light one very audibly, and say, "Here, hold this lighted match until it
begins to burn your fingers, and then quickly throw it down." He holds out his hand slightly (30)
and I place between his thumb and index finger the end of the match which has not been lighted.
I say, "It's burning close. There! IT'S BURNING YOU!" While holding the match he has
worked it outward so that its tip only is held between the very tips of his thumb and finger. At
my exclamation he suddenly throws the match down with a slight facial wince of pain, and for
some time after continues to rub the end of his finger with his thumb. Upon inquiry he says that
his finger was burned a little but not seriously.
After a minute or so I say to the subject, "Presently I shall wake you. After you have
been awake a short time I shall go over by the table and put one hand in my pocket. When I do
that you will go over and adjust the window-shade. You will not remember that I have told you
to do it, but will think merely that the shade needs straightening." I wake him in the usual
manner. After a few minutes I ask him if he remembers what happened while he was in the
trance and he says he remembers nothing after the first few moments of looking at the crystal
until I snapped my finger. This inability to recall trance events is known as post-hypnotic
amnesia. I notice that he continues to rub the "burned" index finger with his thumb. I ask him
why he does it and he says the finger feels smooth as if it had been burned, but it doesn't exactly
hurt him.
I casually walk over by the table and put a hand in my pocket. The subject looks around
uneasily at the window, fidgets in his chair for a few seconds, then goes over and adjusts the
shade. I say, "Why did you do that?" He answers that he noticed the shade was not straight and
he felt uncomfortable until he had straightened it. This is an illustration of posthypnotic
suggestion.
I hold the crystal again before the subject's eyes and say, "This time you will go to sleep
very quickly." In a few seconds his eyes close and he appears to be in deep sleep. I say, "Can
you remember what happened when you were in the trance before?" He recalls clearly
everything that happened. I then wake him in the usual manner, first telling him that when he
wakes he will feel calm, rested, and generally well.
The main subject in Demonstration IV represents about as profound a trance as is ever
obtained. Such subjects are often called somnambules. The distinguishing mark of good
somnambuIes is their inability to recall what took place in the trance, i.e., they show
post-hypnotic amnesia. Under the favorable conditions described above, three or four out of ten
persons usually will show this amnesia. In addition, four or five others (31) will show one or
more of the phenomena, though without amnesia, while two or so may show nothing more than
a slight drowsiness, or perhaps no influence whatever. It is desirable in such experiments that the
first subject worked with should go into the deep trance. This strongly facilitates the tendency to
positive response on the part of the others. That is why the man showing the strongest influence
in the preliminary experiment on waking suggestion was chosen as the first trance subject. Those
individuals showing strong responsiveness to waking suggestion are almost certain to make good
hypnotic subjects.
POST-HYPNOTIC SUGGESTION AS AN INDIRECT TEST OF
POST-HYPNOTIC AMNESIA: Demonstration V: A youth of eighteen or nineteen years is
brought in by my assistant. He has consented to act as subject in a research project. I stand
before him and look directly into his eyes. As he tilts his head backward to look into my eyes I
observe as usual the sign of considerable emotional disturbance in the beating of his carotid
artery. I explain the symptoms to him very much as was done in Demonstration IV. I direct him
to look steadily into my eyes and to think of nothing but sleep, to relax his muscles all over,
even so much that his knees bend a little and his legs scarcely hold him up. After three or four
minutes his eyes close, his head nods forward, and his breathing becomes heavy. I say, "Now
you are falling toward me, you can't help yourself. There you go, there you go!" He sways
forward two or three times, but each time catches himself just before falling. At last he sways
forward like a log and I catch him when well off his balance. Upon inquiry he states, in a drowsy
tone, that he could not help falling forward but that he isn't sound asleep "because I know
everything that is going on." I suspect that he is mistaken and employ the following objective
test. I give him a post-hypnotic suggestion that after waking he shall pick up and examine a book
on my desk when I sit down in a chair, but that he won't recall anything about why he did it. I
wake him as usual with a snap of my finger. He seems rather dazed at first, so I shake his
shoulder a little and say firmly, "Wake up!" Gradually his appearance and actions become
normal. A few minutes later I sit down in the chair. He casually walks over to my desk, picks up
the book, and after glancing at its title lays it down. I say, "Why did you look at the book?" . He
answers that he just happened to notice it lying there and wondered what it was about.
This demonstration shows a second method of inducing the trance which involves optical
fixation. In the hands of the author this technique has proved perhaps the most effective of any,
provided the subject is shorter in stature than the experimenter, so that the former's eyes must be
turned upward. It also illustrates a second time (Demonstration III) that subjects are very prone
to underestimate the extent to which they have been influenced, and shows perhaps the most
satisfactory technique for determining whether a subject is in the deep trance. If a subject is
asked point-blank upon waking from the trance whether he recalls what took place in it, he is
likely to regard it as a kind of challenge and to make very persistent efforts to do so. Such
attempts sometimes succeed after some minutes. In such cases it is rarely possible afterward to
secure a satisfactory amnesia.
It is quite general for subjects to show a slightly dazed appearance, as observed in this
demonstration, upon first being roused from deep trance. As a matter of fact, subjects do not
really wake instantly, as is sometimes thought by superficial observers of a trance demonstration.
On the contrary, it is evident that the influence of the trance persists for some little time after
they have responded to the command to wake by opening their eyes. This is probably why
subjects can be thrown into a trance very much more readily a minute or so after waking from
one than a day or two after.
In this connection it may be added that the author, in a very extensive experience with
hypnosis, has never encountered any greater difficulty in waking a subject than that described
above. The very widespread belief among laymen that there is danger of not being able to wake
hypnotic subjects, once they are in the trance, is a little difficult to understand. It is possible, of
course, that with amateurs, if both the subject and the experimenter believed this myth, this
belief itself might bring about difficulty in terminating the trance, merely through suggestion. As
a matter of fact, subjects, unless definitely commanded to continue sleeping, will ordinarily
wake spontaneously in a short time if left to themselves. Frequently they wake when left to
themselves for a time even when commanded not to. The real problem in hypnotic
experimentation is to keep the subject in a deep trance, not to wake him. (33)
HYPNOSIS INDUCED BY PURELY SUGGESTIVE MEANS: Demonstration VI.
The subject is a graduate student in psychology, aged about twenty-eight years. He has
volunteered to act as subject for a class demonstration. The subject sits in an ordinary chair at the
front of the classroom with his back turned toward the window. He looks straight before him. 1
stand about two feet from the subject, leaning slightly against a very high desk, and look out the
window. 1 begin talking slowly to the subject, in a quiet, soothing, somewhat monotonous tone,
with pauses of several seconds between each statement. "You are going into the deep trance.
Now you feel drowsy. Sleep is creeping up your arms. Now it is creeping up your legs and your
body. Now it is passing into your brain. Your eyes are drooping lower and lower. Your breath is
becoming slower and deeper, like one asleep. Your head is faIling forward upon your chest.
Now your eyes are closed. You are sound asleep." The subject by this time appears to be
sleeping quietly in his chair. 1 give him the suggestion that, after waking, when 1 go to the
window to look out he shall adjust his tie and then make a speech to the class. I then say that
when 1 tell him to start he is to count slowly aloud from one to ten, and that when he gets to five
he will open his eyes, and that when he gets to ten he will be wide awake and will feel fine. 1
say, "All ready, one .... " He begins counting. At five he opens his eyes and from then on
gradually throws off the signs of sleep until at ten he appears to be wide awake. Presently I go to
the window and look out. The subject casually adjusts his tie but makes no move to address the
class, even though 1 remain at the window for some time.
Demonstration VI illustrates the method of inducing a trance by pure verbal suggestion
and without the aid of optical fixation. This method is associated with the Nancy school. It is
also to be observed that the subject carried out the simple post-hypnotic suggestion, but
characteristically failed to make the speech. It is usually difficult to secure the successful
execution of post-hypnotic suggestions which are at all in serious conflict with the natural
inclination of the subjects.
RAPPORT: Demonstration VII. The subject is a young woman of twenty years. She is
a student of history with a very excellent scholastic record. She has asked one of my assistants to
(34)
device when the person desiring to carry out an experiment or treatment lacks the necessary
prestige or skill to induce the trance the first time, and a more experienced person is available to
perform this service. It may be added that this technique is one of the best means of giving an
inexperienced person practice and the necessary self-confidence in inducing the trance.
RESISTANCE TO HYPNOTIC SUGGESTIONS: Demonstration VIII. We have before us a
young man of nineteen who has volunteered to act as a hypnotic subject. I employ the method
utilizing the crystal, as described in Demonstration III. As the subject sits down in the chair he
remarks that he doesn't think he will go into the trance. I do my best, but each time his lids start
to droop he rouses himself. After about ten minutes of this I abandon the attempt, excuse him,
and begin on another subject who turns out to be fairly responsive. After some minutes I chance
to glance up and observe the first subject, sitting a few feet away, with his arm extended over the
back of the chair and a partially consumed cigarette in his hand. His head is tipped forward. His
eyes are closed, and he appears to be fast asleep. Presently the cigarette drops from his fingers.
This rouses him. He picks up the cigarette and extinguishes it. Upon being questioned by an
acquaintance who has observed his behavior he denies firmly that he has been asleep. He is
unable to relate anything of what took place during the five minutes or so preceding the
dropping of the cigarette.
In this demonstration we see manifested a tendency occasionally observed in the case of
subjects who more or less deliberately resist suggestions directed specifically to them, namely, a
tendency to respond to similar suggestions given to another subject after they have been excused.
It is as if the preliminary direct stimulation had built up a considerable impulse to go into the
trance, but that it had not become manifest until, having been excused and therefore thinking
himself safe, the first subject ceased his resistance. Doubtless the stimulations to the succeeding
subject are also influential, as we have observed in Demonstration IV.
NEGATIVE SUGGESTIBILITY: Demonstration IX. The subject is a young woman
of nineteen. Somewhat against her natural inclination, she has (36) been persuaded to submit to
the hypnotic technique by a friend who wishes to use her as a subject. I try the crystal method of
Demonstration III, but after some minutes only succeed in securing a slight drooping of the lids
and some watering of the eyes. As a last resort I adopt the standing technique of Demonstration
V. I gaze steadily into her eyes for three or four minutes, make passes, and give strong verbal
suggestions. At length her eyes assume a slightly glassy stare and I suspect that she has fallen
under the influence while resisting the specific suggestion to close her eyes. I therefore press the
lids down firmly with my fingers and tell her to let them remain closed. I then tell her, with great
firmness, that she is falling toward me, that she can't help herself. Instead she sways so far
backward as nearly to lose her balance. I now tell her that she is falling over backwards; that she
can't help herself. She at once reverses and sways far forward. These suggestions are repeated
several times with the same results. The subject is told to open her eyes. She says she remembers
everything. She does not know, however, that at the falling suggestions she responded exactly
opposite to what was suggested. After being told, she explains it by saying that she had made up
her mind not to be hypnotized and probably responded negatively as a means of resistance.
The negative reactions of the subject in Demonstration IX represent an exaggerated form
of what probably takes place with a considerable number of subjects who do not go into the
trance, or who show only slight influence. Indeed, it may be doubted whether very many
subjects approach hypnosis without some slight subjective reservations. This may be due in part
to the very general belief among laymen that to be able to go into a hypnotic trance is an
indication of mental inferiority.
As a final view in this introduction to concrete hypnotic phenomena, perhaps we can do
no better than to consider what it feels like to be in a trance.
THE TRANCE EXPERIENCE: Demonstration X. We have here two young women
who have each been put into the deep trance several times by my assistant, Miss C. Miss C. puts
the subject into the trance by a simple command to "go to sleep." In a few seconds the subject
appears to be sound asleep. Miss C. asks the subject to write on a sheet of paper, which is
provided, what it feels like to be in a trance. The subj ect writes as follows: "I am (37) sound
asleep. I feel tired. I am relaxed all over. I can hardly hold the pencil. My legs and feet feel
numb. I can't stand up. I can hardly keep my eyes open. I can't see anything but this paper,
pencil, and my hands. I don't seem to know why I do things, but seem to obey without knowing
why."
 The second subject is now hypnotized and is asked to tell what it feels like to be in a
trance. She relates the following in a slow drowsy tone: "Sort of like ... I don't know ... like
nothingness. Just hearing things but being somewhat powerless, and one does not want to wake
up. You're too comfortable. It's sort of like just before you wake up-sort of dozing. You never
feel like talking."
The subjects who made the above observations, while intelligent and well educated, were
psychologically quite naIve and unsophisticated. Fortunately there exists an account of similar
observations written by Doctor E. Bleuler, Director of the Psychiatric Clinic at Zurich and a
psychiatrist of international reputation. Dr. Bleuler had himself done considerable hypnotic
work. People who have learned and practiced the art of hypnotism are not apt, themselves, to go
into a trance. By a piece of good fortune Dr. Bleuler, after a number of attempts, was finally put
into a trance with Liebeault's method by Professor von Speyer. Since Bleuler had little or no
posthypnotic amnesia, he was able later to publish an exact, and rather detailed, report of his
self-observations "My condition then was that of a pleasant, comfortable rest; it occurred to me
that I was not in the least inclined to alter my position, which, under other circumstances, would
not have been actually comfortable. Psychically I was quite clear, observing myself; my
hypnotist was able to confirm all the objective things, which I told him of, later. My conceived
thoughts were not influenced in a different way to the waking condition during the following
suggestions, but in spite of this the greater part of them were realized. I did not fix my parcicular
attention on the hypnotist, but did so on myself alone.
"My friend placed one of my forearms horizontally in the air, and told me that I could
not put it down. I tried to do this directly afterwards with success, but was prevented from
carrying this out completely by a light touch of his hand and by renewed suggestion. I then felt
my biceps contracting against my will as soon as I attempted to move my arm by means of the
extensor muscles; once, on making a stronger effort to carry out my intention, the contraction of
the, flexors (38) became so energetic that the arm, instead of moving outward as I had intended,
moved backward on the upper arm.
"Then my friend said that my right hand was anesthetic. I thought to myself that he had
made a mistake in this, as it was still too soon for such a suggestion, and when he stated that he
had pricked me on the back of the hand I thought that he was trying to deceive me to make me
more confident. I only felt the touch of a blunt object (I thought that it was the edge of my
watch). On awakening, I was not a little astonished to find that I had been pricked. He did not
succeed in producing real anesthesia; only once when he remarked that the hand was as if it had
gone to sleep I felt a tingling sensation for a .;hort time, and only felt a touch as if through a
thick bandage.
"On the following evening I was hypnotized twice, lying on the sofa, by Dr. von Speyer,
and on the following day once by Professor Forel. The experiments mentioned were repeated
with great ease, and, further, an arm was rendered rigid, and certain acts were required of me.
The suggested analgesia often lasted for such a short time that when other suggestions were
given immediately the pricks, which I had only felt as touches while they were being made,
began to pain during the same hypnosis. Painful stiffness of my legs after a long walk, on the
other hand, disappeared permanently after a few suggestions. When the impossibility of carrying
out a certain movemem was made to me I no longer observed the contractions of the antagonists
so frequently. The power over my will appeared to be interfered with; my muscle would not
contract, notwithstanding all my efforts. In the later suggestions my will had become so
weakened that I no longer innervated at times, contrary to my intentions, because the vain
attempt was too exhausting, or because I did not think for the moment of opposing the
suggestion. When I was required to perform an act I was able to struggle against it for a long
time. At length, however, I carried it out, partly from want of will-power to resist it, just as one
gives in to a reflex which costs a great effort to resist. At other times I felt that the movement
was made without any active taking part of my ego, this being especially marked with
unimportant commands, such as the lifting of a leg. I had the feeling on several occasions of
giving in, in order to please the hypnotist. But since I was still mostly clear enough in such cases
during the carrying out to attempt to resist, the uselessness of the latter convinced me of the
incorrectness of my views. I felt every new suggestion, even the command to desist in an act
which I had begun, at first to be unpleasant, and this made the resisting easier for me. I was able
to oppose the order to fetch something outside the room with comparative ease, but could not do
so when the act was divided up into its (39) component parts-e.g., when I received the
suggestion to move one leg, then the other, and so on until the act was accomplished. "I was able
to resist the carrying out of a post-hypnotic suggestion. However, this cost me considerable
trouble, and if I forgot for an instant during talking my resolve not to take any notice of the
plate, which I was supposed to place somewhere else, I suddenly found myself fixing this object
with my eyes. The thought of what I had been ordered to do worried me until I went to sleep,
and when I was in bed I nearly got up again to carry it out, merely to ease my mind. However, I
soon fell asleep, and the action of the suggestion was then lost." (40)

11. CH 3: EXPERIMENTAL PHENOMENA OF DIRECT WAKING SUGGESTION:

CLARK HULL: HYPNOSIS AND SUGGESTIBILITY: APPLETON-CENTURY-CROFTS:
NY: COPYRIGHT 1933: 1961
 In the preceding pages there have been described a variety of the manifestations of
hypnosis and related phenomena as they appear to ordinary observation. These examples should
give a useful preliminary orientation. But mere unaided observation does not provide an adequate
basis for scientific knowledge. We shall accordingly proceed to the description of a number of
simple experiments. These will be concerned with direct waking suggestion, which marks the
transition from the more familiar forms of human behavior to the less commonly observed ones
known as hypnotic. It is hoped that this material will serve to bridge the gap between ordinary
observation aNd the more precise forms of experimentation later to be presented. It should also
serve to divest hypnosis ot the air of mystery which usually surrounds it, by showing it to be
entirely of a piece with everyday human nature.
UNCONSCIOUS MIMICRY: Everyone has at one time or another caught himself
unintentionally performing the actions which he is observing in some other person. When we
watch a contortionist going through his straining activities on the stage we tend automatically to
take similar postures. When we watch a person trying very hard to reach something, we tend
unconsciously to reach. Even the lowly chimpanzee seems to show this tendency, as is indicated
by a remarkable photograph published by Kohler (7, 142). The phenomenon has been called
unconscious imitation, ideomotor action, and empathy. The writer, wishing to secure a graphic
record of this fairly well known but little studied tendency, carried out the following experiment.
Since the process is definitely not voluntary and the (41) imitation not conscious when being
performed, it was necessary to employ a somewhat adroit technique. A young man, unfamiliar
with the purpose of the experiment, was asked to see how still he could stand with eyes closed.
Under the pretext of placing him in the right position, a pin with a tiny hook at its end was deftly
caught into the collar of his coat at the back. From this pin a black thread ran backward through a
black screen to a sensitive recording apparatus. This apparatus traced faithfully on smoked paper
at one-third their actual magnitude the subject's forward and backward postural movements, but
quite without his knowledge. After he had been standing thus for about four minutes an assistant
came apologetically into the room and inquired of the experimenter whether she could take her
"test" at once, as she had an appointment. The subject was asked if he was willing to wait during
the interruption. He readily consented and was permitted to open his eyes, but was cautioned not
to move from his position. The assistant then took up a position in clear view of the subject, and
just far enough from the wall so that she could not possibly touch it. She then proceeded to reach
for the wall with all her might, giving free expression to her efforts by facial grimaces and
otherwise. Meanwhile, the unconsciously mimetic postural movements of the unsuspecting
subject were being recorded in detail. Sometimes subjects see through the ruse and then it
commonly loses much of its effectiveness. But with a naIve and unsuspicious subject there can
usually be seen in the tracings evidence of a tendency to reproduce in his own body the gross
postural movements of the person under observation. If the assistant reaches far forward, the
subject sways forward slightly; if the assistant bends backward, the subject unconsciously sways
backward, though not so far.
A tracing obtained by means of such an experiment is reproduced. From A to B the
assistant (Miss H.) stood quietly in position as a control to establish the normal postural level of
the subject. The small up-and-down oscillations of the line are caused by spontaneous postural
tremors common to all normal persons. At B the assistant began very gradually to reach for the
wall. At point C, when her straining began to be acute, the subject began to respond by swaying
forward, which is shown by the marked upward slope of the (43) line. This continued for about
half a minute, at the end of which time the subject had moved forward something over one and
one-half inches. The stimulus seems at this point to have lost its potency, as he returned to his
normal position in the course of the following ten seconds, even though the assistant continued to
reach as before.
UNCONSCIOUS MOVEMENT AND THE IMAGINATION: A second series of
experiments involving a somewhat different psychological process was carried out by means of
the apparatus and general procedure just described, but with this difference: After the nai've
subject had been placed in position, with his eyes closed and the thread attached to his collar, he
was told to imagine as vividly as possible that he was leaning far forward, but that he must not
perform the movements imagined. At the beginning of the reproduced record, (44) the subject
was standing in a normal manner, with lids closed. At B he had just been told to imagine himself
bending far backward but not to move. Almost immediately at the conclusion of the instruction
his body began to move backward, as is shown by the fall in the line. In less than fifteen seconds
he had swayed backward some two and one-half inches and appeared to be nearly balanced on his
heels, and consequently unable to sway any farther without falling. At 5 he was commanded to
stop the imagination. Recovery began at once and was complete in some eight seconds. After
about three-quarters of a minute of rest the subject was told to imagine himself leaning far
forward but not to move. The tracing may be seen to move sharply upward at once.
11easurements made of the record show that in the course of about thirty seconds the subject had
swayed forward over four inches. After this maximal position had been maintained about seven
seconds, the command to stop imagining was given. The subject returned to his former posture in
approximately five seconds. After a few coarse oscillations the posture became stabilized at
nearly its former level. When questioned at the conclusion of the experiment, the (45) subject
reported that he had definitely inhibited all tendency to execute the movements imagined. The
fact that he had not succeeded in this, contrary to his belief, indicates that he had no appreciation
of the movements which he had been making, and therefore that they could hardly have been of
conscious or voluntary origin.
The results of this experiment are of considerable interest as revealing the nature of the
imagination as a physical process. While by no means conclusive evidence, it is strikingly ill
harmony with the view that imagination may be essentially the feeble execution of acts
customarily performed in connection with the situation being imagined. We shall have more to
say concerning this interesting possibility when we come to consider the nature of hypnosis and
suggestibility. It will suffice for the present to point out that the bodies of most normal persons
tend, under favorable conditions, rather uniformly to execute without voluntary or conscious
intent acts observed attentively or merely thought of. Under the conditions of our experiment the
latter appears to be distinctly more potent in this respect than the former. Very likely the lesser
degree of responsiveness in the mimetic experiment is due in large part to the fact that the eyes
are open, which naturally gives the subject clear indications of his movements and thus leads to
compensatory impulses. The tendency to movement is so marked in the case of the imagination
that it may easily be observed with the unaided eye quite without the aid of apparatus in many
subjects. This may be verified by anyone in a few minutes.
MOVEMENT AND WHISPERED AUTOSUGGESTION: Proceeding a step nearer to
our goal, let us consider a third variant of the experimental technique described above. In this
case the subject, standing with eyes closed, is merely told to whisper the following words
continuously, so that they will be clearly audible to himself: "I am swaying forward. I can't help
myself. I am swaying forward. Forward. Forward. Farther yet. I am leaning forward. More yet,
more, more, still more. Forward," etc. This may be continued as long as desired, though positive
results usually appear within thirty seconds. A (46) typical record secured from a thoroughly
normal subject is reproduced as Fig. 7. From A to B the subject was standing quietly with eyes
closed. At B he began whispering to himself the suggestion that he was falling backward. The
tracing shows that almost at once the subject began to sway in a backward direction. After
forty-eight seconds he was far back on his heels and seemed in danger of falling, so the command
"Stop" was given at C. Despite this command, the backward movement continued for some six
seconds. Measurement of the original record shows the movement to have been fully three and
one-half inches in extent. The return to normal posture was typically more prompt than the
backward movement, as is indicated by the greater steepness of the curve as it rises. The recovery
required only twenty-five seconds. It is noteworthy that this correctional movement shows a
tendency to persistence very much like the continuation of the backward movement after the stop
signal. The over-correction in this (47) case amounted to nearly an inch. This over-compensation
was corrected after some twenty seconds.
Between C and D the subject was permitted to stand quietly. At D he began to whisper to
himself suggestions to fall forward. From this point the curve may be seen gradually to rise. At
the end of a minute the subject was far forward, teetering on his toes and evidently in danger of
falling. Measurement of the tracing shows that he had swayed forward at the level of his
shoulders a total of eight and one-half inches. At E the stop signal was given. Following a brief
after-discharge, the tracing shows a prompt return to normal; the characteristic
over-compensation is seen to follow quickly with a falling of the tracing below the normal level.
This, in turn, persists for nearly a minute, after which normal posture is finally attained.
The behavior obtained in the experiment just described resembles in a remarkable manner
that secured from the second, as may be seen by comparing. This striking similarity in behavioral
pattern is particularly significant because the procedure in the second experiment quite clearly
shows the reaction in that case to have been an imagination phenomenon, whereas the technique
of the third experiment clearly involves suggestion. If the imagination experiment described
above had not been presented here in the midst of a discussion of hypnosis, probably few readers
would think of associating it with suggestion. In the third experiment, on the other hand, the
stimuli which immediately precede the postural movements and which presumably are effective
in evoking them are definitely symbolic; they mean the reactions which they ,"voke, as contrasted
with the situation in the mimetic experiment, where, in an important sense, the stimulus
resembles the reactions evoked by it. This distinction is of considerable significance as a means
of differentiating suggestion from related phenomena. It may be noted, further, that the symbolic
stimuli (words) arise from the activity of the subject's own body; the suggestion is therefore
self-suggestion-i.e., autosuggestion. We have before us, then, a series of experiments which by
almost insensible gradations pass from unconscious imitation through imagination into
autosuggestion, the latter differing in no significant particular from what M. Coue (48) recently
exploited in such a spectacular manner. In the experiment just described we have merely reduced
autosuggestion to an experimental and quantitative basis.
IMPERSONAL HETEROSUGGESTION: We may now consider a fourth variant of
the familiar postural technique in our approach to the phenomena of suggestion. The stimulus in
this case is delivered by a phonographic apparatus specially constructed for the purpose. It
consists of an electrically driven phonograph which (48) is provided with an electrical pick-up, a
seven-tube table-model radio set, D, for amplifying the sound, and a dynamic loudspeaker, A,
which has been mounted on a heavy baffle-board, B. For the production of the sound there is
placed on the phonograph turn-table an aluminum disk upon which have been recorded waking
postural suggestions. These various parts are mounted appropriately on a tall adding-machine
carriage provided with large rubber-tired casters. The center of the loud-speaker is fifty-three
inches from the floor and at such a height that its tones come from but a few inches lower than
the mouth of a person of average height.
In a typical experiment with this apparatus the subject was placed in the usual position,
with his eyes closed, before the postural recording mechanism, the thread as usual being hooked
to his collar. The phonographic apparatus was then quietly rolled up to a position such that the
loud-speaker was about eighteen inches in front of the subject's face. Meanwhile the experimenter
talked quietly to the subject about where and how he should stand. Silence was maintained for a
short time while a normal or control postural tracing was taken. At the end of this period, the
experimenter closed a silent switch and a few seconds later a voice, slightly altered yet easily
recognizable, spoke distinctly from the loud speaker as follows: "Now you will begin to feel
yourself falling forward. You are beginning to fall forward. You are falling forward. You feel
(50) yourself falling forward a little more; a little more. You are falling fonyard. You are falling
forward over on to your toes. You are falling forward. Falling. Falling. A little more. A little
more. A little more. You are falling forward. Forward. Forward on to your toes. Forward, a little
more. A little more. More. You are falling over now. You are falling. You are falling. Falling
over. Falling. Falling. A little more. You can't help yourself. You are falling forward," etc.
The normal postural position is shown by that section of the record (51) begins to show an
upward tendency, the exact moment being somewhat obscured by the postural tremors. From this
point the forward movement was slow but continuous, except for the somewhat exaggerated
postural tremors, as long as the suggestions continued. At B the phonograph was turned off.
Almost at once recovery began and proceeded rapidly, terminating with the usual
over-compensation and gradual secondary recovery.
INDIRECT HETEROSUGGESTION: A fifth variant of this type of suggestibility is
indirect heterosuggestion. This may be illustrated by an experiment utilizing what is known as
Chevreul's pendulum. A subject is asked to hold in his extended hand a thread about a foot in
length, at the other end of which is suspended a small iron bob. Immediately beneath the
pendulum is a short steel bar. The subject may be told that the magnetic properties of the bar are
such that the pendulum will gradually begin to swing in the direction of the lines of force running
from one end of the bar to the other. When held by naive, susceptible subjects, the pendulum will
presently begin to swing in the direction suggested. By rotating the bar the pendulum may be
made to swing in a forward backward direction or from side to side at will. G. H. Estabrooks has
utilized this general technique as the basis for a suggestion test to be used with children. The bob
of the pendulum was suspended in the center of an empty glass tumbler. The score was the
number of seconds before the bob swung far enough to strike the glass.
PERSONAL DIRECT HETEROSUGGESTION: The sixth and last type in this series
is personal direct heterosuggestion. This is exactly like the experiment in which the suggestions
were delivered mechanically by means of the phonograph except that in the present case they are
spoken by the experimenter standing directly in front of the subject. The latter is told over and
over in various ways that he is swaying forward, that he cannot help himself, and so on, as
already described. Since this form of suggestibility is more closely related to hypnosis than those
transitional forms (52) previously discussed, a somewhat more detailed account will be given of
the behavior elicited by it.
The most characteristic and striking reaction to this form of suggestion is the subject's
gradual swaying forward, with occasional transitory reversals, until at last he is completely off his
balance and is caught by the experimenter to prevent his falling or stumbling forward.
Many persons, however, appear to be nearly or quite unresponsive to such suggestions.
The record of such a subject is shown in Figure 12. Despite the fact that suggestions to this
subject were continued for three minutes, no definite forward tendency can be discovered in the
record.
NEGATIVE SUGGESTIBILITY: A third mode of responding to direct personal
heterosuggestion is by a negative reaction. It has long been known that certain personalities are
negative-at least under certain conditions. When the writer first began this form of
experimentation, (54) it was with the expectation of finding perhaps as many as 10 per cent of the
population showing such reactions. As a matter of fact, it is distinctly uncommon to find a normal
subject who shows this in a clear form uncomplicated by positive reactions. Such individuals
occasionally are found, however, and a record displaying this kind of behavior. The subject. a
young woman, reported after the experiment that she thought an attempt was being made to
hypnotize her and that she made a strong resolution not to let this occur. Her verbal report
following the experiment indicated that she had no realization whatever of her negative reactions.
She supposed that she had merely resisted successfully the suggestions given her.
TRANSITORY INITIAL NEGATIVISM: There is a sense, however, in which the
proportion of negatively reacting :;ubjects in this experiment is somewhat greater (55) than the
preceding paragraph might imply. It is quite common for records to show a fairly marked
depression immediately after the beginning of "forward" suggestions. This soon gives place to an
extended rise 'which not infrequently terminates in the subject's actually falling forward. It is
difficult to be certain of the genuineness of this apparent preliminary negative reaction in most
cases because of the presence of the postural tremors which frequently attain a size of comparable
(56) magnitude as the sharp dip which immediately follows the beginning of the suggestion and
immediately precedes the sharp rise at S illustrates this difficulty of interpretation.
Fortunately it is possible to obtain by artificial means records which show the tendency in
a quite unmistakable manner. The method is to instruct a subject who normally shows a strong
positive reaction to resist as strongly as possible the suggestions which are about to be given him.
The first part of the record shuws the normal posture. At RES the subject was instructed to resist
the suggestions which were about to be given. After a half-minute, "forward" suggestions were
begun at F. Immediately after this the subject began a gradual backward movement which
reached its maximum twenty-three seconds later. 1tleasurement of the original record shows this
movement to have amounted to approximately three inches. At this point, however, the positive
suggestions began to triumph. During the next twenty-two seconds the subject swayed forward
fully thirteen inches, when the suggestions were terminated.
THE SUGGESTION OF ARM MOVEMENTS: The matter of transitory initial negative
response to suggestions may be investigated under somewhat more favorable circumstances by
substituting, for the postural movements of the body as a whole, swaying movements of the
horizontally extended arm. In order to minimize postural tremors, the subject sits in an ordinary
chair which furnishes steadying support for the back and shoulders. Fatigue of the extended arm
is largely avoided by supporting it on a freely swinging platform of light wood which is
suspended from a pulley attached to the ceiling. The height of the platform is nicely adjustable to
the height of each subject's arm by means of a hook at the end of the portion of chain descending
from the pulley, which can be inserted into any desired link of the ascending portion. The
platform is shaped roughly like a sector cut from a circle of thirty-inch radius. The thread
connecting the platform to the recording apparatus is attached to the front edge of the sheet-metal
plate which follows the curve of the circle. As the (57) arm moves forward or backward, the
thread follows this arc and at all times maintains a distance closely approximating thirty inches
from the shoulder joint, regardless of the position of the arm or its length. That the arm technique
and accessory apparatus greatly reduce the amplitude of involuntary tremor may be seen by
comparing the initial or control portions, which 'were made by the same subject. Because of this
fact, the preliminary negative response, even though scarcely more than an inch in amplitude, is
clearly discernible.
It can also be observed that after the suggestions had begun there was a decided increase
in the oscillations, which are (58) clearly distinguishable from ordinary tremors. It is likely that
these oscillations are the result of alternate but partial triumphs of the opposing impulses. This
interpretation is corroborated by the fact that the subject appeared to be making powerful efforts
to resist the suggestion; the manifestations took the form of general muscular contractions which
were especially noticeable on the head, face, and neck. The subject later confirmed this
impression verbally. Moreover, he supposed that his efforts had been almost wholly successful
and that his hand had moved not more than three inches, and professed great astonishment when
informed that it had actually moved over ten inches.
POSTURAL TREMORS AND RESISTANCE TO SUGGESTION: The gross tremors
characteristic of the standing postural technique may, however, be made to serve a useful
purpose. While some subjects resist suggestion by making movements in the opposite direction,
there is reason to believe that a much more common type of resistance is associated with a
reduction in the aggregate amount of postural tremor per unit time. It would be difficult to
establish such a relationship by a mere inspection of the postural tracings. Fortunately, it has
become possible recently to summate the amount of postural tremor with precision by means of
the linear oscillometer. The amount of tremor in oscillometer units for each forty-second part of
the record is inscribed beneath the section in question.
The evidence in this matter was encountered as an incidental observation made during an
investigation of the influence of the drug scopolamine hydrobromide upon responsiveness to
waking postural suggestion by Baernstein. The experiment involved, first, a period of
heterosuggestion, then one of autosuggestion, then one of heterosuggestion with preliminary
instructions to resist, and, finally, a period of ordinary heteros)lggestion as at first. A period of
normal undisturbed tremor tracing was taken at the beginning and at the end of the experiment, as
well as between each of the suggestion periods. It (59) happened that of Baernstein's nineteen
subjects eight proved positively responsive to suggestion and eleven showed themselves entirely
resistant. The shifts in the amount of tremor for the two groups of subjects under the several
experimental conditions are shown in Figure 18. There it may be seen that the resistant subjects at
every suggestion test showed a uniform and decided tendency to reduction of postural tremors
below the amount found during the control periods preceding and following. The one occasion
where they were specifically instructed to resist differs in no respect from the other three. (60)
It is significant that the suggestible group resembles the resistant group by showing a
reduction in the amount of tremor during the test in which preliminary instructions to resist were
given. On the remaining three tests there are, however, rises in the curve; these latter may be
disregarded since they presumably represent distortions due to the fact that the forward postural
movements in response to the suggestions were measured along with the tremors. While needing
specific experimental confirmation, these results indicate rather strongly that the resistant subjects
were not merely neutral or passive to the suggestions but were actively resistant, just as were the
suggestible group on the occasion when they were instructed by the experimenter to resist. A
plausible hypothesis to explain the tendency for a voluntary resistance to suggestion to result in a
concomitant reduction in postural tremor would be that the volitional effort involves a general
increase in the tonicity of the postural muscles, which incidentally reduces the amount of postural
tremor. It would seem that the tendency revealed by this experiment, especially because of its
immunity to (61)voluntary compensation, might be made the basis of a useful psychological test
capable of revealing a significant character trait.
SUMMARY AND CONCLUSIONS: We have seen that it is possible by means of
rather simple apparatus and experimental procedures to record with precision responses to various
levels of direct and indirect prestige suggestions. Experiments of this kind yield clear evidence of
unconscious mimicry when the subject is observing another person in a straining posture. Similar
but more marked responses are made when a subject imagines himself taking such a posture,
when whispering to himself that he is taking the posture, when receiving such suggestions from a
phonograph or when receiving them from some other person standing near by. The tracings of the
above reactions are so strikingly similar that they strongly confirm the general a priori probability
that all are of essentially the same nature. This evidence supports the hypothesis long held that
ideomotor action is the basis of prestige suggestion.
The typical reaction to direct suggestion to fall forward when standing is as follows: At
first there is a period of latency during which the subject makes no response. Frequently during
this phase there may be a small but probably significant negative reaction. With continued
suggestion this usually gives place to a clear, positive reaction. At the beginning the positive
reaction is very slow, but as suggestion continues it gradually gains speed until the subject loses
balance. Suggested reactions show a kind of after-discharge; motion does not cease abruptly
when the suggestions end, but terminates gradually. It is very rare to find normal subjects in a
normal environment who show continued negative reaction in the face of persistent repetition of
the suggestion. Some subjects, however, seem to show no reaction of any kind. Careful
measurements indicate, nevertheless, that such subjects as a group markedly reduce the gross
amount of their postural tremors when suggestions are given them. Exactly the same phenomenon
is displayed by normally suggestible subjects when they are specifically instructed to resist. This
indicates the probability that all normal subjects who do not respond positively to direct prestige
(62) suggestion actively resist them, sometimes by reacting in the opposite sense, but more
commonly by a general volountary tonickity of all muscles involved in the act being suggested.
(63)

12. CH 4: SOME RELATIONSHIPS OF EXPERIMENTAL HYPNOSIS AND

SUGGESTIBILITY:
CLARK HULL: HYPNOSIS AND SUGGESTIBILITY: APPLETON-CENTURY-CROFTS:
NY: COPYRIGHT 1933: 1961
HOW is suggestibility distributed in the ordinary population? Do people who manifest a
given degree of one type of suggestibility display the same degree of other forms of
suggestibility: Is suggestibility a single unitary process, or is it merely an aggregation of
processes? How does suggestibility vary with sex: Are women more suggestible than men? How
is it related to age: Are children more suggestible than adults? How is it related to intelligence: Is
suggestibility a mark of a strong or a weak mind? How is it related to personality: Is it a mark of
a weak will or character? How does it vary with respect to mental pathology: Does it indicate a
tendency to hysteria, as has been so widely believed? How is it related to delinquency: Is
negativism, for example, associated with antisocial behavior? How is it related to the action of
drugs: Do certain drugs render persons more suggestible and hence more easily influenced by
others? While the evidence on none of these questions is as full and conclusive as we should like
it to be, there is now available a certain amount of fairly scientific data on most of them. It will
be our task in the present chapter to piece this evidence together as best we may, and to try to
secure from it a coherent picture of suggestion in its relations to other important variables.
THE DISTRIBUTION OF SUSCEPTIBILITY TO DIRECT WAKING
SUGGESTION: How is the susceptibility to direct waking suggestion distributed among the
individuals of the population? To the nontechnical reader it may be explained that most
psychological (as well as biological) traits show a bell-shaped distribution. In studying
distributions a graph is usually plotted in which (64) horizontal distances represent different
degrees of the trait on some convenient scale, and vertical distances represent the number of
individuals possessing each degree of it. An idealized diagram of this kind is shown in Figure 19,
in which the general bell shape of the contour of the distribution is evident. This means, in plain
language, that in most traits such as size, weight, strength of grip, speed of tapping, jumping
ability, and ability to learn typing, drawing, music, reading, arithmetic, Latin, calculus, and
innumerable other things, the great mass of the ordinary population clusters closely around the
middle or average value for each trait. But as values depart from this central tendency in either
direction the number of individuals at each interval grows less and less until at last there will be
found values sufficiently remote from the group average that no individuals will appear at these
points at all. While it is not uncommon for distributions to be somewhat asymmetrical, i.e., to tail
off farther on one side than on the other, it is very rare indeed for distributions in this field to fail
to show the general bell-shaped picture if plotted on a true linear scale.
An examination of a number of distributions of subjects with respect to susceptibility to
suggestion of various kinds shows a somewhat greater tendency to nonconformity with the above
principle than is usually encountered. This dot diagram gives the distribution of responsiveness to
direct postural suggestion in the (65) waking state of sixty-four students at the University of
Wisconsin. Responsiveness was measured by the amount of postural movement induced by
verbal suggestion and recorded by the apparatus shown in Figure 4. This distribution departs
from the usual form chiefly in its disproportionate number of zero and near-zero values. It is
possible that if the method of measuring susceptibility had been adequate to indicate different
degrees of actual resistance to suggestions beyond the mere failure to move forward, this atypical
heaping-up of cases might have been distributed a considerable distance farther to the left and so
have yielded a fairly conventional distribution. In short, it is possible that the atypical distribution
obtained may reflect nothing more than the imperfection of the scale on which it is plotted.
A still more atypical distribution of the degree of susceptibility to suggestion has been
reported by Aveling and Hargreaves. (66) In 1921 these writers carried out an experiment on two
groups of English school children ranging around twelve and thirteen years of age, and
comprising something over a hundred subjects in all. Seven suggestion tests were employed, of
which two, a test of "hand rigidity by suggestion" and one of "hand levitation by suggestion,"
were evidently of the same general nature as the standing postural movement and the arm
movement techniques described above. The procedure with the latter test is typical of both.
Aveling and Hargreaves's description is as follows:
"It was of a similar general character to test No. I (hand rigidity). A preliminary
instruction was given as follows: 'Place your hand on the table so.' (Subject, facing experimenter,
placed his right hand, palm down, upon the table.) 'I am going to lay my hand lightly on top of
yours; and I want you to pay very special attention to the feelings you get from your hand. You
need not tell me anything about them unless I ask you. I just want you to get the feel of them.'
The experimenter then laid his hand firmly but lightly upon that of the subject, so that the tips of
his fingers rested upon its back near the wrist. He continued the instruction : 'You will feel your
hand growing lighter, as if it was losing weight, and not resting on the table. It gets lighter still
now, as if it had no weight at all. Do you feel it:' If the subject answered 'No,' the instruction was
continued in the same sense; if 'Yes' the experimenter almost imperceptibly raised his hand,
relaxing the very slight pressure and at the same time saying: 'Good! It is quite light now. It has
no weight at all. It is leaving the table altogether -coming right off it and rising in the air.'
Reaction times were taken with the stop-watch, from the beginning of the instruction after.”
Aveling and Hargreaves scored the responoes on a crude scale based on the rough
observable states of the child's hand after the experiment:
"1. Hand completely off the table (an inch or so) pushing experimenter's hand up. (Score, + 2)
"2. Hand just off the table (less than one-half inch). (Score, + I)
"3. Part of hand lifted but some other part still on the table. (Score, - I)
"4. Hand entirely resting on table. (Score, - 2)"
The test was administered twice to each subject and the two values thus obtained were
combined to make a final score. The (67) distribution in terms of these scores of ninety-four
children obtained by combining two of Aveling and Hargreaves's groups is shown in Figure 21. A
more complete departure from the usual distribution could hardly be imagined. Aveling and
Hargreaves are inclined to consider the U-shaped distribution as characteristic of this form of
suggestibility, and as differentiating it from the indirect or non-prestige forms of suggestibility
which display the usual or bell-shaped distribution.
It is important in this connection to note that Aveling and Hargreaves's four-step scale
given above, while apparently based on intervals of one unit, has at the middle an interval of two
units, that from + I to - I where a step marked zero ought logically to appear. It is this which
brings about the somewhat paradoxical result of a nine-step range of final scores (68) (+ 4, + 3, +
2, + I, 0, - I, - 2, - 3, - 4) from combining two scores with an apparent total range of only four
steps each. Our chief concern, however, is to point out that this peculiar scoring system results in
an artificial reduction in the number of individuals who would normally fall at steps + 2, + I, - I,
and - 2 of the final scale, i.e., in the middle range where the greatest number might normally be
expected to fall if the usual biological principle of distribution were operative. This fact would
obviously tend to diminish the middle region of the distribution shown in Figure 21, and thus to
explain in part its paradoxical form. It is hardly credible, however, that this factor alone could
have been entirely responsible for the profound deviation of Figure 21 from the conventional
distribution.
Estabrooks, taking Aveling and Hargreaves's work as a point of departure, administered
four suggestibility tests and examined the responsiveness to mildly painful stimuli, and threats of
such stimuli, as indicated by the galvanic skin reaction. Two of the tests are of special interest
because they involved rather strong elements of personal heterosuggestion, though neither,
apparently, was so direct and persistent in nature as the two of Aveling and Hargreaves
mentioned above. The apparatus of Estabrooks's "electricity test" consisted of a medical shocking
induction coil with appropriate electrodes and other electrical devices arranged conspicuously on
a table. The child touched the electrodes and was led to expect that he would receive a faint
shock. If he reported feeling a shock before one was actually given, he was considered to have
shown an indication of sensory suggestibility. The second test was based on a modification of
Chevreul's pendulum, the general nature of which has been described. A rather strong but indirect
personal suggestion was given that the bob would swing. Marked swinging of the pendulum was
taken as an indication of motor suggestibility. Estabrooks publishes graphs showing that both
these tests yield fairly marked U-shaped distributions. These results are regarded by him as
confirming the U-shaped distribution of susceptibility to direct or prestige suggestion reported by
Aveling and Hargreaves. (69)
THE DISTRIBUTION OF SUSCEPTIBILITY TO HYPNOTIZATION: Turning to
susceptibility of a definitely hypnotic variety we find that here also the distributions tend, at least
superficially, to be atypical. Davis, in an attempt to measure susceptibility to hypnosis, arranged
thirty symptoms or signs of the hypnotic trance in an elaborate series or scale such that each item
of the scoring system was supposed to indicate a greater degree of susceptibility than the
preceding one. It is stated by Davis and Husband that "it was only very rarely that the more
difficult suggestions were successful when the simpler ones had failed." These thirty symptoms
were divided into four more or less arbitrary groups, and were given names which have become
current in the hypnotic literature as degrees of hypnotic influence. A fifth group was made up of
those subjects who showed no hypnotic symptoms whatever.
[HYPNOTIC SUSCEPTIBILITY SCORING SYSTEM (DAVIS AND HUSBAND)
Depth Score Objective symptoms
Insusceptible 0
Hypnodial 1
2 Relaxation
3 Fluttering of the eyes
4 Closing of eyes
5 Complete physical relaxation
Light trance 6 Catalepsy of eyes
7 Limb catalepsy
10 Rigid catalepsy
11 Anesthesia (glove)
Medium 13 Partial amnesai
15 Post-hypnotic anaesthesia
17 Personality changes
18 Simple post-hypnotic suggestions
20 Kinesthetic delusions; complete amnesia
Somnambulistic
trance 21 Ability to open eyes without affecting trance
23 Bizarre post-hypnotic suggestions
25 Complete somnambulism
 26 Positive visual hallucinations, posthypnotic
27 Positive auditory hallucinations, post-hypnotic
28 Systematized post-hypnotic amnesias
29 Negative auditory hallucinations
30 Negative visual hallucinations; hyperaesthesias (70)]
Davis and Husband publish a distribution of fifty-five male and female university students
who were scored on the above scale of hypnotic susceptibility. This is reproduced as the first
portion. of Table 1. While presenting marked deviations from the "normal" or bell-shaped
arrangement, as do the results of Life, these data give little indication of the V-shaped distribution
of Aveling and Hargreaves.
A second and closely related study of hypnotic susceptibility is reported by Barry,
Mackinnon, and Murray. They tested seventy-three university students for susceptibility to
hypnosis, and report their scores on a five-point scale. The scale was constructed as follows:
"After the subject had become thoroughly relaxed on the couch and his eyes closed in the guise of
sleep, the following suggestions were given: (i) you cannot open your eyes; (ii) you cannot raise
your arm from the couch; and after (71) straightening the subject's arm in a vertical position, (iii)
you cannot bend your arm; and after interlocking the fingers of both hands, (iv) you cannot
separate your fingers; and finally (v) you cannot speak your name. Before waking the subject,
amnesia was suggested.
"0 = No suggestions carried out. No tendency ;1.t all for them to be carried out.
I = No suggestions carried out, but clear evidence of a difficulty or heSitancy in surmounting
them.
1.5 = One suggestion carried out.
2 = Two or three suggestions carried out.
2.5 = Four suggestions carried out.
3 = All suggestions carried out.
"Amnesia
"0 = No loss of memory and no difficulty in recall. 0.5 = Difficulty in recall, cloudiness, but final
memory.
I = Partial loss of memory.
2 = Complete or almost complete loss of memory.
"As a measure of hypnotizability (Hypnotic Index), the ratings for suggestibility and
amnesia were added together for each subject. Consequently
the hypnotic index might vary from o (no hypnotizability) to 5 (complete hypnotizability)."
The distribution thus obtained is given as the second or "Harvard group" in Table 1.
Since the Wisconsin and the Harvard study each has five steps in its scale, we have ventured to
combine them in the hope that the deviations from a true scale in one might offset similar
deviations in the other, and that the pooled results might yield a truer picture of the general
situation. Taken at face value, i.e., assuming the scale upon which the distribution has been based
to be accurate and truly linear, these results would seem to indicate that susceptibility to hypnotic
suggestion is fairly evenly distributed over the range from no susceptibility whatever to the most
profound susceptibility.
CLINICAL OBSERVATIONS OF HYPNOTIC SUSCEPTIBILITY: Various writers,
usually clinical workers, have in the past reported the percentages of individuals who were
susceptible (72) to the trance. These figures vary widely. Obvious causes for this variation are
such factors as the personality of the experimenter, his technique, his reputation for skill in
hypnotizing, and the basis of selection of the subjects. Doubtless a factor of no inconsiderable
significance is the circumstance that the depth of the hypnotic trance varies over a wide range and
by quite imperceptible degrees, coupled with the fact that up to the present time there has been no
mutually recognized scale by which these degrees might be indicated. This difficulty is
aggravated where, as often happens, writers merely divide their subjects into two groups-those
who are susceptible and those who are not. Bramwell has gathered together reports of hypnotic
susceptibility from numerous authors, a number of whom have attempted to indicate the several
degrees of susceptibility. These latter have been assembled in Table 2, and the mean percentage
values for all five studies are shown in the last column. While the scale is very coarse, the mean
values approach quite closely the theoretical or normal distribution.
In summary of this phase of the matter, then, it may be said that in the present state of the
experimental evidence it is doubtful whether we are justified in regarding responsiveness to direct
verbal suggestion as an exception to the general law of "normal" distribution. It seems much more
probable that the cases of apparent deviations from the bell-shaped arrangement are due to
defective measuring instruments.
It may be added in concluding our discussion of this phase of the problem that Davis and
Husband find no tendency to a (73) lack of continuity in the distribution of their subjects with
respect to this trait. For a very long time there has existed vague but widespread belief in "types"
as regards character, and "stages" as regards depth of the hypnotic trance. While this is by no
means universal (20, 66), there is more or less implicit in this view the notion that there are
critical points on the scales in question at which subjects would be found to cluster in relatively
large numbers, but that at intervals in between these there would be very few or no individuals at
all. The more exact determinations characteristic of modern experimental methods have forced
the abandonment of this view in the field of characterology (10, 32) and, as regards the hypnotic
trance, the results of Davis and Husband tend definitely to clarify this question. They state: "We
have been unable to find any evidence for possible 'critical points' which might indicate progress
of hypnosis to be qualitative, that is, by large steps rather than by gradual increments."
CORRELATIONS AMONG SUGGESTION PHENOMENA: We pass now to the
question of whether people in general are equally susceptible to all forms of suggestion. This
question is of great importance because it promises to assist materially in removing one of the
greatest obstacles to the development of a consistent science of hypnotism, namely, the difficulty
of defining suggestion in a significant manner. It is clear that a mere arbitrary definition, however
elegant to superficial view, must be of little assistance if it includes several phemonena which
prove to be experimentally distinct. One of the most useful methods of determining
experimentally whether two distinguishable phenomena are essentially similar or distinct is to
measure the degree to which a considerable number of persons manifest each of the two traits and
then observe whether persons showing given degrees of one tend, upon the whole, to manifest
corresponding degrees of the other.
For the purpose of investigating problems of this general nature, involving the possibility
of concomitant variation of one tendency with another, mathematicians have devised an
extremely useful index known as the coefficient of correlation. (75)
This coefficient is of such a nature that if high values of one variable go uniformly with
high values of another variable, and low values of the first are uniformly associated with low
values of the second, the correlation will be positive and high, the maximum running up to +
1.00. If high values of the first are associated with both high and low values of the second
according to mere chance, the correlation coefficient will be around zero. And if, as sometimes
happens, high values of the first are associated with low values of the second, and low values of
the first with high values of the second, the coefficient will be negative, the maximum in this
direction being - 1.00. Correlation coefficients ordinarily take the form of decimals. Values from
0.00 to 0.15 are considered as negligible, from 0.15 to 0.35 as indicating a slight tendency, from
035 to 0.60 a moderate tendency, and from 0.60 up to 0.85 as indicating a decided tendency.
Values above 0.90 are rarely obtained from psychological data, and above 0.95 very rarely
indeed. The standard correlation coefficient of Pearson can be approximated for certain purposes
by various other coefficients, the least defective of these being the rank-difference coefficient of
Spearman. It should be noted, also, that correlation coefficients, when based upon a small number
of subjects such as twenty or twenty-five, are not very dependable, since they are exposed to the
accidental influence of the particular individuals who chance to make up the group. As the size of
the group increases up to 100 or 200 or more, the coefficients gradually become more and more
dependable. But even with small groups of subjects the correlation coefficient (r, as it is
conventionally represented) is ordinarily the best index of the tendency to concomitant variability
which it is possible to obtain.
A considerable portion of the experimental work on the intercorrelations among
phenomena commonly classed under the head of suggestion has been concerned with various
forms of non-prestige suggestion, consideration of which must be reserved for a future chapter (p.
350ff.). Here we shall consider, in the main, only those forms of direct verbal suggestion which
most obviously resemble hypnosis in external appearance. Of experimental results in this field
there may be mentioned, first, the fact that Aveling and Hargreaves found their handrigidity test
and their hand-levitation test correlating (76)to the extent of + 0.55, a fairly high value, indicating
that the two processes are rather closely related and quite likely are dependent upon the same
basic mechanism. Further correlation values are reported by Barry, Mackinnon, and Murray (3),
who gave the waking postural suggestion test to seventy-three subjects who had been measured
for hypnotic suggestibility. Three types of instruction were used in the posture test: (A) to
day-dream, (B) not to day-dream but to assume a normal attitude, and (C) to relax and not resist
the suggestions. It was found that the day-dreaming suggestion correlated with the hypnotic index
+ 0.23, the not day-dreaming, normal attitude correlated - 0.06, and the relaxed, non-resistant
attitude correlated + 0.52. A third study is reported by White (I8). This investigator measured the
responsiveness of twenty-two young men and women to a modified form of the postural
suggestion test. He also measured their susceptibility to hypnosis by a rather elaborate point scale
of the general nature of that devispd by Barry, Mackinnon, and Murray and described above (p.
72). These two values correlated to the extent of + 0.75. As a fourth bit of evidence, Jenness
reports (I2) that with eight subjects he found rank-difference correlations of + 0.63 and + 0.33,
according to conditions of the experiment, between response to suggested arm movement (p. 57)
and hypnotizability as indicated by rate of lid closure. Lastly there may be mentioned the
unpublished correlation value in the neighborhood of + 0.60 between postural suggestion and
hypnotizability (as indicated by rate of lid closure) obtained from a group of eighteen male
subjects by Williams and Krueger. To these results may be added the related fact that in the
routine work of selecting subjects who would be susceptible to hypnosis it became a uniform
procedure with Williams, Krueger, and others in the writer's laboratory after September, 1929, to
test all prospective hypnotic subjects with the waking postural technique before attempting
hypnosis and to reject summarily those who did not "fall." In this way was avoided the loss of
great amounts of time usually consumed in attempting to hypnotize the unhypnotizable. It was
rarely found that a subject who would "fall" promptly by the postural technique failed to go into a
fairly deep trance.
The number of subjects employed in some of the individual (77) experiments cited above
was not great enough to render the correlation coefficients decisive, but the striking agreement
among them on this point constitutes a strong objective indication that the response to waking
prestige suggestion, while not identical with hypnosis (since correlation is by no means perfect),
is intimately related to it. So far as these data go, the indication is that hypnosis and the ordinary
forms of direct verbal suggestibility may significantly be included within a single definition.
Unfortunately we have no comparable correlational data between hypnotic susceptibility and the
indirect forms of suggestion of the Binet type (see Chapter XIII). However, the fact that direct
personal (prestige) suggestion as reported by Aveling and Hargreaves (Table 60) correlates to a
relatively low degree with the indirect varieties indicates (but does not prove) that these latter
forms of suggestion may be no more related to hypnosis and hypnotizability than innumerable
other psychological tests which do not happen to bear the name of suggestion. Fortunately the
question at issue is capable of a very direct experimental solution and hence the present
uncertainty should be brief.
SEX DIFFERENCES IN DIRECT WAKING SUGGESTIBILITY: We may now
consider whether there is a sex difference in response to verbal "waking" suggestibility. Only two
studies in this field are as yet available. The writer had forty-one young men and twenty-eight
young women tested by the postural technique for both auto- and heterosuggestibility. All were
members of the same class in abnormal psychology at the University of Wisconsin. The assistant
who performed the experimental work was a young woman of the somewhat commanding height
of sixty-eight inches, but normally feminine in features, voice, and manner. In this study, time
scores were disregarded as practically meaningless, since in many cases the subject did not "fall."
The amplitudes of the responses were measured on the tracings in units of r/50 inch. Since a tall
person, for a given number of degrees of angular sway, will move farther at the level of his
shoulders than will a short person, a correction needs to be introduced at this point. This was
accomplished with the aid of the following formula: (78)
S.A.I. = 100 A over H
where
S.A.I. = suggestion amplitude index
A = recorded amplitude of postural movement in units of r/50 inch
H = standing height of subject in inches
An examination of this table shows that the women, on the whole, are somewhat more
responsive to this form of suggestion, though the critical ratios in both cases are below the figure
of 2.5 necessary to make them of scientific significance if taken entirely by themselves.
Fortunately these data are supported by other evidence, as will presently be shown. It may also be
added in this connection that these results confirmed the impressions of the writer, obtained from
rather extensive explorational research carried out previously.
The fact that the experimenter in the investigation just described was a woman raises a
new but related question. The matter of sex differences has usually been discussed entirely from
the point of view of the subject, as was the case above. It is quite possible. however, that the sex
of the experimenter (79)may be quite as influential. There seems not yet to have been any
systematic attempt to solve this problem, though it is clearly susceptible of an entirely
straightforward solution.2
A second study involving sex differences in direct waking suggestibility of the prestige
type is that reported by Aveling and Hargreaves. They tested seventy-nine English school-boys in
three groups averaging about thirteen years of age, and found an average of 31 per cent showing
definite positive responses on their hand-rigidity and hand-levitation tests, whereas forty-two
girls in two groups of about the same age as the boys showed an average of 63 per cent of definite
positive responses on the same tests. The authors attribute this marked excess of suggestibility on
the part of the girls to differences in teachers and in the general school discipline accorded the
two sexes. The sex of the experimenter seems also to have had an influence here. It is not
possible to determine the statistical reliability of this difference from the published data, but the
gross amount of the difference indicates that it is probably high.
SEX DIFFERENCES IN HYPNOTIC SUSCEPTIBILITY: Closely related to the
matter of a differential sex susceptibility to direct waking suggestion is that of whether there is a
similar sex difference in susceptibility to hypnotic suggestion. Three studies bearing on this
question have been found. Bramwell (5) publishes Liebeault's results on 287 men and 468
women. By assigning somewhat arbitrary quantitative values to his sixfold classification of trance
stages it has been found possible to secure mean susceptibilities for the respective sexes. The
essential facts are shown in the first portion of Table 4. The small difference of 0.04 is in the
direction of the women's being more susceptible, although its critical ratio (0.539) would prevent
it from receiving any serious consideration if uncorroborated by other evidence.
Two modern studies, however, each with a relatively small number of subjects happen to
have employed a fivepoint scale in the measurement of hypnotic susceptibility. Although these
scales were not identical (p. 70ff.), they seemed sufficiently comparable to justify pooling them
to see if a (80) statistically reliable mean sex difference would emerge. The difference of while
small and possessed of the unsatisfactory critical ra 0.867, once more is in the direction of the
women's being suggestible.
THE STATISTICAL PROBABILITY THAT FEMALES ARE MORE
SUGGESTIBLE THAN MALES: In attempting to evaluate the differences found above
between the mean suggestion scores of men and women, important to remember that,
theoretically, there is always a certain danger that such a difference may be due to the accidents
involved in the choice of the individuals to cor the respective groups. It accordingly becomes of
great theoretical (81) and practical importance to know, so far as possible, how great this danger
is and, conversely, how great is the likelihood that the difference found represents a genuine
tendency. Mathematicians have devised certain standard procedures by which a very fair idea
may be obtained as to the reliability or dependability of a difference found to exist between two
contrasted means. This is a function of the ratio of the difference between the two means to the
probable error of the difference (D over P.E.d). The reliability of the difference increases with the
size of this quotient. In general a difference is not considered as dependable in scientific
experimentation, unless supported by other evi-
dence, if the D over P.E.d (critical ratio, as it is sometimes called) is not 2.5 or over. Two means
will differ in a given direction by enough to yield a ratio of this size from mere chance in only
about one experiment in twenty. When the ratio rises to 3.0, the probability of such a result by
mere chance is only about one in forty-seven. When it rises to 4.0, the probability that it is due to
chance alone is only one in 285. When the critical ratio rises to 5.0, the factor of chance shrinks
to only one in 2,832, i.e., the likelihood of its being due solely to chance becomes almost
negligible, as is, indeed, practically the case at 4.0.
As already noted, the critical ratios in the investigations reported above have all been
below the minimum acceptable as significant in experimental work. In a sense this justifies the
conclusion of several of these authors that no sex difference in suggestibility exists. But the fact
that so many independent investigations agree in showing a higher suggestibility in girls and
women greatly increases the probability that the observed mean differences are genuine. Roughly
speaking, other things being equal, a total of four independent studies considered jointly and all
yielding results in the same direction doubles the size of the critical ratio obtained from a single
one of the investigations.
MAGNITUDE OF SEX DIFFERENCE IN SUGGESTIBILITY COMPARED WITH
THAT IN STATURE: A useful objective indication of the size of the sex difference revealed by
these experiments is the size of the decimal (82) obtained by dividing the difference found
between the mean scores of the two sexes by the standard deviation (0) of the male distribution of
the trait in question. The writer's Wisconsin studies in direct waking suggestion yielded ratios of
0.14 for heterosuggestion and 0.27 for autosuggestion. Liebeault's hypnotic susceptibility results
yield a ratio of 0.027. The combined Harvard and Wisconsin results on hypnotic susceptibility
yield a ratio of o. I 1. These four sets of results of the more direct and hypnotic form of
suggestibility average 0.14, i.e., in this type of suggestion the women, on the average, exceed the
men in susceptibility by an amount something like an eighth as great as the standard deviation of
the male distribution of the traits in question.
A notion of the relatively small size of these differences is easily obtained if the above
ratios are compared with a similar one based on the sex differences in standing height. In a
representative sample of 1,079 British men and the same number of British women, it was found
that the men exceeded the women in height by 1.92 times the standard deviation of the
distribution of the height of the males. It thus appears that the average sex difference in
suggestibility (mean ratio of 0.14) is something like one-fifteenth as great as is the familiar sex
difference in standing height. The indication is, then, that women and girls upon the whole are
truly but very slightly more suggestible than are men and boys under the experimental conditions
usually employed, Whether this slender difference will disappear when the sex of the
experimenter and the possibly important factor of selection have been properly controlled is a
question which must await further investigation.
SUGGESTIBILITY AND AGE: The general nature of the relationship of direct verbal
suggestibility to age has long been known. It is summed up in the common observation that
children are more suggestible than are adults. For a confirmation of this time-honored belief by
specific experiment, as well as for a more detailed picture of the situation, we are indebted to
Messerschmidt. She tested approximately twenty-five children at five years and at twoyear
intervals from six to sixteen years of age, using the (83) postural suggestion technique. The mean
index of amplitude of response corrected for height (S.A.I., p. 79) for the several ages is shown
graphically in Figure 22. While the data present com;iderable irregularity, the essentials of the
relationship are probably shown with a fair degree of accuracy. Aside from the general falling-off
of the tendency to suggestibility with advancing years, the most noteworthy characteristic of these
results is the marked rise from year five to year eight.
Such a reversal of a growth curve as these data present is a distinctly unusual phemonenon
in human behavior and calls for comment. In considering the question, the fact may be noted as
possibly not entirely a coincidence that a relationship of an almost identical nature was long ago
observed by Guidi between age and the strong indirect suggestibility evoked by his pseudo-stove
test. Very similar and even more (84) convincing results on a considerable variety of indirect
suggestibility tests are reported by Messerschmidt (p. 375). As a plausible hypothesis to account
for this reversal, it may be supposed that suggestion is based upon a primitive habit tendency (of
responding directly to verbal stimulations) which is useful in most situations but maladaptive in
the special type of situation represented by this suggestion test. Presumably the maladaptivity is
related largely to the fact that if a person responds positively and indiscriminately to all
suggestions made by others, he is likely to be taken advantage of by his associates in that the
energies needed for his own welfare will be diverted to that of those giving the suggestions. The
rise of the curve accordingly represents the acquisition of a working knowledge of the language,
which obviously must proceed a certain distance before its maladaptive possibilities may be
encountered; and the gradual fall observed from about eight years on may be regarded as an
indication of the progress in "unlearning" those particular reactions to verbal stimuli which,
having been established, have proved maladaptive.
HYPNOTIC SUSCEPTIBILITY AND AGE: The question of responsiveness to direct
verbal suggestibility as related to age has been given a certain amount of attention by hypnotists.
Liebeault reports twenty-three cases under seven years and sixty-five cases between seven and
fourteen years of age. None of these children failed to show a certain amount of hypnotic
influence, whereas ages above fourteen showed failures ranging around 10 per cent. Bramwell
summarizes the evidence available to him (including his own extensive experience) as indicating
that children are somewhat more susceptible to hypnosis than are adults. So far as this goes it is in
agreement with Messerschmidt's results. vVe have found no intimation in the literature on
hypnotic susceptibility of the increase in susceptibility to hypnosis during the early years of
childhood. That adequate experiment will reveal this seems certain from the consideration that a
child's responsiveness to verbal hypnotic suggestions must be zero previous to his learning of
language. (85)
SUGGESTIBILITY AND INTELLIGENCE: The well-established fact that
suggestibility, probably of all kinds, declines progressively with age from about the eighth year
until late adolescence doubtless justifies in some sense the use made in the Binet psychological
scale of suggestibility as a test of intelligence. Since general intelligence (scholastic aptitude)
increases with age, a hasty view of the situation might easily lead to the expectation that
suggestibility is negatively correlated with intelligence even for groups of children who are all of
the same age. This expectation is somewhat strengthened by the nature of indirect non-prestige
suggestion, which appears, on inspection, to represent in the main a lack of alertness on the part
of the susceptible subject. But whatever the case may be with indirect suggestion, the available
evidence as regards direct suggestibility points rather in the opposite direction. In Table 5 (Not
available) there have been assembled eight correlation coefficients between direct personal or
prestige suggestion and one or another measure of scholastic aptitude. In nearly all cases these
coefficients are small and of an unsatisfactory statistical reliability. Even so, it can hardly be
without significance that there is not a negative value in the series. 'When it is observed that
correlation coefficients are based on an aggregate of 296 subjects, this array of positive values
should go far to dissipate the somewhat vague but widespread belief that for a person to be
susceptible to hypnosis is an indication of feeble intelligence. The above evidence, on the other
hand, is in perfect accord with the experience of hypnotists that normally intelligent individuals
make the best hypnotic subjects, and that feeble-minded individuals are apt to make poor
subjects.
CHARACTER AND SUSCEPTIBILITY TO WAKING SUGGESTION: It is
difficult at the present time to consider in any fundamental manner the relationship of
suggestibility to the various character traits without taking some notice of certain alleged physical
and psychopathological types at the same time. Barry, Mackinnon, and Murray have assembled
the more important of these supposedly contrasted types. They appear in Table 6, classified as to
what might be expected to be their relationship (87) to suggestibility if a considerable amount of
the somewhat chaotic theorizing in this field were true. A careful examination of this table may
assist the reader to a comprehension of the significance of the evidence presented in the following
paragraphs.
A number of experimental investigations have concerned themselves with the relationship
of verbal suggestibility to character traits. The first of these to be considered is that of
Baumgartner (4). Fifty-six pupil nurses were tested with both heterosuggestion and
autosuggestion by means of the postural technique. In addition these young women were rated by
three superiors on a variety of character traits, among which was that of general suggestibility
itself. The results of this part of the investigation are summarized in Table 7. The values for
autosuggestion parallel very closely those for heterosuggestion. These coefficients are all very
low, and none of them has a satisfactory statistical reliability. The fact, however, that the sign of
the coefficients indicates in general a relationship in the direction that might reasonably be
expected from a commonsense knowledge of human nature increases somewhat the (88)
probability that it represents a genuine tendency. The fact that the two highest values are between
the test scores and estimated general suggestibility increases this probability. The results would
seem to indicate that this test really measures something which enters into the behavior complex
known as general suggestibility, and that it is probably also a significant component of the group
of complex traits called sympathy, sweet temper, and tactfulness, which presumably were not
really differentiated by the character raters from what might be called general amiability. The
correlation values on honesty would suggest that this trait may have fallen within the "halo" of
general amiability. The consistent negative relationship to optimism, if genuine, is more difficult
to understand. Baumgartner's amplitude data were not corrected for the varying heights of her
subjects. This fact, as well as the rather gross inaccuracies involved in even composite character
ratings, would tend to attenuate her correlation values. There is some likelihood, therefore, that
the actual relationships may be considerably higher than those yielded by the experimental data.
The investigation of Aveling and Hargreaves, alluded to above (p. 67), yielded several
correlation coefficients with character traits. These are assembled in Table 8. In general there is
little visible consistency among these results, with the possible exception of positive relationship
to "common sense." (89)
While Aveling and Hargreaves appear to consider this a character trait, most psychologists
would class it simply as intelligence. If it is considered to be the same as intelligence, these
results strengthen the positive relationship already observed between suggestibility and
intelligence.
HYPNOTIC SUSCEPTIBILITY AND CHARACTER: Three independent
investigations have studied the relationship of hypnotic susceptibility to character traits. The
findings of these investigations, together with certain results on related physical traits, are
assembled in Table 9. On the whole there is little consistency. This is particularly striking in the
four sets of results involving extraversion. While some of this apparent lack of consistency may
be due to the fact that different instruments were used in the measurement of the character trait,
the conclusion seems warranted that if there is any relationship between hypnotic susceptibility
and extraversion it is a weak one.
A comparison of the theoretical relations of anthropometric types to hypnotic
susceptibility as shown in Table 6 (not available) with the experimental results shown in Table 9
(not available) are a little disappointing, since the theoretical physical side of the relationship is a
ratio, (90) whereas the experimental results are in height and weight each considered by itself.
The results give little support to any relationship of hypnotic susceptibility and physical
constitution.
HYPNOTIC SUSCEPTIBILITY AND PSYCHONEUROTIC TENDENCY: As
implied by Table 6 (not available), it is but a step from the consideration of the relationship
between hypnosis and personality to the relationship between hypnosis and psychoneurotic
tendency. There have been, in general, two schools of opinion as to the relation between hypnotic
susceptibility and the tendency to mental morbidity. In modern times Charcot and his pupils have
contended that hypnotic susceptibility is substantially a symptom of psychopathology-hysteria in
particular. The Nancy group, on the other hand, has insisted that susceptibility to hypnosis is
entirely a normal phenomenon. This wordy conflict has survived to the present day with an
astonishingly meager accumulation of critical evidence upon which to judge the merits of the two
views. Fortunately two studies relating to this question have recently been reported.
Davis and Husband (6, 179), using the Thurstone Personality Schedule, a device which
gives the subject an opportunitz to indicate in detail whether or not he possesses each of a rather
wide variety of neurotic symptoms, found that their fifty-five subjects showed an average
correlation between hypnotic susceptibility and the presence of neurotic symptoms of + 0.045.
While more evidence will be required to settle this question, this evidence, so far as it goes, is
distinctly in favor of the Bernheim view and opposed to Charcot's view. Once more the Nancy
school seems to be vindicated.
The question, however, may not be so simple as the above experiment appears to imply.
According to the view indicated in Murray's table (No.6), it is possible that only certain neurotic
symptoms (those of hysteria) are associated with hypnotic susceptibility, whereas certain others
(those of neurasthenia) are definitely associated with resistance to hypnotic suggestion.
Presumably both types of symptoms were pooled without distinction in Davis and Husband's
investigation. If the two types existed in about equal proportions, they might be (92) expected to
neutralize each other, a situation which would be entirely consistent with the obtained coefficient
of + 0.045. Barry, Mackinnon, and Murray discuss this matter specifically in connection with an
experiment designed to throw light on the question. They express a tentative opinion that the
supposed bipolar relationship of hypnotic susceptibility to hys'teria and neurasthenia is a genuine
one. Unfortunately for the validity of their observations, less than a half-dozen of their subjects
showed any marked symptoms of either kind. Incidentally, it may be remarked that if this
hypothesis should prove to be correct, it would bring small comfort to those insusceptible to
hypnosis, for the implication would then be that such persons are prone to neurasthenia, just as
those strongly susceptible to hypnosis would be prone to hysteria. By this hypothesis those
susceptible to hypnosis, but in an intermediate degree, would be the only individuals without
special susceptibility to neurotic disorder. We accordingly draw a tentative conclusion from the
evidence at present available that there is no reason to believe that to be susceptible to hypnosis is
an indication of special neurotic susceptibility.
SUGGESTIBILITY AND THE PSYCHOSES: It may be observed in Murray's outline
that manic-depressive insanity is classed as susceptible to suggestion, and schizophrenia as not
susceptible. Ordinary observation of these two types of patients supports this view. It is well
known, particularly of the catatonic variety of dementia precox, that such patients are resistant to
positive suggestions. Indeed, they not only fail to make positive responses, but often show
indications of definite negative responses. For this reason the writer was anxious to determine
their response to the postural suggestion technique. This was finally accomplished by Williams.
Thanks to the splendid cooperation of the authorities of two large hospitals for the insane, access
was given him to a large number of individuals with very carefully verified diagnoses. Of these
he tested fifty-three catatonic dementia precox, twelve paranoid dementia precox, and eighteen
manic-depressives in the manic phase. (93)
Each subject was tested twice with a brief interval between tests. Suggestion was
continued for three minutes. Unless the subjects moved 0.75 inch from their normal position, they
were scored as making no response. The results of the investigation as as summarized in Table 10
(not available) show that the small paranoid group, despite the reputation of such patients for
being suspicious, show no negative responses whatever. Indeed, this group does not differ
materially from normals in this respect. The manics are probably somewhat less responsive and
slightly more negative than are such normals as we have available for comparison (see p. 66£.).
However, neither the paranoid dementia precox group nor the manic group is large enough to
provide a basis for any very confident conclusions. (95)
The high percentage of negative responses with the catatonics is the outstanding result of
the investigation. The remarkable proportion of 59.3 per cent of these patients showed definite
negative (i.e., backward) responses. 'Williams reports that these subjects sometimes sway as far
backward as they can, then right themselves by moving the feet backward, after which the body
may sway backward again, and so on. The record reproduced as Figure 23 is from a fairly typical
subject who gradually shuffled backward by moving the feet. It is important to note that this foot
movement is a phenomenon never observed, at least by the author, among normal subjects.
Williams adds the interesting and significant observation that "some of the catatonics kept
repeating 'Backwards,' in a faintly audible voice, during the entire period of stimulation by the
experimenter." This observation indicates in a remarkably clear manner that with these patients
the negativism is a process of active autosuggestion toward an opposite movement. It is likely
that if this process is gross enough with some subjects to be accessible to ordinary observation, a
refined technique would detect it in many if not all such subjects. It opens up most promising
vistas of investigation not only on the pathological level but on the normal level as well.
In summary of Williams's results, it may be said that they rather definitely confirm that
part of Murray's table (p. 88) which concerns catatonia, but fail to support the view that the manic
phase of circular insanity presents an unusually high (95) degree of susceptibility to
heterosuggestion. In addition they confirm in a convincing manner the a priori expectation that
the postural technique really supplies a means of measuring heterosuggestibility. Lastly they
demonstrate an objective and quantitative technique of symptom recording which may quite
possibly prove of use to clinicians as an aid in differential diagnosis.
PRESTIGE SUGGESTIBILITY AND DELINQUENCY: The relation of
suggestibility to delinquency has long interested sociologists and criminologists. Hypnosis has
sometimes been feared because it has been supposed that subjects, while in the trance, might be
influenced to commit immoral or criminal acts. Possibly in a somewhat similar manner it has
been suspected that at least some forms of delinquency might be due to hypersuggestibility. Only
one study bearing on this important question has been found-that of Landis. He used as subjects
II2 girls at Long Lane Farm, a home for young female delinquents. The subjects ranged in age
approximately from twelve to nineteen years. As is usual in such populations, the majority had
been committed for sex delinquency. All had been there at least three months and some for as
long as five years. Among the many different tests applied to this population was a modified form
of the author's postural suggestion test. Suggestions were terminated after one minute. The test
was given twice, once by a man and once by a woman. With half of the subjects the suggestions
by the man were given first, and with half, those by the woman. The test responses were scored
as follows:
"If the subject swayed forward as much as 8 to 16 em., she was given a score of plus I; if
16 em. or more, plus 2. A swaying backward of 6 to 12 em. was scored as minus I, and of 12 em.
or more as minus 2. Swaying between the limits of plus I and minus I was scored as zero." The
final score was obtained by adding algebraically the scores from the two repetitions of the test.
This procedure yielded a score range from -4 to +4.
The gross results of this investigation are shown in the dot distribution of Figure 24 (not
available). This may perhaps be compared somewhat cautiously with the distribution of
university students shown as Figure 20 (not available). The latter distribution shows that the
college group, if they responded at all, always reacted positively, i.e., they swayed forward.
Landis's delinquent group, in marked contrast, shows about as many negative as positive (97)
re-actions. In this respect they present a rather striking resemblance to Williams's group of
catatonics. In the present inadequate state of our knowledge regarding these complex phenomena,
it is impossible to say why delinquent girls should show this marked tendency to negativism. It is
possible that they possess a native tendency toward negativism and that this in some way
predisposes to delinquent behavior. It is even conceivable that their behavior might be related in
some way to that of catatonic dementia precox patients. All this is highly improbable, however. It
seems much more likely that, either as the result of harsh contacts in their free life or as a
consequence of the restraints of the institution, they have developed a somewhat generalized
tendency to resist all kinds of direct suggestions from other persons, or at least from persons in
the class of authorities. This latter hypothesis harmonizes well with a spontaneous remark made
by one of the young women, after having responded negatively to the suggestion that she should
fall forward: "I won't fall for no damned man. I fell for one once!"
THE INFLUENCE OF DRUGS UPON SUGGESTIBILITY: It has long been
believed by hypnotists that the use of certain drugs such as morphine, alcohol, ether, chloroform,
chloral, hashish, and various bromides, facilitates the induction of the hypnotic trance in
refractory cases. Speaking from a practical point of view, the most commonly used of these drugs
is alcohol. For this reason an attempt was made in the writer's laboratory, late in 1929, to
determine its influence upon waking suggestibility. The alcohol was mixed with a very pungent
brand of ginger ale on experimental days, and on control days the ginger ale alone was
administered. Young men unaccustomed to drinking alcoholic beverages were employed as
subjects. They were given no information whatever as to what they were drinking. Doses of
various sizes were tried and suggestibility was tested by the postural technique at various
intervals after drinking the alcohol and the control dose of simple ginger ale. Unfortunately, the
investigation had to be abandoned while in its (98) explorational stage because of circumstances
beyond the control of the author. The results actually obtained, while in no sense conclusive, left
us with the general impression that the influence of alcohol upon responsiveness to
heterosuggestion, under ordinary laboratory conditions, is slight. There was, however, an
unmistakable increase in the amount of tremor, an indication that the alcohol was producing a
real physiological effect.
The drug scopolamine hydrobromide also has been reported to possess the characteristic
of inducing a state of heightened suggestibility. It may be recalled that this drug, in conjunction
with morphine, has been used to produce the famous "twilight sleep" employed in childbirth.
Possibly because of these suggestion-inducing properties, it has been alleged that criminals under
its influence automatically give a truthful account of their misdeeds when questioned about them.
With characteristic journalistic picturesqueness the drug has been given a wide newspaper
notoriety as "truth serum."
Baernstein carried out an experiment with the purpose of determining in an objective
manner to what extent, if at all, suggestibility is increased by moderate doses of this drug. The
dose was one U.S.P. tablet containing 1/200 gr. of scopolamine hydrobromide dissolved in C.c.
of sterile water. This was injected into the upper arm by means of a small hypodermic syringe. A
control experiment was carried out in which the procedure was in every respect the same as in the
primary experiment, except that the dose was c.c. of sterile water only. At the end of an hour the
subjects showed marked dryness in the mouth and, because of dilation of the pupil, decided
difficulty in reading. Because of this indication that the drug was having its effect the tests of
suggestibility were administered at this time. The usual postural technique and method of
recording were employed. Suggestion tests were given in the following order: heterosuggestion,
autosuggestion, heterosug-gestion with preliminary instructions to resist, and finally ordinary
heterosuggestion like that given first. With minor exceptions each subject was tested after one
drug dose and one control dose. The main group of subjects consisted of nineteen university
students, chiefly students of medicine. Of these only eight proved to be responsive to suggestions
following either the drug or the control dose. (99)
It was found in general that subjects who were resistant to suggestions in the normal state
were also resistant when under the influence of the drug. There was a fairly marked tendency,
however, for those who were susceptible in the normal state to be markedly more susceptible
when under the influence of scopolamine. This ,vas true for autosuggestion as well as
heterosuggestion and appeared as clearly when the responses were scored on the basis of time
required to reach maximum response as on that of amplitude of maximum response. It appeared
still more clearly when both speed and amplitude were combined into a single index by dividing
the maximum amplitude in inches by the time in seconds, i,e" the rate of fall in (100) inches per
second. The results of the two pooled heterosuggestion tests by both methods together with the
statistical reliabilities of the differences associated with the drug are shown in Table I (not
available) As we have had occasion to observe in other connections, the amplitude of reaction of
subjects who do not all give maximal reactions yields the more reliable critical ratio -in this case
5.3 I, which is statistically very satisfactory. Mrs. Baernstein's suggestibility index, which
combines both measures, seems to be a still better measure, as it yields the higher critical ratio of
7.25. By this latter method everyone of the eight subjects showed a heightened susceptibility to
suggestion after taking the drug.
SUMMARY AND CONCLUSIONS: We may now pass in brief review the results of
the somewhat numerous experiments concerning relationships of suggestibility to other
fundamental factors and conditions of human nature. We have seen that while perhaps a third of
normal subjects show little or no direct verbal or prestige suggestibility, it is very rare indeed for
them to manifest overt negative response. The distributions of various groups on the several
scales of suggestibility and of hypnotizability which are available have yielded somewhat various
as well as erratic contours. A judicious examination of all the evidence leads to the conclusion,
however, that the seeming deviations from the normal or bellshaped distribution are more likely
the result of defective scales for measurement than of any inherent constitutional tendency.
It has been found wherever the question has been investigated that all forms of direct
verbal or prestige suggestion correlate positively and to a rather high degree with each other and
with susceptibility to hypnosis, whereas there is some indication that they correlate to a low or
zero degree with indirect suggestion. Thus, direct suggestion is indicated to be closely allied
(though not identical) with hypnosis, but indirect suggestion to be probably distinct in its basic
determination.
The indication is that women and girls are probably more (101) susceptible to direct suggestion
and hypnosis than are men and boys, but only very slightly so-the average difference being only
about one-fifteenth as great as that between the average standing height of the two sexes.
The experimental evidence seems to indicate that in childhood suggestibility is at a
minimum before The acquisition of language habits and that it rises rapidly from five until about
eight years of age, after which it declines steadily until adult years are reached.
The popular belief that suggestibility is a mark of stupidity or lack of intelligence appears
to be wholly an error so far as direct or prestige suggestion is concerned. Considerable
experimental evidence indicates that, if anything, there is a slight positive relationship between
direct suggestibility and intelligence.
The attempts experimentally to discover relationships between direct suggestibility and
various character traits have, for the most part, been indecisive, though there is a little indication
of a weak positive relation to a group of traits characterized by general amiability.
Such evidence as has been found bearing on the correlation of hypnotizability and a
tendency to mental morbidity is that the two are quite unrelated, which supports Bernheim's side
of the old controversy with Charcot.
That catatonic dementia precox patients are negatively suggestible has been fully
substantiated by experiment, which confirms both clinical observations and the a priori belief that
the postural technique really yields a measure of suggestibility.
Juvenile delinquent girls in a reformatory have been shown to be far more negatively
suggestible than ordinary young women, though the significance of the findings for delinquency
is not clear.
As regards drugs, alcohol does not appear materially to increase suggestibility.
Scopolamine, on the other hand, seems to increase suggestibility in persons who normally show
positive responses, but not in those who normally do not give positive response. (102)

13. CH 5: THE RECOVERY OF LOST MEMORIES IN THE HYPNOTIC TRANCE: CLARK

HULL: HYPNOSIS AND SUGGESTIBILITY: APPLETON-CENTURY-CROFTS: NY:
COPYRIGHT 1933: 1961
IN Chapter IV we have considered the experimental evidence concerning the more
important relationships of hypnotic susceptibility. We must now plunge into the midst of the
experimental phenomena of the hypnotic trance itself. A large and important group of these
concern memory. One of the most remarkable of the latter is the alleged capacity of subjects in
the hypnotic trance to recall events which are completely lost to the ordinary waking memory.
Such special facility of recall is called hypermnesia. The phenomena have very great interest in
themselves and, in addition, possess special significance as bearing on the nature of the hypnotic
state. Various experimental aspects of this problem of hypermnesia will accordingly occupy us in
the present chapter.
Losses of Memory: Two types of lost memories have been utilized in the study of
hypnotic hypermnesia. In one type, hypnosis is used as a device to bring about recall in cases in
which there has been an amnesia, i.e., an abnormal or pathological lapse of memory. In the other,
it is used to recover memories of very early childhood which might reasonably be presumed to
have been lost through the ordinary processes of forgetting. While superficially quite distinct,
these two types of lost memories have frequently been brought into very close relationship,
particularly by the psychoanalysts. This appears quite clearly in the opening paragraph of an early
article by Sigmund Freud, written in conjunction with Joseph Breuer, before the former gave up
the use of hypnosis in his treatment of mental disorders:
"Instigated by a number of accidental observations, we have investigated for a number of
years the different forms and (105) symptoms of hysteria in order to discover the cause and the
process which provoked the phenomena in question for the first time, in a great many cases years
back. In the great majority of cases we did not succeed in elucidating this starting point from the
mere history, no matter how detailed it might have been, partly because we had to deal with
experiences about which discussion was disagreeable to the patients, but mainly because they
really could not recall them; often they had no inkling of the causal connection between the
occasioning process and the pathological phenomenon. It was generally necessary to hypnotize
the patients and reawaken the memory of that time in which the symptom first appeared, and we
thus succeeded in exposing that connection in a most precise and convincing manner."
Many cases of amnesia have been studied and reported in voluminous detail, though we
are still far from a satisfactory understanding and agreement as to their nature. A trifling amnesia
in the form of a temporary failure of recall of some perfectly well-known memory is an
occurrence familiar to nearly everyone. A person may be a little fatigued or worried. He attempts
to recall a name. He may try his best, but to no avail. On a later occasion, without any intervening
rehearsal, the name may be recalled with ease.
Extensive amnesias are very apt to be associated with shocks of some kind. These may be
gross physical injuries, such as severe blows on the head or the shock may be an emotional
disturbance or even a moral conflict. The resulting loss of memory may cover only the few
minutes immediately preceding an injury which produced unconsciousness; on the other hand,
cases not infrequently appear in which the patient can recall nothing of what took place during a
period of several months or even years. He may be able to recall neither his name nor anything of
his former life. Such cases are believed usually to be hysterical in nature. That the memory traces
are really intact during the amnesia is shown by the fact that recall can usually be effected,
without any relearning, merely by putting the subject into the hypnotic trance and suggesting to
him that he will be able to recall everything in a normal manner. Further indication that the
memory traces are intact in such cases lies in the fact that bodies of learned material are relearned
with remarkable ease even without resort to hypnosis. (106)
AN EXAMPLE OF TRAUMATIC AMNESIA: A somewhat dramatic example of the
recovery of a traumatic amnesia through hypnosis is given in the following report by Mr. Albert
A. Sames, who made the observations under the writer's direction while a member of a seminar
on hypnotism:
On Thursday afternoon, the 8th of March, Mr. B rented a car, and at 2 :00 P.M. called for
Miss K. They rode around, had something to eat, and at 4:30 P.M. drove to B's quarters to get a
hat, coat, and blanket. It was snowing hard at this time, with a high wind and poor visibility.
After B had gotten his coat, etc., they drove away. Sometime around 5: 30 they were hit by a
train on a crossing at College Hill. The car was overturned and B was thrown out through the top,
landing some distance from the car. Help was called by some one, and Miss K was taken to the
hospital, where she soon regained consciousness, and asked about B. Up to that time no one had
thought that there had been another person in the car. They returned to the scene of the accident,
got E, and took him to the hospital. Miss K had a compound fracture of the pelvic girdle. B had a
few bruises on his head and spine. He was released from the hospital the following Tuesday
afternoon (March 13). I talked with him that evening, and found he had, with one exception (he
remembered crossing a railroad track) a complete amnesia, extending from 4:30 Thursday, when
he drove away from his quarters, until Friday noon following, when he awoke at the hospital.
One evening about a week previous to the accident I had gotten B into fairly deep
hypnosis, showing amnesia for events during the trance. On Tuesday evening, when I discovered
the amnesia covering the accident, I tried to hypnotize him, but only succeeded in making him
somewhat sleepy. My failure was due, I think, mainly to the fact that his eyes watered excessively
(I was using the optical fixation method), and because his head and back were aching. He seemed
restless and nervous.
The following Thursday evening, at about I I: 15 I again tried to get him into a trance. He
was seated in a chair and gazed upward at a bright bit of metal fastened to a piece of wood that I
held. Meanwhile I stroked his head and suggested sleep. His eyes gradually closed and his head
fell forward, but he did not seem to be in a deep sleep. I continued stroking his head, and
suggested that when I had counted ten he would fall into a deep sleep. When I had done this he
was in the trance, and I had him move from the chair to the bed and lie down. I then began to
question him, writing down his replies as he spoke. Following is the conversation which took
place: (107)
Q. Do you remember when you were out riding last Thursday afternoon, and stopped here to get
your hat and coat about 4:30?
A. Yes.
Q. Where did you go from there?
A. Along the lake.
Q. Where did you go then?
A. Turned off.
Q. Where then?
A. There were houses there.
Q. Where next?
A. We went away from the houses.
Q. What did you do then?
A. Rode some more.
Q. What next?
A. On another road.
Q. What did you do then?
A. Turned again.
Q. What then?
A. We were on a road with trees.
Q. What next?
A. We stopped to fix the windshield.
Q. What then?
A. We talked.
Q. What did-you do next? A. My hands were cold.
Q. What next?
A. Started the car.
Q. What did you do then?
A. Rode around some more.
Q. What next?
A. We were going down a hill.
Q. What then?
A. The road was slippery.
Q. What then?
A. It was snowing-some snow blew in my face.
Q. What next?
A. Heard a noise.
Q. What then?
A. Dorothy hollered.
Q. What next?
A. She put her hand on my arm.
Q. What then?
A. It sounded like a train.
Q. What next?
A. Looked like a train coming.
Q. What then?
A. I tried to stop or turn. (108)
Q. What next?
A. Something hit us.
Q. What next?
A. I don't know.
Q. Yes, you do. Tell me. What happened then?
A. The nurse was standing by the door. There wasn't any one else around. I was sick at my
stomach.
Following B's reply above about hearing a noise, until the end of the conversation, there
was a recurrence of the original excitement and consternation, marked by constriction in the
muscles of the throat, and by apparent inability to get any words out. I quieted him by assuring
him that all this had happened to some one else, and that he was merely telling me the story. He
calmed down enough to talk, and I got the rest of it out of him. I then gave him the usual
suggestions about feeling well, and not worrying when he should have wakened. I then awakened
him.
SOME TYPICAL OPINIONS REGARDING HYPNOTIC HYPERMNESIA: The
view that memories completely inaccessible to normal recall may be recovered in hypnosis has
had a very wide currency, chiefly among physicians dealing with psychopathology. For the most
part, the use of hypnosis for this purpose has been interpreted largely in terms of that hypothetical
psychic entity known as the subconscious mind and a hypothetical state of the nervous system
called dissociation. Thus Breuer and Freud write: " ... we had to emphasize the fact that the
recollection of the effective psychic trauma is not to be found in the normal memory of the
patient but in the hypnotized memory. The more we occupied ourselves with these phenomena
the more certain became our convictions that the splitting of consciousness so striking in the
familiar classical cases of double consciousness exists rudimentarily in every hysteria, and that
the tendency to this dissociation ... is the chief phenomena of this neurosis."
Prince, in connection with the description of one of his hysterical cases, reports: (109)
" ... she in hypnosis asserted [that] fear afterward had continued to be present more or less
persistently, although she was not conscious of the fact when awake ... " Passing to the more
normal types of loss of memory, we may cite Moll, who remarks: "Events of the normal life can
also be remembered in hypnosis even when they have apparently been long forgotten."
McDougall states, "It is true that in hypnosis, especially in its deeper stages, the power of recall
of seemingly forgotten incidents, especially those of early childhood, is greatly increased; ... "
Wingfield writes, "In the deep stage the memory may extend farther back, often very much
farther than it does in the normal condition, and a i;ubject may be able to recall events of his
earliest childhood. This extension of memory is not limited to the deep state, though it is then
more marked ... In one case ... I made a lady of thirty-two remember herself in long clothes. "
Bramwell says, "The improvement of memory as to remote events is still more
interesting; and this I have frequently demonstrated in the following manner: Certain subjects
were first questioned in the normal state as to the earliest events they could remember, when it
was generally found they could recall nothing which had happened before the age of five or six.
They were then hypnotized and, starting from the first event in their lives they could recall, it was
suggested that they should revive the memory of earlier and earlier incidents. Some of the
subjects related what they stated had happened at the age of two, and one described a children's
party given on the first anniversary of her birthday."
THE PROBLEM OF ESTABLISHING THE VALIDITY OF HYPNOTIC
HYPERMNESIAS: Reports of such remarkable power of recall as the two last cited naturally
arouse a considerable degree of skepticism as to the genuineness of the memory. There is always
the possibility (110) that the subject under the influence of suggestion has fabricated the report.
Even the fact that psychotherapy based on the supposed memory has been of value to the patient
is hardly satisfactory evidence of historic accuracy because it is always possible that a
pseudo-memory might be quite as efficacious in such a case as a genuine one. As a rule, writers
who report such results are more or less aware of this difficulty and usually seek by means of the
testimony of relatives or documents such as diaries to substantiate the accuracy of the supposed
memory as to fact. Such evidence is rarely or never under experimental control and hence is
usually unsatisfactory as scientific material. The same, of course, holds for the supposed hypnotic
recall in cases of obviously traumatic amnesias such as the one reported by Mr. Sames. Luckily it
is not necessary to depend entirely upon such uncertain data. In the case of standard and
well-known verbal material memorized at an early age, the nature of the original memory is
easily established and with quite satisfactory accuracy. We are fortunate in having a quantitative
study based on such evidence.
 AN EXPERIMENTAL TEST OF THE POWER OF HYPNOSIS TO RESTORE
LOST CHILDHOOD MEMORIES: Stalnaker and Riddle (8) determined the relative ability of
twelve male university students to recall in the trance and the waking state the words of verse or
prose which had been learned at least one year previously. All subjects showed such hypnotic
phenomena as negative hallucinations, analgesia, and catalepsy. All were completely naIve as
regards the problem under investigation. They were warned, however, that they must not reread
or relearn in any way the selections in question, as this would invalidate the experiment. The
authors are convinced that no difficulty was encountered from this source. On a given
experimental occasion the subject wrote out all he could recall of the selection, in one state;
immediately thereafter he did the same in the other state, so that a comparison of the power of
recall in the two states might be made. Some days the writing would be done first in the trance,
and on others it would be performed first in the waking state. In a given fourday period, if the
first day began with the trance, the second (111) and third days began with the waking state and
the fourth day began with the trance, and so on-the ordinary A-B-B-A order. Only one selection
was recalled on a given day. With one or two exceptions each subject served on eight different
days. In this way a total of ninety-two trance-waking recall comparisons were made. The written
recalls were compared with the original sources and scored for the number of words correctly
recalled. Of the ninety-two recollections thus performed in the two states, eighty-seven, or nearly
95 per cent, showed a superior score in the hypnotic condition. The results by subjects are
summarized in Table 12 (not available). There it appears that every subject without exception
shows an advantage in recall when in the trance state. The mean percentage advantage in recall
for the trance state is 64.86. This value is probably unduly influenced by the very large
percentage value of Subject No.6. A median value falls (112) around 37 per cent excess recall in
the trance. Or, if we consider the gross total recall in the two states of all subjects pooled
together, we find that the trance exceeds the normal state by 53.4 per cent-probably the fairest
single value for representing the situation. Stalnaker and Riddle report the extremely high critical
ratio of 5.3, though they do not make it entirely clear to which value this ratio applies. It is
evident from a mere inspection of Table 12, however, that we have here an extremely stable
tendency. The experiment goes far to justify in a controlled and quantitative manner the claims of
clinical observers that hypnosis is somehow able to bring about a recall of more or less remote
memories which is more effective than waking recall.
SOME QUALITATIVE OBSERVATIONS OF HYPNOTIC RECALL OF EARLY
MEMORIES: Stalnaker and Riddle made some interesting qualitative observations which tend
to explain some, at least, of the advantage in power of recall observed in the trance state. They
found a noticeable tendency for subjects in the trance state to reconstruct or partially fabricate
material which they were unable to recall in exact detail. As an example of this, one subject wrote
as the second stanza of "The Village Blacksmith" the following:
"The smithy whistles at his forge
As he shapes the iron band;
The smith is very happy
As he owes not any man."
In the waking state this material was entirely omitted.
The first two stanzas of Longfellow's poem, "The Village Blacksmith," are as follow:
Under a spreading chestnut-tree
The village smithy stands;
The smith, a mighty man is he,
With large and sinewy hands;
And the muscles of his brawny arms
Are strong as iron bands.
His hair is crisp, and black, and long,
His face is like the tan;
His brow is wet with honest sweat,
He earns whate'er he can,
And looks the whole world in the face,
For he owes not any man. (113)
In this connection it should be stated that somewhat different suggestions were given the
subjects before they attempted recall in the waking state as contrasted with suggestions given
before they made the attempt in the trance state. The instructions for the waking state were:
"You said that you had memorized at some time in the past 'The Village Blacksmith.' Now
I want you to take this pencil and paper and write out this selection just exactly as you memorized
it. Write as much of it as you can. When you have done your very best, bring me the paper."
Those preceding the recall in the trance were: "I want you to think back to the time when
you memorized 'The Village Blacksmith.' You can remember exactly when you learned it. It
comes back to your mind. You can remember the piece vividly, without any strain or worry.
"I am going to give you a pad of paper which I want you to hold in your left hand, and
when I put the pencil in your right hand you will find you can open your eyes. Then you will
write the piece just as you memorized it. As soon as you have completed writing as much as you
can remember of the selection, you will put the pencil down, your eyes will again become heavy
as they are now, and you will rest just as you now are resting. Remember, even when your eyes
are opened you will remain in just the state you are now in."
It is common in attempts to secure very early memories from hypnotic subjects to give
elaborate suggestions that the person is again a child, and thus to attempt to make him relive the
portion of his life in question. The first paragraph of the above trance instructions to recall shows
a clear, though rather mild, tendency in this direction. Because of this the results of the
Stalnaker-Riddle experiment may with some plausibility be assumed to be typical of the reality
behind the familiar clinical reports.
It is possible that the marked tendency toward fabrication noted above may be inherently
characteristic of the trance. It is conceivable, however, that it may be due merely to a greater
suggestive pressure put upon the subject in the trance state (as typified by the differences in the
formula just considered) and that an equal tendency to fabrication would result if an equal
pressure were put on the subject in the waking state. Strictly (114) speaking, a legitimate
comparison could not be made with respect to this tendency in the trance and waking states unless
the instructions or suggestions were made identical in the two states. Until results are available in
which this has been done, there must be a certain suspension of judgment.
Moreover, there is strong indication in the Stalnaker-Riddle results that the standard of
material which the subjects were willing to offer as satisfactory recall in the trance was
considerably lower than that characteristic of the waking condition. It is conceivable that this
difference in standard of certainty or accuracy may be entirely responsible for the increased
amoun1 of recall characteristic of the trance state, though it seems a little surprising that it should
produce such a large difference as that reported. The uncertainty regarding this factor can be
cleared up only by further experiment in which the standard of excellence of recall is somehow
equalized in the two cases.
Meanwhile, the demonstration of the radically greater tendency to fabrication in the trance
state under the conditions usually obtaining in such experiments should make us extremely
cautious about accepting as authentic alleged recollections not accessible to the subject in the
waking state, particularly when they relate to matters of very early infancy.
TWO HYPOTHESES CONCERNING HYPNOTIC HYPERMNESIA: It is very
generally believed by psychopathologists that the memory traces in the case of hysterical
amnesias are not particularly weaker than those which mediate ordinary waking recall. It is
supposed, rather, that the trauma or emotional complex, whatever its nature, somehow inhibits or
blocks the recall; that it interferes with the action of excitatory tendencies which would otherwise
bring about perfectly normal recall. On this hypothesis, then, hypnosis merely removes the block
or inhibition in the case of hysterical or traumatic amnesias and thus permits the existent
excitatory tendencies to function in a normal manner.
It is possible, however, to frame an alternative, or at least a supplementary, hypothesis.
According to this second supposition, the greater facility of recall alleged in cases such as are
(115)
cited by Wingfield and Bramwell, as well as the results reported by Stalnaker and Riddle, may be
accounted for by assuming a genuine lowering of the threshold of recall in the trance. If such a
general lowering of the recall threshold takes place automatically in the trance, on some primitive
physiological basis, it should operate to facilitate the recall of recently learned material as well as
of relatively remote material. Fortunately, we have in the very carefully controlled work of Huse
an experimental test of this lower-threshold hypothesis.
An Experiment Concerning the Power of Hypnosis to Facilitate the Recall of
Recently Learned Material: The Huse experiment was set up on the assumption that if just
barely learned material, such as paired associates, was "illlowed to deteriorate through the lapse
of time until only a part of it could normally be recalled, then a considerable number of the
reactions inaccessible to recall would be just barely below the threshold. If, then, the threshold
should be lowered in some way, e.g., by means of hypnosis, some of these normally just barely
subliminal excitatory tendencies would be able to evoke active recalls. These recalls added to
those normally available should yield a substantially larger average recall score in the trance than
in the normal state. If, on the other hand, the trance state should consistently fail to show such an
advantage, other things being equal, the indication would be that hypnosis, as such, does not
lower the threshold of recall.
The nature of the memory material employed is shown in Figure 25. The nine characters
in the left-hand column were exposed, one at a time, by an automatic exposure machine (Fig. 26)
for a period of five seconds each. In the middle of this exposure period the name of the character
was spoken by the experimenter and repeated by the subject. 'When learning on any given pair
had progressed to a point such that the ~ubject could recall the name at the presentation of the
character, he did so and was credited with the success. In case the subject did not respond
correctly by the middle of the exposure \nterval for a particular character, the experimenter gave
the (11) name again. This prompting process was continued until the subject was able to go
through the series of nine characters twice without error. In order that the learning should be
purely that of paired associates and not complicated by accessory serial learning based on
position of the pairs in a particular sequence, photographic duplicates of the characters were so
arranged on a long band running over the drum of the machine that the nine characters of each set
of memory material were presented to the subject in eight different orders. Each of the (117)
eight complete presentations of the nine characters, in whatever order, was called a presentation
cycle.
All of the original learning involved in the experiment was performed in the normal or
waking condition. Four sets of material such as that shown in Figure 25 (not available) were
learned in close succession by each subject on each experimental day. In order to secure a
measure of the relative capacity of the trance and the normal state for the recall of faint
memories, this material was presented to the subject for recall twenty-four hours after learning.
On this occasion the characters of a given series were presented in three cycles without any
prompting whatever. A perfect score on one set of material would thus be 3 x 9, or 27 correct
responses. For the four sets of material on a given day a perfect score would be 4 x 27, or 108
correct reactions.
HUSE'S PRECAUTIONS AGAINST CONSTANT ERRORS: It is to be observed that
in an experiment of this kind the greatest care must be exercised that the final averages are not
distorted from their true values by a constant error caused by some factor not under
investigation-in the present case, some factor other than hypnosis.
Five of the more important of these possibly disturbing factors are:
1. Differences in retentiveness characteristic of the individual subjects used in the experiment.
This difficulty arises from what are commonly known as individual differences.
2. Differences in a given subject's reactions from one experimental occasion to another.
3. Differences in retentiveness between the earlier and later stages of training. Such differences
an~ commonly known as practice effects.
4. Differences in the ease or difficulty of retention characteristic of the several sets of memory
material.
5. Differences between the trance and waking states arising from the fact that a subject in the
trance state can usually recall material learned in a previous waking state, whereas a subject in the
waking state usually cannot recall material learned in a previous trance state.
In the Ruse experiment these various dangers of error or distortion were met as follows:
The matter of individual (118) differences in recall was equalized (controlled) by determining the
power of recall of faint memories by each subject in each of the two states under consideration so
that a superior natural recall capacity could have an equal chance to function in both the trance
and normal states. The variability of a subject from time to time can be minimized (controlled) by
taking repeated measures and finding the central tendency. A total of eight determinations were
made on each subject in each state. The possible distortions from practice effects were equalized
(controlled) by scheduling recall in the normal state for the first and fourth experimental days,
whereas the trance recalls took place on the second and third experimental days. In order still
further to equalize this tendency, the subjects were divided into two groups, one half being tested
as just described, the other
(119) half taking the primary recalls in the trance state on the first and fourth experimental days
and the primary waking recalls on the second and third experimental days. This is, of course, the
double application of the A-B-B-A order. In order to equalize differences in the difficulty of
recall characteristic of the several series of memory material, the latter were so rotated among the
experimental days and experimental groups of subjects that each set of material was utilized an
equal number of times for each type of both primary and secondary recall.
 The difference in the capacity of subjects in each of the two respective states to recall
material acquired in the other was not taken into consideration in the original set-up of the
experiment. In order more fully to equalize both the changes in the subject's capacity to recall
from period to period and, at the same time, the differences in the difficulty of the sets of
memory material, the subjects were tested on each occasion for recall both in the normal and in
the trance state, half the times with the one preceding and half the times with the other. When the
experimental work was completed it was found that the improvement through practice in recall
during the first state carried over so as to facilitate recall in the second state, e.g., trance recalls
when preceded by normal recalls were appreciably better than when not so preceded. It was
found from a further statistical examination of the experimental results, however, that in the
trance-normal recall combination the advantage of the second position was much less than in the
normaltrance combination. This had been suspected on the principle that events taking place in
the trance are generally supposed to be inaccessible to a following waking state, whereas events
taking place in the normal condition are generally supposed to be fully accessible to a following
trance state. In view of this situation it is evident that a na'ive comparison of the secondary trance
recalls with the secondary normal recalls would have given the trance recall an appearance of a
spuriously low threshold, and that an inclusion of these results in the final computations would
therefore introduce a gross constant error. Accordingly, both sets of secondary recalls were
discarded in the ultimate comparison of the power of the two states to recall faint memories, the
final conclusions being based on the primary recalls only. (121)
A number of somewhat complicated procedures necessary to protect the results of this
otherwise simple experiment from various possibilities of distortion and miscarriage are shown
systematically in Table 13.3 (not available)
RESULTS OF THE HUSE EXPERIMENT The basic results of the Huse experiment
are given in Table 14 (not available), and in still more summary form, together with the
differences between contrasted values, probable errors, and critical ratios, in Table 15 (not
available). Actually these results show a slight mean tendency for recall to be better in the normal
state than in the trance state. However, none of the differences has a sufficiently high critical ratio
to entitle it, unless supported by other evidence, to serious consideration. So far as the results of
the Huse experiment are concerned, then, the indication is definitely against the hypothesis of a
special lowering in the hypnotic trance of the threshold of recall for faint memories.
A SECOND STUDY CONCERNING THE HYPNOTIC RECALL OF FAINT
MEMORIES: Fortunately there are two additional experimental studies which fully corroborate
Miss Huse's findings. The first of these to be considered is reported by Mitchell. Two subjects,
previously well trained in memory experimentation, were employed. They were required to
memorize sets of ten three-place numbers. These were exposed for two seconds each by means of
a McGeoch automatic memory drum, an apparatus which accomplishes substantially the same
things as the one shown in Fig. 26. Sixty-four sets of such numbers were learned in the normal
state by each subject and seven minutes later were relearned. The sets were divided into four
equal groups on the basis of the conditions of the retention interval and the state in which the
relearning took place, as shown in Table 16 (not availabe). It will be noted that procedures A and
C yield a comparison between the power of the trance and that of the waking state to reinstate
very recently acquired material, while in a similar manner conditions Band D perform the same
comparison. With these latter two series, however, we have the interesting additional
complication that to the amount of loss due to ordinary forgetting during the seven minutes there
is added the loss known to result from the interference (retroactive inhibition) produced by the
learning of the extra series immediately following. In order to protect the results from distortion
by practice effects, one subject (Beh) was put through the experiment with the four procedures,
one each day, in the order
ABCDDCBABADCCDAB
Subject Fus carried out the experiment in the order
DCBABADCCDABABCD
Mitchell reports the results of the experiment in terms of the amount of "saving" on the
relearning. This is a standard (123) quantitative method in memory experimentation. Suppose it
should require a total of 102 presentations or repetitions of the memory material for a subject to
complete the original learning of the four number series on a given day, and that it should require
a total of only forty-two repetitions to relearn the same material seven minutes later. There
would, then, be a total saving for the four series of sixty repetitions. Mitchell's results for the two
subjects by this method of scoring are shown in Table 17 (not available). An inspection of this
table shows at the outset that the recall in both states when the original learning was followed by
other learning was distinctly poorer than when followed by light reading. This phenomenon has
long been known as retroactive inhibition. Coming to our immediate problem, we find that the
first subject tends, in the experiment as a whole, to show a somewhat better capacity for
reinstatement in the waking state than in the trance state. The other subject tends to show a slight
advantage in the trance state. The average of both subjects shows practically no difference in the
case of relearning after the reading interval. In the case of relearning after seven minutes of
inhibitory learning, the average is rather in the direction of a better waking recovery of the loss.
This experiment furnishes no support for the view that hypnosis favors the recovery of faint
memories. (124)
A THIRD STUDY OF THE HYPNOTIC RECALL OF FAINT MEMORIES: We
may conclude this discussion with the presentation of some results reported by Young (II), who is
the first person, so far as we have been able to discover, to carry out adequately controlled
experiments on a large scale in the field of hypnotism. Among numerous other problems he
investigated the hypnotic recall of both remote and recent memories. His attempt to induce
hypnotic hypermnesia for childhood memories seems to have been distinctly successful with one
of the two subjects employed. Of primary interest, however, are his experiments on hypermnesia
for recent memories.
Young reports a test on two subjects of their power to recall objects in an adjoining room
which both had casually observed, though their attention had never been called to its furnishings.
The trance showed no advantage in power of recall over the normal condition. In a second
experiment involving recall of recently learned material, each of fifteen hypnotic subjects learned
eight sets of twenty-five paired associates of adjectives linked with nouns. In the following
experimental session they were tested for recall of this material in the trance and the waking
states. It was found that these subjects showed, on the average, an advantage of 5 per cent while
in the hypnotic state. In a parallel experiment, however, four non-hypnotic control subjects who
voluntarily relaxed in a kind of pseudotrance showed a "trance" advantage of 7 per cent. Clearly
these experiments yield no evidence of a lowered threshold of recall in the trance state for
recently acquired memories.
IMPLICATIONS OF THE EXPERIMENTS ON RECALL OF RECENTLY
LEARNED MATERIAL: We have, then, including the major investigation of Huse, three
wholly independent experiments, all of which indicate that the threshold for memories of recently
acquired material is no better but, if anything, a little worse in the trance state than in the normal
waking condition. On the other hand, if we take the clinical reports at face value, there is a great
mass of evidence not only to the effect that the hypnotic trance is able to facilitate recall of
memories which have been inhibited by traumas of various kinds, but also that it greatly
facilitates the recall of ordinary memories of early childhood. The clinical evidence concerning
this latter tendency is strongly substantiated by the controlled quantitative experiment of
Stalnaker and Riddle. This later evidence seems, superficially, to indicate something like a
lowered threshold for remote memories. That hypnosis should produce a lowered threshold for
remote memories and no such lowering for relatively recent memories is distinctly paradoxical.
It is conceivable, of course, that ordinary forgetting, when based on the acquisitions of
childhood or when extended over a long period of time at any age of life, may possess a
considerable element of inhibition or blocking of existent excitatory tendencies as distinguished
from the simple weakening and disappearance of excitatory tendencies which is usually assumed
to be the cause of ordinary forgetting. This hypothesis would account very well for all of the
observed phenomena with the possible exception of the evidence reported by Mitchell (p. 124)
tending to show that memory material which had suffered a great deal of retroactive inhibition
was recovered no more readily in the trance than in the waking state. This phase of Mitchell's
experiment should be repeated on a larger scale to establish the involved relationship with
certainty.
It may be, however, that the inhibitions involved in hypnotic hypermnesias are of the
emotional type rather than of the retroactive character found in the Mitchell experiment, i.e., that
the two forms of inhibition are essentially distinct. Considerable illumination on this question
might be obtained by a carefully controlled experiment of the Huse type, but one in which strong
emotions are somehow associated with the memory content. This should be performed with both
meaningful and nonsense material. If such an experiment should show a dear and strong tendency
for the hypnotic recall to be superior, it would tend to indicate that hypnotic hypermnesia is due
to the removal of inhibition rather than to the lowering of the threshold of recall. It would thus
throw great light not only on the nature of hypnosis but on the nature of the memory and
retention processes as well. (126)
SUMMARY AND CONCLUSIONS: The experimental evidence shows rather
definitely that recently acquired memory material is recovered no better in hypnosis than in the
waking state, which probably indicates that hypnosis does not lower the threshold of recall. There
is some striking experimental evidence which, while not absolutely convincing, tends strongly to
confirm the clinical observations that hypnosis facilitates the recall of childhood and perhaps
other remote memories.
(127)

14. CH 6: EXPERIMENTAL ASPECTS OF POST-HYPNOTIC PHENOMENA: CLARK

HULL: HYPNOSIS AND SUGGESTIBILITY: APPLETON-CENTURY-CROFTS: NY:
COPYRIGHT 1933: 1961
WE HAVE examined in the last chapter such evidence as is available regarding hypnotic
hypermnesia, a phenomenon widely regarded as intimately related to dissociation. In the present
one we shall continue the consideration of the supposed facts of dissociation as displayed in
post-hypnotic amnesia and post-hypnotic suggestion. We shall begin with the consideration of the
former, since it is a memory phenomenon directly contrasted with hypnotic hypermnesia.
A. Post-hypnotic Amnesia: Discovery: Post-hypnotic amnesia is a remarkable inability
shown by many hypnotic subjects to recall in the normal state events which took place during
hypnosis. The phenomenon was discovered by the Marquis de Puysegur, in 1784, in connection
with experiments on his famous subject Victor. When this subject awoke from his dreamlike
condition, he manifested a complete amnesia for everything which had taken place while it lasted.
Since it was already well known at that time that spontaneous somnambulists or sleep-walkers
have such amnesias, the peasant's condition was quite naturally called artificial or induced
somnambulism. Because of the wholly erroneous belief that Victor, while in this state, possessed
marvelous powers of prescribing medical treatment, Puysegur's artificial somnambulism had a
tremendous local vogue. So many patients came to him that he personally could not treat them
all, and he resorted to Mesmer's old device. He magnetized an elm tree on the village green, to
which the patients could have free access. Fortunately we have what (128) appears to be a
trustworthy contemporary account of the behavior of these subjects.
"Cloquet, an eye-witness, has given us some valuable information on the subject. He says
that the patient's eyes were closed, and there was no sense of hearing, unless it was awakened by
the master's voice. Care was taken not to touch the patient during his crisis, nor even the chair on
which he was seated, as this would produce suffering and convulsions, which could only be
subdued by the master. To rouse them from the trance, the master touched the patient's eyes, or
said, 'Go and embrace the tree.' Then they arose, still asleep, went straight to the tree, and soon
afterwards opened their eyes. As soon as they returned to a normal condition, the patients retained
no recollection of what had occurred during the three or four hours' crisis." The essential point to
be noted in the present connection is that upon recovery there was a complete amnesia for the
events occurring during the crises, exactly as had been observed in Victor, with whose case these
subjects were doubtless familiar.
Since the publication of Liebeault's book on hypnotism (1866), post-hypnotic amnesia
not only has been a recognized symptom of hypnosis, but has taken a specific position in the
hierarchy of its stages or degrees. Subjects showing noticeable amounts of this, in accord with
both the tradition and the logic of Puysegur, are called somnambules. Thus in the hierarchy of
degrees of hypnosis propounded by Liebeault, the two highest, his fifth and sixth, were
characterized by a mild amnesia and a complete amnesia respectively. Bernheim's somewhat
more elaborate classification of degrees of hypnosis makes this phenomenon a leading
characteristic of his sixth, seventh, eighth, and ninth degrees. In general most writers since
Liebeault have considered post-hypnotic amnesia as the most significant single symptom of
hypnosis, and as marking the most profound degree of hypnosis which it is possible to attain.
RECALL BY HYPNOTIZED SUBJECTS OF EVENTS OF PREVIOUS
TRANCES: There is no indication in the sources available to the present author that either
Puysegur or his associates knew of the capacity of subjects while in the trance to recall events of
previous (129) trances, even though, for the subject in question, the latter events are inaccessible
to recall in the normal state. As late as 1843 even Braid, who was familiar with certain of the
French works on the subject, seems quite definitely to have missed this point. He writes: "There
is also another remarkable difference. It is stated that although natural somnambulists cannot
remember, when awake, what they were engaged in when asleep, they have a vivid recollection
of it when in that state again; but I have found no parallel to this in the somnambulism induced by
hypnotism. By this I mean that they cannot explain what happened during the former
somnambulic state, but they may approximate to the words and actions which had formerly
manifested themselves, provided they are placed under exactly similar circumstances. For the
extent to which peculiar manifestations may be brought out by manipulating the head and face, at
a certain stage of hypnotism, ... examples are given of memory as regarded events which
happened during the waking condition, whilst they seemed to have no recollection of what
happened during a former state of hypnotism."
It is a little curious that Braid should have failed to discover this exceedingly obvious
phenomenon, since, as shown by the above quotation, he was familiar both with the parallel
phenomenon in sleep-walking and with the supposed kinship of hypnosis to it.
Although we do not know who may have discovered that trance events amnesic to the
waking condition are normally accessible to succeeding trance states, this particular fact has long
been an accepted commonplace among writers on the subject. Binet and Fere state: "The
hypnotized subject seldom remembers, on awaking, the evems which occurred during his
hypnotic sleep. On the other hand, when he is asleep his memory embraces all the facts of his
sleep, his waking state, and of previous hypnotic sleeps."
Moll remarks: "But ... in some cases, chiefly in the deepest hypnosis, memory of even the
hypnotic proceedings cannot be recalled on awakening ... On the other hand, the subject
remembers in hypnosis all that has happened in previous hypnoses ... Wolfort relates the case of a
woman who remembered in the magnetic sleep all that had taken place in a magnetic sleep
thirteen (130) years before, although in the meantime she had never recollected it."
Wingfield writes: "If a good hypnotic subject, capable of passing into the somnambulistic
state, is put to sleep, ... and is aroused, we find that he has no recollection whatever of what he
was doing during his sleep. If, however, he be sent to sleep again, the memory of the events of his
previous hypnosis will have returned ... "
The memory phenomena of somnambules as outlined in the above quotations are shown
diagrammatically in Figure 27 (not available).
This diagram represents the conventional view and corresponds to the facts usually
encountered. Actually, however, the situation is more complicated. If a subject chances to fall
into a state of true sleep while in the trance (probably the so-called lethargic state, p. 205), events
taking place in the neighborhood of the subject will not be recalled on waking for the obvi0us
reason that they will never have been registered. Again, if a definite suggestion has been given
that the subject will not be able to recall in a subsequent trance the events of the one in which the
suggestion is given, the suggestion will very likely be realized. Indeed, almost any memory
phenomenon that is physically possible to the subject will be encountered if specific suggestions
to that effect are given. Thus the standard posthypnotic amnesia usually is not obtained unless
specific suggestion to that effect is given, and it is doubtful if it ever would be obtained if definite
suggestion to the contrary were given, though this latter point appears not yet to have been
submitted to experiment. (131)
HYPNOSIS AND SUGGESTIBILITY EXPERIMENTAL APPROACH TO THE
PROBLEMS OF POST-HYPNOTIC AMNESIA: Such considerations as the above serve to
raise the fundamental questions of (1) what memory processes actually are subject to
post-hypnotic amnesia and (2) the profundity of the amnesia where existent. Certain casual
observations seem to indicate that post-hypnotic amnesia may prove to be a fairly superficial
phenomenon. It is a well-known fact, for example, that if, just before a subject is waked from the
trance, a suggestion is given him that he shall remember everything that took place while he was
under hypnosis, he will have no amnesia for that particular trance; at least there will be none
evident to ordinary observation. This, coupled with the fact that a subject in one trance state can
recall events of previous trances without any intervening rehearsal whatever, shows very clearly
that the amnesia is not a phenomenon of retention. The memory traces must have been existent
throughout the waking amnesic period; otherwise the registration (learning) would need to be
reenacted before recall could occur in the succeeding trance state.
In this connection it must be noted that, from the present point of view, memory has three
fairly distinct aspects. First, there must be the process of registratton or learning. Second, there is
the physiological state of retention during which the subject is neither thinking of nor rehearsing
what was registered. Third, there is the process of recall, whereby the organism performs some
kind of action, functionally related to the original registration process, which indicates retention.
Perhaps the most obvious and logical method of testing for recall may be illustrated by an
experiment in which paired (132) associates have been memorized. The first or stimulus member
of each pair is presented in systematic order to determine whether this evokes in the subject the
response originally associated with it, that is, the second member of the pair. This may be thought
of as the method of simple reinstatement, since the subject in recall reenacts the original learned
behavior. A second method closely related to the first, but delicate enough to detect memory
traces which are too weak to evoke any recall whatever, is to have the subject relearn the
material. 1£ there is any retention at all, even of a quite subliminal magnitude, the second
learning will require less time than was originally needed. This may be thought of as the method
of reinstatement by relearning. A third method for testing recall, and the one usually employed in
the older hypnotic experimentation, is merely to ask the subject upon awaking what he did while
in the trance. It is important to note that in this latter case the subject's acts in making a verbal or
symbolic report of the action of the trance usually will resemble the latter in no detail whatever.
Instead the report will ordinarily consist of speech movements (movements of the vocal cords,
tongue, lips, etc.), whereas the acts reported (e.g., going to a window and closing it) may have
involved the arms or legs. This form of recall distinctly is not a literal reinstatement. It may
accordingly be spoken of as the method of symbolic reinstatement.
THREE PROBLEMS IN POST-HYPNOTIC AMNESIA ATTACKED BY
STRICKLER: These considerations give rise to a number of specific quantitative questions
concerning post-hypnotic amnesia which were attacked experimentally by Strickle. Suppose there
were available for experimental use a group of subjects who show 100 per cent post-hypnotic
amnesia by the method of symbolic reinstatement traditionally employed in this field. What
percentage of amnesia will such subjects show where a detailed and systematic reinstatement type
of recall is employed? What (133) percentage of amnesia will be found when it is measured by
the more elaborate and searching relearning method? How much, if any, of the inhibitory
tendency superficially existent in posthypnotic amnesia will cling to the material twenty-four
hours after it has been thoroughly relearned in the waking state?
STRICKLER'S EXPERIMENTAL PROCEDURE AND METHODS OF
CONTROL: In the conduct of his experiment Strickler employed two laboratory rooms: a rather
large room with several tables and chairs, and a smaller room where the apparatus was situated.
The subject reported each day at the same time to the large room, and there was put into the
hypnotic trance. He was then gently led into the apparatus room, seated before the memory
machine, and instructed to learn the presented material as quickly as possible.3 The nonsense
material was learned in this way until it was repeated twice in immediate succession without
error. The subject was then told to close his eyes, after which he was gently led back into the
original room and seated in the same chair exactly as when he first went into the trance. He was
permitted to sleep thus for five minutes and then was awakened and directed to walk about and
thoroughly arouse himself. After ten minutes of walking about he was taken into the apparatus
room and presented with the characters, one at a time, in the same manner as during the learning
process, and was directed to give their names if any such occurred to him.
Under such circumstances the subjects invariably said that they had never seen the
characters before, thus showing a perfect symbolic amnesia. However, as the characters, one after
the other, appeared at the window in the detailed reinstatement test, some of the subjects would
occasionally respond with the appropriate syllable. In such cases they stated that the name seemed
to come to mind from "nowhere" and was not accompanied by any recollection that the character
or syllable had ever been encountered before. Following this reinstatementrecall test, prompting
was begun (p. 116) and the subject (135) relearned the material exactly as if he had never seen it
before. Twenty-four hours later both the reinstatement-recall test and the relearning test for the
same series were repeated. All this experimental procedure is summarized systematically in the
first column of Table 18.
SUMMARY OUTLINE OF PARALLEL EXPERIMENTAL PROCEDURES EMPLOYED BY
STRICKLER (I4)
TABLE I8
Trance procedure Control
procedure
1. Trance induced 1. Normal
waking state
2. Learned series in trance 2. Learned series
awake
3. Continued in trance five minutes 3. Trance induced for
five-minute sleep
4. Ten-minute recovery period

4
.
T
e
n
-
m
i
n
u
t
e

r
e
 c
o
v
e
r
y

p
e
r
i
o
d

5. Relearn in waking state 5. Relearn in waking

state
6. Second relearning twenty-four hours later 6. Second relearning twenty-four hours later
The control series was designed to repeat the procedure sketched above with comparable
memory material and with all the experimental conditions exactly the same except that hypnosis
was not involved in the original learning. In the control series, therefore, the subject was led into
the apparatus room, in the waleing state, where he learned the characters just as he had in the
trance series. When the characters were mastered to two perfect successive repetitions, he was led
back into the large room, seated in the chair, put into a trance, and, as in the trance series,
permitted to sleep five minutes. He was then wakened, and directed to walk about and arouse
himself for ten minutes, exactly as before. From here on the control series paralleled exactly the
regular trance series. The control series is summarized systematically in the second column of
Table 18.
The memory apparatus used by Strickler operated substantially like that shown in Figure
26. The bands of characters, mimeographed recording sheets, methods of prompting, etc., were
exactly the same as those employed by Huse and described above . Four university students were
employed as subjects in the investigation. These individuals served in a total of 102 experimental
periods, during fifty-one of which a series of characters was learned in the trance state, while
during each of the remaining fifty-one a parallel series was learned in the waking state. (135)
AMOUNT OF POST-HYPNOTIC AMNESIA AS MANIFESTED BY
REINSTATEMENT-RECALL: The amount of post-hypnotic amnesia as measured by
reinstatement-recall is shown for each of the four subjects in Table 19. An examination of the
third column shows that the subjects vary to a certain extent in the amount of detailed recall
amnesia. The average amount for the entire group is slightly over 97 per cent. This means that on
the average only about one character in thirty-three was read correctly, which is a very high
degree of amnesia. Thus the detailed reinstatement amnesia turns out to agree very closely indeed
with the results obtained by the symbolic recall test employed by the classical hypnotists.
In cases of this kind the question always arises as to whether the trance is the entire active
factor in producing this amnesia. How much loss of memory, for example, would have taken
place had there been no trance whatever? It is the function of the control series to enable us to
answer such questions exactly. The results of the control series show that there is a (136) very
appreciable amount of loss in reinstatement-recall even in so short a period as fifteen minutes. It
varies with the different subjects from 10 per cent to 27 per cent, with the mean at a little over 16
per cent.
The presence of this 16 per cent of normal loss at once raises the question as to how
much, if any, of the 97 per cent loss following the trance was due to the trance, and how much to
true forgetting. If the forgetting could have been reduced to zero, would the amnesia following
the trance also have been reduced by 16.35 per cent (i.e., so as to show only 81 per cent)?
Experimental evidence providing an answer to this question is entirely lacking. Not only this, but
we are utterly without assured guiding principles even for constructing an hypothesis. The
questions involved are, however, of very much more than passing interest. Are the natures of
post-hypnotic amnesia and normal forgetting due to lapse of time sufficiently alike so that they
will combine in a summative manner, as just suggested, or does each operate quite independently
of the other? The problem should yield quite readily to experimental attack, and its solution
might throw considerable light on the nature of posthypnotic amnesia and hypnosis in general.
AMOUNT OF POST-HYPNOTIC AMNESIA AS MANIFESTED BY
RELEARNING: We pass now to the consideration of the degree of post-hypnotic amnesia as
shown by the relearning method. The relearning score was obtained by dividing the mean number
of promptings required to relearn after the trance by the mean number required to do the original
learning of the series in question. This was done on the assumption that if the amnesia were
complete it would require exactly as long to relearn the series as it did to learn it originally, in
which case the relearning or amnesic score would be 100 per cent. The results for the four
subjects are shown in Table 20 (not available).
A strikingly different picture meets us in this table from that in Table 19. Here we find, in
place of a practically perfect amnesia, one of scarcely more than 50 per cent. A second contrast is
found in the amount of normal forgetting: it has dropped almost to negligible proportions, so that
it matters little (137) whether it is deducted from the gross post-hypnotic amnesia score or not. As
in Table 19, there appears to be considerable variability among the subjects. It must also be
remembered that we have here an exceedingly small sample of hypnotic subjects. Even so, the
results are sufficiently consistent to indicate that the post-hypnotic amnesia ordinarily met with,
which appears superficially to be a complete wiping-out of memory, is by no means complete.
This is all quite in harmony with what has been observed in the reacquisition of former
knowledge and skill by patients subject to traumatic amnesia.
STRICKLER'S RESULTS ON THE PERSISTENCE OF AMNESIC INHIBITION
AFTER RELEARNING: We have, finally, to consider the results from relearning the
twice-learned material after a period of twenty-four hours. The data from the several subjects are
shown in Table 21 (not available), both for the reinstatement-recall method and for the relearning
method. An examination of these results reveals a situation of considerable psychological
significance. A comparison of the first two columns shows that everyone of the four subjects
gave a (138) greater failure score on the reinstatement-recall in the trancelearned than in the
waking-learned series, the average number of failures in the former being nearly twice as great as
in the latter. Likewise a comparison of the second two columns shows that everyone of the four
subjects required more promptings to relearn the originally trance-learned series, the average
number of promptings in the former case again being almost twice as great. Despite the small
number of subjects, the probable error of the difference between the two recall means (neglecting
the correlation between the two columns of data) is only 0.412. This is contained 6.79 times in
the difference, which indicates a very satisfactory statistical reliability. The probable error of the
difference in the case of the relearning is 0.846, whereas the difference is 3.26, nearly four times
as great, which again indicates a satisfactory statistical reliability.
The significance of the above results is greatly increased by the fact that, except for the
influence of the trance and its associated amnesia, the ordinary psychological expectation (139)
would have anticipated an exactly opposite result. Ebbinghaus long ago showed that overlearning
of memory material improves the retention. We have already seen that the trance-learned series
required a very considerable amount more of relearning after fifteen minutes than did the
waking-learned series. Computations from Strickler's original data show that the trance-learned
series had actually received 36 per cent more learning than had the control series. It would seem
that this really large amount of what, except for the trance, would have been overlearning, should
have favored the trance-learned series. But even if this did not take place, one might expect, since
at the termination of the fifteen-minute relearning both series were equally well learned in a
strictly objective sense (and in the waking state), that the trance-learned series would have
survived the twenty-four hour forgetting period as well as the control series.
That this did not take place calls for explanation. If these results are confirmed by
subsequent experiment, the indication will be tolerably clear that trance learning and waking
learning, while combining to a certain extent, do not combine in the ordinary arithmetical
manner. Here two and two do not equal four. When we seek for the explanation of this, we find
ourselves largely at a loss. Perhaps as plausible an hypothesis as any is that post-hypnotic amnesia
is in reality a kind of inhibition (possibly related to those Pavlov has investigated), and that this
inhibition is overridden by the relearning but recovers sufficiently after twenty-four hours to
reveal its presence. This problem presents an inviting field for further experimentation. Indeed,
the entire investigation should be repeated, with more subjects learning more series and with the
latter systematically rotated to avoid constant errors from difference in memory materials.
ARE STYLUS-MAZE HABITS SUBJECT TO POST-HYPNOTIC AMNESIA? In
the investigation just summarized we have a comparison under post-hypnotic waking conditions
of symbolic recall, together with two different m~thods of measuring the amnesic influence on
literal reinstatement tendencies. The activity reinstated was oral speech. It happens that, while
oral speech is (140) the basis of the most of our symbolic behavior, it need not necessarily be
symbolic; apparently in the quantitative part of Strickler's investigation it was not. Nevertheless
the intimate relation of oral speech to the symbolic, together with our great ignorance concerning
the symbolic mechanisms, makes it desirable to know the role played by post-hypnotic amnesia
in an activity much further removed from symbolism. This is found in Coors's study of the
post-hypnotic memory for maze learning.
The maze is a device originally introduced into psychology for the purpose of studying
the learning processes of animals. It consists of a complicated pathway leading from a starting
point to a goal, where some reward such as food is usually placed. Scattered through this pathway
at frequent intervals are blind alleys. The first few times a hungry rat runs the maze he will enter
a considerable number of these blind alleys. With successive repetitions, however, these entrances
become fewer and fewer until at length the animal may make several runs in succession without
any errors at all.
Several adaptations of the maze have been made for the study of human learning. One of
the most valuable of these is the stylus maze. This maze usually takes the form of a complicated
pathway grooved out of the surface of a rectangular plate of some hard substance such as metal or
fiber. The size is such that all the parts of the maze are within convenient reach of a person sitting
before it as it rests on a table. The stylus usually consists of a wooden cylinder about the size of a
large fountain-pen, at the lower end of which is an enlargement or flange. Beyond the flange a
small metal rod projects from the handle about an inch. This serves as the tracing point with
which the subject follows the pathway. The stylus is usually held by the subject in very much the
same way as a pen or pencil, but in a nearly vertical position (Fig. 28) (not available). The
wooden flange serves to keep the fingers from sliding down into contact with the maze, which the
subject ordinarily is not permitted to see either during the learning or previously. To this end the
maze is often placed beneath a hood. The subject thrusts his preferred hand into the maze
chamber through a curtain at one side of the hood, and the experimenter observes the subject's
behavior from the opposite side. (141)
METHODS AND RESULTS OF THE MAZE EXPERIMENT: Coors definitely
attempted to parallel with the stylus maze the technique employed by Strickler on rote learning.
To this end the subject was hypnotized in one room and then led into an adjoining room where
the maze was located. Here, while still in the trance, he went through the maze time after time
until it was learned. At each trial the experimenter recorded by number on a special
mimeographed form the blind alleys entered. In addition, the time required for each trial was
taken with a stop-watch. The maze was considered learned when the subject went through it twice
in succession without error. At the conclusion of the learning the still entranced subject was taken
back into the first room and seated in the chair exactly as when he was originally hypnotized.
Here he was left undisturbed for five minutes, after which he was wakened and given exercise for
ten minutes in order to thoroughly rouse him. Then he was taken back to the maze room and
directed to learn the maze. This procedure is summarized in the first column of Table 22.
[TABLE 22: SUMMARY OF TRANCE AND CONTROL PROCEDURES IN COORS'S
INVESTIGATION OF THE EXTENT OF POST-HYPNOTIC AMNESIA FOR THE STYLUS
NIAZE]
Procedure with hypnotic group Procedure with
control group
1. Trance induced
1. Pseudo-trance procedure
2. Subject taken into adjoining room 2. Subject
taken into adjoining room
3. Maze learned in trance 3. Maze
learned in normal state
4. Subject taken back to first room 4. Subject
taken back to first room
5. Subject remains quietly in trance five minutes 5. Subject sits quietly with
closed eyes for five minutes
6. Subject wakened and exercised for ten minutes 6. Subject opens eyes and
exercises for ten minutes
7. Subject relearns maze in waking state 7. Subject relearns
maze still in waking state
Certain differences between the conditions of verbal rote Learning and those of learning
the stylus maze necessitated a number of deviations from a strict parallelism between the
procedures of Strickler and Coors. One of the most important of these is the fact that in rote
learning a rather large amount of fairly comparable material may be assembled without much
difficulty, whereas only a small number of stylus mazes are (142) available. Moreover, these
mazes are known to differ rather widely as to difficulty. It was accordingly impracticable in the
maze study to have each subject run his own control series as had been done in the Strickler
study. Instead, the seven mazes available for the investigation were all learned in a certain order
by each hypnotic subject. The control series was obtained by having the same mazes learned in a
comparable manner but in the normal state by an entirely different group of individuals.
A subject of the control group was first placed in the same chair as were the hypnotic
subjects. Here certain words were spoken and gestures made which somewhat resembled those
used to induce the trance, but which had no hypnotic effect. The subject was then taken to the
maze room, where he learned the maze in exactly the same manner that the hypnotic subjects did.
At the completion of the learning he was returned to the original chair, where he sat quietly with
closed eyes for five minutes, after which he exercised for ten minutes. He then returned to the
experimental room and relearned the maze. The procedure with the control subjects is
summarized in parallel with that of the hypnotic group in the second column of Table 22 (not
available).
In so far as the two processes are comparable, the results (143) of the maze investigation
agree rather closely with those on rote learning. The percentage amounts of loss upon relearning
in the waking state fifteen minutes after a preceding trance learning are shown for the six
hypnotic subjects in Table 23 (not available). The parallel results for the twenty control subjects
are shown in Table 24 (not available). Subtraction of the relearning loss of the control group
from that of the trance group, assuming that true forgetting and post-hypnotic amnesia combine
additively, should give us approximately the amount of loss due to post-hypnotic amnesia alone.
By this method we find that the hypnotic group shows an average excess loss of 45.27 per cent
(P.E.n = 3.7) (144) when measured by number of trials required, and an average excess loss of
46.42 per cent (P.E.D = 9.85) when measured by errors. It is evident that under the conditions of
this experiment manual habits of the stylus maze variety are subject to post-hypnotic amnesia,
though the amnesia is by no means complete.
ARE SPECIFIC PRACTICE EFFECTS IN ADDING SUBJECT TO
POST-HYPNOTIC AMNESIA? Speaking of the hypnotic subjects ordinarily called
somnambuIes, we have seen that those who show perfect post-hypnotic amnesia as tested by
symbolic recall also show on the average scarcely more than 50 per cent amnesia when tested by
the relearning method. The results in this respect do not seem to differ materially in the case of
stylus maze learning from that of memorized nonsense paired associates. There are, however,
several other types of learning which may conceivably yield different results and which therefore
require investigation. On one of these we are fortunate in having available three independent
experimental studies. These investigations are concerned with whether or not post-hypnotic
amnesia is operative on practice effects. From the point of view of the learning mechanisms
involved, practice effects may be of two fairly distinct kinds. The first one which we shall
consider concerns the results of practice on specific excitatory tendencies which, at the beginning
of practice in the trance state, are all fully formed in the sense that they are definitely functioning
above the threshold of recall. Practice in such cases results in a more or less radical increase in
the facility of operation of the excitatory tendencies, but presumably not in the formation of any
new ones. This reveals itself as an increase in the speed of functioning of the previously
established excitatory tendencies. The extent to which post-hypnotic amnesia is operative on this
form of learning has been investigated by Patten (n).
It is obvious that learning of this nature demands a radically different experimental
procedure from that employed in the rote learning and the stylus maze learning previously
considered. The influence of practice is typically revealed in what is known as a practice curve.
As practice proceeds, the efficiency (145) of performance progresses. If the principle of
post-hypnotic amnesia is operative in such a case, the gain in efficiency resulting from a period of
hypnotic practice should not function in the following waking states. That is to say, the speed of
waking performance after a period of hypnotic practice should be exactly the same as the speed
of performance in the waking state preceding the hypnotic practice. If, however, the posthypnotic
amnesia is not at all operative in this function, we should expect that the practice curve of the
waking state following a period of hypnotic practice should be entirely continuous with the latter.
Patten had available for subjects seven college students who showed complete
post-hypnotic amnesia as tested by the customary symbolic recall. In addition he employed seven
students, who were not hypnotized at any time, as a control group. As experimental material it
was necessary to choose something which every college student knows, but which would still be
susceptible of a very considerable amount of improvement in respect to speed. Excellent material
for this purpose was found in a form of complex mental addition which has been widely used in
psychological laboratories. In this experiment the subject was given any two-place number, such
as 86. To this number he would add 6. To the resulting total he next added 7, to that total 8, to
that 9, then 6 again, then 7, then 8, then 9, and so on as rapidly as possible. At the end of thirty
seconds the subject was interrupted by the experimenter, who called out some new number to
which the subject began adding 6, 7, 8, and 9, as before. The subject spoke only the totals as he
obtained them. Thus in case he started out with 86 he would call out in succession the values 92,
99, r07, r6, 22, 29, and so on. After 100 had been reached, the hundreds were dropped from the
totals as indicated in the above series. The experiment was continued for five minutes on each
occasion, and yielded ten sets of additions. For recording the performance in the test, special
mimeographed sheets were prepared. These contained in parallel columns the totals
corresponding to the various starting numbers as far as a subject would be likely to go in thirty
seconds. In cases where a subject made an error at . some place during a thirty-second period,
naturally all of the values from there on would differ from those on the (146) mimeographed
sheet, even though all might be perfectly correct after the error. In such cases the experimenter
jotted down at the side of the corresponding column the totals as actually given by the subject.
The control subjects came to the laboratory at a given time on each of eighteen successive
days, and added as rapidly as possible for five minutes as described above. The average
performance of these subjects for each of the successive days is plotted as the upper or broken
line in Figure 29. An inspection of this composite graph shows a very clearly marked practice
effect throughout the entire period. The group of seven hypnotic subjects also practiced adding
five minutes each day as rapidly as possible for a period of eighteen days. In order (147) to secure
a reliable measure of their speed of performance preceding the trance practice, the hypnotic group
added during their first six days in the waking state exactly as did the control group. On the
second six days, that is, days seven to twelve inclusive, the members of the hypnotic group
performed their practice in the trance state. On the remaining six days, however, they practised in
the normal or waking condition in order to secure an adequate measure of the post-hypnotic
waking speed of adding. The results with the hypnotic group were averaged and plotted in a
manner exactly parallel to those of the control group. If the tendency to post-hypnotic amnesia
were roo per cent operative here, the curve for the six days of waking practice following the
trance practice should begin very close to where the waking practice curve left off preceding the
trance practice. This means that the practice curve for the six waking days following the trance
practice should be practically a duplicate of the trance curve. If, on the other hand, post-
hypnotic amnesia is not at all operative on this activity, the curve for the last six waking days
should begin where the trance curve leaves off, and should constitute a normal continuation of the
latter, rather than a duplicate. The nature of what a normal continuation should be is indicated by
the shape of the curve from the control group.
The results as actually obtained are shown by the solid line in Figure 29 (not available).
There it will be observed that the practice curve during the six waking days following the period
of hypnotic practice is entirely continuous with the latter. It is true that the first performance of
the second waking series (the thirteenth day) is slightly lower than the last trance performance.
This is, however, in all probability nothing more than a sampling error, since we find an almost
identical drop at the corresponding point in the control series. The indication seems to be quite
definite that so-called hypnotic somnambules, unless specific suggestions are given to the
contrary, show no tendency to amnesia in later waking states for practice effects acquired in the
hypnotic trance.
It may be noted that the rate of adding and the amount of improvement through practice in
the trance state as shown by a comparison of the middle sections of the two graphs of Figure 29
(not available) differ in no obvious way from parallel initial adding (148) rate and practice
improvement in the normal condition. Presumably, also, the considerably slower rate of adding
by the hypnotic group is merely a manifestation of the chance inclusion of more slow adders in
this sample. This hypothesis is supported by the fact that a closely parallel study by Life (8) with
similarly sized groups showed an exactly opposite tendency in this respect.
ARE GENERALIZED PRACTICE EFFECTS IN MEMORIZING SUBJECT TO
POST-HYPNOTIC AMNESIA? There remains to be considered the influence of post-hypnotic
amnesia upon another type of practice effect. This form of practice influence is distinctly more
generalized than that observed in Patten's investigation. Instead of involving the facilitation of
excitatory tendencies already existent at the opening of the experiment, this second form of
practice C0ncerns an increase in the generalized capacity for the establishment of such excitatory
tendencies. The activity employed was the memorization of paired associates with the same
apparatus and materials as were employed by I-Iuse (p. II7). It is well known that, as series after
series of such material is learned by subjects hitherto untrained, the rate at which successive
series are mastered increases very markedly and with considerable uniformity. Curves may be
plotted either in terms of the number of promptings required to master each successive series to
two perfect reproductions, or in terms of the number of presentations of the material as a whole
which is required to produce the same degree of perfection in the learning. Since both criteria
show progressive decreases as practice goes forward, such graphs become genuine practice
curves. If, now, a period of hypnotic practice be interposed between two periods of waking
practice in the same function, it will be possible to determine, exactly as was done in Patten's
experiment, whether the generalized improvement in speed of learning which results from the
hypnotic practice does or does not fail to manifest itself in the succeeding waking learning.
The report of such an investigation is contained in an unpublished study by Life. She
employed as subjects six university students, all of whom showed stable post-hypnotic amnesias
as tested by ordinary symbolic recall. She employed a control group, also composed of university
students who were not hypnotized. In general the results of this investigation are somewhat
erratic, presumably because the various sets of nonsense material were not rotated in such a way
as to equalize their inherent differences in difficulty. However, the results, such as they are,
corroborate the outcome of Patten's investigation in considerable detail: The practice curves of
the control and experimental groups parallel each other rather closely even in their eccentricities.
The speed of learning of the hypnotic group appears to be no different in the trance state from
that in the waking condition, and the amount of improvement through practice in the trance
appears to be no different from corresponding improvement of the non-trance control group.
Lastly, and of primary importance in the present connection, the waking performance of the
hypnotic group immediately following the trance practice shows no indication of a tendency to
revert to the pre-trance practice level, as might be expected if hypnotic amnesia were operative.
ARE CONDITIONED REFLEXES SUBJECT TO POST-HYPNOTIC AMNESIA?
In the investigation of the phenomena of post-hypnotic amnesia we arrive, finally, at the question
whether it is operative in the case of the conditioned reflex, one of the most primitive of all
learning and memory processes. The fundamentals of the problem are really very simple: 'Will
subjects manifesting the classical signs of artificial somnambulism show in a supervening waking
state a loss of conditioned reflexes set up in the trance state?
An experiment performed by Scott throws a little light on this intriguing problem. He
stimulated subjects while in the hypnotic state by the sound of a buzzer followed after one-fifth of
a second by an electric shock. After an average of fourteen or fifteen of such paired stimulations,
the buzzer was sounded alone. When this was done everyone of his ten subjects gave reactions
weaker but otherwise substantially like those previously given to the shock, i.e., they manifested
the conditioned reflex. The normal human reaction to a strong electric shock is (150) rather
extensive and varied. The subject will normally jerk his hand off the electrodes which deliver the
shock, and, in addition, there are widespread disturbances set up in the respiratory, circulatory,
and secretory systems. Scott employed an elaborate apparatus by which there were recorded
automatically on smoked paper not only the movements of the subject's hand but his breathing,
his heart action, and the electric conductivity of the skin-the so-called galvanic skin reflex. He
reports numerical data, however, from only the hand movements and the breathing.
The results from all Scott's subjects whose records were finally measured are presented as
composite graphs in Figures 30 and 41. These graphs indicate that with both hand withdrawals
(151)
and breathing changes the subjects show on the average not only a marked tendency to
conditioned reaction when tested in the trance state at once after the conditioning stimulation, but
also a clear tendency of the same kind when tested in the waking state shortly after. This latter
reaction, of course, bears directly on the central issue of the present chapter. These graphs show,
in short, that conditioned reflexes set up in the trance persist in the following waking state. If
complete post-hypnotic amnesia were operative, this would, of course, not take place.
Unfortunately, Scott's data do not enable us to go further and state whether or not a partial
tendency to amnesia exists. It is clear from the graphs that the amplitude of the conditioned
reaction is considerably weaker in the waking state than when tested shortly before in the trance
state. If no other factor were operative, this would indicate a partial amnesia similar to that found
by Strickler. It happens, however, that a conditioned-reflex tendency known as experimental
extinction would also have produced exactly such a weakening. In general, each time a
conditioned stimulus (buzzer) is given without reinforcement (i.e., without shock) the strength of
the reaction tendency is known to grow progressively weaker. The twelve unreinforced buzzer
stimulations (tests) given at the end of the trance would naturally have considerably weakened the
conditioned reaction tendency, as should also the seven tests given in the following waking state.
Unfortunately, the progress of the weakening process in the respective states is not sufficiently
smooth and regular to enable us to judge with much confidence by an inspection of the curves, as
was done in the case of Patten's experiment above, whether there is any real break in the
continuity of the extinction process at the point of transition from the trance to the waking state.
There is, however, no special indication in the graphs of such a break, and Figure 41 (not
available) presents a distinct appearance of continuity.
To make certain whether there is any shrinkage in the amplitude of the conditioned
reactions in excess of that produced by experimental extinction it would be necessary to perform
a special control experiment. A fairly satisfactory experiment of this kind would involve
conditioning a large squad of subjects in the normal state and then testing them twelve and seven
times respectively, with pauses and all other circumstances exactly (152) like the corresponding
conditions imposed on Scott's subjects. A comparison could then be made of the relative amount
of shrinkage exhibited by the two groups. By using sufficiently large numbers of subjects,
provided appropriate statistical treatment were given the data, an accurate determination could be
made as to whether there is any tendency toward greater weakening in the amplitude of the
conditioned reactions on passing from the trance to the waking state than for similar tests all
made in the waking state. If a statistically reliable difference of this kind should be found, it
would indicate the existence of a tendency to post-hypnotic amnesia for at least this particular
form of conditioned reflex. It would also indicate something of the magnitude of the tendency.
Naturally, final decision in this matter must await the performance of the experiment. Meanwhile
the most we can say in the light of Scott's results is that the discovery of any considerable degree
of post-hypnotic amnesia in this field, apart from specific suggestions to that effect, is distinctly
improbable.
IS POST-HYPNOTIC AMNESIA A PRIMITIVE PHYSIOLOGICAL
MECHANISM DISTINCT FROM SUGGESTION? It is a question of the greatest significance
for an understanding of the nature of hypnosis whether there ever is posthypnotic amnesia
without some kind of suggestion to bring it about. There is no doubt that some subjects who have
been profoundly influenced by the ordinary hypnotic suggestions show post-hypnotic amnesia
without ever having received any direct suggestion to that effect from the experimenter. It is
never possible, however, to be certain that the subject has not heard in some obscure way that
when waking from the hypnotic trance people cannot recall the trance events. But even if that
contingency were ruled out, there remains the fact, as pointed out by Bramwell, that the subject
when waking from true sleep invariably is oblivious to all that has transpired during the sleeping
interval. This, it would seem, should be quite enough to account for an occasional spontaneous
autosuggestion that there will be an oblivion to the trance events after waking from that state.
The experimental results summarized in the preceding pages (153) agree very well with
this general hypothesis. The experiments indicate that the results of a specific training in adding
and of a generalized training in memorizing almost certainly are not subject to post-hypnotic
amnesia. If post-hypnotic amnesia were a primitive physiological tendency, it is hard to see how
this could be. And while not so conclusive, the experimental results involving the effects of
specific practice on rehearsing memory material and on the conditioned reflex are entirely
consistent with this view. On the other hand, the fact that a substantial amount of amnesia was
found in the results of the ordinary learning procedure of Strickler and of Coors is quite to be
expected from the fact that the subjects had become habituated, through the influence of direct
suggestion administered previous to the experiment, to being ignorant in the waking state of
events taking place in the trance. That this amnesia, even though merely the result of suggestion,
is of a fairly deepseated nature is indicated by its apparent persistence: it is still evident after a
complete relearning of the amnesic material in the normal state.
No doubt, if a different technique had been employed, somewhat different results might
have been obtained in some or all of the investigations summarized above. If, for example, in the
investigations of Strickler and Coors specific suggestions had been given to the effect that the
subject would not be able to relearn the material in the waking state, it is entirely likely that the
suggestion would have been realized. Some color is given to such a view by a recent paper of
Wells (IS) In a similar manner it may be anticipated with some confidence that if the
investigations of Patten and Life were to be repeated with the specific suggestions, given just
before the termination of the trance, that the subjects after a certain amount of improvement in
trance practice would revert to their pretrance practice status upon resuming practice in the
waking state, this also would be realized more or less completely.
As the result of these various considerations, then, we arrive at the conclusion that
post-hypnotic amnesia is not a primitive physiological mechanism but that, instead, all of the
phenomena are probably produced by suggestion in some form. However, as we have already had
repeated occasion to notice, the evidence is still uncertain on a number of points, and until this
uncertainty is remedied by adequate experiment the above conclusion must be held subject to
revision.
SUMMARY: Subjects ordinarily classed as somnambules, but who have received no
suggestions of any kind regarding post-hypnotic amnesia if the experiments, have been tested for
post-hypnotic amnesia a number of different memory functions. These subjects uniformly deny
any recollection of trance events, i.e., as tested by general symbolic recall, amnesia is 100 per
cent. By detailed specific recall this amount of amnesia is reduced for nonsense monateria1
probably to about 97 per cent. By the relearning method the amnesia falls to approximately 50
per cent. Mantlal habits learned in the stylus maze show by the relearning method an amnesia
also of about 50 per cent. With specific training in arithmetical addition and general training in
memorizing nonsense material the amount of post-hypnotic amnesia is reduced to zero. The
experimental results, while inconclusive, probably also indicate that post-hypnotic amnesia is not
operative in the case of conditioned reflexes.
B. POST-HYPNOTIC SUGGESTION: The second major post-hypnotic phenomenon
which has been connected with the theory of dissociation is post-hypnotic suggestion. This
appears to have been discovered very early in the history of hypnosis, along with post-hypnotic
amnesia and the other major hypnotic phenomena. Moll, referring to posthypnotic Suggestion,
writes: "The old mesmerists observed some cases of it. In 1787 Mouillesa\\x ordered a lady in the
hypnotic state to pay a visit to a certaitJ. person next day; the command was exactly obeyed (Du
Prel). Kluge Noizet, and Schopenhauer mention other cases. POSt-hypnotic suggestion has been
much studied lately, particularly- by Liebeault, Richet, Bernheim, Delboeuf, Gurney, and Ford,"
A typical and authentic picture of post-hypnotic suggestion is sketched by Bernheim:
(155) "I wish to speak of the possibility of inducing in somnambulists by means of suggestion,
acts, illusions of the senses, and hallucinations which shall not be manifested during the sleeping
condition, but upon waking. The patient hears what I tell him in his sleep, but no memory of what
I said remains. He no longer knows that I spoke to him. The idea suggested arises in his mind
when he wakes, but he has forgotten its origin, and believes it is spontaneous. Facts of this kind
have been observed by A. Bertrand, Gen. Noiset, Dr. Liebeault and Chas. Richet. I have repeated
these observations successfully many times in a large number of hypnotic cases and am
convinced of their accuracy."
As indicated by the quotation from Bernheim, post-hypnotic suggestion presents two main
aspects-the subjective and the objective. A concrete example of the former would be the
suggestion to a subject that at a certain time after waking from the trance he will hear the clock in
the ncar-by tower strike the hour of eight. The subject is likely to report hearing the bell and may
even be induced to count aloud the individual strokes as they are hallucinated. Because of the
relative inaccessibility of subjective phenomena to experimental investigation, the more exact and
quantitative work has been directed toward the objective processes. As yet, however, but two
such studies have appeared in this rich field. Both of these arc concerned with changes in the
strength of post-hypnotic suggestion tendencies to action with the lapse of time, more or less
analogous to the curve of forgetting in memory investigations.
While the quantitative investigation of the durability of posthypnotic suggestion is of
fundamental significance for a comprehensive knowledge of hypnotism, this problem has an
added interest arising from the attempts which have been made to utilize this form of suggestion
in clinical practice for the treatment of stammering and various behavior difficulties. The general
nature and proportions of the curve of diminution of this phenomenon with the passage of time
have become of special significance in this connection because it has been found that striking
improvements in symptoms observable during the trance too often disappear disappointingly soon
after its (156) termination, and in spite of the use of vigorous post-hypnotic suggestion.
There are two conditions under which the progress of diminution in the strength of
post-hypnotic suggestion may be traced: one which simulates, in a controlled way, the
circumstances under which it ordinarily declines in clinical practice; and another which conforms
to the simpler and more abstract requirements of pure science, whereby a curve of the tendency
may be plotted as an uncomplicated function of time. The former condition involves repeatedly
testing the strength of the same posthypnotic suggestion over a considerable period. Obviously
this repeated testing involves the question of practice effects as well as that of fading-out of the
tendency from mere passage of time. The second condition, however, should yield the pure and
uncomplicated curve of the weakening of post-hypnotic suggestion. We are fortunate in having a
controlled experiment on each of the two conditions, the former by Kellogg, the latter by Patten
(IO).
DURABILITY OF POST-HYPNOTIC SUGGESTION UNDER CONDITIONS OF
REPEATED TESTING: The post-hypnotic suggestions employed by Kellogg were that the
subject while reading the even-numbered pages of a book would breathe twice as fast as usual,
but when reading the odd-numbered pages he would breathe only half as fast as usual. The book
in question was a small volume by Edna St. Vincent Millay, upon each page of which appeared a
single sonnet. It was assumed that, in general, there would normally be no difference in average
rate of breathing on the two types of pages. But if, after the post-hypnotic suggestion had been
given, a mean difference in rate between the two should appear, this would indicate ina
convenient manner the existence of the influence of post-hypnotic suggestion, and the relative
difference in rate would indicate the magnitude of the influence. Moreover, as fast as the strength
of the post-hypnotic suggestion weakened with the passage of time, the natural physiological
tendency to a constant rate in speed of breathing would progressively restore the rate of the
respective segments of the breathing record to an equality. Accordingly the ratio of the (157)
mean breathing rate while reading the odd pages to that while reading the even pages should
serve as a useful objective index on the basis of which to trace the course of the declining
strength of post-hypnotic suggestion.
The subjects were hypnotized but once during the experiment. At this time the
post-hypnotic suggestion described above was given, as wel! as some practice in carrying out the
suggestion. At the same time post-hypnotic suggestions were given with a view to diverting the
subject's waking suspicions from the real purpose of the experiment. Five subjects showing
complete post-hypnotic amnesia, as tested by symbolic recall, were employed in the
investigation. They were each tested fifteen minutes after waking from the trance, then after one
day, eight days, fifteen days, twenty-two days, and more or less irregularly at longer intervals, the
last test being at ninety-nine days from the original giving of the suggestion.
The apparatus involved primarily a Sumner pneumograph placed about the chest of the
subject. This was connected by a rubber tube to a recording tambour, the stylus of which was in
contact with the smoked paper on the kymograph. In addition to the record of the breathing thus
obtained, time was recorded (158) in five-second intervals by an electromagnetic signal marker
connected with a laboratory clock situated in a distant room, and a third line, controlled by a key
under the experimenter's hand, indicated the instant of shift from odd to even page and vice versa.
One of Kellogg's typical records is reproduced as Figure 31 (not available).
The numerical results from the five hypnotic subjects, together with results from four
non-hypnotic subjects run under substantially parallel instructions, are shown in Table 25 (not
available). An inspection of this table shows that the subjects are rather variable in their reactions
and, after the twenty-second day, the data are rather fragmentary. In order to pool the results and
thereby eliminate some of the influence of chance factors, each subject's results were reduced to a
percentage of his initial (practice) performance. These results, when averaged for each period
after waking, yield fairly consistent values. These averages are represented graphically for the
hypnotics and the control subjects separately in Figure 32 (not available). (159)
This composite graph shows at a glance the substantial outcome of the investigation. The
non-hypnotic control subjects show at first a fairly marked rise in the strength of the tendency
above its original practice value, after which it remains relatively constant. The true
post-hypnotic amnesia, on the other hand, displays a marked and fairly continuous weakening
during the first three weeks of the post-hypnotic period, after which the loss appears to be much
slower. Apparently a single administration of post-hypnotic suggestion is sufficient to maintain in
many subjects a disposition to arbitrary action strong enough to override perceptibly a mild
physiological tendency (that of uniformity in breathing rate) for a period of at least two months.
The interpretation of the results from the normal control group is difficult because these
subjects were not given strictly parallel treatment in the sense of that employed by Williams. It is
quite possible that if this were done the difference between the curves of the two groups would be
decidedly less marked, though there is evidence, as we shall presently see, that there is probably a
genuine tendency for responses to a single waking command to be more potent in controlling
behavior and to disintegrate less rapidly than do the results of post-hypnotic suggestions.
DURABILITY OF POST-HYPNOTIC SUGGESTIONS WHEN
UNCOMPLICATED BY PRACTICE EFFECTS: While the results of Kellogg's investigation
probably giye a fair indication of the course of the decline in strength of post-hypnotic
suggestions where the occasion for carrying out the acts suggested occurs repeatedly, it has the
defect that the natural rate of decline is complicated by prac:tice effects. Moreover, there is no
way of knowing, apart txom experimental investigation, whether the effect of repeated
performance would be to retard the loss of the post-hypnotic tendency or to hasten it. With the
experimental results of Kellogg before him, Patten set himself the task of plotting the pure curve
of post-hypnotic suggestion uncomplicated by practice effects.
It is clear that the possibility of distortion of the curveo! diminution of post-hypnotic
suggestion by practice may be met (161) by giving a post-hypnotic suggestion to a relatively
large number of subjects and then testing each subject on but one occasion; one group of subjects
may be tested after one day, another after three days, another after five days, another after ten
days, and so on. In general, except for sampling errors, the group tested after ten days should
show the same mean strength of reaction tendency as the group tested at five days, unless the
mere passage of time has produced an effect. Conversely, the mean scores from distinct groups
tested after varying lengths of time should (except for sampling errors) show the true quantitative
course of the tendency in question, quite uncomplicated by practice effects. This laborious and
difficult but effective method was adopted by Patten. He tested a total of eighteen totally amnesic
subjects on one occasion each as follows: two after a zero interval, two after one day, two after
three days, two after five days, two after ten days, two after fifteen days, two after twenty days,
two after twenty-five days, one after thirty days, and one after thirty-three days.
Patten presented to his subjects in the post-hypnotic (waking) state series of thirty words,
one at a time, by means of an exposure apparatus of the general type used by Ruse and shown in
Figure 26. Each list of thirty words contained one word which appeared twice. In order to secure
the subject's attention to the words of the list, he was instructed in the normal state to watch for
this repeated word so as to be able to report which it was at the conclusion of the experiment.
This was the subject's conscious task. In addition, and without relation to the foregoing, each list
contained three names of animals. In the preceding trance state the subject was given the
post-hypnotic suggestion that the moment he saw one of these words he would depress promptly,
but with om conscious knowledge, his right forefinger. The finger in question was placed within
a small stirrup and was concealed from the subject's view by a screen. The stirrup was connected
by means of appropriate cords with an ink recording device which accurately registered both the
time and the amplitude of the subject's reactions.
Patten, like Kellogg, also ran a control group of non-hypnotic subjects who were simply
given a parallel verbal command. Of the sixteen subjects in this latter group, two were (162)
tested after one day, two after three days, two after five days, two after ten days, two after fifteen
days, two after twenty days, two after twenty-five days, one after thirty days, and one after
thirty-three days.
As an index of the strength of suggestion, Patten took the mean amplitude of the finger
responses made to the three animal words in his lists, assigning a zero amplitude where no
reaction whatever occurred. He publishes a graph of the mean values for each time-interval
group. Because of the extremely small size of the several groups these values, considered
individually, present a somewhat erratic appearance. It has been possible by a careful reading of
an original graph, kindly (163) furnished by Professor Patten, to pool days 0, I, and 3, days 5, 10,
and IS, days 20 and 25, and days 30 and 33, thereby reducing the distortion due to meager
samples. The four mean values thus obtained have been plotted for both the experimental and the
control group (Fig. 33) (not available). In general this graph shows a fairly smooth and consistent
diminution in the strength of post-hypnotic suggestion during the first month of its duration. The
curve is one of negative acceleration, i.e., the fall is comparatively rapid at first and then
gradually slower, as is the very general rule with behavior functions. It is probably significant
that at the end of a month the tendency is hardly more than a fifth its original strength, a
materially greater diminution than that shown by Kellogg's subjects after a similar period. The
control group also shows a loss, though far less in amount than the hypnotic group, quite as
Kellogg's results have prepared us to expect. The results of the two studies of post-hypnotic
suggestion thus corroborate each other as to the general shape of the curve of loss; a5 to the fact
that an appreciable amount of the tendency, when opposed by a very weak physiolo~ical
counter-tendency, survives for at least a month; and as to the fact that post-hypnotic suggestions
are markedly weaker than parallel instructions given in the waking state.
THE INFLUENCE OF REPEATED PERFORMANCE ON THE STRENGTH OF
POST-HYPNOTIC SUGGESTIONS: The characteristically lower values both for hypnotic
and for control subjects in the Patten study than in that of Kellogg lends some presumption in
favor of the hypothesis that practice may tend to strengthen a post-hypnotic suggestion instead of
weakening it. Fortunately, Patten made a special investigation of this question. After several of
Lis subjects had completed their single test for the experiment just described, he continued giving
them repeated tests daily, using new words on each occasion. The results of two of these subjects
are representative of the entire group. They were tested at the zero interval originally and then
daily for thirteen days. The composite graph of these results shows that the vigor of the response,
while slightly variable, displays no tendency; whatever (164) to fall but, if anything, a slight
tendency to rise. Patten believes that with daily practice post-hypnotic suggestion might persist
indefinitely without renewal of the suggestion. However this may be, it is evident that the
repeated performance of the posthypnotically suggested act characteristic of clinical practice
would seem to be favorable for maintaining its strength. This presumably accounts for the slower
rate of weakening displayed by Kellogg's subjects over that shown by those of Patten.
SUMMARY: We have seen two techniques by which it is possible to measure the
strength of a post-hypnotic suggestion and to trace its course as it declines with the lapse of time.
It has been found by these methods that the curve of diminution in the strength of post-hypnotic
suggestion is most rapid at first and becomes progressively slower as time goes on, in this respect
resembling the curve of ordinary forgetting. Control groups given parallel instruction in the
waking state probably show a decline, but one decidedly less in amount. Thus a post-hypnotic
suggestion, at least for commonplace action, appears to be dtcideclly less effective as a
motivating device than ordinary waking instruction. Lastly, the evidence is fairly clear that daily
repetition of post-hypnotically suggested acts tends strongly to prevent their normal disintegration
for some weeks at least. (165)

15. CH 7: HYPNOSIS AND THE DISSOCIATION HYPOTHESIS: CLARK HULL:

HYPNOSIS ANDSUGGESTIBILITY: APPLETON-CENTURY-CROFTS: NY: COPYRIGHT
1933: 1961
WE have had more than one occasion to observe (p . 18if.) that hypnotism throughout its
entire history has been closely associated with medical practice. In this association it has too often
played the role of a substitute for medicine in the treatment of human ills. Since hypnosis
probably possesses no influence beyond the purely psychological, it was inevitable that in the end
it should become associated especially with mental disorders. This connection has profoundly
influenced the interpretation of hypnotic phenomena. It is our task in the present chapter to
examine the evidence bearing on a widely accepted hypothesis which owes its origin to this
relationship, viz., that hypnosis is essentially a state of dissociation. Our consideration of
dissociation at this time should be somewhat aided by the fact that the two preceding chapters
have already familiarized us with a number of phenomena intimately related to dissociation: the
supposed recovery of lost memories in the hypnotic trance, post-hypnotic amnesia, and
post-hypnotic suggestion.
JANET'S CONCEPTION OF DISSOCIATION: Pierre Janet gives perhaps as clear an
exposition of the dissociation hypothesis as anyone. Quite typically he illustrates his account with
cases from his clinical experience. One of his patients was a young girl of twenty years, called
Irene. This poor girl had for a long time cared for her mother who was dying of consumption, and
at the same time had worked as a seamstress to provide the bare necessities of their lives. Death
came at last, with its characteristic gasping, blood-vomiting, and other ghastly accompaniments.
The girl, terrified, tried frantically to bring the breath back to the lifeless body. In the attempt the
corpse fell to the floor. After frantic (167) efforts the girl was finally able to lift the body back to
the bed. It is scarcely surprising, after all these frightful experiences, that Irene should have
developed a severe neurosis.
The neurosis took the form which Janet calls monoideic somnambulism. When in the
somnambulistic condition, Irene was likely to rehearse the whole gruesome episode of her
mother's death. Like Lady Macbeth in the famous sleep-walking scene of Shakespeare's play, she
appeared to be oblivious to everything about her. If spoken to, she would not answer. Even
though her eyes were open, they appeared to see nothing but what was relevant to her dreamlike
preoccupation. However, when the somnambulistic episode had ended and she had regained her
normal comprehension of her surroundings, she remembered nothing of what she had just been
doing and saying -there was a complete amnesia for the entire episode. It developed, upon further
investigation, that in the normal state she also had a complete amnesia for the incidents connected
with her mother's death. She had, indeed, no recollection at all of that tragic event except as it had
been told her.
It is important at this point to note that amnesia is fundamental in the dissociation
hypothesis. It was assumed that the act of normal recall is the result of association. Quite
naturally, then, the failure of the power to recall events which normally should be recalled, and
which at one time perhaps were really susceptible to recall, was regarded as an interruption or
destruction of the normal associative tendencies. It consequently was called dissociation.
According to this hypothesis, the painful episode in the life of Irene is regarded as having given
rise to a group of memories which, while closely associated with one another, are separated, i.e.,
dissociated from, the main body of her memories. Janet has attempted to illustrate the condition
of Irene by the diagram reproduced as Figure 34 (not available). In explanation of the diagram he
says: "Allow me to represent to you this system of psychological facts, which constitutes an idea,
by a system of points connected together by some lines forming a sort of polygon. The point S
represents the sight of the face of the dead mother, the point V is the sound of her voice; another
point, M, is the feeling of the movements made to carry up the body, and so on. This polygon is
like the system of thoughts which was developed in the mind and in the brain of our patient Irene.
Each point is connected with the others, so one cannot excite the first without giving birth to the
second, and the entire system has a tendency to develop itself to the utmost. But at the same time
in healthy minds these systems pertaining to each idea are connected with an infinitely wider
system of which they are only a part,-the system of our entire consciousness, of our entire
individuality. The remembrance of the mother's death, even the affection Irene feels for her
mother, with all the memories that are connected with it, forms only a part of the whole
consciousness of the young girl with all its memories and other tendencies. Let this large circle,
P, near the little polygon represent the whole personality of the girl, the memory of all that
happened in her previous life. Normally, in good health, the little system must be connected with
the large one, and must in great part depend on it. Generally the partial system remains subject to
the laws of the total system: it is called up only when the whole consciousness is willing, and
within the limits in which this consciousness allows it. (169)
"Now, to picture to ourselves what has taken place during somnambulism, we may adopt
a simple provisional resume. Things happen as if an idea, a partial system of thoughts,
emancipated itself, became independent and developed itself on its own account."
DISSOCIATION AND THE CONCEPT OF THE SUBCONSCIOUS: Careful study
of other hysterical phenomena brought to light evidence which seemingly pointed not only to the
existence of isolated, self-integrated groups of memories or ideas, but suggested that these
dissociated aggregates could, and frequently did, live a life of their own and might in very rare
instances somehow become a kind of secondary personality. Sometimes the secondary personality
might even attain a temporary dominance over the patient's behavior, producing characteristic
changes in his character. When not dominant, the secondary personality (dissociated memory or
habit system) managed somehow to maintain a fairly active existence, though the patient usually
had no articulate or conscious knowledge of the fact. Since, according to the current psychology
of Janet's day, the ability to give verbal report was regarded as a test of consciousness, and since
these memories were ordinarily not able to activate the symbolic speech processes in such a way
as to be reported, the dissociated memory complex came to be regarded as somehow submerged
and beneath the conscious. Such memories were accordingly said to be subconscious. Sometimes
they were said to reside in the subconscious, a kind of dim psychological realm where all sorts of
strange and bizarre happenings might take place. Curiously enough, these dissociated habit
systems, even though subconscious, apparently were able to influence the overt behavior of the
patient, sometimes to a marked degree. Unfortunately this influence too often led to more or less
serious individual and social maladjustments. In short, these eccentricities of behavior constituted
the symptoms of a psychoneurosis. (170)
PRELIMINARY EXPERIMENTS BASED ON THE DISSOCIATION
HYPOTHESIS: Now, under such circumstances, it clearly is most important for
psychopathology that the characteristics of this subconscious activity be discovered. The proper
procedure would be, of course, to do this by experiment. But few persons would have the
temerity to provoke a neurosis deliberately in order to study its history, however desirable the
resulting knowledge might be. It was found, however, that what appeared to be dissociations of a
closely similar but harmless nature could be induced experimentally by hypnosis. This was
naturally regarded as opening up the entire realm of the subconscious to experimental
investigation, and it was soon discovered that subconscious integrates produced in this manner,
even though strictly unconnected with the speech systems of the patient, might become tolerably
articulate through the activities of an accessory language system, namely, that of handwriting. By
a rather simple experimental procedure the hand could be induced to write fairly elaborate scripts
automatically, i.e., apparently without any conscious knowledge on the part of the subject as to
what was being written. This is known as automatic writing. It was discovered by this procedure
that the dissociated or subconscious system of ideas was often capable of fairly complex
intellectual achievements. Among these may be mentioned acts of judgment, the composition of
brief original letters, and fairly complicated arithmetical computations. These results were
somewhat paradoxical, since such acts normally are performed only when the subject is clearly
conscious of the processes in question. Indeed, it has often been thought that such processes could
not take place without "conscious contro!." It was therefore inferred that the dissociated system
must be in some sense or other genuinely conscious. It followed from this as a logical necessity
that in such cases there must exist a kind of double consciousness, two simultaneous
consciousnesses somehow coexistent in association with the same body. As a means of
emphasizing this simultaneity or coexistence, the secondary or dissociated consciousness was
frequently called the co-conscious.
This form of experimentation, like so many other (171) innovations, seems to have
originated with Pierre Janet. The subject, in this instance a somnambule by the name of Lucie,
was first hypnotized. During the trance the post-hypnotic suggestion was given that when, in
saying over a series of letters, the experimenter would give in immediate succession two letters
alike, the subject would execute unconsciously some trivial act to show that the succession of
similars had been recognized. After awaking the subject, the experimenter murmured "A, C, D,
E, A, A," etc. Immediately upon the pronunciation of the two A's, Lucie gave the sign, thereby
indicating to Janet the performance of an unconscious judgment of resemblance.
On another occasion Janet gave Lucie the post-hypnotic suggestion that upon awaking
from the trance she was to write some kind of a letter. After waking Lucie took part in a casual
conversation. While this was going on her hand proceeded automatically to write the following
quaint note: “Madam "I shah not be able to come Sunday as I had intended. I pray you will
forgive me. It would give me great pleasure to come to you but I can not accept for that day.
"Your friend, "Lucie. "P. S. Best of regards to the children, please." Presently, when Janet
showed her the letter, she was unable to understand it, and believed that the doctor had copied her
signature.
A third experiment involved computation. Lucie was given the post-hypnotic suggestion:
"You are going to multiply by writing 739 by 42." When she awoke, she carried out the
arithmetic in detail. At the same time Lucie told Janet of her day's work and did not stop speaking
once while her hand calculated correctly.
PRINCE'S ATTEMPT TO TEST THE DISSOCIATION HYPOTHESIS: Some
months after Janet's publication of the experiment£ sketched above, Morton Prince published
some very similar ones (8). His chief subject was a woman with a dual personality. One
personality is designated as A, the other as B. (172) State A had an amnesia for the events
experienced by B, but B had normal recall for the events of both A and B. State B, however,
could be hypnotized. The hypnotic state is represented by b. After waking from the trance B had
an amnesia for b, its own trance state. Personality B made vigorous claims to Dr. Prince that
when state (or personality) A was conscious, i.e., dominating the oral speech mechanism,
personality B or b was actively and vividly conscious. Thus b was at such times, in a literal sense,
co-conscious. Of all this, state or personality A professed to have no knowledge whatever.
Prince seems to have regarded these remarkable allegations with a certain degree of
skepticism. He accordingly carried out a number of experiments with the avowed purpose of
testing the subject's claims. One of the experiments took the same general form as Janet's
computational experiment already described, except that the induced alternation of the
personalities seems largely to have replaced hypnosis.
"Accordingly it was agreed with b that when co-conscious she would do a particular sum
in arithmetic while A's consciousness was engaged in another task. The figures to be added ...
were not to be given until A was present. ... It was accordingly agreed that A's attention should be
engaged by having her write some verses with which she was familiar on a sheet of paper, on the
upper and lower margins of which the required figures for the sum would be unobtrusively
written. Co-conscious b was to perceive these figures and make the computation .... A was kept
entirely in the dark regarding the nature of the experiment, and was not even informed that an
experiment as such was being made .... It was further arrangecl with b that as soon as A had
completed the verse she was to be changed to b, who was to give the answer immediately ...
before she could have time to make the computation from memory after the change to b, in case it
had not been done co-consciously."
The number 53 was written on the upper left-hand corner of a sheet of paper eight and
one-half by eleven inches in size, and the number 61 in the lower right-hand corner. This was
presented to the subject when in the A state, with the request that she write on the paper the
stanza, "Tell me not in mournful numbers," etc. This she proceeded to do, repeating aloud the
words as she wrote them, making oral comments on the (173) poet's thoughts, on her bad
memory, and so on. Prince regards this as of importance because of its indication that state A was
alert and fully dominant throughout the writing process and, by implication, that b was
subconscious throughout the period. Moreover, there appeared to be little evidence of
interference with the conscious process by the subconscious computation supposedly going on
simultaneously. "Only once (the first experiment) she stumbled a bit while vrriting, as if there
was a momentary inhibition of her thoughts by the subconscious process." \Vhen the writing of
the stanza was completed, the patient was changed to b. "Immediately on appearing, b exclaimed,
almost shouted: which is correct." A number of variants of the above procedure are said to have
yielded substantially similar results. When questioned while in the dominant condition, the
personality which had performed the calculation in the co-conscious status reported in
considerable detail the characteristic incidents of the computational process, thus indicating a
detailed memory for it. These various evidences led Prince to the conclusion "that such
perceptions, interpretations, calculations, and translations could have been made by pure
physiological processes without thought is inconceivable and not substantiated by anything that
we know of physiological processes." He concludes, in short, that his "experiments to determine
co-conscious (subconscious) ideation" had yielded a conclusive affirmative.
TWO SIMULTANEOUS INDEPENDENT CONSCIOUS PROCESSES
REPORTED BY BURNETT: Seventeen years after the publication of the investigations of
Janet and Prince, Burnett reported in voluminous detail what amounts substantially to a
continuation of the same type of experimentation by a strictly hypnotic technique. His
experiments are different chiefly in that his three subjects were thoroughly normal as contrasted
with the hysterical individuals employed by Janet and Prince. Burnett's procedures varied
somewhat from one experimental session to another, but the following is typical of those which
he regarded as furnishing strict proof of the simultaneous existence of two independent conscious
processes: The subject is hypnotized and told that (174) after waking he is to multiply
subconsciously a pair of numbers, the product to be written automatically in the waking state but
without the subject's knowledge. One of the pair of numbers to be multiplied is the number of
taps the experimenter makes while the subject is looking at a book. The second of the pair is to be
that on the right-hand corner of the page fixated by the subject at the moment when the
experimenter asks for the book. At this point the subject is awakened from the trance by a
prearranged signal. Upon preliminary questioning, the subject now insists that he is awake and
does not recall anything from the trance just concluded. The subject is then handed a copy of
AEsop’ss Fables, and told to find a page containing three fables. While the search is in progress
the experimenter taps three times. He then takes the book from the subject w]len the latter has it
open at page 41 and asks him to repeat some verses which he knows well. The recitation is
performed easily and without obvious distraction. Meanwhile the hand, presumably behind a
screen, is writing over and over the number 123, which is the correct product of the
multiplication. Burnett believes that we have here evidence that two tasks have been performed
simultaneously, each quite independently of the other. As evidence that the computation was
subconscious, the subject at the very moment when the writing is in progress insists in reply to
questions that his right hand is doing nothing and that he is fully awake. Actually he appears to be
very sleepy but resists motor suggestions, which fact would tend to indicate that he has not
relapsed into the trance, since he normally obeys such suggestions when hypnotized.2 Upon being
put back into the trance, the subject is able to recall the fact of automatic writing just performed
and the nature of the multiplication which led to it. Thirteen experiments of this general nature
were performed, each of which was regarded as offering satisfactory proof of a dissociated
co-conscious activity. In Burnett's phraseology, the experiments indicated a "splitting of the
mind."
As a kind of control series, Burnett performed six additional experiments in which he
gave the same subjects simultaneous (175) tasks to perform, both to be carried out consciously.
An example of these tasks is seen in the fourth of the series, in which the subject was asked to
write the words of a poem well known to him and to read aloud simultaneously from !Esop's
Fables. The technique employed in these controls did not yield any definitely quantitative
measures, though Burnett was of the opinion from observing the behavior of the subjects in the
two contrasted series that their performances of two simultaneous tasks were distinctly less
efficient when both tasks were conscious than when one was subconscious.
In summarizing the investigations of Janet, Prince, and Burnett, it is well to remember
that there are implicit in them two fairly distinct questions: (1) Is it possible by means of hypnosis
to bring about an intellectual process within a subject's body and yet have the subject remain in
professed ignorance of its existence while it is going on? (2) Is the dissociation, which is
supposed to effect the results indicated in, of such a nature that two distinct intellectual processes
may go on quite or nearly independently of each other, i.e., is the wall or barrier between the
conscious and the subconscious (or co-conscious) of such a nature that one process does not
interfere with the other under conditions in which interference normally takes place? There seems
little room for disagreement with the conclusion of the authors on the first question, i.e., that
post-hypnotic suggestion rather easily brings about a condition in which a person will perform
fairly complex intellectual tasks and yet will deny all knowledge of the process at the very
moment that it is taking place. Burnett's work is particularly convincing on this point. But in
answer to the second question the experiments cited above are not nearly so conclusive.
Characteristics of Adequate Experimental Procedures for Testing the Dissociation
Hypothesis: It should be noted at this point that there are two distinct ways in which two tasks
may interfere with each other. The first form of interference occurs when the attempt is made to
perform two tasks at the same time; the second, when they are performed in succession. The
former obviously applies to a (176) great variety of processes, as everyone knows. The latter
tendency appears to be limited to the learning processes-especially those in which the material
learned in succession is very much of the same kind. Because a following learning activity
disturbs the retention and subsequent recall of material previously learned, the successive form of
interference is known as retroactive inhibition. The implications of the principle of retroactive
inhibition for the present problem have only recently been recognized. As already pointed out,
Janet, Prince, and Burnett have utilized the more obvious simultaneous method and seem more or
less aware of the theoretical implications of an experimental outcome showing interference; this
is especially true of Burnett. Unfortunately, the experimental technique employed by these
authors was quite inadequate either to prove or to disprove the dissociation hypothesis. We are
fortunate, however, in having two later investigations, one by each of the two methods, which are
fairly well controlled. We shall consider first the investigation by the method of simultaneous
interference, leaving the problem of successive interference for a later section. Owing to the
importance of the technique involved, it will first be desirable to consider rather carefully certain
principles of experimental methodology.
There are two distinct experimental approaches to the testing of the dissociation
hypothesis from the standpoint of simultaneous interference. One approach is: first, to determine
the amount a subject can perform within a given period of a task (A), such as consciously reading
aloud at top speed while undisturbed by any other distracting task or activity; secondly, to
determine the rate of performance of another activity (B), such as continuous addition, the latter
to be performed subconsciously at top speed but without any simultaneous conscious task; and,
finally, to determine the amount performed of tasks A and B simultaneously when the subject is
in a dissociated condition, task A being performed consciously and task B, subconsciously. Now,
if dissociation is complete in the sense of a functional independence such as might be expected in
the minds of two different persons or in the completely separated portions of a disembodied
consciousness such as is contemplated by metaphysical idealism, task A and task B should each
be executed quite as rapidly when performed (177) simultaneously as when each is performed at
a different time. But if there should prove to be a very considerable loss in efficiency in one or
both tasks when simultaneous performance is attempted, the findings would be fatal to the
hypothesis of complete dissociation.
This brings us to a second approach to the problem of simultaneous interference.
Dissociation may be conceived as being partial instead of complete. The implication of such an
hypothesis would be that a certain amount of mutual interference might be expected from
simultaneous conscious and subconscious (or co-conscious) activities, but that this should be
considerably less in degree than when the attempt is made to execute both tasks consciously. The
magnitude of this difference, if any, would indicate the extent of the dissociation. The test is
made by determining (Part I) the amount of interference when tasks A and B are performed
simultaneously, A consciously and B subconsciously, as compared with that resulting from (Part
II) a simultaneous execution in which A and Bare both performed consciously. If the interference
observed in Part I is significantly less than that found in Part II, the hypothesis is verified; if not,
it is disproved.
Part I of the procedure just outlined may be considered as the main experiment, and Part
II as the control experiment. As so often has happened in the history of hypnosis, Janet and
Prince performed Part I but neglected the control experiment. Burnett, however, recognized the
scientific necessity of performing both parts of the experiment and attempted to perform them.
Unfortunately his procedure did not secure quantitative or numerical measures of the efficiency
of performance in either condition, which defect alone would prevent decision of a purely
quantitative question such as that before us. But even if he had done so, the tasks given in his
experimental and control series were so far from comparable that quantitative results would have
been quite meaningless for the present problem. For example, in his control experiment sketched
above, the writing of a poem from memory while continuously reading aloud from a book is
offered as a parallel to a subconscious solution of a simple arithmetical problem while engaged
consciously in a casual conversation. There is not the remotest reason for believing these pairs of
tasks to be of equal (178) interference potentiality. On the surface the attempt to do continuous
writing while continuously reading aloud might well be expected to produce far more indications
of mutual interference than would the other two processes, since casual conversation is by no
means continuous in its objective manifestations. Because of this presumptive inequality a
constant error, leading to the possibly incorrect belief that less interference took place in the state
of dissociation, might easily have arisen. Therefore, we can give no weight to Burnett's
impression that dissociation eliminates or greatly reduces mutual interference of simultaneous
tasks.
MESSERSCHMIDT'S EXPERIMENTAL TEST OF THE DISSOCIATION
HYPOTHESIS: A fairly well-controlled experimental test of the dissociation hypothesis was
carried out by Messerschmidt. Her procedure was based on the simultaneous type of interference
and conformed substantially to the method outlined in the preceding section. Partly because the
number of different sub-determinations in her experiment is rather large and might therefore
prove confusing to the reader, and partly because the technique employed offers a convenient
means of summarizing the important methodological principles just referred to, the essentials of
Messerschmidt's procedure have been presented systematically in Table 26 (not available) .
An examination of the procedure indicated for the main or dissociation experiment
outlined in Table 26 (not available) will show that Messerschmidt utilized two combinations of
simultaneous tasks: in one combination subconscious addition was performed simultaneously
with conscious oral reading; in the other the subconscious addition was performed simultaneously
with oral addition. The object of employing two simultaneous combinations rather than one, as
strictly demanded by the method, was to secure evidence concerning the relative mutual
interference in the dissociated condition of two degrees of similarity in the simultaneous tasks.
The expectation natural to a physical psychology would be that the more alike two simultaneous
processes are, the greater the degree to which they probably would involve the same portions of
the body in their execution, which (179)
would result in a greater mutual interference. Thus two forms of simultaneous addition, even
though the overt performance of one form is manual and the other is oral, should display more
mutual interference than would attempted simultaneous reading and adding. But if, on the other
hand, dissociation is a state permitting of genuine functional independence of two simultaneous
processes, no such difference in degree of interference should be found.
TABLE 26
A CONDENSED OUTLINE OF MESSERSCHMIDT'S PLAN OF EXPERIMENTAL
PROCEDURE IN HER INVESTIGATION OF THE ALLEGED INDEPENDENT
OPERATION OF CONSCIOUS AND SUBCONSCIOUS PROCESSES
I. THE MAIN OR DISSOCIATION EXPERIMENT. Two simultaneous tasks, the one performed
consciously, the other subcon, sciously, both as rapidly as possible for a period of five minutes.
Combination of tasks (A):
(1) Oral reading of Edman's Human Traits (conscious) .
(2) Serial addition of 7, 8, and 9 by automatic writing (subconscious)
Combination of tasks (B):
(1) Oral serial addition of 6,7, and 8 (conscious)
(2) Serial addition of 7,8, and 9 by automatic writing (subconscious)
II. SINGLE-TASK CONTROL EXPERIMENT. Work performed as rapidly as possible for five
minutes.
(A) Oral serial addition of 6, 7, and 8 (conscious).
(B) Oral reading of Edman's Human Traits (conscious).
(C) Written serial addition of 7, 8, and 9 (consC1:ous).
(D) Written serial addition of 7, 8, and 9 (subconscious)
III. CONSCIOUS SIMULTANEOUS-TASK CONTROL EXPERIMENT.
Work performed as rapidly as possible for five minutes.
Combination of tasks (A):
(I) Oral reading of Edman's Human Traits (conscious) .
(2) Written serial addition 01 7,8, and 9 (conscious).
Combination of tasks (B):
(I) Oral serial addition of 6, 7, and 8 (conscious)
(2) Written serial addition of 7, 8, and 9 (conscious). (180)
The subconscious additions were performed by automatic writing in which the writing
hand, script, etc., were hidden from the subject's view by means of a small screen. The subject
was instructed in the trance state that after waking she should, at the signal, "Ready," add
cumulatively and as rapidly as possible to a given two-place number such as 37 the digit 7, then
8, then 9, then 7, then 8, then 9, etc., in recurring cycles continuously until the signal "Stop" was
given. The totals thus arrived at by the subject were to be written quite unconsciously. The
conscious oral additions were performed in substantially the same manner except that the digits to
be added were 6, 7, and 8, and the totals were to be spoken aloud. The totals arrived at by the
subject in the latter procedure were checked by the experimenter on a convenient mimeographed
form. The score by each procedure was the number of correct additions performed in five
minutes. The score on the reading test was the number of words pronounced within the same
period.
To parallel the main experimental series, just described, Messerschmidt ran two distinct
control series (Table 26). In the first control experiment (Section II of the table), she merely
determined the speed at which each of the three activities mentioned above could be performed
consciously and alone. In the second control experiment (Section III of the table), she determined
the speed of the several activities when performed in the same two simultaneous combinations as
were employed in the main or dissociation experiment but with the difference that both the
simultaneous tasks were performed consciously.
Messerschmidt employed three young college women as subjects. All showed a perfect
post-hypnotic amnesia and a strong positive response to post-hypnotic suggestion. The three
subjects carried out the procedure indicated in Table 26 as the (180) main or dissociation
experiment a total of eleven times, and that of the single-task control experiment an equal number
of times. Only one subject eX) served in the conscious simultaneous-task control experiment with
a total of only three experimental periods.
MESSERSCHMIDT'S RESULTS: The mean results of the three subjects are shown in
parallel columns for the main or dissociation experiment and for the single-task control
experiment, in Table 27. A mere glance at the values for the different subjects in the body of this
tabl~ shows that the simultaneous execution of tasks in the dissociated condition enormously
reduces the score from that which is obtained by the conscious performance of but a single task.
The mean values from the group of subjects as a whole show that the number of words read drops
from 1,II2.6 to 653.0, and the number of additions of 6, 7, and 8 drops from 152.3 to 58.3. The
interpretation of the data obtained from the subconscious written addition of 7, 8, and 9 is
hampered by the fact that portion D of the single-task control experiment was not performed, so
that we do not know the efficiency of this activity in the subconscious state as a single task. If,
however, we assume as a first approximation that the rate of subconscious addition of 7, 8, and 9
written under a screen would not differ materially from the same activity performed consciously,
we find it dropping from 158.9 to 46.6 when competing with conscious oral reading, and to only
21.8 when competing with. conscious oral addition of a very similar nature.
The results are even more striking in indicating a high de" gree of mutual interference of
conscious and subconscious pro-cesses when the efficiency of the execution of the various tash as
performed simultaneously is expressed in percentages of the corresponding single-task capacity.
These values are shown separately for the several subjects in Table 28. There it may be seen that
when the conscious and subconscious processes are rather dissimilar, the average simultaneous
efficiency of the two is only 42.15 per cent, while the corresponding average where the conscious
and subconscious processes are very similar drops to only 26. I 5 per cent of single task
efficiency. Indeed. (182) the above values show that there is considerably more mutual
interference than would result if the two competing processes frankly alternated, since in that case
the average efficiency value should be reduced to only 50 per cent. These results are, of course,
diametrically opposed to the hypothesis of a functional independence of simultaneous conscious
and Subconscious processes.
Incidentally, Table 28 throws interesting light on the relative capacity of conscious and
subconscious processes to survive when placed in direct competition. Averaging the results for
simultaneous similar and dissimilar tasks, we obtain a mean conscious efficiency of 47.8 per cent,
whereas the average subconscious efficiency under the same circumstances amounts to only 20.1
per cent. While these results doubtless reflect to some degree the peculiarities of the particular
processes involved, they probably indicate in some sense a distinct feebleness and lack of vitality
on the part of subconscious processes.
We have seen (Table 28) that simultaneous conscious and subconscious processes are by
no means functionally independent. There remains the secondary question of whether there is any
independence whatever. This should be shown by the "Conscious Simultaneous-Task Control
Experiment." As (184) indicated above, Messerschmidt's results on this portion of the experiment
are meager, since based on only one subject. Such as they are, however, they suggest that when
both processes are conscious there is even less mutual interference than when one of them is
subconscious. If these results are found by subsequent experiment to be substantially correct, the
indication will be that whatever else so-called hypnotic dissociation may be, it is not a functional
independence between two simultaneous mental processes.
THE DISSOCIATION HYPOTHESIS AND RETROACTIVE INHIBITION:
Having canvassed the available evidence concerning the mutual simultaneous interference of
supposedly dissociated processes, we turn now to the question of whether such processes show a
successive interference. As already pointed out (p. In), it has long been known that if one
memorizing activity follows rather soon after another, the first will suffer an appreciable amount
of interference. This will show itself as a reduced efficiency in recall, when this is tested after a
certain rest interval, as compared with the effectiveness of recall of comparable control series not
followed by learning activity. It is clear, however, that if a dissociated process, such as hypnosis
has been alleged to be, is completely insulated from the conscious or normal waking processes as
two disembodied minds might be insulated from one another, learning in one state could not
influence learning previously performed in the other state and therefore retroactive interference
could not take place.
Concrete consideration of this question reveals two distinct combinations which might
conceivably yield very different interference tendencies, since the waking state in artificial
somnambules is normally dissociated from the trance states as indicated by the amnesia, whereas
the trance state normally is not dissociated from the waking state as judged by the same criterion.
Therefore one experimental combination would be to have the original learning performed in the
waking state and the inhibitory learning shortly after in the trance state. The other combination
would be to have the original learning (185) performed in the trance state and to have the
inhibitory learning follow soon after in the waking state. In view of this asymmetrical amnesic
relationship, it might reasonably be expected (assuming the dissociation hypothesis) that the first
combination would show a greater immunity to mutual interference than the second.
Bearing on this phase of the question there is but one piece of experimental work, a recent
study by Mitchell. Her two subjects showed complete post-hypnotic amnesia as indicated by
inability to recall trance-learned memory material in an immediately following waking state. The
memory material employed consisted of series of ten three-place numbers. These numbers were
presented to the subjects by means of the memory machine shown in Figure 26. The machine was
set to make the exposures of the individual numbers for a period of approximately two seconds
each. The drum of the machine went round and round until the material could be correctly recited
by the subject, at which point the learning was terminated. Each revolution of the drum was
counted as a "triaL" Mitchell (186) investigated each of the two possible phases of this problem
as outlined above and ran a parallel control experiment on each. A schematic outline of her
procedure in all four portions of the experiment is presented in Table 29 (not available).
Mitchell reports her results in terms of the percentage of saving on the relearning of a
series subjected to the possibility of inhibition over the effort required to learn the series
originally. The percentage score was computed by the following formula: It is clear from the
structure of this formula that the savmg is scored on a lOa-point scale, and that a small value
means much inhibition and a large value means relatively little inhibition.
A condensed summary of Mitchell's experimental results is shown in Table 30 (not
available). Each entry represents the mean from four experimental determinations of retroactive
inhibition on as many days. A glance at this table shows that upon the whole these two subjects
produced scant evidence for the dissociation hypothesis. On the waking-trance combination the
subjects show an average saving of only 19.5 per cent, whereas the control (waking-waking)
shows an average saving of 48 per cent. If dissociation were such as to prevent all retroactive
inhibition of one learning process by the other, the saving on the waking-trance series should
have been something approaching 100 per cent. The same general results appear with the
combination of trance-waking, which shows a saving of only 11.0 per cent against 23.0 per cent
for its control. Dissociation thus
(187) appears to have produced an increase in the retroactive interference, rather than a reduction.
Because of the paucity of data here presented, it is impossible to say whether this
increased interference in the supposedly dissociated conditions is significant or not. That the
dissociated conditions should show equal interference would not be surprising, but that they
should show an amount of interference greater than that when the two states are of the same
nature is difficult to understand. Some additional color is lent to this possibility, however, by the
fact that a very similar indication was presented by Messerschmidt's second control experiment.
But one thing, at least, is as clear as such limited data could make it: the experiment gives no
support whatever to the hypothesis that the type of hypnosis known as artificial somnambulism
involves a functional independence from the activities of the normal waking state. (188)
SUSCEPTIBILITY TO HYPNOTIC "DISSOCIATION" AND RESISTANCE TO
SIMULTANEOUS-TASK INTERFERENCE: A third quantitative controlled experimental
attack on the problem of functional dissociation was made by Barry, Mackinnon, and Murray (I).
They reasoned that if hypnosis is an induced state of dissociation, people who are susceptible to
hypnosis must be more readily dissociable than individuals who are relatively resistant to
hypnosis. It was supposed, further, that persons easily dissociable by the hypnotic technique
should tend also to dissociate spontaneously into two functionally independent states if they were
trying to perform two difficult simultaneous tasks, and that in so far as this dissociation took
place there should be a diminution in the amount of mutual interference between the pair of
processes. These authors accordingly made a rather elaborate determination of the hypnotic
susceptibility of eighteen subjects. Each of these individuals was also measured in his capacity to
perform each of eight tasks or tests singly, i.e., one at a time. They were then required to perform
the tasks in pairs simultaneously, and the amount of resulting mutual interference was
determined.
The pairs of tasks were as follows:
1. Tapping test and opposites test
2. Threading needles and solving problem of reversed clocks
3. Inserting pegs in holes and checking numbers
4. Sorting cards and making up words from a gr:oup of letters From these subjects there were
then chosen two groups. The nine hypnotically susceptible were placed in one, and the five least
susceptible in the other.
The average interference scores for each of the eighteen subjects is shown in the
right-hand half of Table 3 I, (not available) and in adjoining columns are shown by asterisks the
individuals of the two contrasted groups. The mean interference index of the hypnotic group is
58.13, whereas that of the non-hypnotic group is 70.84. The difference of 12.71 points is in the
direction called for by the hypothesis. The P.E.D (4.01) is rather large, but the critical ratio of the
difference is 3.17, a value frequently accepted as indicating a tendency. In this connection it may
be pointed out, as a possible explanation of the tendency indicated by the above difference, that
the hypnotic group appear to have (189) even a greater advantage on the performance of the
single tasks than they have in immunity to mutual interference on the double tasks. The critical
ratio of the difference between the two groups in the latter case rises to 3.79. The indication
would seem to be that the reduced double-task interference on the part of the hypnotically
susceptible subjects is a function of their greater ability to do the tasks rather than of any
dissociation mechanism.
Finally, despite the inconclusive outcome of this investigation, it deserves commendation
for the courageous manner in which a psychiatric dogma has been submitted to experimental test.
The more general such experiments become, the sooner will psychiatry (as well as psychology
itself) cease to be a welter of inconsistent and unsubstantiated hypotheses. (190)
SUMMARY AND CONCLUSIONS: We may now summarize the experimental results
bearing on the dissociation hypothesis as applied to hypnosis. Post-hypnotic amnesia is a fact as
well established as anything in hypnosis. Furthermore, there can scarcely be any doubt that
complex intellectual processes such as continuous addition may go on and the subject orally deny
all knowledge of the activity at the very time it is taking place. It is difficult to understand how
these phenomena could exist without some very special suspension or interference of the normal
association processes. In this limited sense it would seem that the concept of dissociation has
some reason for existence.
In view of the clinical interests of Janet and Prince, together with the notions of the nature
of mind prevalent at the time, it is not surprising that this solid basis of fact should have led to the
extension of the principle of dissociation to include something quite different, i.e., a functional
independence of the processes dissociated. The significance of the clever experimental approach
of Barry, Mackinnon, and Murray is not entirely clear. So far as it goes, it possibly points in the
direction of dissociation as a functional independence, though a quite different explanation seems
more likely, and the authors quite properly decline to draw from it conclusions in favor of the
dissociation hypothesis. The results of Messerschmidt and Mitchell, on the other hand, suggest
rather strongly that the whole concept of dissociation as functional independence is an error. It is
to be hoped that the situation is now sufficiently clarified that the near future will see a series of
well-controlled, large-scale investigations which will completely remove the uncertainties which
at present becloud this extremely important problem. (191)

16. CH 8: HYPNOSIS CONCEIVED AS SLEEP: CLARK HULL: HYPNOSIS AND

SUGGESTIBILITY: APPLETON-CENTURY-CROFTS: NY: COPYRIGHT 1933: 1961
SOME TYPICAL VIEWS CONCERNING HYPNOSIS AND SLEEP: MESMER
seems never to have associated in any way his theories of animal magnetism with the phenomena
of sleep. This association appears first to have been made by the Marquis de Puysegur, in 1784,
as the result of observations made on his famous subject, Victor. Because the behavior of this
youth after being magnetized resembled that of a sleep-walker, Puysegur regarded his state as an
artificially induced somnambulism. From that day to the present, the concepts of sleep and what
later came to be known as hypnosis have been intimately related. Probably the majority of the
writers on the subject, up to the present time, have tended to regard the hypnotic trance as
differing in no way from natural sleep except in degree or in the artificial means by which it is
brought about.
In 1813 the Abbe Faria gave demonstrations of what he called "lucid slumber." As a part
of his technique of inducing the state he employed the emphatic command, "Sleep!" Liebeault's
views, as stated by Bernheim, are as follows: "Ordinary sleep does not differ from hypnotic sleep.
The one is, like the other, due to the fixation of the attention and of the nervous force upon the
idea of sleep."
On his own behalf, Bernheim is quite as emphatic. He says: "I have tried to show that suggested
sleep differs in no respect from natural sleep. Some of the phenomena of suggestion can be
obtained in natural sleep, if one succeeds in putting (193) one's self in relationship with the
sleeping person without waking him." 1
Forel is of the same opinion. He says: "The relationship of hypnosis to normal sleep is
unmistakable, and I agree with Liebeault when he says that the former is only distinguishable in
its essence from the latter by the fact of the connection between the sleeper and the hypnotist."
Binet and Fere, who oppose the Nancy school in so much, seem to be in substantial
agreement with Bernheim and his followers on this point. They say: " ... at other times it
[hypnotic sleep] does not differ from natural sleep, and it is then termed slight hypnotism.
Between natural sleep and the most profound hypnosis it is possible to establish an unbroken
chain of intermediate states which it is somewhat difficult to distinguish from each other."
As an example of a more recent opinion, and one emanating from the psychoanalytic
movement, there may be cited a statement from Schilder and Kauders: "The fact that sleep may
be produced by suggestion through hypnosis is beyond dispute. The identity of the sleep
produced by hypnosis and spontaneous sleep is evident also from a number of more specific traits
... It is possible to secure contact with any good sleeper during the progress of his sleep ... He will
receive orders and usually has amnesia with regard to the conversation; also he will carry out
post-hypnotic orders."
James Braid (1843), in the subtitle given to his classical work on hypnosis, calls the state
"nervous sleep." The relation of the state to sleep is further emphasized by the derivation of the
term hypnotism, which Braid coined. Nevertheless, Braid is very clear and decided in the
expression of his opinion that the resemblance between hypnosis and natural sleep is merely
superficial. He says (5, 126): "The most striking proofs that it is different from common sleep,
are the extraordinary effects produced by it ... In passing into common sleep, ... the limbs become
flaccid from cessation of muscular tone and action ... In the hypnotic state, induced with the view
to exhibiting what I call hypnotic phenomena ... [the limbs are] maintained in a state of tonic
rigidity for any length of time I have thought it prudent to try ... In passing into natural sleep,
anything held in the hand is soon allowed to drop from the grasp, but, in the artificial sleep now
referred to, it will be held more firmly than before falling asleep ... "Of the circumstances
connected with the artificial sleep which I induce, nothing so strongly marks the difference
between it and natural sleep as the wonderful power the former evinces in curing many diseases
of long standing, and which had resisted natural sleep ... "
Bramwell may be cited as a more recent writer who also regards hypnosis and natural
sleep as essentially different. He says (6,310): "Delboeuf said: 'I put the subjects to sleep; or,
more correctly speaking, they believe that they have been asleep.' This conception of the
condition is, I think, the true one. In deep hypnosis the subjects believe that they have been asleep
because on awaking they are unable to recall what has happened."
BASIC PRINCIPLES FOR AN EXPERIMENTAL APPROACH TO THE
PROBLEM: The phenomena both of hypnosis and of natural sleep are not only elusive but a
number of the few established facts concerning them have only recently become generally
known. It is, accordingly, scarcely a source of surprise that great differences of opinion as to their
relationship should have existed even in the recent past. It will be our task in the present chapter
to bring to bear on this question the results of the modern experimental approach in an attempt to
decide the issue. The procedure in its essentials is simple. It is to find a few basic and objectively
measurable phenomena which clearly differentiate natural sleep from the waking condition, and
then determine experimentally whether the trance state yields results essentially like those
obtained in the sleeping state. If the hypnotic condition is found to differ in one or more
important respects from true sleep, the two states will thereby be differentiated.
Two such differentize are available to us from the ordinary observation and everyday
practice of life. One of these is the (195) fact that upon waking a sleeper is unable to recall the
events which took place in his presence during the period of sleep. This is the test which is
usually applied in deciding whether we ourselves have been asleep. Another differentiating
criterion is the fact that a sleeper does not give his normal responses to stimuli of moderate
intensity, such as spoken or whispered commands.
A third method of differentiating between sleep and the waking state comes from the
laboratory. It is based on the fact that certain reflexes show very marked reduction or even
complete abolition during sleep. This fact was first reported by Lombard, in 1887. While engaged
in an investigation of the patellar reflex with himself as subject, he noticed one morning, while
feeling very drowsy, that he gave a number of abnormally low reflex kicks. The relationship has
been fully confirmed by numerous experiments during the years since Lombard's original
observation was made. Among those who have contributed to the confirmation are Bowditch and
Warren, Noyes, Lee and Kleitman, and Tuttle. Bowditch and Warren, as well as Tuttle, made an
additional observation of the greatest practical significance in this connection. It was that many
subjects, contrary to what might naively have been supposed, tend to fall asleep with remarkable
facility when stimulated regularly and continuously with heavy blows on the patellar tendon.
BASS'S ATTEMPT TO DIFFERENTIATE HYPNOSIS FROM TRUE SLEEP:
With these facts available it was a relatively straightforward procedure for Bass to set up an
experiment the results of which promise to clarify a confusion that has existed nearly a century
and a half. The major apparatus problems involved had already been solved by Tuttle. The
subject, a trained hypnotic, lay down on the modified barber's chair shown in Figure 35 (not
available). One leg was slightly raised by a padded support (I) placed beneath the knee so as to
facilitate the action of the reflex. The apparatus mounted on the adjustable support at the right
automatically delivered a blow of the hammer C to the patellar tendon at D every 9.68 seconds.
The action of (196) the hammer was controlled by the large cam B, which, in turn, was activated
by the constant-speed motor A through an appropriate system of gears. When in use the motor
and gear system were enclosed in a felt-lined box, which reduced the sound to a soothing drone.
The dotted position of the hammer at C' shows approximately the position from which it fell in its
pendular swing preceding its delivery of the blow. The cam was so constructed that it caught the
hammer at its first rebound from the blow and at once began to lift it for the next blow.
The automatic recording of the individual knee-jerks was accomplished through the stout
line H which extended from the heel of the subject's shoe to the vertical lever J. As J was pulled
to the right against the weak spring K, its upper end moved a hinged arm L to the left. One end of
L rested on the broad strip of continuous-roll white paper R and held a glass capillary-pointed
recording pen, the top of which just shows beyond the roller S. The paper was drawn at a
constant rate beneath the pen by a pair of rubber-surfaced rollers, the lower of which was
activated by the silent adjustable-speed phonograph motor Q. A typical record of kicks obtained
in this way is shown, greatly reduced, as the lowest line of Figure 36 (not available).
It was desired to secure a second and corroborative objective indication of when the
subject was in true sleep. To this end there was placed on the end of the roller shaft at P a special
cylinder which made an electric contact at intervals of approximately ninety seconds. This contact
produced a single faint but easily audible tap on an electric bell, a stimulus at which the subject,
in the waking state, was instructed to press the button at M. The incidence of the bell stimulus
and the responses of the subject by voluntarily pressing the button were separately recorded by
means of glass pens attached to electromagnetic markers. These reactions are indicated by
notches in the two upper lines shown in Figure 36 (not available).
Lastly, there was placed around the subject's chest the tube-like Sumner pneumograph (N,
Fig. 35), which communicated by means of the tube N' to a tambour provided with a recording
pen, the top of which is shown at O. A somewhat atypical (because rather shallow) respiratory
tracing given by this method is shown as the fourth line of Figure 36 (not available).
At each experimental period each subject was given 100 strokes on the tendon (A) in the
waking state, (B) in the trance state, and (C) in the natural sleeping state. These states were
rotated systematically so as to avoid constant errors due to the sequence.
Seven young men were used as subjects. Since they had all served in at least one other
hypnotic experiment, they were accustomed to laboratory surroundings. All showed complete
post-hypnotic recall amnesia except two, and these showed a partial amnesia. All showed prompt
and vigorous catalepsy by rigid clenching of the fist at a whispered suggestion while in the trance.
Moreover, at the conclusion of each experimental period the subject was questioned regarding his
impression of the depth of his sleep. Three of the subjects (Co, Ri, and St) always reported a
sound sleep in the sense that they were unable to recall what took place during a part of the
period. The remaining subjects reported various degrees of drowsiness or light sleep and,
occasionally, sound sleep. There is evidence, however, that these latter subjects were often in a
deeper state of sleep than they realized, since they frequently showed in ability to recall events of
a period during which they reported themselves as having been merely drowsy. The majority of
the seven subjects served through six experimental periods each, the total amounting to
forty-three. (199)
BASS'S RESULTS ON THE KNEE-JERK CRITERION: An examination of Figure
36 (not available) gives an excellent impression in brief space of the outcome of the experiment.
The record opens at the eighty-sixth patellar stimulus of the sleeping series. There was a trace of
response on this kick and another at the one hundred and third stimulus, but otherwise there was a
complete abolition of the kick. At W the subject was wakened, and at once the kicks appeared in
full strength. The amplitude of the kick is seen to have varied somewhat, and at one stimulus it
entirely failed to appear. At T the subject was put into the trance. Owing to the frequent
hypnotization of the subject, this took place instantly. It may be seen that the kicks do not
disappear, as would be expected if the trance were true sleep. Instead, they undergo no change
whatever beyond what results from the ordinary variability characteristic of this reaction. Parallel
with the above it may be noted that the first two bell stimuli shown at the left of the record, which
took place while the subject was asleep, were followed by no voluntary pressure of the key. On
the other hand, all those bell stimuli taking place in both the waking and the trance states were
reacted to without exception.
In all, the reactions to 12,900 patellar stimulations were recorded, measured as to
amplitude, and tabulated. These amplitudes have been averaged in groups of five, and the results
of all seven subjects pooled so as to secure a picture of the process in each state uncomplicated by
accidental factors and individual peculiarities. From these averages were plotted composite
curves of the amplitude of the kicks as obtained in the three states throughout the series of 100
stimulations. These are shown in Figure 37 (not available).
All three curves manifest a fairly marked tendency to fall throughout the progress of the
100 stimulations. Our main interest, however, is in the fact that the curves of the waking and
trance states follow a practically identical course, whereas that for sleep differs radically from
both. At the beginning, as might be expected, the sleeping curve is nearly as high as the others,
since recording began within one or two minutes after the giving of the directions to try to sleep.
Thus the curve (200) really represents, particularly at its beginning, the transItlon from the
waking to the sleeping states. The fact that the process of falling asleep during the sixteen
minutes available proceeded no further than deep drowsiness in many cases makes the results of
the experiment especially convincing as to the differentiation of hypnosis from natural sleep. A
technique sensitive enough to differentiate mere drowsiness from the waking state surely ought to
differentiate the trance from the waking state (201) if, as has been supposed, the former is true
sleep. It also makes the results of Bass's experiment particularly relevant to those theories which
regard hypnosis as a kind of drowsiness transitional between waking and sleeping.
A detailed statistical presentation of the results of the patellar stimulation in the three
states is shown in Table 32. The mean values shown were derived from the second fifty
stimulations of each subject so that the sleeping condition would have had time to become fairly
pronounced. It may be seen that the difference between the mean amplitude of the knee-jerks in
the waking and hypnotic states is only 0.31, which is quite insig~ nificant, since the critical ratio
is only 0.49. The difference between the trance and true sleep, on the other hand, is marked.
Without exception the individual subjects show a lower mean value in sleep. The difference
between the means of the second two columns of data is 7.60, which has the convincing critical
ratio of 10.83. The difference between the means for sleep and for waking amounts to 7.29,
which has the eminently satisfactory critical ratio of 13.11. (202)
BASS’S RESULTS ON THE VOLUNTRARY-ACTION CRITERION: A parallel
tabulation was made of the number of times that each subject pressed the key to response to the
bell signal in each state. From these scores it was possible to plot a set of curves showing the
course of the responses throughout the sixteen minutes or so of each of the three states, exactly
parallel to those of Figure 37 (not available). They are shown in Figure 38 (not available). Here
we (203) observe in an entirely distinct response a picture essentially identical with that observed
in the patellar reflex. The waking and trance reactions follow practically the same course, though
the efficiency in the later minutes of the trance seems to be a little lower than that of the waking
condition at the corresponding point. The sleeping curve, however, while naturally beginning at
the same point as the other two, falls sharply and fairly continuously throughout the period. This
voluntary reaction to the bell stimulus thus also clearly distinguishes both the trance and the
waking states from that of sleep.
It was not possible to compile from Bass's published report the mean values for his
subjects during the last half of the experimental periods to parallel the values shown in Table 32.
Bass does, however, give a table showing the mean values for the entire period of stimulation.
This is reproduced as Table 33. There it may be seen that the waking and hypnotic states yield
closely similar mean values, whereas the true sleeping state differs radically from both. The
hypnotic state, however, shows slightly lower efficiency (something less than 2 per cent) on (204)
the average, than the waking state. The critical ratio of this difference is 2.9, which suggests that,
though small, it may indicate a genuine tendency. The differences between the waking and trance
states on one hand and true sleep on the other are enormous in spite of the fact that these data
include the scores taken from subjects during transition from the waking state into that of sleep.
The critical ratio of the difference between waking and sleep is 28.7, and that of the difference
between trance and sleep is 12.6-both highly satisfactory.
We may say, then, on the basis of two entirely distinct criteria, that of the knee-jerk
(which is primarily physiological) and that of the voluntary response to a signal (which might be
classed by some as primarily psychological), that hypnosis is sharply and definitely differentiated
from true sleep. Or, to put the matter in neurological terms, the results show that sleep involves
an inhibition of both the lower and the higher centers of the nervous system but that hypnosis
shows scarcely a trace of inhibition of either. This latter fact possesses an added significance
because of its be'aring on an aspect of Pavlov's hypothesis concerning the nature of hypnosis,
which is to be considered later in the chapter.
IS THE LETHARGIC STATE OF HYPNOSIS ORDINARY SLEEP? Bass's
experiment not only establishes in an objective manner some of the differentive distinguishing
hypnosis from sleep, but certain incidental observations made during its progress enable us to
understand why those writers who have regarded hypnosis and sleep as identical may have been
led into error. It happened that there was one subject (Co) who fell into a natural sleep with
remarkable facility, and who also snored quite noticeably when in that state. It was observed that
sometimes this subject would snore a little during the last few of the kick stimulations of the
trance series. It was also noticed that at such times he tended to miss pressing his key at the bell
stimulus, and that the amplitude of the knee-jerk was somewhat diminished. All these
phenomena, of course, pointed strongly to true sleep. As a result of these observations a special
experiment was set up to make doubly certain that on such occasions the trance had really given
place to natural sleep. (205)
The subject, without knowing the purpose of the experiment, was placed in the chair and
hypnotized by Mr. Bass, and the experiment proceeded as usual except for one thing. The present
writer was put in rapport with the subject, and proceeded to read to him in a quiet, monotonous
tone a series of simple fairy-tales by Hans Christian Andersen. As the reading progressed, Mr.
Bass stood near by and recorded on the polygraph paper, in line with the recording points, the
number of each page as it was begun. After the reading had continued about a half-hour, it was
noticed that the subject was failing to respond to the bell and that his knee-jerk had practically
disappeared. This was allowed to continue for a short time. The writer then began questioning the
subject regarding the stories, beginning with the first one read and proceeding through to the last.
Meanwhile the recording of the experimental reactions continued without interruption.
The responses of the subject, considered in conjunction with the other evidence, indicate
in a most objective and convincing manner that during the reading he had passed from the trance
into a true sleep, and that, upon being roused from his sleep by the writer's questions, he had
resumed his trance. As evidence of this it may be noted, first, that immediately upon the
beginning of the questioning both the knee-jerk and the response to the bell returned to normal.
Secondly, the subject was able to report in considerable detail the stories of "Ib and Christian,"
"The Golden Swan," "Story of the Oak," and "The Little Mermaid." He remembered just a little
about "Thickheaded Jack," but could recall nothing about "Daisy" or "The Story of the Swan." It
is to be noted that this questioning was performed during the trance itself, so that there can be no
involvement of post-hypnotic amnesia in the subject's failure to report. In other words, this
inability to recall is not an amnesia at all, but a simple failure of registration characteristic of
sleep and a criterion employed by all people from the remotest times as a test of sleep. Lastly
there may be mentioned the fact that upon being de-hypnotized the subject reported a complete
amnesia for everything that transpired after he was put into the trance, which establishes the fact
that when he was aroused by the writer's questions he passed directly from a true sleep into the
hypnotic state. (206)
There is really nothing surprising or paradoxical about a subject passing into natural sleep
while in the trance, particularly if his posture happens to be one favorable for sleep and he be left
undisturbed. These conditions, together with the quiescence usually imposed on the subject by the
suggestions, are exactly those which normally induce sleep. As a matter of fact, the transition
from hypnosis into sleep and from sleep into the trance without an intervening waking interval
has been repeatedly reported, though without accompanying objective evidences which are
necessary to render interpretation unambiguous. For example, Jacoby (I4, 167) states: "On the
other hand, it is possible to transform sleep into hypnosis or hypnosis into sleep. In the former
case a rapport. may be subsequently attained, in the latter case the rapport at once ceases."
The point at issue is thus not whether one state may pass into the other, but whether the
two states are essentially identical except as to their origin. That the spontaneous transition from
the hypnotic state to true sleep can be traced and indicated by three objective criteria (knee-jerk,
voluntary pressure, and recall of story) is as clear and emphatic an indication that there is a
genuine difference between them as could easily be imagined.
Granted that true sleep may supervene upon the trance, this would seem to explain in a
very simple and natural manner the so-called lethargic state of hypnosis which is contrasted by
certain writers to a second or alert state. The extreme lethargic state, according to this view, is
nothing but natural sleep and is not hypnosis at all. Only the alert stage is, properly speaking,
hypnotic. But, since the transition to natural sleep is a gradual passage through successive stages
of drowsiness, there would, on this hypothesis, naturally be intermediate stages where hypnosis
and true sleep would be mixed or blended, just as is the case in the transition from the ordinary
waking state into sleep. Accordingly, there is no paradox in the view that the hypnotic technique
may result in two contrasting extremesone of alertness and the other of lethargy. That such states
exist appears not only to be true, but quite natural and entirely to be expected. (207)
CAN HYPNOTIC PHENOMENA BE INDUCED IN TRUE SLEEP? Lastly there
must be considered a type of observation which seems to have played an important role in
convincing many writers of the identity of hypnosis and sleep. Reference to this matter is
contained in the quotations from Bernheim and from Schilder and Kaudel's, given above. It is the
alleged fact that, with care, rapport may be established with a "good sleeper," after which there
may be obtained all of the phenomena characteristic of the trance. Bernheim (2, 188) describes an
experiment of this kind as follows: "In reality, ordinary sleep does not differ from hypnotic sleep.
To be subject to my influence it is sufficient that a person goes to sleep voluntarily and naturally
before me, with his thoughts fixed on me. I recently found a poor phthisical patient asleep. I had
never hypnotized her. Touching her lightly with my hand I said, 'Do not wake, sleep, continue to
sleep, you cannot wake.' In two minutes I lifted both arms. They remained cataleptic. After
having said that she would wake up again in three minutes, I left her. At the time indicated she
waked, and I went back to talk to her. She did not remember anything. Here then was a natural
sleep, during which I was able to put myself into relationship with the sleeping subject, and that
alone constituted the hypnotic sleep. How was it possible to establish the relationship? I suppose
that the patient began to wake, but that my command to continue sleeping prevented the complete
awakening. The patient passed into the sleep called hypnotic, that is, she was in relation with
me."
There can be no doubt whatever as to the facts as objectively reported. Indeed, our own
experiment just described fully confirms and corroborates them. The real question is whether the
subject is still asleep in any physiological sense after hypnotic rapport has been established. It is
highly probable that in most cases such experiments have been conducted with sleepers who
either had already been trained to the trance or were well informed of hypnotic phenomena.
When the experimenter began talking to the subject in the attempt to establish (208) rapport, it is
natural to suppose that this simply wakened the subject from her sleep and that she passed
directly into the trance as she had expected to do when stimulated in that manner in the waking
condition not immediately preceded by sleep. The trouble lies in na'ively supposing the subject to
be still in a natural sleep just because she remained relatively quiescent and with her lids closed.
That such an error should have been made at the time of Liebeault or Bernheim is, of course, not
in the least surprising in view of the lack of knowledge of the physiology of sleep prevalent at
that time.
It would be an easy matter, by means of a simple adaptation of Bass's technique, to test
Bernheim's implied hypothesis. The question is simple. If the subject is in a deep, natural sleep
when the hypnotist begins the attempt to establish rapport, as shown objectively by the abolition
of his knee-jerk and consistent failure to respond to signals such as Bass's bell, will the knee-jerk
and the response to an auditory signal still remain inhibited after the supposed rapport from
natural sleep is established? The fairly confident prediction is here ventured that at the time the
subject begins responding to suggestions in a normal manner his knee-jerks and voluntary
reactions to the bell stimulus will also have returned to normal. If this takes place it will indicate
that rapport was not established during the sleep, but only after the subject had wakened. It is
naturally impossible to be certain of the outcome of an experiment in advance of trial, but there is
so much presumption against the view that hypnotic rapport can be established during natural
sleep that before anyone can safely advocate it further he can hardly afford to neglect the securing
of experimental evidence so readily available.
PAVLOV'S HYPOTHESIS CONCERNING THE NATURE OF HYPNOSIS AND
SLEEP: So far, in the consideration of the relation of hypnosis to sleep, the writers cited have
based their views on no clear theory as to the nature of sleep. We come now to the examination of
a view which purports to link hypnosis with sleep, and on the basis of a specific physiological
mechanism-inhibition. This important hypothesis has been put forward by L. P. (209) Pavlov, the
Russian reflexologist, on the basis of years of intensive research on dogs. His views on the matter
may perhaps be stated most authentically in his own words (I9, 604): "As the result of many
preliminary guesses and tests we have arrived at the following thesis: Inhibition, ordinary sleep,
and hypnosis are one and the same process ... Every conditioned stimulus, as soon as it is used
alone, without being accompanied by an unconditioned one, leads, early or late, to a drowsy state
or to sleep ... The chief merit of the physiology of conditioned reflexes consists in the fact that it
has undoubtedly proved that inhibitory processes have the property of spreading (irradiation) ...
Inhibition is partial sleep, or sleep distributed in localized parts, forced into narrow limits; true
sleep is a diffused and continuous inhibition of the whole of the hemispheres ... From this point
of view the phenomenon of hypnosis may be easily understood. It represents one of the different
steps in the process of irradiation over the mass of hemispheres-the partial sleep of the
hemispheres."
At first sight the lay reader may be somewhat surprised to learn that a theory of hypnosis
has been evolved from the observation of dogs. This apparent paradox is lessened when it is
recalled that many hypnotists up to the time of Charcot believed that muscular rigidity or
immobility was a specific characteristic of at least one "state" of the hypnotic trance. To this fact
must be added the observation, first made by Kirscher in 1646 (I6, 199), that hens may be thrown
artificially into a kind of cataleptic state. The technique is, in general, to hold the animal in a rigid
position for some time and then very gently relax the restraint. Under such conditions many
animals such as hens, rabbits, frogs, lizards, salamanders, etc., which might normally exhibit
considerable spontaneous activity, especially in the direction of flight, may remain quiescent for
minutes and even hours. Presumably because of the analogy to hypnotic catalepsy, this tonic
immobility was called animal hypnosis. Pavlov was familiar with these views. Accordingly, when
in the course of his numerous experiments on the conditioned reflex he occasionally found a dog
which became rigid and immobile in its skeletal musculature, what more natural than that he
should think of animal hypnosis? This interpretation of his observations was supported by the fact
that this state of canine catalepsy was ordinarily encountered under experimental conditions
which usually resulted in sleep. This coincidence agreed with the widely accepted view that
hypnosis and sleep were essentially identical. There was the additional coincidence (I9, 607) that
the unresponsiveness of the dogs in this state resembled the lack of responsiveness of hypnotic
subjects to persons with whom they are not in rapport.
While it is easy to see how Pavlov, who seems to have had little or no experimental
contact with human hypnosis, might have been impressed by the above considerations, we
ourselves, with the background of the more recent developments in the scientific aspects of
hypnotism, find them distinctly unconvincing. In the first place, catalepsy is not, as Charcot and
his school supposed, a symptom of hypnosis as such. Catalepsy is never observed in the writer's
laboratory unless specific suggestion is given that it will take place. By this criterion, touching the
end of the nose with the tip of the right index finger might also be regarded as a symptom of
hypnosis as such, since in the hypnotic state it also can be induced quite as readily as catalepsy.
Moreover, catalepsy can be induced with remarkable readiness in subjects in the normal waking
conditions previous to the giving of any other suggestions whatever. Not only this, but the
technique of inducing catalepsy in hens, frogs, salamanders, etc., which consists essentially of
strict restraint very gradually relaxed, is quite different from that of specific suggestion employed
with humans. The implication of all this is, of course, that the supposed identity between animal
catalepsy and human hypnosis is wholly erroneous; the two phenomena have nothing in common.
The situation is no different in the case of rapport, which is also cited by Pavlov. Rapport
is a phenomenon of hypnosis only in the sense that it may be obtained if it is suggested either
directly or by some indirect means; but this is equally true of visual, auditory, or cutaneous
anaesthesia. Indeed, rapport differs from the latter only in that it is a selective anaesthesia, an
anaesthesia mainly to suggestions from individuals other than (211) the person or persons with
whom the subject is in the so-called rapport. Experimental work to determine the exact status of
the apparent amesthesia of rapport needs very much to be performed (I2, 399). Pending the
performance of the experiments, we must base judgment as best we can on observations and
related experiment. The conclusion thus arrived at is that, as Sears found with cutaneous
amesthesia, the an;esthesia is by no means the complete insensibility apparent to casual
observation but, rather, a kind of pseudo-insensibility, at least in part. It seems likely, on the other
hand, that the insensibility of Pavlov's dogs in the cataleptic state generated by the conditioned
reflex technique may have been a genuine one more like that of Bass's subjects for faint sounds
during true sleep.
Thirdly, Pavlov's hypothesis regards hypnosis as a transitional stage between the waking
condition and true sleep. This is definitely controverted by Bass's experimental results, since they
indicate a sharp and definite differentiation between the trance state and even drowsiness or
incipient sleep.
There are in addition some experimental findings by Scott (22) which may turn out to
offer a radical refutation of the Pavlovian views in so far as they concern hypnosis. The relevancy
of Scott's results to Pavlov's hypothesis respecting hypnosis depends, however, upon the
assumption that the dogs when in the supposed hypnotic catalepsy were insusceptible to the
establishment of conditioned reflexes. Unfortunately, so far as the present writer has been able to
discover, Pavlov has made no explicit statement in regard to this matter. There is considerable
indirect reason to believe that this is the case, however: In addition to the muscular rigidity
characteristic of the supposed canine hypnosis, several peculiar changes are said to come over
conditioned reflexes which have previously been established. For the most part these changes are
of an inhibitory nature, which presumably would militate against the acquisition of new
positively conditioned tendencies. Moreover, Pavlov states explicitly that the tendency for the
animals to fall asleep during experiments involving considerable inhibition of the type alluded to
in the above quotation actually did prevent conditioning. In this connection Pavlov says: (212) " ..
in the early period of our work this interfered with our researches, for, being not yet thoroughly
familiar with the technique, it was impossible to obtain in some animals the reflex we required,
on account of the development of sleep ... The interference by sleep in the case of thermal stimuli
was indeed so persistent and upset the work to so great an extent that in the early period of our
research I had real difficulty in finding collaborators who would agree to work with these
stimuli."
Lastly, if complete somnolence (fully irradiated inhibition) completely abolishes
susceptibility to the acquisition of conditioned reflexes, it seems reasonable to assume that partial
somnolence (which hypnosis is here supposed to be) would manifest at least a partial destruction
of the tendency to conditionability.
ARE CONDITIONED REFLEXES ACQUIRED WITH NORMAL FACILITY
DURING HYPNOSIS? Scott employed nineteen subjects in his conditioned reflex experiment.
Of these, ten were subjected to the conditioning technique when in the hypnotic state, and nine,
the control group, when in the waking condition. The conditioned (supposedly neutral) stimulus
was a somewhat muffled buzzer. The unconditioned or reinforcing stimulus was an electric shock
sufficiently sharp and painful to make the subject jerk his hand off the electrodes which delivered
it. Strong shocks such as were employed by Scott also produce rather widespread disturbances in
other physiological processes. Of such processes Scott secured automatic records of the breathing
by means of a Sumner pneumograph, and of the arm pulse and blood-pressure by means of the
Erlanger-Meek recording sphygmomanometer. A record of the changes in the electrical activity
of the skin was secured by means of a delicate D' Arsonval galvanometer, which threw a beam of
light upon a translucent scale. A sliding pointer mounted close to the screen could easily be made
to follow the point of light. The pointer was so connected with a vertical recording device that all
of the movements of the former were faithfully recorded to scale on the smoked paper of an
extension kymograph. A photograph showing the general arrangement of Scott's apparatus is
reproduced as Figure 39 (not available). One of his more (213) successful recordings is
reproduced as Figure 40. Because of frequent apparatus difficulties with the sphygmomanometer
and the galvanic skin reflex, Scott reports results only for the finger reaction and for respiration.
In the actual conduct of the experiment, the subject was first placed in the apparatus and
the buzzer was sounded a few times to determine whether it really was a neutral stimulus, i.e.,
whether it would evoke a reaction at the outset of the experiment. The subject was then taken into
an adjoining room of the laboratory, where he was seated in a comfortable chair and put into the
hypnotic trance. vVhen he was judged to be in the deepest state possible, he was gently led back
into the experimental room and again placed in the apparatus. Again the buzzer was sounded
several times at suitable intervals to determine its neutrality in the trance state. The stimuli were
then given in pairs at approximately one-minute intervals, the buzzer preceding the shock by a
fifth of a second. After ten or so such reinforcements, a single (unreinforced) buzzer stimulus was
given occasionally to see whether the conditioning tendency had become strong enough to evoke
a clearly marked conditioned reaction. vVhen this phenomenon occurred, reinforcement was
terminated and the buzzer was given alone a dozen times at intervals similar to those used during
reinforcement. The subject was then led gently back to the chair in which he was originally
hypnotized, piaced in exactly the same posture as he had been just before being hypnotized,
allowed to rest a short time, and then awakened. Since all the hypnotic subjects of this group
showed complete post-hypnotic amnesia, they had no recollection of the conditioning process just
completed. Finally, after a total absence of fifteen minutes, the subject, fully awake, was taken
back into the experimental room and placed in the apparatus and the conditioned reaction was
tested six more times by sounding the buzzer alone at approximately one-minute intervals as
already done at the end of the trance.
SCOTT'S EXPERIMENTAL RESULTS ON CONDITIONABILITY DURING
HYPNOSIS: The smoked paper tracings of the finger reactions were scored by measuring the
deflection of the finger line from the (215) pre-stimulus level in fiftieths of an inch.
Unfortunately, Scott's results are marred for purposes of statistical treatment by the fact that the
number of stimulations in the several parts of his experiment varied more or less from subject to
subject. Despite this, fairly consistent composite graphs of the amplitude of the conditioned and
unconditioned reactions were plotted. Such a graph for the finger reaction is shown in Figure 30
(not available). An inspection of this graph shows that at the outset of the experiment the buzzer
was entirely neutral with respect to the finger reaction, since no reactions of this kind followed
this stimulus, either in the waking or the trance state. During reinforcement with the shock,
however, the reaction was marked, though highly irregular. At the conclusion of the conditioning
process the buzzer, when given alone, produced reactions of very considerable average
amplitude, though, as is usual in such cases, distinctly smaller than those given to the shock.
These are, of course, conditioned responses. Lastly, fifteen minutes later, when the subject had
returned to the normal waking condition, the buzzer alone still evoked a distinct finger reaction.
The decided weakening of the conditioned reaction in the waking state from the strength
displayed in the preceding trance state was presumably due, at least in the main, to the action of
experimental extinction, which normally weakens conditioned reactions when frequently evoked
without reinforcement.
The extent of the conditioning of the respiratory process required more elaborate
measurement. The typical respiratory reaction to a shock is a fairly marked increase in the total
amount of breathing per unit time, together with a general irregularity or unevenness in the
respiratory rhythm. Scott reports data from a variety of measurements, but since all yield
substantially similar results only one will be discussed here. The method employed in this
particular case was to measure in fiftieths of an inch the depth of each inspiration for a period of
fifteen seconds preceding and following each stimulus. The mean of those preceding the stimulus
is thus the standard. If the buzzer at a given point of the experiment has any special excitatory
tendency on the breathing, the mean inspiration amplitude after stimulation will be larger than
that before, but if not, it will be about the same. Accordingly, if the mean (216) inspiration value
after stimulation is divided by that before, and the ratio comes out 1.00 there will be no indication
of any conditioning tendency. But if the ratio is more than unity, there will be indicated, except
for sampling deviations, a tendency to conditioning; and the more the ratio exceeds unity, the
stronger the tendency.
A composite curve based on such ratios from seven subjects is given in Figure 41. An
inspection of this graph shows that at the outset of the experiment (in the waking state) the
subjects showed a slight but perceptible tendency toward respiratory reaction to the buzzer,
thereby proving it to be not strictly neutral. In the succeeding trance test practical neutrality
appears, presumably as the result of negative adaptation resulting (217) from the test stimulations.
Following the reinforcement, but still in the trance state, the subjects show at buzzer stimuli the
very marked increase in depth of inspiration of 41 per cent, which is nearly half as great as that
produced by the shock itself. These are conditioned reactions and parallel the conditioned finger
reactions mentioned above. Fifteen minutes later in the waking state the conditioned tendency is
still present, though weakened. While extremely irregular, both segments of the graph show a
marked indication of a progressive fall with successive stimulations, exactly as would be
expected from the influence of experimental extinction.
We have clear evidence, then, that human subjects are susceptible to conditioning when
in the hypnotic state. We must now raise the further question whether they are as susceptible to
conditioning as when in the normal waking state. The control group mentioned above was run by
Scott in an attempt to answer this question. The comparative susceptibility to conditioning of the
two groups is shown in Table 34 (not available). There it appears that the trance subjects, without
exception, showed susceptibility to conditioning, whereas little over half of the waking subjects
showed it. Moreover, of those who showed conditioning, the trance subjects required only about
half as many reinforcements as the waking subjects. On the surface the indication would seem to
be that subjects not only acquire the conditioned reflex in the trance state as readily as in the
waking condition, but that the trance state, instead of interfering with the process, may be
positively favorable to conditioning.
Aside from the small number of subjects employed in the Scott experiment, there must be
noted a serious limitation to (218) the conclusions to be drawn from it. This arises from the fact
that presumably a number of the individuals in the control group were not even susceptible to
hypnosis, whereas all those in the trance group were. In other words, it is possible that the
hypnotic group were conditioned more readily because they were the kind of individuals who are
susceptible to hypnosis rather than because they were actually in the trance when subjected to the
conditioning procedure. A number of casual observations in the laboratory suggest that there may
be some truth in this conjecture. The matter is of such importance for an understanding both of
hypnosis and of the conditioned reflex as to warrant an early and thoroughly controlled
investigation to clear up the ambiguity. But even if hypnotics as a group should prove more
readily conditionable, it would hardly improve the credibility of Pavlov's hypothesis.
There remains to be considered a rather different interpretation of Pavlov's inhibition
hypothesis concerning the nature of hypnosis. It might be urged that hypnosis in man is
characteristically different from that in dogs owing to the relatively dominant influence of the
higher centers in the former, and that the partial inhibition which in dogs affects primarily the
lower centers, in man affects chiefly the higher centers. Since (it might be assumed) the
conditioning process involves chiefly the lower centers, the above hypothesis would lead us to
expect hypnosis in dogs to interfere greatly with conditioning; but in man, since the lower centers
in hypnosis are released from the normal control exercised by the higher centers, we should
expect an even greater facility of conditioning than in the waking state, exactly as Scott found.4
Fortunately we have evidence which bears upon this alternative interpretation. Bass, as we have
already seen, has reported that subjects show practically no disturbance in voluntary reactions to
faint signals when first put into the trance. There seems little doubt that such activity involves the
functional integrity of the higher centers, though we must, perhaps, leave the ultimate decision
(219) in this matter to the neurologists. To these results may be added the facts that subjects in
hypnosis learn nonsense syllables (r3, 23), learn complex stylus mazes (7), and perform complex
additions (I8), which could hardly be accomplished without a substantial functional integrity of
the higher centers. Pavlov's inhibitory hypothesis concerning the nature of human hypnosis
accordingly appears to be quite as effectually refuted under the present interpretation as under
that originally put forward by Scott.
A PHYSICO-CHEMICAL APPROACH TO THE PROBLEM OF HYPNOSIS AND
SLEEP: Lastly there must be mentioned a study by Hoff and Wermer which purports to identify
hypnosis and true sleep by physicochemical methods. A number of experiments have shown that
the kidneys greatly reduce the secretion of urine after an injection of an extract from the pituivuy
gland. Other experiments having shown that this inhibitory action of pituitrin is in abeyance
during sleep, Hoff and Wermer set up an experiment to determine whether hypnosis would also
prevent the normally inhibitory action. The pituitary injection was given at 7:15 A.M. Between
7:30 and 8:00 A.M. the subject drank 500 C.c. of weak tea. At 8:00 A.M. he was placed in what
was regarded as deep hypnosis-a state manifesting catalepsy and analgesia, and was given
instructions to urinate every hour, though any suggestion of urinary pressure was avoided. It is
reported that under these conditions the inhibition of urinary secretion, normally characteristic of
this drug, was absent exactly as in the case of true sleep. It is stated that parallel experiments were
performed to determine the subject's normal waking secretory tendencies both with and without
the drug, and that these results are in harmony with the view that hypnosis in this
physico-chemical sense is similar to true sleep.
These results are not necessarily in conflict with those of Bass, though they tend
somewhat in that direction. Their interpretation must depend to a considerable extent on the
nature of the control series run. It is easy to harmonize the two sets of results if we may assume
that the subject during his (220) prolonged hypnosis actually was sufficiently quiescent to fall
into at least a light slumber, the so-called lethargic state of hypnosis. An investigation designed to
establish any physiological identity between hypnosis and sleep must make quite certain that the
subject actually does not fall asleep. We have no such assurance in the published report of the
experiment of Hoff and Wermer. Such a controlled experiment would not be difficult to set up. It
would only be necessary to repeat Hoff and Wermer's experiment and at the same time record the
knee-j erks and the voluntary reactions to weak signals as was done by Bass. If our hypothesis is
correct, the renal secretions during hypnosis will be found to resemble those of true sleep only
during those experimental periods when the subject is shown by the objective indicators actually
to be more or less completely asleep. Meanwhile we must reserve judgment.
SUMMARY AND CONCLUSIONS: In concluding this discussion we seem forced to
the view that hypnosis is not sleep, that it has no special relationship to sleep, and that the whole
concept of sleep when applied to hypnosis obscures rather than clarifies the situation. The
physiological criterion of the knee-jerk shows the trance to be allied, rather, with the waking
state, and to be clearly differentiated not only from sleep but from incipient drowsiness as well.
The psychological test of voluntary reaction to a signal shows practically identical results.
Moreover, the same techniques which distinguished hypnosis from sleep were able definitely to
trace objectively the gradual passage from the hypnotic trance into true sleep. This wholly natural
transition accounts very simply and logically for the so-called lethargic state of hypnosis, which
has often been observed and contrasted with the so-called alert stage. Thus the extreme lethargic
state is not hypnosis, but true sleep; only the alert stage is hypnotic. Lastly, evidence has been
presented which indicates not only that conditioned reflexes may be set up during hypnosis, but
that this may perhaps be accomplished with even greater ease than in the waking state. This
probably disproves Pavlov's hypothesis that hypnosis is a state of partial sleep in the sense of a
partial irradiation of inhibition. On the other hand, (221) the remarkable facility of subjects for
falling asleep while receiving regular heavy blows on the patellar tendon seems to support
Pavlov's theory of the nature of sleep itself. The alternative interpretation, that Pavlov's
hypothesis assumes hypnosis to be an inhibition of the higher rather than the lower centers, is
effectively refuted by Bass's results showing that voluntary action to a signal is not appreciably
disturbed by hypnosis. The results of Hoff and Wermer cannot be evaluated until the experiment
is performed under conditions which provide objective evidence that true sleep was not involved
where only hypnosis was supposed to be. (222)

17. CH 9: HYPNOTIC SUGGESTIBILITY AND THE TRANSCENDENCE OF VOLUNTARY

CAPACITY: CLARK HULL: HYPNOSIS AND SUGGESTIBILITY: APPLETON-CENTURY-
CROFTS: NY: COPYRIGHT 1933: 1961
A. PROCESSES PRIMARILY MOTOR: WE have already had repeated occasion to
observe the strong tendency to attribute remarkable or supernormal powers to persons during
hypnosis. Happily, the belief in the ability of subjects while in the trance to see without the
normal use of the eyes, to obtain the normal pharmacological effects of drugs from the mere
contact of the sealed container, to give strange but characteristic responses to the presence of
certain metals, and to manifest marvelous sensitivity to magnets has now been very generally
abandoned. Nevertheless, the view persists that hypnotized individuals may display certain
powers which transcend those possible to normal voluntary effort.
SOME PERSISTENT CLAIMS CONCERNING SUPERNORMAL
PERFORMANCE DURING HYPNOSIS: It has been reported and widely credited that in the
state of hypnotic catalepsy a subject becomes so immune to ordinary muscular fatigue that the
arm can be held out horizontally for long periods with comparatively little sagging. Other writers
have contradicted this, stating that the extended arm will sag during hypnosis, just as in the
normal state, but that its descent is smooth and gradual, whereas the descent in the non-trance
state is accompanied by marked involuntary tremors. It has been reported that subjects in the
trance can perform feats on the ergograph which are quite impossible in the normal state. By
some it has been alleged that temperature changes amounting to several degrees may be brought
about by direct suggestions. It has been claimed that hypnotic suggestion (224) can cause the
blood to leave or return to a certain part of the body, such as the hand, so that it will become
alternately pale or red at the whim of the hypnotist. It has even been said that bleeding from open
wounds could be stopped and later resumed by similar suggestions. It has been asserted that under
the influence of hypnotic suggestion subjects become hypersensitive so that they can see and hear
stimuli normally quite beyond their powers. On the other hand, it is asserted that under suggested
am-esthesia individuals may undergo even major operations without either pain or shock.
The great ease with which such claims have attained currency doubtless may be attributed
in part to the natural human craving for the marvelous, but other and more creditable sources of
interest have not been lacking. One such source lies in the fact that, from the first,
experimentalists in this field have been haunted by the fear of being deceived by the simulation of
hypnosis on the part of their subjects. Clearly, if certain phenomena can be evoked in the trance
which it is not possible to attain through voluntary effort in the non-trance state, this fact
immediately becomes of importance as an objective means of detecting simulation and thus,
indirectly, a means of protection against this danger. An additional consideration which makes
such possibilities of importance, but one which has received little attention, is the likelihood that,
if hypnosis is really able to evoke supernormal processes of any kind, a sagacious investigation of
such phenomena might lead to the discovery of hitherto unknown physiological mechanisms. It is
conceivable, for example, that it might lead to an effective control of certain autonomic processes
ordinarily very feebly influenced by voluntary effort. Apart from the contribution to science of
such a sequel, there is the additional possibility that the ready control of such a physiological
process might prove of practical therapeutic value in the treatment of nervous, mental, and even
somatic disorders.
While it cannot be denied that claims of the transcendence of physiological normality
smack strongly of the miraculous, it is to be observed that as a rule they are not put forward as
magic but soberly, the cause being attributed to some hypothetical but as yet unrecognized
physiological principle. In many cases the reports of the investigations in question have (225)
apmarked. Parallel differences in the accompanying respiration during the two states are also
reported. Charcot did not attempt to record the actual descent of the arm, as did Rieger, but only
the tremors. The arm was extended with the palm of the hand downward. Upon the back of the
hand was mounted a pressure chamber, which evidently supported upon its thin rubber membrane
a weight. Even a very slight movement in the vertical plane, if quick, would produce a marked
change in the air pressure within the system, owing to the natural inertia of the weight which
would cause it to lag behind the cup attached to the hand. A weight so supported might be
expected, when disturbed, to display pendular oscillations which would be recorded by the
sensitive recording lever. Some indications of this are seen in published records obtained by
means of the apparatus.
Charcot's conclusions in this matter seem to be based on hypnotic records of arm tremor
and respiration taken on an hystero-epileptic patient. These were compared with the parallel
record of another person (presumably normal) who attempted to maintain the same cataleptic
attitude in the nontrance state. The arm record of the hystero-epileptic was a perfectly smooth
line, whereas that of the normal man showed great agitation, particularly after the experiment had
gone on for some time and fatigue had had an opportunity to develop. The breathing of the two
subjects showed a corresponding difference. That of the hystero-epileptic was shallow and
regular, whereas that of the normal man was relatively deep and irregular, particularly after the
experiment had continued for some time.
Commenting upon the results of this experiment and others of a somewhat similar nature,
Binet and Fere characteristically remark: "On this point the demonstration is complete. It is
almost certain that no individual in the waking state, unless affected by a nervous state allied to
hypnotism, could imitate the distinctive signs by which hypnotism is manifested. The dread of
simulation, which has dominated the whole history of animal magnetism, has now become a
completely imaginary danger, if the experimenter is adroit and cautious." (228)
Such a display of uncritical cocksureness on the part of a person who later attained the
scientific eminence of Alfred Binet is all but inexplicable, for it would be difficult to imagine a
more thoroughly bad experiment. Charcot might at least have secured control records from his
hystero-epileptic when in the non-trance state. This would have removed the gross danger that the
difference observed was due to the constitution of the subjects rather than to the states in which
they were. It is possible, of course, that he believed, as might be understood from the above
quotation from Binet and Fere, that the lack. of tremor was due to the hysterical constitution of
the subject, so that he could not serve as his own control even in the nontrance state. But in that
event the lack of tremor should have been attributed to hysteria and not to hypnosis; for Charcot,
despite his belief in the intimate relation between hypnosis and hysteria, surely did not hold that
hystericals were in a state of hypnosis at all times.
Nevertheless we find Moll, who is usually one of the most judicious writers in this field,
remarking, evidently with the Charcot experiment among others in mind: "At all events,
experience teaches us that suggestion in hypnosis can bring about muscular phenomena which
cannot be produced voluntarily ... If a person holds out his arm for a long time without trembling
to any extent, this may be held to exclude fraud to a certain extent ... A heavy weight placed in a
hypnotic's hand will often be held longer and more steadily than it would be possible for a
waking man to."
It is only just to add that the context from which the above quotation was taken indicates
that Moll had himself performed some of these experiments, though we have no reason to believe
that they were any more carefully performed than those of Rieger and Charcot. The same may
also be said of McDougall, a still more recent writer, who states: "And an arm may be maintained
in such a position, requiring considerable muscular work for its maintenance, for as many as
thirty minutes ... If the maintenance of the position is due merely to the complacence of the
conscious subject ... the limb is less perfectly plastic and it soon reveals, by tremors and a
tendency to drop, that the patient is feeling fatigue." (229)
WILLIAMS'S INVESTIGATION OF HYPNOTIC CATALEPSY: We turn now to
the latest available investigation of this venerable problem-that of Williams. The technique and
methods employed by Williams present in many ways an agreeable contrast to those just
reviewed. He employed eight normal subjects, each of whom was tested for cataleptic arm
extension eight times-four times in the trance and four in the non-trance state. In order to protect
the experimental results from possible practice effects or other temporal distortions, the tests were
administered in the order N, T, T, N, T, N, N, T. The subjects saw neither their arms nor the
records made by them. They were given no intimation as to the nature or purpose of the
experiment.
The important matter of the suggestions administered to the subjects during the
experiment is best given in the words of Williams's report: "The S's were put into the trance by
having them look into one of E's eyes. Verbal suggestions of being in a deep trance were also
used when S's eyes had closed. Cataleptic rigidity of the arm was induced by verbal suggestions
that the arm was becoming stiff and rigid. The arm was simultaneously lifted to the level of the
shoulder by E, and the thread was then attached to the cuff about S's wrist. In the normal state the
arm was similarly lifted, but without the repeated suggestions of catalepsy. A voluntary catalepsy
was induced by giving the single instruction, 'Hold your arm there rigidly.' This instruction was
also given after the arm had been lifted in the trance cataleptic state. This was the only instruction
given, so that conditions were identical when the arm had been made cataleptic and lifted in the
trance state and had been similarly lifted by E with voluntary catalepsy in the normal state."
The method of recording the behavior of the arm is shown in Figure 42. The arm was
extended rigidly and horizontally in the lateral plane of the body. Around the wrist was placed a
light leather cuff to which could be attached readily the thread fastened to the recorder. In order
to eliminate all serious distortion of the record from lateral movements, the thread ran upward
several feet, passing over an extremely free-running pulley, and thence down to a slender rod
sliding easily in a vertical tubular sleeve. To the lower end of this (230) rod was attached a stylus
which maintained a very light but steady contact with the smoked paper on a ten-inch kymograph
drum. The device was extremely sensitive and recorded even minute tremors. Two typical
records, one by a steady and the other by an unsteady subject, are shown in Figure 43. The rise in
the tracings indicates the gradual descent of the arms. In addition, a respiration tracing was made
at the top of the record and a signal line and time line were traced at the bottom. This method of
recording made it possible for Williams to investigate in a wholly objective and quantitative
manner the claims of both Rieger and Charcot.
Because of the natural tremors it is not sufficient, in measuring the extent of the arm's
descent after a given period of extension, to determine its exact position at a particular instant. It
is desirable, rather, to know its average position during a certain interval. This may be done with
precision by first erecting two vertical lines (separated by the desired time interval) from the
horizontal line marking the arm's original height, until they intersect the arm tracing. The area
bounded by the two verticals, the base line, and the irregular arm tracing may then be measured
with precision by tracing its outline with a polar planimeter. This area, when divided by the
length of the base, yields the average distance of the segment of arm tracing from the base line
for the period in question. The rulings and measurements of such an area are shown in part A of
Figure 43. The vertical lines are separated by six five-second intervals, as shown by the time line.
The area of the figure as measured by the planimeter was 8.29 sq. in., and the base was 2.12 in.
The former divided by the latter yields 3.91 in., or the average height of the oscillating arm line
above the base. This latter point is marked by the broken horizontal line. In this manner 'Williams
measured the average position of the arm during each half-minute recorded for each of his
experimental periods.
The amount of tremor or oscillatory irregularity in the descent of the arm was measured
with precision by the author's linear oscillometer. This device (Fig. 44) is designed to measure the
vertical oscillations of a line moving in a generally horizontal direction, by simply following the
line with a special tracing point provided by the instrument. The same(231) half-minute sections
used for height measurement by the planimeter were traced by the oscillometer for the
measurement of tremor. Such a section, three ten-second intervals as shown by the time line, is
marked by the arrows in part B of Figure 43. Unfortunately this instrument does not distinguish a
general drift in an upward or downward direction from true tremor. As a consequence, Williams's
actual readings combine the descent of the arm with the true oscillations. Since the upward
movements of the traced line are measured by the instrument separately from the downward, it
would have been possible to make a correction for this distortion. However, since it turned out
that the subjects as a group showed about the same rate of arm sag in the two states, the relative
amount of tremor is represented with little distortion by the uncorrected values.
 WILLIAMS'S RESULTS ON RATE OF ARM DESCENT AND AMOUNT OF
TREMOR: The results of the measurements of arm descent for the eight subjects during the
respective thirds of each subject's mean performance in the trance and waking states are shown in
Table 35. An inspection of this table reveals that four of the subjects showed an average tendency
for the arm to fall faster during the waking state, but the other four showed an even stronger
tendency for the arm to fall faster during the trance .. The average of all eight subjects shows a
slightly more rapid fall during the trance, but this does not possess a significant statistical
reliability. These results of Williams, therefore, support Charcot and oppose Rieger, Moll, and
McDougall.
The results of the tremor measurements are similarly presented by thirds of the extension
periods in Table 36 (not available). An inspection of this table shows the marked individual
differences already noted in rate of arm descent. A comparison of the amount of tremor in the
two states reveals the fact that on the final third of the extension period five subjects show more
oscillation in the normal state, two show more in the trance, and one shows practically no
difference. The average of all eight subjects shows during the final third of the extension period a
slight excess of tremor in the non-trance condition. This, however, has the unsatisfactory critical
ratio of 1.51. The difference (232) of thirteen units on the first third is in the direction of greater
tremor in the trance. This difference, also, has an unsatisfactory statistical reliability. The tremor
in the middle section averages exactly the same amount in both states.
Williams reports a qualitative observation which is probably of significance in the
interpretation of his quantitative experimental results. This is that his subjects stated uniformly
that they suffered a great deal of pain in their arms during the waking experiments but none
whatever during the trance experiments. Rieger reports very similar observations in connection
with his experiment. Williams states explicitly that no intentional suggestion to this effect was
given to his subjects. Nevertheless, it is difficult to believe that the suggestion did not (233) reach
them in some obscure and indirect fashion. But however it came about, analgesia evidently was
present. We shall have occasion later to observe under controlled experimental conditions the
power of suggested analgesia to inhibit the usual physiological responses to normally painful
stimuli. In view of such facts, together with the fact of the heightened susceptibility to suggestion
characteristic of the trance state, it is a little surprising that Williams's group as a whole did not
show a slight but genuinely greater resistance to fatigue in the trance state, though no such gross
differences as reported by Rieger and Charcot could have been anticipated. In this connection it is
greatly to be regretted that 'Williams did not (234) measure his respiration records for irregularity
of breathing so as to secure a check on Charcot's claims in this respect. Unfortunately the author
had not yet devised this method of measuring breathing reactions at the time Williams's
experiment was performed.
It is possible that certain subjects, Re and W for example, may have had a tendency
actually to fall asleep during the trance. If this were known to be the case, it would easily account
for their rapid rate of arm descent during that state. Because of this possibility, any repetition of
this experiment ought to be accompanied by some precedure like that employed by Bass (p. 196)
to make sure whether the subjects are fully awake at all times; those subjects who tend to fall
asleep during the experiment should be excluded.
RESISTANCE TO ERGOGRAPH FATIGUE DURING HYPNOSIS
(NICHOLSON'S EXPERIMENT): Closely related to the problem of the influence of hypnotic
suggestion upon the power to resist postural fatigue is that concerning its power to facilitate
resistance to the fatigue resulting from muscular action under load. An experimental investigation
of this problem has been reported by the late N. C. Nicholson, from the physiological laboratory
of the Johns Hopkins Hospital.
The Mosso ergograph is an instrument designed to measure the amount of work a subject
can perform with one of his fingers. To limit the activity as far as possible to the finger in
question, the arm is held firmly to a rigid support by a system of adjustable metal shoulders and
strong bands. The distal phalanx of the finger catches a loop attached to a wire which, when
pulled forward by flexion of the finger, lifts a weight. Interposed between the loop and the pulley
is a kind of car mounted on polished rods. Attached to the car is a stylus which records faithfully
on the smoked paper of a horizontally placed kymograph drum the exact distance of each pull. By
summating the length of these pulls and multiplying by the weight lifted, it is possible to
calculate with some precision the amount of work done within a given period, or before the finger
(235) musculature becomes so exhausted that it can no longer lift the weight.
Nicholson first practised his seven subjects in the non-tran state with this instrument daily
for two weeks, in order determine the normal work capacity of each. He then d termined the
normal period of recovery which must elap before a second performance judged as good as the
first cou be accomplished. This was found to be between one and tv hours. With these norms
established, the subjects were hy notized and their work capacity redetermined. Throughout t:
hypnotic work periods (but not the normal periods, apparentl: suggestions were given "of
increased energy and muscul power with absence of fatigue sensations." The weight er ployed in
all experiments was three kilograms, and the lif timed by the ticks of a metronome, were made at
intervals two seconds.
Unfortunately, Nicholson gives no quantitative data concer ing the results of his
investigation, but his report contai reproductions of a number of his ergographic tracings from 0
subject. The others are said to have given similar results. general, Nicholson reports that with this
three-kilogram weig his subjects would reach practical exhaustion in the non-tran state after about
ten minutes of work. He states, howevi that "during the hypnotic sleep the capacity for work seem
practically endless." One of the more striking variants of 1 experiment was to let the subject work
to practical exhausti, in the non-trance state and then suddenly put him into t trance without
interrupting the work. A record made by his subject in such an experiment is reproduced as the
upper tracing of Figure 45 (not available). Part A shows a typical ergograph tracing the general
shape of which is well known. At B the subject was hypnotized and at once the amplitude of the
lift returns was nearly to its original level and, most remarkable of all, continues at this level
without perceptible diminution for six minutes. To make this experiment still more convincing,
control experiment was performed in which the subject again worked in the waking state until
practical exhaustion, wh the same suggestions of encouragement, etc., were given as the
experiment just described, with the one difference that t subject was not hypnotized. The record
of this control experiment is reproduced as the lower tracing in Figure 45 (not available). The
suggestions of increased power in this latter case were followed by a slight rise in the lifts and by
a marked increase in duration but the amount of the increase in performance is clearly not
remotely comparable to that resulting from the same suggestions during hypnosis. Nicholson
concludes from his investigation as a whole that suggestions given in the hypnotic condition are
far more effective than when given in the non-trance state. He says: "Again and again in the
literature the statement is made that suggestions given while the subject is fully awake are as
effective as when given during the hypnotic sleep. The records obtained ... do not seem to bear
this out ... ; it would seem reasonable to conclude from them that suggestions given during the
hypnotic sleep are much more effective than the same suggestions given in the absence of
hypnosis. Hence we may say that wherever suggestion is indicated as a therapeutic agent
hypnosis should be the method of choice."
This, of course, is but another way of saying that hypnosis is a state of heightened
responsiveness to suggestion. In the remainder of the present chapter and in the two following we
shall see repeated evidences of this, though not in such striking degree.
WILLIAMS'S INVESTIGATION OF HYPNOTIC SUGGESTION AND
ERGOGRAPHIC FATIGUE: The unexpectedly large differences in responsiveness to
suggestions of strength and accomplishment in the trance and non-trance states reported by
Nicholson, together with certain indications of unsatisfactory technique in the conduct of the
experiment, led 'Williams to repeat a portion of it. He had a subject in the normal state lift a
weight of four and one-half kilograms eighty times per minute. After the subject had continued
his work for half an hour without noticeable weakening, the experiment was stopped as the
subject asserted that he "could keep it up all night." It will be noted that this weight was
considerably heavier than that used by Nicholson, and was being lifted over twice as frequently,
per unit time, as was done by Nicholson's subjects. While by no means conclusive, such an (237)
outcome suggests the possibility that Nicholson's subjects were not properly motivated in the
non-trance state. If this were true, it would, of course, mean that the deviation from normal
performance would lie in the low performance in the waking state rather than in the magnitude of
the performance in the trance state. Williams therefore undertook a repetition of the Nicholson
investigation with improved conditions which sought to eliminate certain possible sources of
error.
Williams employed five subjects. Four of these showed posthypnotic amnesia and one
showed marked cataleptic rigidity. Williams's ergograph possessed, in addition to the tracing
device forming part of Nicholson's instrument, a mechanism which registered the total height that
the weight had been lifted, thus giving a precise quantitative measure of the amount of work
done. The weight lifted was five kilograms, and the metronome setting the rhythm of the pulls
was placed at eighty ticks per minute, which presented the subjects with a task definitely more
severe than that in Nicholson's experiment. Only one record was taken each day, the hour being
constant for each subject. Before the actual beginning of the experiment, each subject was
practiced in the normal state until his muscles had hardened so that no soreness followed the
work. The trance and normal records were alternated in a manner designed to neutralize any
general practice or other temporal effects. So far as the experiment permitted, the suggestions
were the same for both states, as may be seen by the following formula given in the hypnotic
state: "I want you to pull this weight and keep time with the metronome. I will start the
metronome so that you get the rhythm (but do not wake up. You can now open your eyes but you
are sound asleep. Now open your eyes and see your record. All right, eyes wide open but sound
asleep). There will be no fatigue of your arm and finger and you can make a good record.
Ready,-go,-one,-two."
The words enclosed in parentheses were not spoken when the subjects were in the
non-trance state. Typical records of four-day cycles by two different subjects are reproduced in
Figure 46 (not available). Three of the subjects worked through three such cycles each, and two
worked through four cycles each. The mean results for each subject and for the (238) respective
states are summarized in Table 37 (not available). A brief inspection of this table shows that
while the five subjects differed considerably among themselves, they all showed an advantage in
favor of the trance. With one exception, however, the percentage differences were small. The
average of the percentages is 16.08, which has a critical ratio of 2.8. The means of the average
work scores yield a difference of 0.25 kilogram-meters, which amounts to an advantage of 12.4
per cent. This difference has a critical ratio of 4.46. The results of Williams's more meticulous
experiment thus confirm Nicholson's results qualitatively in showing the trance state to possess a
heightened susceptibility to suggestion, but they disagree radically as to the amount.
Pursuing the matter of checking Nicholson's experiment still further, Williams performed
a number of experiments of the type represented by Nicholson's tracings shown in Figure 45
above. The outcome of this part of Williams's project may be seen by an inspection of Figure 47.
In general it may be said that Williams's results very definitely do not confirm the (239)
marvelous capacity of the hypnotic state to transcend physiological normality in restoring
exhausted muscles to approximately complete freshness and vigor. It is true that upon shifting
instantly from one state to the other there is marked evidence of a revival of muscular power, but
this does not appear to be noticeably different when passing from the normal to the trance state
than when passing from the trance to the normal state. Neither do these differ greatly from the
partial revival resulting from renewing suggestions of increased strength in the same state, as
shown by the second pair of records. Such temporary heightening of capacity to perform on the
ergograph as the result of any stimulation of an emotional character has, of course, long been
known to investigators in the field of work and fatigue.
It is impossible to account with any certainty for the quantitative differences between the
results reported by Nicholson and by Williams. The meticulous care with which Williams's
investigation was conducted raises considerable presumption that his results represent a fairly
close approximation to the truth of the situation. To this presumption is added the fact that the
results as reported by WilIiams do not demand a transcendence of physiological normality but, on
the contrary, are quite within the range of what might be expected from the increased influence of
suggestion, both of encouragement and of anaesthesia to fatigue pains, when given in the trance
over similar ones given in the non-trance state.
HADFIELD'S INVESTIGATION OF HYPNOTIC SUGGESTION AND THE
STRENGTH OF GRIP: The investigations reported in the preceding pages of the present
chapter have been concerned with muscular performances which involved fatigue and the painful
accompaniments of aching muscles when the activity took place in the nontrance state. There
remains the problem of the influence of hypnotic suggestion of power, etc., upon the upper limit
of momentary muscular contraction where no fatigue or pain element is involved.
A typical experiment of the older and uncontrolIed variety is casually reported by
Hadfield to illustrate an (240) inspirational work on the psychology of "power." He asked three
men to "submit themselves to test the effect of mental suggestion on their strength." The strength
was measured by means of a hand dynamometer, an instrument which registers on a convenient
dial the strength of hand grip. The three are said to have averaged 101 pounds per grip in the
normal condition. Next, when in the hypnotic trance, they were suggested to be "very strong."
Under these conditions they averaged 142 pounds per grip. Again, when in the same state, it was
suggested that they would be "weak," under which conditions their grips averaged only
twenty-nine pounds. There is nothing particularly surprising in the reaction reported as obtained
under suggestion of weakness. But if the lOI pounds really represents the maximum voluntary
contraction in the waking condition, to increase this by forty-one pounds represents a genuine
supernormal performance. However, if Hadfield's subjects knew the purpose of the experiment,
as seems probable from certain phrases of his report quoted above, his results at once lose all
scientific value so far as concerns any question of the hypnotic transcendence of physiological
normality.
YOUNG'S CONTROLLED INVESTIGATION OF HYPNOTIC SUGGESTION
AND STRENGTH OF GRIP: In marked and agreeable contrast to the type of hypnotic
experimentation just considered is an investigation of the same problem by P. C. Young. His
experimental work was carried out in the Harvard psychological laboratory in 1922, and enjoys
the distinction of being the first well-controlled large-scale experimental study to be reported in
the history of hypnotism. Like Hadfield, Young measured his subjects' strength of grip by means
of a dynamometer. Twelve subjects were tested repeatedly in both the normal and the hypnotic
state, the two states alternating in the A, B, B, A order so as to minimize the influence of practice
and any other distorting temporal effects. On every experimental (241) occasion each hand was
tested five times. Before each grip, in both states alike, the suggestion to "grip the apparatus as
tightly as possible" was given. The subjects worked in ignorance of their scores. In all, 616 such
measurements were made, half in each state. Young's published report states that the averages of
the grip scores for the two states were exactly alike.
SUMMARY AND CONCLUSIONS: Several of the investigations reviewed in the
present chapter are, by reason of one defect or another, of only historical interest. Of the more
recent authors, Williams finds no significant difference between the trance and the waking state
either in ability to hold the extended arm in a horizontal position or in the amount of tremor when
so doing. Nicholson found a great, though undetermined, amount of advantage in the trance state
in power to resist muscular fatigue. Williams's more carefully controlled experiment confirms
this finding, but the advantage of the trance state shrinks, on the average, to something in the
neighborhood of 12 or 16 per cent, according to the method of computation. Young reports no
difference between the two states in respect to the power of momentary grip of a dynamometer.
In the consideration of these apparently conHicting facts, as well as evidence to be
presented in the two succeeding chapters, it is important to note that there are involved two
distinct questions which have not always been clearly distinguished. They are:
1. Does the hypnotic state, as such, produce a supernormal capacity as to muscular power,
resistance to fatigue, and so on?
2. Is the hypnotic state a condition of heightened suggestibility such that performances beyond
the usual voluntary range can be induced if suggestions to that effect are given?
The evidence at present available seems to indicate with some definiteness that the former
question should receive a negative reply, but that the latter should be answered affirmatively.
The superficial inconsistency between the negative results of Williams's arm-extension
experiment and the positive results of his ergographic experiment probably disappears on these
(242) assumptions. The arm-extension experiment should not have shown any trance advantage
because no strong or special suggestions were given that the arm should stay up indefinitely or
that there would be no fatigue pains. In the case of his ergographic experiment, however, there
was a definite suggestion that the subject would suffer no fatigue and that he would make a good
record; and we find the trance in this case yielding a clearly superior performance. It seems
likely, therefore, that if the arm-extension experiment should be repeated with suggestions of
superior strength in holding the arm horizontal and of absence of fatigue, the trance state might
show a moderate but fairly consistent advantage. It is difficult to account for Young's negative
finding in the matter of grip, though it would be worth while also to repeat his experiment, with
more vigorous suggestions that the subject would manifest superior strength than Young seems to
have used. Exactly the same suggestions should, of course, be given in the waking tests. If,
however, this experiment should fail to yield a moderate advantage for the trance, it would
probably mean that the suggestions against fatigue may be the essential active factor in the
protracted activities involving fatigue and pain. (241)

18. CH 10: HYPNOTIC SUGGESTIBILITY AND THE TRANSCENDENCE OF

VOLUNTARY CAPACITY (Continued): CLARK HULL: HYPNOSIS AND
SUGGESTIBILITY: APPLETON-CENTURY-CROFTS: NY: COPYRIGHT 1933: 1961
B. PROCESSES PRIMARILY RELATED TO THE SENSES: AT THIS point we pass
from the consideration of the supposed hypnotic transcendence of voluntary capacity primarily in
the field of the motor processes to that primarily in the field of the sensory or receptor processes.
In this connection we shall consider first the matter of suggested hyperesthesia (increased acuity
of the senses); of selective anaesthesias (insensitivity to certain arbitrary classes of stimuli) ; of
suggested analgesia (specific insensitivity to pain); and of suggested hallucinations (reactions as if
to non-existent stimuli), with particular reference to certain investigations concerning the
influence of hypnotic suggestions on processes normally supposed to be under the control of the
autonomic nervous system.
HYPNOTIC SUGGESTION AND HYPERESTHESIAL: During the long history of
hypnosis, beliefs in the greatest variety have been held as to the supposed supernormal sensitivity
to faint stimuli of subjects while in the hypnotic trance. The following passage from the sober
Braid may serve as an example:
“When we consider that in this process we have acquired the power of raising sensibility
to the most extraordinary degree, and also of depressing it far below the torpor of natural sleep;
and that from the latter condition, any or all of the senses may be raised to the exalted state of
sensibility referred to, almost with the rapidity of thought, by so simple an agency as a puff of air
directed against the respective parts.” (245) Bramwell, a relatively recent writer, may also be
cited as typical in this respect. He believes (4, 89 ff.) that hypnotic suggestion is able to improve
markedly the senses of vision, hearing, smell, and temperature and the discrimination of lifted
weights and of two points on the skin. The most of his opinions are based on the usual clinical
observations, but in the cases of lifted weights and the two-point threshold he reports a few
experimental results. Unfortunately, the common failure of clinical workers, namely, the failure
to run adequate control experiments in order to determine exactly what the same individuals
could do in the non-trance state at the same stage of practice, renders these results quite valueless.
Despite the widespread belief in hypnotic hyperesthesia, there has been during the long
history of hypnosis but one properly controlled experimental investigation of the subject, so far as
the present writer has been able to discover. The experiment in question was reported by P. C.
Young and was part of the same project as the investigation of strength of grip described above.
The sense investigated was cutaneous sensitivity to faint pressure. The stimulating mechanisms
employed were two short hairs, each glued at one end to a skewer. The procedure of stimulation
was to hold the hair by means of the skewer in a position perpendicular to the skin and apply
sufficient pressure to make the hair bend a constant amount. Under these conditions, one hair
exerted a pressure of 570 mgm. and the other a pressure of 220 mgm. The subject merely said
"Yes," when he felt a touch. Care was taken that no other clue save that of the pressure of the hair
should inform the subjects when they were being stimulated. Since it is more of an achievement
to feel the weak than the strong hair, the successful detections of the former were given twice the
weight of the latter. Special suggestions of a greatly heightened sensitivity to the slightest touch
were given in the hypnotic state. Each of the fifteen subjects was tested repeatedly in both the
hypnotic and the waking state. On the average, each subject was stimulated 765 times, half in
each state, the entire group totaling 13,340 stimulations. The final score of each subject was the
ratio x 100 of the number of his successes in the trance divided by the number achieved in the
waking state. (246)
The results of Young's investigation are summarized in Table 38. The mean of the twelve
subjects is 93.4 + 2.36, which indicates that hypnotic suggestion, instead of heightening their
sensitivity, seems, upon the whole, rather to have had an effect in the opposite direction. If we
exclude the three subjects judged by Young as being definitely lethargic, there still remain ten
individuals who might with full propriety be expected to show supernormal sensitivity if hypnotic
suggestion is capable of producing such an effect. Yet, of these, only two attain even a modest
superiority in the trance, and the selected group as a whole yields a mean score of 98.2 + 1.93.
This means that, even with the elimination of those individuals who under hypnosis become
sleepy and lethargic, there is no tendency whatever for subjects under the influence of hypnotic
suggestion to display a tactual sensitivity transcending that of the normal waking state. Young
reports, however, that all of his subjects (249) while in the trance expressed the belief that they
possessed greatly augmented powers of perception. It is probable that this observation indicates
one of the major sources of the long prevailing belief in the power of hypnotic suggestion to
induce gross hypersensitivity.
HYPNOTIC AND POST-HYPNOTIC APPRECIATION OF TIME: Closely related
to the question of hyperesthesia of the ordinary senses is that of the supposed increase in accuracy
of the time sense in the hypnotic and post-hypnotic conditions. Bramwell cites experiments
performed by Delboeuf in 1886, and describes at length fifty-five attempts made by himself with
a Miss D, in 1895. The technique of the latter investigation is typical of both. The subject was
told while in the trance that at the expiration of a certain number of minutes, such as 320 or 4,453
or 20,190, she should mark a cross on a piece of paper and then write down the exact time at
which this occurred. If the time notation on the piece of paper turned out actually to be 4,453
minutes, say, from the time when the suggestion was given, the trial was counted as correct. Of
the fifty-five trials, only nine are said to have been incorrect and these were in error by from two
to five minutes only. Several other subjects are reported to have yielded results of a substantially
similar nature.
Bramwell believes that his results indicate a genuine transcendence of the normal
voluntary capacity to appreciate time. Unfortunately we have here once more an experiment
without the proper accompanying control procedure, which alone would make interpretation
possible; i.e., Bramwell neglected to determine the precision of his subject's appreciation of time
under parallel conditions when motivated by ordinary waking instructions. Moreover, the
experiment is complicated by the investigation's failure to determine the ability of the subject to
perform, within the period elapsing from the giving of the suggestion to the time set for its
execution, the arithmetical computations presumably necessary for its success. Instead, he seems
tacitly to have assumed that such computations were beyond the subject's capacity, at least in the
waking state, because early in the experiment she appeared not to be able to (248) perform them
within a brief set interval. In any case, the posthypnotic capacity to perform computations is
distinct from that of time estimation, and the two ought not to be confused in an experimental
situation. Delboeuf's results possess the same limitations as those of Bramwell, and therefore need
no comment.
Fortunately we possess in an investigation reported by Stalnaker and Richardson what
purports to be a well-controlled experiment. Nine subjects were tested as to their power to judge
short intervals of time during ten experimental periods each. The subject was merely instructed to
say "Time" at the conclusion of what he believed to be the correct interval following the starting
signal given by the experimenter. The actual times thus marked off were measured by means of a
ten-second stop-watch. In all, 1,080 such determinations were made, half in the trance and half in
the normal waking condition. The judgments were equally divided with each subject between the
intervals of one, two, and three minutes respectively. The sequences of these time intervals and of
the two states were rotated in a manner intended to eliminate constant errors from the final
results. These are shown in Table 39 (not available).
This table presents a strikingly different picture from that reported by Delboeuf and by
Bramwell. Of the three durations judged, the one-minute interval was estimated better in 'the
waking state and the other two more accurately in the trance (249) state. The critical ratios (given
in the last column) show that none of these differences is statistically significant, i.e., that all are
probably due to chance. It is entirely typical of such experiments that, despite the lack of real
advantage in the trance, the subjects in that state believed themselves remarkably exact in their
judgments as compared, in their opinion, with the corresponding judgments made in the waking
state.
To the negative results of Stalnaker and Richardson may be added the unpublished results
from a minor experiment performed in the Wisconsin laboratory by Ruth Eken. This experiment
compared the ability of persons to judge sixteenminute time intervals in the trance and the waking
state respectively. Her results agree with those of Stalnaker and Richardson in showing no
statistically significant advantage for either state.
In the light of the above evidence we are probably justified in the tentative conclusion,
despite the apparent clinical observations to the contrary, that hypnosis really does not heighten
the power to judge the passage of time, but only produces in the subject an illusion of greater
capacity to do so.
HYPNOTIC ANRESTHESIA AND SURGERY: It may be recalled that in the
quotation from Braid mention is made of the power of suggestion in the trance state to depress
sensitivity as well as to heighten it, just as Hadfield (8) speaks of the capacity of hypnotic
suggestion to weaken the power of muscular contraction as well as to strengthen it. In the motor
field such a loss of power resembles paralysis. Spontaneous functional paralyses closely
resembling those producible by suggestion have long been known and studied as symptoms of
hysteria. The parallel functional loss of sensory capacity is usually called ancesthesia. Sometimes
the loss of the sense of pain receives the special name of analgesia. Spontaneous functional
amesthesias have also long been recognized as symptoms of hysteria and employed in the
diagnosis of this complex group of disorders. Ironically enough, the presence of spontaneously
anresthetic spots on the skin was once supposed (250) to be a valuable distinguishing
characteristic of witches and was utilized in their detection.
In the case of the inhibition of the power of muscular contraction, there is clearly not
involved the same type of transcendence of physiological normality that a supernormal
performance implies; nothing is easier than a submaximal effort. In a sense the same may be said
of an apparent blunting or depression of sensitivity by suggestion, as contrasted with a suggested
hyperesthesia. Suggested anresthesiamay possibly be nothing more than the influencing of the
speech apparatus by means of suggestion in such a way that it will deny the fact of stimulation
rather than affirm it. If carried to its logical limit, however, the lack of response to stimulation
might conceivably imply a genuine physiological dissociation between the stimulus and the
response. This may best be seen, perhaps, by considering the case of very strongly injurious
stimuli. Normally such stimuli evoke such overt reactions as shrinking or withdrawing from the
stimulus, grimaces, cries, tears, and so on. Reactions of this type are, however, more or less under
voluntary control, provided the intensity of the stimulus is not too great. That a dissociation 1 of
some kind between stimulus and the normal response may be brought about by means of
appropriate sug~ gestions of anresthesia or analgesia in the hypnotic state, even in the case of the
most violent stimuli, has long been believed, and apparently with some reason.
On March 3, 1837, a committee of the French Academy of Medicine examined the claims
of Berna, a young magnetizer who asserted that by means of hypnosis he could render one of his
subjects insensible to pain. Berna surrounded the experiment with certain arbitrary restrictions
and apparently would permit the use of rather mild pin-pricks only. Moreover, he had the face of
the anresthetic individual partially covered so that it was not possible for the members of the
committee easily to observe whether the subject showed facial expressions of pain or not. Under
the circumstances it is not (251) surprising that the committee reported negatively on the claims
of hypnotic anaesthesias. Despite this negative official pronouncement, however, hypnotic
ancesthesia gradually began to be used in a practical way as a means of relieving pain in medical
practice. Indeed, the examination of Berna's claims by the committee of the French Academy of
Medicine seems to have been brought about indirectly by the report of a painless tooth extraction
by M. Oudet.
On April 4, 1845, James Esdaile performed in a native hospital in India his first operation
during which the patient was under the influence of hypnotic anaesthesias. In the course of a year
he had performed over a hundred such operations. Before leaving India, Esdaile performed
thousands of painless operations, about 300 of which were major ones, nineteen being
amputations. This aid to surgery was especially welcome at that time because chemical
ancesthetics had not yet come into use. Esdaile asserted that during the operations his patients not
only remained perfectly quiet but failed to show the ordinary physiological signs of pain such as
changes in pulse, dilation of pupil, and so on, which are characteristic of operations in the
non-trance state. It happened that the use of chloroform in surgery was introduced in India while
Esdaile was in the midst of his surgical experiments, and the discovery of ether occurred not long
after. As a result, this phase of hypnosis, which might otherwise have seen a great development,
was destined to become little more than a scientific curiosity. However, reports occasionally
appear in America at the present time of minor operations being performed painlessly by this
means. In Germany and Austria the practice would appear to be somewhat more common,
particularly in cases of childbirth.
Some of the American cases, at least, do not agree very well with Esdaile's claims. This is
true of one instance of a supposedly painless tooth extraction reported to the author by Dr. Milton
H. Erickson, the physician who performed the hypnosis. Though striking, the case could hardly
be cited as a perfect example of ancesthesia. The patient, an adolescent girl, whimpered and
flinched decidedly while the tooth was being extracted. Upon being dehypnotized, however, the
subject professed a complete amnesia for the trance proceedings, and smilingly expressed surprise
when told that her teeth were out, even (252) though at the time her mouth was very bloody from
the effects of the recent ordeal. This has the appearance of amnesia rather than of ancesthesia.
There is some reason to believe, however, that there may be a larger element of amnesia involved
in the ordinary chemical anaesthetic than is usually believed.
But in addition to these overt and more or less voluntarily controllable reactions, the
human behavior laboratory has discovered a number of more or less subtle forms of reaction of
which the subject is very dimly or not at all aware, but which" when recorded by suitable
apparatus, not only show responses to injurious stimuli but show them in amounts which are
easily measurable. Among the activities most readily available for such purposes are respiration,
the pulse, and changes in the electrical condition of the skin. It will be recalled in this connection
that Esdaile long ago referred to the pulse and the size of the pupil as signs of pain which were
notable by their absence when operations were performed in a state of hypnotic anaesthesias. No
doubt such observations as he was able to make at the time were more or less haphazard and
unsystematic. Recently, however, some careful quantitative work has been performed to
determine such truth as may exist in these claims of complete lack of responsiveness to painful
stimuli by subjects when under hypnotic ancesthesia.
ELECTRICAL REACTIONS TO PAINFUL STIMULI DURING HYPNOTIC
ANAESTHESIA: The response most extensively studied in this connection is the galvanic skin
reaction (the so-called psychogalvanic reflex), though the number of subjects experimented on
and reported previous to 1930 is small. In most cases but a single subject was employed, and the
psychological techniques, as (253) well as the methods of recording, differed considerably.
Moreover, in no case does there appear to have been any systematic measurement of the
amplitudes of the reactions. Possibly in part, at least, as a result of these various circumstances,
the several investigators report widely differing results. About equally numerous are reports of
complete dissociation, on the one hand, and complete normality of reaction, on the other. Reports
of similar investigations of spontaneous hysterical anresthesias present substantially the same
picture.
As an example of one of the better studies of this type we shall consider that of Levine.
He employed two hysterical female subjects, one of whom could be put into a very deep state of
hypnosis during which what appeared to be profound cutaneous anaesthesia was produced. The
other subject had what appeared to be an equally profound spontaneous cutaneous anaesthesia of
the arms and legs. With both subjects (eyes closed) a sterile venipuncture needle thrust through
the skin on an anaesthetic area evoked no observable reaction, and the subject professed complete
lack of knowledge that the event had taken place. This, of course, is the classical picture of
hysterical and hypnotic anresthesia in its extreme form. Control experiments on the hypnotic
subject in the waking state evoked marked gasping, withdrawal, and facial reactions characteristic
of pain. Records of the galvanic skin reactions to these stimulations in the various states, together
with the onset of the stimulus, etc., were secured by means of a delicate string galvanometer
provided with a photographic attachment. The galvanometer was placed in circuit with a pair of
non-polarizable electrodes which were attached to the palm and back, respectively, of one of the
subject's hands.3 Typical galvanic reactions to venipuncture needle pricks when under the
influence of hypnotic anresthesia and when in the normal condition are reproduced as Figure 48
(not available). The results of this investigation are summarized in Table 40 (not available).
Levine concludes from his experiments with these two subjects that anresthesia, either as
induced by hypnotic suggestion (254) or as evolved spontaneously as an hysterical symptom,
does not produce a dissociation between painful stimuli and the normal electrical reaction
observed in the skin under such stimulation. In judging the significance of these results, it should
be remembered that Levine, like his predecessors in this field, appears to have been looking for
some kind of all-or-none relationship, and it seems not to have occurred to him to make
systematic measurements to determine whether the amesthetic condition might not have been
associated with a certain amount of weakening of the reaction. It is true that his published records
offer slight basis for such a conjecture. It must be remembered, on the other hand, that a tendency
of moderate amount can be neither proved nor disproved by the mere inspection of a small
number of records. Accidental elements of unknown nature within the immensely complex
mechanism of man are such that tendencies of moderate strength may be either (255) simulated or
entirely masked in any individual record. It is only when careful objective measurements have
been made and the data subjected to statistical treatment which will permit the accidental
elements to neutralize each other that a really convincing comparison can be made between an
experimental (aesthetic) series and its (normal) control. We are fortunate, however, in possessing
a study, that of Sears (23), which satisfies these conditions rather well.
SEARS'S INVESTIGATION OF REACTIONS TO PAINFUL STIMULI DURING
HYPNOTIC ANALGESIA: Sears employed as subjects seven male university students. They
were hypnotized by a combination of optical fixation and suggestions of relaxation, sleep, etc. All
appeared to go into the deepest trance, as shown by profound analgesia and posthypnotic
amnesia, which, in turn, were evidenced by verbal report and general appearance when the
conventional tests were applied. The specific anaesthesias for the purposes of the experiment was
induced by the following suggestions: "I am going to take all the sensation out of your left leg.
You will be able to feel absolutely nothing that touches it; all the feeling will be gone.
Now I am sliding my hand along the calf of your leg. Each time I do this my hand will feel
lighter and lighter. All the sensation is going. It's going! It's going! Now you can scarcely feel my
hand; it is just like feathers brushing your leg.4 Now all the feeling is gone. You cannot tell when
I touch your leg and you can feel absolutely no pain when I prick it."
The pain stimulations employed were made by means of the instrument shown in Figure
49 (not available). This consisted essentially of a cylinder (B) partially projecting from a metal
tube. Beyond the cylinder in the tube was placed a compression spring. At the opposite end of the
cylinder was placed the curved and pointed metal piece (A). The stimulus was administered to the
calf of the subject's leg by pressure, with the cylinder and direction of pressure perpendicular to
the surface of the skin. This applied the point A at an angle of about forty-five degrees (256)
which, while decidedly painful, did not rupture the skin. Pressures of twenty ounces were
delivered for periods of approximately one second. Uniformity in respect to the amount of
pressure was attained by forcing the metal tube toward the leg until the calibration mark at (b)
Was just about to pass from view. Over the opposite end of the metal tube was one end of a
rubber tube which communicated at the other end with a recording tambour. When the cylinder
(B) was forced into the tube by pressing the point (A) against the skin, this increased the air
pressure within the rubber pneumatic system which, in turn, caused an automatic registration on
the smoked paper by the stylus of the tambour. The record produced by such a stimulation is seen
as the elevation, C', in the line C of Figure 51 (not available).
It was regarded as a possibility that the phenomenon of hypnotic anaesthesia might be
little. more than a kind of naive simulation. Another possibility considered, one perhaps scarcely
distinguishable from the first, was that hypnotic anaesthesia accomplishes little more than a
control of the speech apparatus and the ordinary voluntary muscles is such a way as to cause them
to deny the incidence of the Stimulus. Because of such considerations, Sears measured
simultaneously reactions to painful stimuli at various levels of behavior with respect to the degree
of voluntary control. His first criterion of analgesia was the subject's verbal report (usually a
spontaneous statement) as to whether or not he felt pain when pricked by the stimulator. No
subjects were used who did not show this with practical uniformity. A second criterion, of a
somewhat different nature but clearly on the voluntary level, was obtained by recording a
component of the facial grimace normally made to painful stimulations. Respiration. was
recorded as another re-' action to pain, but one under joint voluntary and involuntary (257)
control. As reactions to pain characteristic of the presumably non-voluntary type or at least with a
minimum of voluntary control, records were taken of the pulse and the electrical condition of the
skin-the "psychogalvanic reflex." All reactions except those of vocal denial were recorded by
special automatic devices on the smoked paper of a kymograph.
SEARS'S METHODS OF RECORDING AND MEASURING REACTIONS TO
PAINFUL STIMULI: The component of the grimace chosen for recording was a sharp upward
movement of the skin near the outer corner of the eye. This was recorded with the aid of the
apparatus represented in Figure 50 (not available). This is essentially a receiving tambour with
the metal stylus so modified at its end that it could be attached to the subject's skin by means of
an adhesive substance. When the skin at the point of attachment would rise, this would lift the
rubber dam stretched over the cup T. The resulting change in the air-pressure within the rubber
tubing would produce a parallel movement of the stylus of the recording tambour at the other
end. A response recorded in this manner is shown at B' of line B, Figure 51 (not available). In
scoring the results of this portion of the investigation such responses were counted simply as
present or absent, regardless of their amplitude.
It may be noted that this apparatus is responsive to head (258) movements as well as
grimaces. To minimize the former, the head was held in a special padded head-rest and the
subject was instructed to cooperate by holding as still as possible. It is likely, however, that
occasionally slight head movements were included in the score of grimaces. It is doubtful
whether any vicious distortion of the results would be produced by this, however, because
presumably both reactions are of substantially similar nature so far as voluntary control is
concerned.
The breathing was recorded on the smoked paper of the (259) kymograph by means of the
Sumner pneumograph (Fig. facing p. 212) (not available). A typical respiratory response to the
prick is shown in the portion of line E, Figure 5 I, just beyond the middle. The characteristic
respiratory response to a painful stimulus is a deepening of the respiration associated with an
irregularity of the rhythm. The total amplitude of the breathing is conveniently measured by
tracing the respiratory line with the stylus of the linear oscillometer. The variability of amplitude
of breathing is easily measured by means of the same instrument. In this latter case, however, the
stylus of the oscillometer merely traces a course determined by the lower limits of each
respiratory cycle, indicated by the broken line connecting these points on section E of Figure 51
(not available). It is clear that for perfectly even breathing the broken line will be quite
horizontal. When traced by the oscillometer such a line will register zero oscillation, whereas if
the breathing is irregular the broken line will oscillate considerably, which will register in
appreciable amounts on the dials of the instrument. In order to secure an indication of the effect
of a particular stimulus in the case of both types of measurements of the breathing, it was
necessary to perform them over a period of twenty seconds preceding the stimulus and one of
twenty seconds following it, and subtract the former from the latter. These differences are the
"gains" shown in the two sections of Table 41 involving respiratory reactions. Normally there
should be a positive "gain" following the stimulus in the case of both the total oscillation and the
variability of amplitude. If the suggested anesthesia were completely effective, however, there
should be no change of any kind either in amplitude or variability of amplitude. If, however, the
suggested anaesthesia should be only partially effective, there should be a certain amount of gain,
but not so much as shown by stimulation of the normal leg.
The pulse activity was recorded on the smoked paper of the kymograph by means of the
Erlanger-Meek recording sphygmomanometer. The cuff of this instrument is shown on the arm of
the subject in Figure 39. A fairly typical transcript of a record secured by it is shown in line D of
Figure 51. A little-studied reaction revealed by this instrument is a fairly marked reduction in the
amplitude of the pulse tracing following a stimulus of the type here employed. This is clearly
evident (260) in the record shown in Figure 51 (not available). Just as was seen to be the case
with respiration, the amplitude change is accompanied by an irregularity from pulse to pulse. The
pulse tracings were accordingly measured by means of the linear oscillometer both for total
oscillation and for variability, the same as in the case of respiration. The periods measured
preceding and following the stimulation, however, were in this case only seven and onehalf
seconds in duration. In harmony with the above statements, the "gain" in pulse amplitude should
normally be a negative value, while the variability in amplitude should be positive. That, upon
the whole, this is the case may be seen by consulting the appropriate sections of Table 41. The
effect of a genuine tendency to an<£sthesia in either case, however, would be to shift the "gain"
shown on the normal leg in the direction of zero.
Changes in the electrical condition of the subject's skin were secured from the index and ring
fingers of the right hand, which were placed in Ostwald non-polarizable electrodes. The
electrodes were wired in circuit with a delicate D' Arsonval wall galvanometer which threw a
beam of light upon a translucent screen. As the electrical conditions of the subject's skin changed,
this spot of light would move to the right or the left across the screen. Normally the spot of light
displays a continuous slow drifting movement, but two or three seconds after the subject has been
stimulated by the prick, the beam of light makes a relatively rapid movement, first slightly in one
direction, then a considerable distance in the other direction, and finally, after twenty seconds or
so, back approximately to its original position. By a simple mechanical arrangement, an assistant
followed this spot continuously with a pointer connected with a recording device, and the
resulting movements were traced on the smoked paper along with the records of the other
processes. Such a tracing is shown at G in Figure 51 (not available). Since it is impossible to
distinguish with certainty the reactions to the specific pain stimuli from the movements due to
spontaneous internal processes, Sears hit upon the happy device of measuring the total oscillation
of the line for periods of twenty seconds preceding and following the stimulation and subtracting
the former from the latter. The final difference obtained by massing a (262) considerable number
of measurements of this kind may safely be regarded as representing changes due to the prick
stimulus rather than to organic spontaneity. Thus, in a wholly objective manner, quantitative
determinations were secured of the extent of galvanic reactions to painful stimuli. The "gain" in
oscillometer units following the stimulus in this case should be a positive value, and any
influence of suggested anesthesia should be manifested by a reduction in this amount.
RESULTS OF SEARS'S MAIN EXPERIMENT ON HYPNOTIC ANALGESIA: It
should be noted at this point that, for precise determinations of such effects as are here involved,
a number of precautions must be observed which were disregarded in the investigations
previously performed in this field. In the first place, there is the basic question of whether the part
of the body which is under the influence of suggestive anesthesia is any less sensitive to pain than
a symmetrical part (normal) of the same subject's body at approximately the same time. Unless
such a difference can be shown there can be no evidence of local anesthesia. In the present
investigation the left leg was in all cases the one suggested to be anesthetic. An area of two
square inches on the calf was selected for stimulation. In order to secure a practically
simultaneous measure of the normal sensitivity of the subject on a part where anesthesia had not
been suggested, an equal number of exactly similar stimulations were given to a corresponding
area on the calf of the right leg. As a further precaution designed to prevent any such temporal
error as negative adaptation from distorting the results, the stimulations of the two legs were
delivered in multiples of four in the A, B, B, A order.
The results of this main experimental procedure are summarized in Table 41 (not
available). An examination of this table shows at the outset that the subjects differ more or less in
their median scores: in most cases one or more subjects present fairly marked differences from
the tendency of the group as a whole. While it would be rash to assume that the subjects do not
differ from each other in capacity for hypnotic anEsthesia, it is probable that the most of this
apparent inconsistency is due to the (263) mixture of spontaneous reactions with the results
evoked by the injurious stimulus. Such chance distortions as these, however, tend to be
neutralized by the pooling of results from a number of different subjects. Consequently the row
of means in this table presumably presents a distinctly more trustworthy picture of the situation.
An examination and comparison of these values leads to the conclusion that while hypnotic
ana:sthesia does not result in the complete dissociation (or inhibition) of all reactions to a painful
prick, it very nearly or quite abolishes certain reactions and reduces the amount of the reaction to
a certain extent in all. Moreover, in all cases except that of pulse oscillation the statistical
reliability of the difference between the anaesthetic and the control leg has a critical ratio of
practically 3 or above. The general harmony of the nature of the difference in the case of pulse
oscillation with the differences shown by the other measurements, as well as the complete
harmony of these latter results with each other, entitles the pulse-oscillation difference to a degree
of confidence in excess of the bare statistical reliability represented by the critical ratio of 1.74.
With the general fact of a positive influence of hypnotic anaesthesia before us, we are in
position to consider the relation of the volitional character of the several reactions to the (264)
magnitude of this influence. The results of the entire investigation are briefly summarized from
this point of view in Table 42. There it appears that the reactions classed as wholly and as
partially voluntary are approximately alike in the extent of the diminution of the reaction as the
result of suggested anesthesia. The three measures classed as non-voluntary, on the other hand,
present an appreciably smaller amount of reduction. Incidentally, the exceedingly small amount
of reduction of the galvanic skin reaction (22 per cent) harmonizes fairly well with the results
reported by Levine, since his method of merely inspecting a moderate number of records with no
massing of results might easily have permitted such a difference to pass unnoticed. These results
as a whole agree in a general way with the hypothesis that hypnotic suggestion is operative to a
considerable extent on the volitional level, but the fact that both pulse and the galvanic skin
reaction show very appreciable effects points to some other and more deep-lying mechanism as
also operative.
 SEARS'S TWO CONTROL SERIES: Despite the care with which Sears's main
experiment was set up and the controls contained within it, two special control experiments were
necessary. The first was desirable because of the possibility that, since the subjects had received
the suggestions of anxsthesia in the left leg in the trance, this. tendency might persist more or less
into the waking condition. He accordingly ran a control series in which the respective calves were
stimulated alternately in A, B, B, A order in the non-trance state with no suggestions whatever
regarding anesthesia.
Secondly, it was important to know to what extent, if any, the subjects were able to
repress by voluntary effort the usual physiological reactions to pain in one leg while permitting
normal reactions to stimulations of the other. If such a repression could take place, it would tend
to indicate that the hypnotic influence, if any, was acting on the volitional mechanisms rather
than directly on the physiological processes in question-an extremely important matter to be
determined. These considerations naturally led to a second or "voluntary" control (265)
experiment. In this the subjects in the normal waking condition were instructed to repress or
conceal so far as possible all reactions to pain when the left leg wp ~ stimulated, but to make no
such effort when the right leg VI as stimulated.
Sears's first or normal co ltrol series as shown in Table 43 is chiefly of interest in demonst
'ating that when anaesthesia is not suggested the left leg yields s\ bstantially the same reactions to
the prick stimulus as the right. There is very little, if any, tendency for the anGesthesia of the left
leg during hypnosis to persist into the waking condition. That the differences are probably due
merely to sampling errors is indicated by the low reliability coefficients for all six differences
between means.
The results on the "voluntary" control experiment (Table 44 (not available)) seem on the
surface to show that the diminutions in the reactions to injurious stimulation, as shown by the
main experiment, indicate a capacity quite beyond the power of voluntary simulation. The critical
ratios of all six differences indicate that they also are probably the result of mere sampling errors,
just as was the case with the results in the normal control experiment. A closer examination of the
situation, however, gives rise to some doubt as to the significance of these results. Sears states,
doubtless with truth, that the facial flinch can ordinarily be inhibited at will. Nevertheless,. the
mean flinch values as shown by this table indicate practically no influence of voluntary inhibition
whatever. The fact that in this portion of the experiment such an inhibition did not take place is
accordingly extremely paradoxical and throws doubt upon the significance of the lack of
differences shown in the remainder of the table. Clearly, this portion of the experiment ought to
be repeated in such a way as to provide assurance that the subjects really exert themselves to the
maximum to suppress the signs of pain when the left leg is being stimulated. Until this
fundamental fact has been determined, only tentative conclusions concerning the relation of the
volitional mechanism and hypnotic analgesia will be possible.
SELECTIVE ANAESTHESIA AND RAPPORT: Hypnotic anaesthesia may take the
form of the complete abolition of a sense, apparently resulting in a condition such (267) as total
blindness, or it may be limited to an apparent blindness to certain specific things or classes of
things. As an example of the second form of anaesthesia, a subject may be told that Mr. X has
gone out of the room, that he is no longer present. A good subject will usually "accept" such a
suggestion even though Mr. X may not have moved. If, later, the subject is asked to count the
people in the room, pointing to each as he is counted, Mr. X will be entirely overlooked even
though he may be in plain view. This is known as selective anaesthesia.
 A special form of selective ana;sthesia which has played a great role in the history of
hypnosis is the phenomenon known as rapport. This term has been used to designate the fact that
occasionally a subject who has been hypnotized by a given experimenter will not respond to
suggestions given by others and, indeed, will appear to be quite oblivious to communications
from everyone else. This was particularly true earlier in the history of hypnosis. The term implies
a connection established between the hypnotic subject and some other person, usually the person
who has performed the hypnosis. There is, of course, nothing strange or mysterious in the ability
of one person to have communicational connection with another such as would give rise to the
need for a special term, like rapport, to emphasize it. The remarkable thing about the
phenomenon demanding explanation is, rather, the seeming ana;sthesia to stimulations emanating
from persons other than those en rapport.
This paradoxical emphasis of the obvious seems to have grown out of the widely
prevailing notion that hypnosis is a state of sleep. In this latter condition, of course, for a subject
to be responsive to communications from anyone at all would indeed be surprising. By a subtle
process, all too common in the history of our subject, the theoretical preconceptions of the
experimenters appear to have been absorbed by the individuals en;ployed as subjects and, through
suggestion, to have resulted in their displaying the selective ana;sthesia in question. If the
hypnotists had held the view that all subjects when in the hypnotic state should walk on their
heels we would, no doubt, have observed most of the subjects walking in this manner, and
heel-walking would doubtless have been regarded as an inherent characteristic and indispensable
mark of the trance (269) condition. It is true that the phenomenon of spontaneous rapport appears
to have been met with in pathological (hysterical) somnambulism or "sleep-walking," though it is
possible that in the latter case also this has been the result of the unconscious acceptance by the
patient of a theoretical presupposition held by the physicians in charge. At all events, most
modern writers on the subject believe that rapport is not an inherent phenomenon of hypnosis but
is encountered only when the subject has received in one form or another a suggestion that he
shall be responsive solely to communications and suggestions from some particular person or
persons. The phenomena thus find their true characterization as a special case of suggested
selective anesthesia. Accordingly there seems about to be terminated one more of the chapters in
the history of hypnotism, the content of which is largely an account of the slow and halting
process of correcting an original misconception. The term rapport seems destined for oblivion
along with caloric and other useless nonentities, and the more quickly this takes place the better.
REACTIONS ASSOCIATED WITH HYPNOTICALLY SUGGESTED
HALLUCINATIONS: At this point we pass from suggested anaesthesia, where the subject
appears not to sense what is really stimulating him, to suggested hallucinations, where through
suggestion the subject behaves as if sensing stimuli which do not exist. Hallucinations are among
the easiest to obtain of all hypnotic phenomena. Descriptions of two typical hypnotically induced
hallucinations have been given above. With the exercise of a little care by the experimenter, many
subjects in the hypnotic state will act out responses to suggested pseudo stimulations in the most
realistic manner: they will make sour faces when drinking water at the suggestion that it is
vinegar; they become gay from water turned to wine by suggestion; they shiver at suggested cold
and fan themselves at suggested heat; they cry out and writhe at suggested pain. This, of course,
is all very dramatic and convincing to the casual observer who is searching merely for thrills. But
here, just as in the case of suggested anaesthesia, there arises the question of whether all (270)
this is genuine or whether it is merely superficial acting and naive simulation.
 The introspective approach here, as usual, is out of the question because suggestion is just
as operative on the introspective (speech) reactions as on any other. We are accordingly driven,
just as in the case of hypnotic anaesthesia, to the experimental observation of reactions which are
supposed normally to be beyond voluntary control. An example may make this clear. Suppose
that it has been suggested to a subject in the hypnotic state that he shall see a bright light. If at the
moment he reports seeing the light his iris should contract sharply as it normally does when a
bright light really enters the eye, other factors such as fixation remaining constant, this would
constitute a more significant indication of the physiological reality of the hallucination than the
mere verbal report, since the latter may so easily be simulated. The author knows of no
systematic and controlled experiments of this kind on the pupillary reactions to hallucinations. He
has, however, made rough observations on a number of hypnotic subjects who indicated by the
ordinary verbal and other signs that they experienced such hallucinations. The results of these
observations were entirely negative; there were no indications of pupillary responses to
suggestions of strong light, though the subjects asserted positively that they saw the light. It
would be unsafe, however, to conclude from such rough experiments that the phenomenon may
not be detected by careful measurements on susceptible subjects. In the first place, it is said that
occasional individuals possess voluntary control of the iris muscles. Secondly, Cason and
Hudgins have shown that the iris action to light can become conditioned to auditory stimuli. This
being so it would not be surprising if the reaction had occasionally become conditioned to words
such as light and dark and their more complex verbal equivalents which would be employed in
giving the suggestions.
A second experimental phenomenon which has appealed to investigators in this
connection is the tendency of painful stimuli to produce marked, but normally unconscious,
electrical changes in the skin. The na"ive subject cannot simulate this, at least by direct volition,
because he does not know what to simulate. The question usually put in this connection is: Will
(271) these physiological reactions characteristic of pain take place in cases of hallucinated
painful stimuli? While no careful quantitative investigation of this problem is known to the
writer, repeated rough trials in his own laboratory confirm similar reports of other casual
investigations that such galvanic reactions do take place; the galvanometer's response following
the hallucination of a pin-prick appears to be substantially like that resulting from a genuine
pin-prick. We must hasten to add, however, that in the normal waking condition it has been
repeatedly observed that the threat of a pin-prick also yields a galvanic reaction substantially like
that of a real pin-prick. Here the reaction has evidently become so conditioned to the words,
whether used as suggestion or threat, that it takes place automatically. That this reaction is so
much more readily evoked by suggestion than the iris reaction is probably due to the fact,
repeatedly observed in the writer's laboratory, that the galvanic skin reaction is extremely easy to
condition, whereas to condition the iris is relatively difficult.
COLD-SORES INDUCED BY HYPNOTICALLY SUGGESTED
HALLUCINATION:
The general problem of the nature and extent of various physiological reactions to suggested
pseudo-stimuli has attracted a considerable amount of attention among medical men, especially in
Austria and Germany. Their experiments are often exceedingly elaborate, particularly on the side
of chemistry, physiology, and bacteriology. We shall begin our consideration of this material with
a fairly detailed examination of one of the more striking studies, that by Heilig and Hoff (n). This
concerns the induction of herpetic blisters (cold-sores) by means of suggested hallucination.
Three psychopathic women were employed as subjects. The typical experimental procedure was
as follows: The experimenter recalled to the subject's mind when in the deep hypnotic state an
extremely unpleasant emotional experience connected with her particular neurosis. Thereupon the
patient showed signs of great excitement, such as flushing, tossing about, and groaning in fear. At
this juncture the psychiatrist stroked the patient's lower lip and suggested a feeling of itchiness
such as the patient had often felt when a (272) cold-sore was starting. Calming suggestions were
then given, after which the patient was dehypnotized. During the following twenty-four hours or
so the patient reported an itching on the lower lip, at the end of which time there appeared a slight
swelling on the lip in question. At the end of forty-eight hours numerous small herpetic blisters
appeared, which gradually merged into a large blister. After a few days this dried up, forming a
scab, and finally healed. The contents of the fresh blister, when transmitted to the cornea of a
rabbit, caused a herpes in the rabbit on the third day, thus proving the technical genuineness of
the cold-sore. This picture, with unimportant variations, holds for all three patients.
It is important for a proper understanding of the significance of the facts just recounted to
note that the authors state explicitly that "it was not possible to induce herpetic blisters by the
mere suggestion that they were starting, even when this suggestion was fully detailed and very
impressive, unless unpleasant experiences of the type described were suggested at the same time.
The presence in the nostrils of the herpes bacteria and the suggestion that the cold-sore was
coming were not alone sufficient to make it develop." The point is that it was necessary, in
addition, to reduce the physiological resistance of the patient by the emotional ordeal. That the
patient's resistance was actually reduced by this was ascertained by determination of her opsonic
index both before and after the emotional crisis.
In the interpretation of Heilig and Hoff's experiment it is necessary to note that, despite
the many precautions taken, there is superficial uncertainty as to whether the cold-sores were due
to the emotional upset alone, or to the combination of this with the suggestion that a cold-sore
would develop. To clarify this matter fully it would be necessary to induce a similar emotional
upset in the hypnotic state unaccompanied by suggestions of herpetic blisters. If this should be
followed by the appearance of cold-sores, the indication would be tolerably clear that the specific
suggestions and resulting hallucinations of coming herpetic blisters had nothing whatever to do
with their appearance in the experiment described above. As a matter of fact, there are contained
in the report definite indications that the latter is the case, since (273) at least two of the subjects
are reported as regularly having herpes labialis after excitements and strong feelings of fear. The
hallucination of itching on the lips, etc., may accordingly be dismissed from serious consideration
as possessing any physiological or somatic significance in the production of the blisters.
There remains, however, ample evidence in the lowering of the opsonic index and the
subsequent appearance of real coldsores that the emotions aroused in the trance were
physiologically genuine and not a superficial simulation. This is a problem more or less distinct
from that of the hallucinated itching of the lip. It must be remembered, however, that the
emotional subject-matter employed with each patient was itself very genuine, and there is
consequently strong a priori presumption that if the attempt were made to arouse similar
emotional crises in the normal waking state this also would have been followed by the lowered
opsonic index and the appearance of herpetic blisters. If the experiment should be performed and
the blisters should appear, it would mean, of course, that hypnosis itself had nothing to do with
the appearance of the herpes and the whole experiment would immediately become pointless so
far as hypnosis is concerned. Heilig and Hoff, despite the many admirable features of this
investigation, display the almost universal experimental weakness of clinicians in their failure to
perform this critical control experiment. Until this is done we shall not be able to say whether
hypnotic suggestion with such subjects differs significantly from ordinary waking suggestion in
its capacity to arouse emotions.
DIGESTIVE REACTIONS EVOKED BY HYPNOTICALLY INDUCED
HALLUCINATIONS: For the most part physiological investigations of the type now under
consideration have concerned themselves with whether various suggested hallucinations possess
the same digestive and assimilative characteristics as do the real situations which the
hallucinations simulate. In a brief review of the more important of these investigations we may
mention first that by Heilig and Hoff .This study is concerned with gastric secretion and duodenal
peristalsis. Five hospital (274) nurses were employed as subjects. These individuals were given
meals which were normally neither especially liked nor disliked. With some of the meals the
suggestion was given that the food was one that the subject greatly relished; with others, that the
food was one which the subject greatly disliked. When partly digested, the food was pumped
from the subject's stomach and subjected to chemical analysis. It was found, in general, that the
food partially digested under conditions of suggested relish showed a distinctly greater amount of
acid secretion than did a normal control meal, and that the food partially digested under
suggestions of disgust showed less acid than the control The significance of these results lies in
the fact that exactly such differences in acidity of gastric secretion are characteristic of real relish
and real disgust.
In a related experiment, parallel contrasted suggestions were given during the eating of a
barium meal. The motility of the stomach was then observed by means of the Roentgen screen. It
was found that suggestions of relish greatly strengthened the vigor of peristalsis, whereas
suggestions of disgust tended to cause peristalsis to cease or even to produce the beginnings of
antiperistalsis. In both of the above experiments, therefore, we have objective evidence of the
physiological genuineness of the suggested attitude toward the food. In neither of them, however,
was a control experiment performed to determine the influence of similar suggestions given in the
non-trance state.
In this connection may also be mentioned the study by Heyer, in which he investigated the
influence of hypnotic suggestions calculated to produce a pleasant, non-apprehensive state of
mind upon peristalsis which had been pathologically enfeebled. He employed as subjects four
female patients with a history of psychic trauma. They were all suffering from weak peristalsis,
constipation, and general loss of appetite. These digestive symptoms were supposed to be related
in some way to the traumas. Heyer reproduces presumably typical X-ray photographs showing
the state of the patients before and after the suggestive treatment. This is said to have produced, at
least (275) temporarily, a much more normal tonus and motility, with corresponding relief from
physical symptoms.
A second and somewhat related study by Heilig and Hoff (Ia) reports the influence of
suggested moods upon the quality and quantity of urine secreted. Ten subjects, mostly hospital
nurses, were employed. Pleasant suggestions such as, "You are feeling fine," "You are in a good
humor," etc., or unpleasant suggestions such as "You are sad," "You are depressed," etc., were
given them. It is reported that quite uniformly the pleasant suggestions were followed by reduced
secretions of water, chlorides, and phosphates, whereas unpleasant suggestions were followed by
excess secretions of water (diuresis) which was accompanied by a parallel loss in body weight.
These urinary effects are the same as are known normally to accompany the genuine moods
corresponding to those suggested. The experiment accordingly points to the physiological
genuineness of the mood evoked by suggestion. No control experiment was performed to
determine the relative influence of similar suggestions administered in the waking state.
Passing to the question of the physiological genuineness of suggested hallucinations of
consuming food and water, there may be mentioned, first, a study by Langheinrich (I7) concerned
with hallucinations of eating. The number of subjects employed in this investigation is not stated,
but the context seems to indicate that there were at least three. Under deep hypnosis it was
suggested that the subject was either (I) drinking bouillon or (2) eating butter. It had been
ascertained previously that these foods were not repugnant to him. After the suggested eating of
each kind of food the contents of the duodenum were removed by means of a tube. It was found
that the secretion following the hallucination of drinking bouillon was thin and bright yellow in
color. It was therefore assumed to be bile which had come directly from the liver. The
corresponding secretion obtained after the hallucination of eating butter was a dark and viscous
bile which had presumably come from the gall-bladder. The quantity of the secretion also is (276)
said to have differed characteristically for the contrasted food suggestions. The butter suggestion
was followed by a mean secretion of 9.2 mm. per five-minute period, the bouillon sugges-. tion
by a mean secretion of 4.4 mm., whereas a secretion of only I mm. was obtained when no food
suggestion of any kind was given. It is also reported that the albumin-digesting power of the
secretions following the food suggestions was greater than that obtained when no food
suggestions had been given. No control experiment in waking suggestion was performed.
An experiment related to the one just described was carried out by Delhougne and
Hansen. The subject in this experiment is said to have been a very childlike, impressionable,
easily suggestible man. Nineteen experiments were performed, all in deep hypnosis. In three of
these, actual meals were eaten and the nature of the digestive secretions determined. In the
remaining sixteen experiments, parallel determinations were made of the secretions following
purely hallucinated meals of three contrasted chemical constitutions. It is reported that with the
real meals the amounts of the several ferments contained in the digestive secretions were found to
differ characteristically according to the nature of the food eaten. Exactly parallel differential
secretions are said to have followed the suggested meals: a suggested meal rich in albumin
induced an increased secretion of pepsin and trypsin with stasis of the remaining ferments; the
suggestion of a fatty meal of chocolate and almond paste induced an increase in lipase secretion
with stasis of the remaining ferments; and a suggested carbohydrate meal induced an increase in
the distaste values with stasis of the others. No waking control experiments were performed.
A second related experiment involving water metabolism has been reported by Marx (20).
In deep hypnosis an empty glass was held to a subject's lips and it was suggested to him that he
was drinking large quantities of water and enjoying it. It was found regularly that about twenty
minutes after the suggestion there was a distinct increase in the amount of urine secreted, which
was accompanied by a decrease in its specific gravity. The amount secreted during two hours
previous to the suggestion varied from 80 to 100 units, whereas the amount secreted during about
an hour following the suggestion rose to 400 or 500 units. During the same experiment the
specific (277) gravity fell from a pre-suggestion value of 1,025 to a post-suggestion value of
1,002. There was also observed a 7 per cent thinning of the blood following the suggested
drinking of the water. On the basis of a previously performed experiment it is stated that the
above results correspond closely to those following actual drinking of water, with the exception
that the thinning influence on the blood was considerably less after the suggested drinking. A
hypnotic control experiment in which no suggestions of drinking were made is said to have
shown no significant changes either in diuresis or in the thinning of the blood. No waking
suggestion experiment was performed.
The above experiments, while of significance for certain purposes, possess the
characteristic defect from our present point of view that they uniformly fail to determine how
much, if any, of the alleged results of hypnotically induced hallucinations would also result from
substantially similar suggestions administered in the waking state. We are fortunate, however, in
possessing two studies in which such control experiments were performed. These latter
investigations throw interesting light upon the results from the experiments just described.
The first of the waking control investigations is reported by Bath (I). On the basis of
experiments with some twenty-five subjects, it was determined through the actual eating of food
that a leucocytosis usually appears twenty minutes or so after food has been eaten. For example,
it was found that following the administration of milk the leucocytosis amounted to 45 per cent;
after a mixed meal it measured 34.3 per cent; and after the consumption of bicarbonate of soda it
was found to be 25 per cent. Our chief interest, however, lies in the control experiment which
followed. In this the subjects were merely shown food. This alone produced a leucocytosis of
68.9 per cent. It is also reported that one subject who was both blind and anosmic received in the
waking state merely the suggestions of the sight and odor of food. The amount of leucocytosis
resulting in this case is not stated numerically, but is reported to have been large. (279)
The second study in this group of waking control investigations is published by Luckhardt
and Johnston. These authors report experiments performed on a law student at the University of
Chicago. They suggested to him the hallucination of eating a meal, and the subject responded
with realistic motions as if actually eating. The suggested hallucinations were followed by an
increase both in volume and in acidity of the digestive secretions which was comparable with
corresponding results obtained when an actual meal had been eaten. One control experiment is
reported in which the subject was hypnotized but no foods or food suggestions were given. The
somewhat surprising observation was made that the mere induction of the trance produced an
increase in secretions quite comparable to the actual consumption of foods.
Luckhardt and Johnston performed a second control experiment which is especially
pertinent to the problem now before us. In this case they merely talked to their subject about
appetizing food when he was in the normal waking condition. It was found that this
gastronomical conversation alone was sufficient to cause a rise both in the acidity and in the
volume of the digestive secretions. Luckhardt and Johnston publish representative tables of their
results. One of these shows a set of experimental data when a real meal was eaten, and one shows
data obtained following a typical waking conversation about food. A comparison of these tables
indicates that, striking as the results of the hypnotic hallucinations of eating may be, they
probably do not differ materially from those of an ordinary conversation about the same foods as
were suggested in connection with the hypnotic hallucination.
All of the results concerning digestive secretions recounted above may appear much more
natural when we remember (279) how a hungry person's mouth waters when he sees or smells
appetizing food. It thus turns out, as has so often been the case in the history of hypnotism, that
results supposed to be supernormal and to be peculiar to the hypnotic state prove upon
investigation to be wholly normal, and even commonplace, performances quite as characteristic
of the waking state. They appear remarkable only because the attempt is rarely made to observe
them in any state.
 DIFFERENTIAL SECRETIONS PRODUCED IN ANIMALS AS CONDITIONED
REFLEXES: It is extremely fortunate that it is not necessary for us to depend entirely upon
casual observations for non-suggestion phenomena with which to compare the experimental
findings in hypnosis briefly sketched in the preceding sections. It happens that for many years
Pavlov, the Russian physiologist, has been investigating experimentally with dogs the capacity of
stimuli to acquire the power of evoking a typical secretory reaction, that of salivation. By merely
permitting a hitherto neutral stimulus, such as the sound of a buzzer or a bell, the ticking of a
metronome, the odor of camphor, the touch of a cold or a warm object on the skin, etc., to affect
the animal's sense-organs while it is actively salivating, the neutral stimulus quite readily and
uniformly acquires the capacity to evoke the salivary secretion. This much, of course, we might
easily have anticipated from observing our own behavior when hungry and in the presence of
food. In the suggestion experiments the words of the experimenter presumably are merely
performing the function served by the arbitrary sounds, temperatures, etc., in the conditioned
reflex experiments.
We are doubly fortunate in that Pavlov's experiments also furnish a parallel to the more
remarkable phenomenon of differential gastric secretions for different types and quantities of
suggested foods and drink reported above by Langheinrich, Delhougne and Hansen, and Marx. It
seems that a dog can be made to salivate to two types. of neutral (unacquired) stimuli. The dog
will normally respond with a moderate amount of thick saliva possessing much mucin when meat
powder has been placed in its mouth; when a dilute acid is introduced into the (280) mouth,
however, a copious flow of thin saliva with little mucin will take place. Pavlov reports numerous
experiments in which one stimulus is able to evoke one of the reactions, and a distinct stimulus
will produce the other. For example (2I) 184), Dr. Vassiliev, one of Pavlov's pupils, administered
a thermal stimulus of 47 C. in conjunction with the placing of meat powder in a dog's mouth
and, on other occasions, a second cutaneous thermal stimulus of 0 C. with the introduction of
acid into the same dog's mouth. After repeated paired stimulations it was found that the cutaneous
stimulation of 0 C. would evoke a copious secretion of thin saliva, the so-called defense
reaction, and a stimulation of 47 C. would evoke a secretion of thick saliva rich in mucin, the
so-called alimentary reaction. The parallel of this experiment to the suggestion experiments of
Langheinrich and of Delhougne and Hansen is sufficiently close to require little comment. There
is accordingly no special a priori reason to doubt the suggestion results reported by the authors
cited above; neither is there reason to believe that anything unique or beyond the range of
physiological normality has taken place even though specific and internal secretion, the nature of
which is quite unknown to the subject, is found to be a component of the behavior associated with
a hypnotically suggested hallucination, attitude, or mood.
SUMMARY AND CONCLUSIONS: We may now take a brief backward glance at the
evidence which has been assembled concerning the more sensory aspects of the power of
hypnotic suggestion to produce a transcendence of voluntary capacity. Young's study showed that
cutaneous sensitivity probably is not improved by suggestions of hyperesthesia. It is noteworthy,
however, that his subjects when in the hypnotic state believed themselves to have a greatly
increased sensory acuity. Stalnaker and Richardson showed that subjects in the trance, when it
was suggested that they would display heightened sensibility to time intervals, manifested no
increased power of judging time but-exactly as in Young's study-uniformly indicated their belief
that their sensitivity was greatly enhanced.
We thus come to the contrasted matter of sensitivity lowered (281) by suggestion. We find
reason to believe that hypnotic anesthesia is a genuine phenomenon in the sense that painful
stimuli will often be endured without the usual outward signs of suffering. For this reason
analgesia was doubtless useful in surgery before the discovery of the chemical amesthetics.
Sears's careful measurements of various voluntary and non-voluntary reactions show that
moderately painful stimuli applied under hypnotically suggested anaosthesia produce complete or
nearly complete abolition of verbal reports of pain, of facial grimaces ordinarily associated with
pain, and of two different experimental indicators of pain normally manifested by respiration, a
semivoluntary process. However, it was found that the presumably non-voluntary pulse activity
shows a clear reaction to painful stimuli but that the amount of the reaction is reduced through
suggested anaosthesia by a half or three fourths. Neither is the galvanic skin reaction obliterated,
though it is reduced under the influence of suggested anaosthesia by something like a fifth of its
normal amount. It thus appears that hypnotically suggested anaosthesia produces a nearly perfect
abolition of the ordinary signs of pain in the case of voluntary and semivoluntary processes but
shows a reduction ranging from 20 to 75 per cent, only, on non-voluntary processes. This
differential effect of hypnotically induced analgesia probably indicates that the process in
question is in some way significantly related to the mechanism of volition. Here also must be
considered the phenomena traditionally called rapport, which appear upon analysis to be nothing
whatever but a selective anaosthesia resulting either from direct or from indirect suggestion.
The matter of suggested hallucinations may be summarized rather simply by saying that
some remarkable differential physiological responses have been obtained with susceptible
subjects by suggesting to them pseudo-stimuli, particularly such as relate to food and drink.
Normal waking control experiments in this field are very scarce, but such as are available, taken
together with the Russian experimental work on conditioned salivary reactions, seem to indicate
that the observed phenomena probably do not differ materially from those producible in both
animals and men by means of the conditioned reflex technique. This raises a certain amount of
presumption (282) in favor of the view that the non-voluntary differential secretions and
metabolic effects obtained from suggested hallucinations may be essentially conditioned-reaction
phenomena.
Owing to the fact that in most cases comparable control experiments in the non-trance
state were not carr;ed out, it is impossible at the present time to make a final decision concerning
whether the effects observed under hypnotic suggestion are any more pronounced than the effects
following parallel suggestions administered in the waking state. Certain facts, some of which will
be presented in the next chapter, raise some a priori expectation that a moderate tendency in this
direction may be anticipated from suitable experimental procedures. (283)

19. CH 11: HYPNOSIS AS A STATE OF HEIGHTENED SUGGESTIBILITY: CLARK HULL:

HYPNOSIS AND SUGGESTIBILITY: APPLETON-CENTURY-CROFTS: NY: COPYRIGHT
1933: 1961
At least since the time of Braid's later activity (1847), it has been generally recognized by
writers on the subject of hypnosis that persons display while in the hypnotic trance a heightened
responsiveness to direct suggestion. Bramwell (3) 283), who had seen certain of Braid's later
works not generally accessible, gives it as Braid's view that "Such suggestions [verbal] acted
more powerfully during hypnosis." Binet and Fere (2) 364) assert, "In slight hypnotism, in the
states described as fascination, magnetic sleep, etc., subjects appear to be peculiarly liable to
suggestion." Bernheim (I) 141) writes, "Hypnotism ... facilitates suggestion when it can be
induced; but other suggestions may sometimes succeed when the suggestion of sleep is
inefficacious." 1 Forel (7,60) remarks, "Sleep or slumbering is produced by suggestion. But as
soon as this is present, the suggestibility is increased by sleep, as long as the latter does not
become lethargic." Sidis (IS) 70) states, "With the Nancy school we agree that suggestion is
all-powerful in the hypnotic trance; the hypnotic trance is, in fact, a state of heightened
suggestibility, .... " Moll (I2, 264) says, "We have seen that susceptibility to suggestion is the
chief phenomenon of hypnosis .... A certain degree of susceptibility to suggestion is normal."
Bramwell (3, 333) reports it as his own observation that, "In hypnosis, to gain increased
suggestibility it is often only necessary to repeat quietly some recognized formula once or
twice ... " Lastly, Wingfield (I7) 8) states, "To put the matter in another way, successful
suggestion will induce hypnotic phenomena, and the phenomena in their turn induce an increased
suggestibility." However different may be the theoretical bias of the various writers, (289)
and however varying may be their interpretation of the phenomenon, there appears to be no
disagreement regarding the fundamental fact that whatever else it may be, the hypnotic trance is a
state of heightened susceptibility to suggestion.
It is wholly characteristic of hypnotism as a science that despite the long period during
which the phenomenon of hypersuggestibility has been widely recognized, no serious attempt at a
quantitatively controlled verification of the principle has, been made. We shall later have
occasion to note a similar experimental sluggishness in the case of habituation as related to
hypnosis. But the modern scientific temper will not be satisfied with mere belief and opinion; it
demands objective evidence from controlled experiments that subjects actually are more
susceptible to suggestion when under hypnosis than when in the normal state. And if there really
is an increased suggestibility while in the trance, this must be to a certain degree or quantity. How
great is this? Is the difference, for example, a relatively slight and unimportant one, or is a person
many times more susceptible to suggestion in the trance than in the waking state? How does the
degree of this supposed hypersuggestibility vary from one subject to another? How does it vary
with the degree of hypnotic habituation of the subject? Are all the processes classed as
suggestion, e.g., the indirect forms of suggestion such as Binet's progressive weights as well as
suggested movement, paralysis, and the like, susceptible to this influence and equally so? Many
other questions of a similar nature at once suggest themselves. Unfortunately, there is as yet
quantitative experimental evidence bearing on only a few of the questions listed above. The two
preceding chapters have, to a certain extent, been concerned with this problem, though in an
indirect manner. The present chapter will consider the question specifically and in such detail as
the experimental evidence warrants.
TWO EXPERIMENTAL APPROACHES TO THE PROBLEMS OF
HYPERSUGGESTIBILITY: It must be observed at the outset that quantitative problems
involving hypersuggestibility may be attacked by two related but significantly different
techniques. By the first method, the process used as an indicator of the presence of (286)
hypersuggestibility is the same as that which was originally involved in the generation of the
condition. An example illustrating the applicability of the approach is seen in the fact that a
subject will usually go into a trance, as indicated by rate of lid-closure, with special rapidity
immediately after having been aroused from a previous trance. Because this manifestation of
heightened suggestibility involves the influence of one suggestion procedure upon the response to
a second procedure leading to the same act as the first, we shall for convenience refer to it as
homoactive hypersuggestibility.
 The second technique of measuring or detecting the presence of hypersuggestibility
involves the power of one suggestion process to heighten the susceptibility to some second and
different suggestion process. An example of this form of hypersuggestibility is seen in the
influence of the suggestions which lead to hypnosis, manifested objectively by lidclosure, in
increasing the subject's susceptibility to suggestion of forward movement of the head or of the
horizontally extended arm. Because the action here is one of suggestion process indirectly upon a
second suggestion directed to the evocation of a different act, we shall, for convenience, refer to
this second phenomenon as heteroactive hypersuggestibility.
Certain problems may be approached more effectively by the first method, certain others
by the second, and some with about equal readiness by both. It should be especially noted that
these two varieties of hypersuggestibility are related to each other in somewhat the manner that
direct training is related to the transfer of training. Because of this difference the results by one
method (where both may be employed) may not entirely agree with those by the other. We must
accordingly be cautious in generalizing from experimental results obtained by one method alone.
Because of certain complications arising from the similarities existing between hypnotic
hypersuggestibility and practice effects, it will be simpler to begin the consideration of the
problems of hypersuggestibility from the heteroactive angle, which happens to be the one from
which they are usually viewed. (287)
RELATIVE SUGGESTIBITY TO POSTURAL SUGGESTION IN THE TRANCE
AND WAKING STATE: The quantitative problem of hypersuggestibility was broached
experimentally in 1929 by Hull and Huse. In essence the procedure employed was very simple. It
resolved itself into an experimental determination of whether subjects on the average are any
more responsive to postural suggestions in the trance than in the normal waking state. It was
merely necessary to measure repeatedly the length of time in seconds required to evoke the
maximal postural reaction in each state, average the results, and then observe which condition
yields the shorter times and by how much. The actual procedure, however, was complicated by
the necessity of protecting the final averages from certain distortions otherwise due to such
factors as the results of practice, possible perseveration effects, and so on.
"The basic experimental procedure of the ... investigation was to induce by suggestion t'vo
falls fairly close together in one of the two states and then two in the other state, with about 1.5
min. separating the second fall of the just two from (288) the first fall of the second two. In any
pair of falls, the second suggestion was given approximately ½ min. after the fall resulting from
the first suggestion. Every subject served on four different experimental days. On two of these
days the waking suggestion was given first and on two the trance suggestion preceded. We shall
arbitrarily call the days on which the wakIng suggestion came first, A; and the days on which the
trance suggestion preceded, B. The record reproduced in Fig. [52] (not available) shows a
complete tracing made on an A day. In order to minimize any constant errors which might result
from the time factor, half of the subjects were experimented on in the A B B A order and the
other half in the B A A B order. The experimental programs for the two groups of subjects are
shown in detail in Table [45] (not available). (289)
"The method by which the suggestion time is secured from the records may be best
eXplained by referring to Fig. [52 J. The upper line is the time record, with a notch marking the
lapse of every 5-sec. period. The second line is a record of the forward-backward postural
movements, the sharp peaks representing the 'fall' or culmination of the response, after which the
suggestion is discontinued. The lowest line records by means of notches the exact point at which
the suggestive stimulation begins and approximately where it ends. The former is marked by 'A'
and the latter by 'B.' After the record had been shellacked, a perpendicular white line was drawn
from 'A' across the time line. A second perpendicular white line was drawn across the time line
from the point of maximum reaction. The distance between these two lines represents the
suggestion time. Thus the first suggestion time of the record shown in Fig. [52] (not available) is
9 sec., the second is 26 sec., the third 12 sec., and so on. The suggestion times for all subjects are
given in detail in Table [46] (not available). Those of the record shown in Fig. [52] (not
available) appear in the June 4th entry for Subject No.6." (8)
We may now seek in Tables 46 and 47 the answer to our primary question as to whether
subjects in general are more susceptible to postural suggestion in the hypnotic trance. The
procedure of the present experiment and the mode of tabulation makes it possible to obtain two
fairly independent comparisons of the suggestion time in the t\VO states. One is based on the
time of the first waking suggestion as compared with that of the first trance suggestion. The other
is based on the time of the second waking suggestion as compared with the second trance
suggestion. Even a casual inspection of Table 46 shows that, except for occasional single
measures, the tranc~ state yields a decidedly more rapid response to this type of suggestion than
does the waking state. This is shown without ambiguity by means of the contrasted columns. The
average suggestion time required to evoke the first waking fall was 26.90 sec., whereas that
required for the first trance fall was 11.97 sec., only 44-5 per cent as long. Again, the average
duration of suggestion required to evoke the second waking fall was 20.42 sec., whereas that for
the second trance fall was only 9.12 sec., or 44.6 per cent as long. The close percentage
agreement between these t\vo determinations from entirely distinct sets of measurements is
reassuring.
An even clearer view, perhaps, is presented by Table 47 (not available), (290) which
gives the results of Table 46 in condensed and summary form. A glance down this table shows
that, taking each person's scores in each state as a whole, everyone of the eight subjects shows a
clear difference in speed of response to postural suggestion in favor of the trance state. The
statistical reliability of this difference yields a critical ratio of 4.9, which is extremely
satisfactory. The indication is that resistance to this particular form of suggestion as measured by
time of

Williams's Experiment Concerning Relative Suggestibility in the Trance and the Waking State
Because of the central importance of the question involved as bearing on the essential nature of
hypnosis, several other investigations have been directed to the same end. All of them have been
patterned in a general way after that of Hull and Huse. The most extensive of these investigations
was that of Williams, which embraced not only suggestion of the postural type but included
susceptibility to indirect suggestion as well (p. 35Iff.). One of the experiments reported by
Williams (I6) was in all respects comparable to that of Hull and Huse, including the number of
subjects employed. A condensed view of the outcome of this experiment is shown in Table 48.
There it may be seen that the same general results were obtained as were reported by Hull and
Huse, though the trance effect is not response is reduced by 55.5 per cent, or something more
than half. To put the matter in terms of susceptibility, and less exactly but more in conformity
with conventional ways of thinking, we may say that the Hull-Huse experiment furnishes a
preliminary indication that people tend to be about twice as suggestible in the hypnotic as in the
waking state. (292)
 WILLIAMS'S EXPERIMENT CONCERNING RELATIVE SUGGESTIBILITY IN
THE TRANCE AND THE WAKING STATE: Because of the central importance of the
question involved as bearing on the essential nature of hypnosis, several other investigations have
been directed to the same end. All of them have been patterned in a general way after that of Hull
and Huse. The most extensive of these investigations was that of Williams, which embraced not
only suggestion of the postural type but included susceptibility to indirect suggestion as well .
One of the experiments reported by Williams was in all respects comparable to that of Hull and
Huse, including the number of subjects employed. A condensed view of the outcome of this
experiment is shown in Table 48 (not available). There it may be seen that the same general
results were obtained as were reported by Hull and Huse, though the trance effect is not (293)
pronounced. Two of the eight subjects deviate by showing a slight speed advantage for the
normal state. The mean speed of "falling" in the trance state is reduced below that in the waking
condition by 32 per cent. The statistical reliability of the difference between the mean times for
the two states is satisfactory, since the critical ratio is 4.16.
CASTER AND BAKER'S DETERMINATION OF THE DEGREE OF HYPNOTIC
HYPERSUGGESTIBILITY: A third experimental determination of this fundamental
relationship has been reported by Caster and Baker. They tested ten subjects in both states on four
successive days. In general they followed the Hull-Huse technique, with the exception that the
activity chosen as an indicator of suggestive responsiveness was the forward movement of the
horizontally extended right arm. In order to minimize complications due to fatigue, the arm was
supported on the specially constructed free-swinging platform shown in Figure IS. This was
suspended from the ceiling by a chain, as already described A thread connected the rounded end
of the platform to the same type of recording device as was employed in the two previous studies.
In this experiment very special precautions were taken to keep the words, tempo, and emphasis of
the suggestions constant at all times. In order to eliminate the possible unconscious influence of
the sight of the subject's responses upon the expression in the voice of the person giving the
suggestions, two experimenters were employed, and the duties of the two were so divided that the
suggester did not (294) see the subject at all dUring the critical phases of the experiment. The
main measurement was the duration of suggestion required to move the end of the arm-support a
distance of eight inches. Because of the relative absence of tremor by this technique, Caster and
Baker attempted also to measure the suggestion latency, i.e., the duration of suggestion before
any response at all took place.
The results of both sets of measurements are given in Table 49 In general the work of
Caster and Baker confirms the results of the two studies already summarized, though the outcome
by the criterion of latent time is somewhat blurred, presumably by reason of the tremors which
would make it difficult to determine just when response to suggestion really began. ( 295) But
with the suggestion time, where conditions are more nearly comparable with the former studies,
we encounter a picture almost identical with that already seen. Of the ten subjects, eight show a
shorter mean suggestion time in the trance. The difference between the means of the reactions in
the respective conditions has a critical ratio of 4.00, and is accordingly statistically reliable. The
suggestion time is reduced from that of the normal state by 45 per cent, a value about midway
between that obtained by the two studies just considered.

THE PROBLEM OF HYPERSUGGESTIBILITY INVESTIGATED WITH THE

AID OF PHONOGRAPHIC STIMULATIONS: Lastly there may be mentioned in this
connection the experiment reported by Jenness. This study is unique in being the first in the
history of hypnosis proper where all of the suggestions were given by means of the phonograph.
There should accordingly be no possibility of a constant error entering these results through the
unconscious influence on the suggester's intonation, emphasis, etc., which might conceivably
arise from his expectation of a certain experimental outcome. The hypnotic trance was induced by
means of the suggestions delivered by Estabrooks's commercial hypnotic record, while the
suggestions designed to show heteroactive effect were delivered by means of a phonograph disk
specially prepared for the purpose. These latter suggestions were such as to cause the horizontally
extended right arm to sway forward. The first portion of the suggestions presented by this record
is as follows:
"Your arm is resting on the support, so as to free it from fatigue. Listen carefully to what I
say and pay attention to the suggestions as they are given you. Imagine how it would feel if your
arm were to move forward in an are, so as to extend straight before you. Forward, forward,
forward, forward, forward. Just as if your arm were to move forward, forward, forward, forward,
forward. Moving forward, forward, forward, forward, forward. All the time your arm is moving,
forward, forward, forward, forward. Imagine that your arm is moving forward, forward, forward,
forward, forward."
The modes of the arms suspension and of recording it movements were substantially the
same as those (297) used by Caster. With a view to correcting a possible tendency for the effect
of the trance when preceding the waking tests to persist into the latter and thus reduce the amount
of difference between the two states below its true amount, Jenness tested his subjects for waking
arm suggestion on entirely different days from the corresponding tests given in the trance.
Unfortunately his method of alternating the two states during the eight experimental days with
each subject permitted practice effects sufficient to distort somewhat the values for the individual
subjects, though the fact that he reversed the order with half of them presumably saves the final
average effect from any serious distortion from this source. Jenness's results, in so far as they bear
on the present problem, appear in the (297) first two columns of Table 50 (not available). This
table shows that six of the eight subjects required a shorter arm-suggestion time while in the
trance. The mean reduction of hypnotic arm-suggestion time below that in the waking state is
13.2 sec., which has the satisfactory critical ratio of 3.8. This represents a trance reduction in the
suggestion time of 15.9 per cent, a difference much smaller than that obtained in the other three
investigations.
Taking all four studies together, we find a total of thirty-four subjects experimented on.
Of these, twenty-eight, or 823 per cent, showed a greater or less degree of heteroactive
hypersuggestibility when in the trance state. Since the number of measures made on each subject
was rather meager, it is likely that some, at least, of these negative cases may be the result of
inadequate sampling of the individual subject's behavior. It is likely, however, that occasional
individuals do not show the phenomenon. A weighted average from all thirty-four subjects shows
that the time of suggestion response in the trance is reduced from the normal level by a total of 38
per cent. Accordingly one of the most widely held views concerning hypnosis is confirmed,
though the amount of the hypersuggestibility thus revealed, while considerable, is probably far
less than the classical hypnotists would have supposed had the question ever occurred to them.
It must be remembered that the heteroactive influence upon responses to suggestions of
postural movement is only one of many possible ways of securing quantitative evidence of
hypersuggestibility. We may anticipate with confidence that the picture will change to a certain
extent, possibly a great deal, as other methods are employed. In the pages immediately following
we shall consider the evidence bearing on this problem from the homoactive point of view.
It is probable that some significant factor in Jenness's technique produced this marked
deviation in amount from previously obtained results. Jenness is inclined to the view that it may
have been caused by the impersonal nature of the phonograph records. Jenness also reports that
the hypersuggestibility of the trance state was much more marked early in the experiment than
later and concludes from this that hypnotic hypersuggestibility may be reduced through practice.
Comparable graphs from the published results of Hull and Ruse (p. 288) and from two tables by
Williams (I6) were plotted with a view to testing this hypothesis, but none of these graphs
showed any indication of such a tendency. (298)
SOME PROBLEMS CONCERNING THE RISE AND DECAY OF HYPNOTIC
HYPERSUGGESTIBILITY: In the development of an experimental science the determination
of the laws or systematic relationships between variables is almost a matter of routine and is
ordinarily performed as soon as adequate techniques have been developed. Thus the rate of
movement of a falling body, as dependent upon the time since its release from rest, was one of
the first determinations of experimental physics. The plotting of the curves of learning and of
forgetting as a function of time by Ebbinghaus may almost be said to have marked the beginning
of the experimental psychology of the higher mental processes.
It is our special concern here to point out that an analogous question arises concerning the
rate of increase in the hypersuggestibility characteristic of hypnosis during the process of
inducing the hypnotic trance. Does this function have a uniform slope such as is represented by a
straight line, or is it characterized by a curve of negative or of positive acceleration? Does the
hypersuggestibility continue to increase throughout the duration of a quiet trance, or does it begin
to decrease immediately upon the cessation of the specifically trance suggestions? May it be that
it increases for a short time as a kind of after-discharge, following which it gradually falls, or
does it remain at a constant level throughout? How is the degree of trance hypersuggestibility
affected by activities and suggestions of various kinds during the trance? What proportion of the
trance hypersuggestibiJity remains immediately after the eyes have opened at the formal
termination of the trance? What is the shape of the curve according to which this diminishes? Is it
of positive or of negative acceleration? Is it abrupt, or does it fall to zero only after several hours
or days ? We shall begin with a consideration of the group of problems concerned with the
perseveration of hypnotic hypersuggestibility.
VIEWS OF CERTAIN WRITERS ON THE QUESTION OF POSTHYPNOTIC
PERSEVERATION OF HYPERSUGGESTIBILITY: That hypnotic subjects retain a
considerable portion of their hypersuggestibility for a certain period following the formal (299)
termination of the trance has been intimated by a number of authors, though this phenomenon
seems to have been far less clearly recognized than the hypersuggestibility characteristic of the
trance itself. The present writer has been unable to find any reference to this phenomenon in the
available writings of Braid, but Binet and Fere seem to imply something of the kind in the
foIlowing statement: "After awaking, the subject is stiII sensitive to suggestion; this fact has long
been known, and is mentioned by Braid among other writers." By the word "sensitive" do they
reaIly mean "hypersensitive?" Again in connection with the description of an experiment
involving something which may have resembled a post-hypnotic suggestion, they remark, "This
experiment shows that when a subject remains under the influence of a suggestion after awaking,
he has not, whatever be the appearance to the contrary, returned to his normal state." This
statement is similarly ambiguous from our present point of view because acts performed by
post-hypnotic suggestion constitute a special case, as is shown by the fact that they are usuaIly
foIlowed by waking amnesia of the acts in question. Bernheim also is ambiguous on this point.
He says (I, 18), "Some subjects remain sleepy when they wake up. If the operator waves his hand
once or twice before the eyes, he may dispel the drowsiness." MoIl, while somewhat clearer,
gives us by no means a satisfactory statement. He writes (I2, 47): "After deep and long hypnosis a
temporary state of drowsiness often supervenes, in which certain hypnotic phenomena continue.
The latter condition, however, occurs exclusively, or almost exclusively, in the case of hysterical
patients." Wingfield remarks somewhat paradoxicaIly and in a context which renders his meaning
uncertain (I7, 109): "The effects of suggestion frequently persist long after the suggestion has
been removed ... the result was the same when the trial was made during deep hypnosis, or
immediately after waking." BramweIl seems to have completely missed this point. He states (3,
95): "Under ordinary circumstances, the instant hypnosis is terminated all the phenomena which
have characterized it immediately disappear."
It is not surprising, perhaps, that considerable difference of opinion should exist among
workers employing merely clinical methods. Thus to superficial observation most subjects, (300)
Hypnosis and Suggestibility (300) immediately on opening the eyes at the suggestion to awake,
differ little from their state preceding the onset of the trance. It is, however, an easy matter to
show in an entirely objective manner that this is by no means the case. The proof is supplied by
the fact that when a subject has just been aroused from a trance he wiII go back into the trance (as
shown by lid-closure and other less tangible signs) in about half the time required when he has
not been in a trance for some days. This fact has been demonstrated repeatedly and systematically
in the author's laboratories during the last few years. This, it wiII be noted, is a manifestation of
homoactive hypersuggestibility as contrasted with the heteroactive variety considered in the
preceding sections.
KRUEGER'S INVESTIGATION OF THE POST-HYPNOTIC DECAY OF
HOMOACTIVE HYPNOTIC HYPERSUGGESTIBILITY: It will be convenient to begin the
present exposition with an account of Krueger's investigation of the relative degree of
hypersuggestibility at various intervals foIlowing the formal termination of the hypnotic trance.
The technique of the experiment is based on the simple and logical assumption that if there is a
perseveration of hypersuggestibility after the formal termination of the initial or primary trance,
this should be manifested objectively by an appreciable diminution in the suggestion time
required to induce lid-closure in a foIl owing or secondary trance. Presumably this diminution
should be less after protracted perseveration (inter-trance) periods than after short ones. Six
inter-trance intervals were chosen for measurement. Because curves of diminution in psychology
usually show a more rapid change in direction at their beginning, these intervals were placed
closer together immediately following the trance than later. They were: five seconds, thirty
seconds, one minute, five minutes, twenty minutes, and one hour.
Five subjects were employed. Each subject was tested for the rate of lid-closure resulting
from a uniform hypnotic technique twice on each of the six time intervals. In order to safeguard
the results from distortion by practice effects, three of the subjects were given the tests in the foIl
owing order: five seconds, sixty seconds, twenty minutes, sixty minutes, five (301) minutes,
thirty seconds, sixty minutes, five minutes, thirty seconds, five seconds, sixty seconds, and twenty
minutes. The two remaining subjects took the tests in the reverse order. If the groups of subjects
had been equal in number, this procedure would have completely equalized the number of trance
repetitions preceding each of the six perseveration intervals (counting the primary trance) at
twelve. Since the second group of subjects included one less individual than the first, the balance
was not quite perfect. This gave the intervals of twenty minutes and thirty seconds slightly more
practice than average, and the intervals of five seconds and sixty minutes slightly less. As a final
protection against constant errors due to practice effects in the summary table, ratios are
substituted for gross time scores.
The subject was allowed before arousal from the primary trance (the one preceding the
perseveration interval) to lie quietly for two minutes after complete lid-closure. During the
secondary or test trance (the one following the perseveration interval), the subject was allowed to
contin1,le quietly one minute after lid-closure was complete, before arousal. The time required
for lid-closure in both primary and secondary trances was taken with a split-second stop-watch.
The trance was induced by means of the optical fixation method combined with (302) verbal
suggestions of relaxation and sleep. During the induction process the subject rested in a
semireclining position in a comfortable Morris chair. He also remained quietly in this position
during the perseveration or retention interval.
`A condensed summary of the results of the investigation as a whole is given in Table 51. This
table was derived by first dividing the number of seconds required to induce lid-closure in the
secondary trance by the corresponding time of the primary trance preceding it. Each value in this
table is an average from the two measurements made on each subject at each point. An index of
hypersuggestibility may be obtained by subtracting any of these values from unity. Thus a table
value of 0.45 corresponds to a hypersuggestibility of 0.55, i.e., 55 per cent; a table value of 0.95
would represent a hypersuggestibility of 0.05, or 5 per cent.
Possibly the most significant thing about this table is the fact that all of the thirty entries
are below the value of unity. This means that the secondary trance time, even after so long as an
hour, averaged less than that of the primary trance, indicating a markedly stable tendency to
perseveration of homoactive hypersuggestibility. Further inspection of the table reveals a
considerable amount of variability, as might be expected from the fact that each entry represents
only two determinations. There is, however, a well-marked tendency for the ratios to become
larger as the interval between the primary and the secondary trance increases. This is shown with
greater clearness by the arithmetical means at the bottoms of the columns. These are shown in
graphic form by Figure 53. While by no means consistent, the general tendency of these mean
values is clear.
`In order still further to reduce the element of chance, the six columns of data were thrown
together in pairs, thus increasing the number of measurements represented by each mean to
twenty. The combined five- and thirty_second intervals yield a mean ratio of 0.46; the combined
one- and five-minute intervals, one of 0.61; and the combined twenty- and sixty-minute intervals,
a value of 0.75. The statistical reliability (critical ratio) of the difference between the ratios of
0.46 and 0.61 is 5.55, and that between 0.61 and 0.75 is 4.52. It accordingly seems well
established that there is a progressive weakening (303) in the hypersusceptibility to lid-closure
under the influence of hypnotic suggestions with the increase in the length of the interval after the
primary trance. That subsequent investigation will also verify the general shape of the curve, i,e.,
the more rapid fall of this hypersusceptibility in the period immediately following waking from
the primary trance as indicated by Figure 53 (not available), is strengthened by the fact that this
general tendency is practically universal in psychological processes when plotted as a function of
time.
 Two defects in the technique employed in this investigation need to be pointed out. One is
that the fatigue from the subject's gazing at the bright fixation object may have tended to produce
a part of the increased facility of lid-closure in the secondary trance tests of the shortest retention
intervals. This effect could hardly have been great, however, since the primary trance time as
shown by Krueger's published table does not much exceed twenty seconds. In future experiments
of this (304) kind, this difficulty can easily be avoided by discarding the fixation object and
using merely verbal suggestion for inducing the trance. A second defect lies in the fact that the
quiet resting in a comfortable Morris chair for some time may itself predispose subjects to
positive suggestibility, especially when this is protracted as was the case with the sixty-minute
retention interval. Just how much such an influence amounts to, if any, can only be determined by
running a control experiment in which the subject is hypnotized after resting an hour in a semi
reclining position but without the primary trance at the beginning of the period. Such a parallel
control experiment would have been run by Krueger with the same subjects, but the time
limitations under which the experiment happened to be carried out rendered this impossible. For
the above reasons, as well as because of the small number of subjects and the small number of
determinations on each, the results of Krueger's study must be considered as more or less
preliminary, though it is doubtful whether the more exact experiments which the future will bring
forth will materially change the general picture sketched in a rough way by Figure 53 (not
available).
The Course of Hypersuggestibility During the Induction and Continuance of the
Trance: We may now pass to a consideration of the changes in hypersuggestibility which take
place during the process of induction and subsequent duration of the hypnotic trance. This
problem, like the one just considered, might be investigated by both the homoactive and the
heteroactive technique, the one employing the trance itself as an indicator, the other utilizing as
an indicator some non-trance activity such as postural movement. Only the homoactive method
has been employed thus far. The one bit of experimental evidence available on this point was
reported by Krueger. His method, in general, resembled that of the parallel experiment just
described, except that in this case a complication arose. In the preceding experiment concerning
the perseveration of hypersuggestibility after waking from the trance, it was simple enough to
reinduce the trance and note the reduction in time required. But how shall one measure the time
of trance induction by the (305) homo- active technique when the subject is already in the
trance( Krueger met this difficulty by waking the subjects after the trance had lasted various
lengths of time, and then rehypnotizing them at once. This procedure was based on the
assumption that the amount of hypersuggestibility persisting for a constant brief interval after
waking would vary in magnitude according to the amount of hypersuggestibility of the
immediately preceding trance state. The constant interval between the two trances was fifteen
seconds. The trance process was allowed to continue for six different periods before being
interrupted by the waking of the subject. One interruption took place midway in the
trance-induction process, another at the instant the trance was induced (lid-closure), and the
remaining ones after the trance had lasted quietly for one, two, five, and ten minutes respectively.
In the primary trance, suggestions were terminated two minutes after final lid-closure, which was
taken as the beginning of the trance. The secondary trances always were terminated after one
minute. A minimum of twenty-four hours always elapsed between the experimental periods.
Four subjects were employed in this investigation. All had previously been used in
hypnotic experiments. They were accordingly (306) in an advanced state of training (p. 341),
which tended to reduce distortions from practice effects. Each subject was tested twice on each of
the six intervals. As a means of still further reducing distortions from practice effects, the order of
administering the twelve tests received by each individual was for half of them as follows:
one-half lid-closure time, one minute, five minutes, ten minutes, two minutes, lid closure time,
ten minutes, two minutes, lid-closure time, one-half lid-closure time, one minute, and five
minutes. The other half of the subjects took the tests in the reverse order, thus equalizing the
aggregate amount of practice preceding the tests at each of the six points of interruption. Lastly,
in the final table ratios of the secondary to the primary trance time replace the gross time scores.
The condensed results of this investigation are shown in Table 52 (not available). A
composite curve plotted from the means of the columns is shown in Figure 54. It may be noted at
the outset that these results corroborate in a striking manner the results of the experiment
described above as to the amount of homoactive hypersuggestibility existent immediately (307)
following waking from the trance. It will be recalled that the average of the five- and the
thirty-second retention interval in that experiment was 0.46, which corresponds to a
hypersuggestibility index of 0.54 (1.00-0.46). The average of the means from the five full trance
awakenings after fifteen seconds in the present experiment is 0.43, which corresponds to a
slightly higher hypersuggestibility index of 0.57 (1.00-0.43), an excellent agreement between the
two studies and a fair agreement with three of the studies of heteroactive hypersuggestibility (4,
8, 16) .
The shape of the dotted portion of the curve in Figure 54 indicates that, so far as the
present data are able to answer the question, the rise of hypersuggestibility at the onset of the
hypnotic trance follows a course of negative acceleration. Approximately five-sixths of the
hypersuggestibility present at the moment of lid-closure is already present at the mid-point of the
suggestion period.
Once the trance (lid-closure) has been induced, the curve seems to rise a little at one
minute and to fall somewhat at two minutes, after which there is a gradual rise until the end of the
ten-minute period. What takes place beyond this point can only be conjectured on the basis of
general observation. The fact that subjects frequently wake spontaneously if left undisturbed
suggests that, in such cases at least, the curve would gradually fall to a certain leve1.4 It is
doubtful if any of the irregularities in the trance part of the curve may be considered seriously as
indicating the true form of the curve. Perhaps its most remarkable characteristic is its relatively
high position at the end of ten minutes. This fact suggests that, barring spontaneous waking from
the trance, subjects may maintain a relatively high level of hypersuggestibility throughout a
trance period perhaps as long as an hour and even longer. At any rate, the indication is fairly
definite that the hypersuggestibility level during the continuance of the trance does not fall at
anything like the rate following the formal termination of the trance. (308)
Three Hypotheses Concerning the Nature and Origin of Hypnotic
Hypersuggestibility: In the preceding pages we have discarded, as non-characteristic, one after
another of the phenomena once supposed to be distinctive of hypnosis. Among these may be
mentioned catalepsy, rapport, and post-hypnotic suggestion. We have now before us perhaps the
last important claimant for this distinction. This is that the hypnotic state is clearly to be
differentiated from the non-hypnotic condition, not as being merely a state of susceptibility to
suggestion, for that is notoriously true of the normal condition as well, but rather as being a state
of heightened susceptibility to suggestion. The experimental results already given in the present
chapter, together with the more or less indirect evidence contained in the two preceding chapters,
indicate rather conclusively that hypnosis actually manifests this tendency at least with the
majority of subjects. Nevertheless, the question of whether hypersuggestibility is a physiological
tendency genuinely inseparable from the hypnotic trance must still be faced. This point is a
peculiarly critical one owing to the fact that if hypersuggestibility must also be discarded as
distinctive of the hypnotic trance, which must be done if the inquiry receives a negative answer,
we should be forced by logical necessity to abandon the concept of hypnosis as a significant
entity. In the consideration of this question three prominent possibilities may be noted. It will be
convenient to present them as formal hypotheses:
1. Specific Physiological Hypothesis. This hypothesis assumes that there may be a
specific physiological tendency (the nature of which is not specified) for positive response to any
direct prestige suggestion to induce within the responding organism a more or less generalized
increase in the normally existent tendency to respond positively to any and all other forms of
prestige suggestion.
"Having responded to one suggestion, the organism is at once in a state of heightened
susceptibility to further suggestion. According to this line of reasoning, the hypnotic trance
becomes only a special case of the general law. According to the usual procedure for inducing the
trance, the closure and ultimate catalepsy of the eyelids, along with certain other (309)
phenomena, are first evoked on the basis of waking suggestion ... these responses to waking
suggestion would naturally produce a state of heightened susceptibility to further suggestion
which is ... characteristic of the trance state." If this hypothesis should find verification, it would
seem reasonable to conclude that hypnotic hypersuggestibility is an essential rather than a
fortuitous characteristic of hypnosis, and hypnosis as hypersuggestibility may continue to be
regarded as a genuine entity.
2. Indirect Physiological (Relaxation) Hypothesis. It is a very general custom of
hypnotists to give suggestions of relaxation while inducing the trance. Moreover, it is likely
where such specific suggestions are not given that the subjects receive indirect suggestions to the
same end from the direct suggestions of being tired and of going to sleep which almost always are
given. At all events, subjects in deep hypnosis normally present a picture of rather complete
relaxation. The present hypothesis assumes that this relaxation has the effect more or less
completely of suppressing the spontaneous activity of the symbolic or thought processes. With
this suppression should disappear the control normally exercised by symbolism over the lower
levels of activity. This should leave the latter more completely exposed to the influence of
suggestive stimuli from outside sources, as has been shown in the preceding pages to take place.
In so far as this hypothesis may -find experimental veri-fication, hypnotic hypersuggestibility
must be regarded as merely a fortuitous associate of hypnosis, just as rapport and catalepsy
already have come to be regarded, and hypnosis may ac~ cordingly cease to be considered as a
distinctive entity.
3. Social Suggestion Hypothesis. It is an almost universal belief-among the American
people, at least-that once a subject has yielded to hypnosis (lid-catalepsy, the assumption of a
sleeping posture, etc.) he is completely within the power of the hypnotist, and that so long as the
trance persists he cannot resist any suggestions that the latter may choose to give. The widespread
impression that crimes may be induced by means of hypnotic suggestion is but one manifestation
of this belief. It is even believed by large numbers of well-educated persons that the tendency to
hypersuggestibility may become both fixed (310) and generalized by one or two hypnotic
procedures, with the result that a permanent weakening of the "victim's" resistance to suggestion
may lay him open to the eXploitation of unscrupulous persons. The present hypothesis assumes
that this widespread belief itself constitutes a powerful suggestion supplementary to that
administered by the hypnotist, and is quite sufficient to produce, after the trance has been induced
by ordinary suggestion, a marked facilitation of response to further suggestions.
In so far as this hypothesis finds verification it will be fatal to the claims of
hypersuggestibility as being an essential characteristic of hypnosis, and hypnosis should tend, to
this extent, to cease to be regarded as a scientific entity.
It should be observed that none of the above hypotheses is necessarily in conflict with any
of the others. So far as a priori considerations alone are concerned, it is quite possible that all
three may be true, each contributing to a common observed result.
THE ROLE OF HYPOTHESIS, DEDUCTION, AND EXPERIMENT IN
EVOLVING SCIENTIFIC THEORY: There is a statement attributed to a well-known
American psychologist to this effect: "An experiment without an hypothesis is blind. An
hypothesis without an experiment is dead-or ought to be."
If the above principle had been more generally observed throughout the history not only
of hypnotic theory but of psychological theory in general, the current offerings in each would be
less voluminous and at the same time more illuminating. On the contrary, the hypnotic
experiments of the past too often have had little or no relation to the prevailing theory, and the
explanatory hypotheses (incorrectly called theories) too often have failed to call forth attempts at
experimental verification.
The first characteristic of a satisfactory explanatory hypothesis in science is that it shall
really explain all the phenomena which it purports to explain. The test of whether it does this
(311) is whether or not the phenomena to be explained can be deduced from the hypothesis by a
strictly logical procedure. But the fact that an hypothesis can do this is not necessarily convincing
evidence that it is true, especially where the phenomena to be explained are neither numerous nor
complex. Within limits it is always possible for an ingenious person, if permitted to choose his
variables with some freedom, to construct more or less by trial and error an hypothesis which will
fit simple situations. The real test comes at the next step in the evolution of an hypothesis into a
theory. This step consists in passing from the deduction of those phenomena already well known
to the deductive prediction of the existence of phenomena never yet observed. These deductive
predictions naturally set problems for the experimentalists, and at this point the laboratory makes
its contribution. If experiments consistently verify the deductions, the hypothesis progresses to
the status of a theory or a law; if not, it passes (or should pass) out of the realm of the living
science into the history of science. Ordinarily chance alone will not yield very many experimental
agreements with deduced predictions before an exception will be encountered. It follows from the
above considerations that several agreements may be required to produce a dependable
probability of truth, whereas it takes only on clear-cut negative case wholly to discredit an
hypothesis.
With these preliminary observations we shall proceed to state some of the deductions of
the several hypotheses listed in the preceding section. For the most part the deductions to be
presented will be those with relevant evidence bearing on their credibility. It should be noted,
however, that in several cases the hypothesis was framed before the performance of the
experiment which produced the evidence.
DOES HYPNOTIC HYPERSUGGESTIBILITY PROGRESSIVELY DECREASE
DURING A QUIET TRANCE? Deduction 2: If the hypersuggestibility of the trance is the
physiological result of response to the direct suggestion of lid catalepsy, sleep, etc. (Hypothesis
I), it should follow that, in a case where the subject lies for some time in a quiet trance (312) and
actively responds to no suggestion, there should be a progressive diminution of the
hypersuggestibility with temporal remoteness from its point of origin, just as appears to be the
case upon the termination of the trance (n). Indeed, the two cases ought not to be materially
different in this respect.
Krueger's study of hypersuggestibility during a ten-minute quiet trance is relevant here.
So far as his results go, they are distinctly opposed to the hypothesis; there is little or no
indication in Figure 54 of any diminution in hypersuggestibility at ten minutes. Unfortunately,
Krueger's experiment did not follow the quiet trance long enough to enable us to judge what
might have happened at the end of an hour, say. The fact that most subjects will "wake"
spontaneously if left alone for a period suggests that under certain conditions there may yet be
found a weakening of hypersuggestibility in the quiet trance with the passage of time.
WILL POSITIVE RESPONSE TO NON-HYPNOTIC WAKING SUGGESTION:
Deduction 2. If the induction of the hypnotic trance is only the special case of a very general
tendency of all positive responses to direct prestige suggestion to generate a heightened
suggestibility to any closely following prestige suggestion of whatever nature, then the
relationship between hypnosis and waking suggestion should be reciprocal. Concretely, a
previously positive response to a waking suggestion to postural movement, say, should have the
effect of shortening the time required to induce the hypnotic trance, just as the experiments
reported above have shown that a previous positive response to suggestions of lid-closures, etc.,
in the induction of the trance leads to facilitation of the response to closely subsequent
suggestions of postural movement.
We are fortunate in having two experimental investigations bearing directly upon this
deduction from our major hypothesis. The first investigation was reported by Caster and Baker as
a component of their study described above (p. 294). They took with a stop watch the suggestion
times required to (313) produce lid-closure in ten subjects, (1) preceding positive response to arm
suggestion and (2) following such response. Under these conditions the hypothesis demands that
the trance induction time following the arm movement should, upon the whole, be less than that
preceding the response to arm suggestion. Presumably any special facilitation of this kind would
be maximal immediately after the arm movement and would diminish progressively in magnitude
as the time interval between the two increased. Passing now to the facts, we find Caster and
Baker reporting that seven of their ten subjects showed a mean trance suggestion time to a greater
or less degree shorter when following response to arm suggestion than when preceding it, though
for the most part the differences observed were not large. The small differences found may
reasonably be attributed to the fact that the trance induction did not take place until fifteen
minutes after the arm reaction.
A second study, by Jenness, offers a somewhat more favorable opportunity for the
hypothetical tendency to manifest itself. This study also was a component of an investigation
already reported. Jenness took lid-closure times with a split-second stop-watch, using nine
subjects. In general his method was substantially the same as that employed by Caster and Baker,
with the exception that the interval elapsing between the response to arm suggestions and the
beginning of the suggestions leading to trance induction was only three minutes as contrasted
with fifteen. Eight of Jenness's nine subjects showed a lessened lid-closure time following the
arm suggestions. The mean scores of eight of these subjects who went into the deep trance are
shown in the two last columns of Table 50 (not available). The average lid-closure time of the
whole group is 63 sec. when lid-closure precedes the arm suggestion, and only 46 sec. when
following. This amounts to a diminution of 27 per cent in trance suggestion time when following
postural suggestions. The critical ratio of 3.52 shows this to be a statistically reliable difference.
So far as the relevant material at present available goes. then, this deduction from our major
hypothesis is substantially confirmed by experiment. (314)
WILL NAIVE SIMULATED RESPONSES TO SUGGESTION SHOW TRACES
OF HYPERSUGGESTIBILITY? Deduction 3: For reasons somewhat similar to those for
Deduction 2, it might be expected with a certain amount of plausibility that successively
simulated responses to suggestions of the types employed in the experiments of Rull and Ruse,
Williams, Caster and Baker, and Jenness might also show the phenomenon of heightened
responsiveness to a second suggestion over that to a closely preceding one.
The evidence bearing on this deduction was obtained by Williams in an attempt to run a
radical control experiment based on the general procedure of the Rull-Ruse type. The plan of the
experiment was to put a squad of eight subjects through the Rull-Ruse technique in all details,
with the one exception that the subjects should merely go through the motions of entering the
trance and of making positive responses to postural suggestions-all quite voluntarily. In order to
increase the certainty that the subjects were not hypnotized, only men were chosen who had been
previously rejected because they had shown by actual trial that they were subject neither to
hypnosis nor to waking suggestion.
Just before the beginning of the first experimental session the subjects were taken into the
author's private laboratory individually and instructed substantially as follows: "For some time we
have been carrying out in this laboratory some very careful experiments on the psychology of
hypnotism. In this work we are always in danger of having our experiments spoiled by employing
subjects who pretend they go into the trance but who really are not hypnotized. You can easily
see that it is very important for us to have a test which will enable us to tell whether a subject is
really in the trance or not. We believe that we now have a test by which we can tell from the
behavior of a subject whether he is in a trance, but we cannot be sure until the test has been tried
out on some people whom we know are simulating. In order to make our test conclusive, we must
know not only that the subjects are pretending to be in the trance, but that they actually are not in
the trance. In order to be sure that the subjects have not been influenced by suggestions, we have
chosen you and a small number of others because you have shown in a (315) preliminary
experiment that you are entirely immune to the influence of suggestion. Now, in this experiment
Dr. Williams will suggest to you that you will do certain things; for example, he will suggest that
when: you are standing blindfolded you will fall forward toward him and that you cannot help
yourself. When he does this you are to resist a short time and then respond to his suggestions
voluntarily but pretend, in whatever way it seems most natural to you, that you are really doing it
in response to his suggestions. At other times he will try to put you into a trance; he will tell you
that your eyes are closing and that you cannot keep them open. When he does this, you are to
wait a reasonable length of time and then close your eyes voluntarily, but act as nearly as you can
the way you think a person would if he were actually going into the trance. But you must be very
sure that you do not go into the trance. One indication that the subject really is going into the
trance is that he feels rather drowsy. But the most certain way of telling is as follows: After the
subject's eyes have closed, Dr. Williams gives him some further suggestions, finally telling him
that he cannot possibly open his eyes no matter how much he tries. When he tells you this you are
to open your eyes instantly, if possible, just to be sure that you can do so. You are then to close
them immediately and go on with the experiment as if nothing had happened, but note very
carefully when you do this whether they tend to stick together or not, or whether you have any
special difficulty in deciding to open them. If any of these things happen, it indicates that you
have been influenced by the suggestion. Note these things very carefuJly so as to tell me all about
it when I talk with you after the experiment is over. "Please do not read up about hypnosis or talk
to anyone about it, because we want to see what a person unacquainted with hypnosis will do in
the experiment. Do not try to think up any special plan, but do just what you think would be
natural at the time you do it."
Immediately following the last experimental session, the subjects were questioned with
considerable care by the author, to determine, so far as possible by such means, whether they had
observed in themselves any tendency to fall into the trance. All denied any trace of such a
tendency. This report was fully confirmed by Dr. Williams's observation of the men during the
experiment. To secure a further check on this critical matter, they were returned to Dr. Williams's
laboratory and submitted to the hypnotic technique with instructions from the author not to
simulate. Here, again, no signs of susceptibility (316) were observed. The men were also
questioned with care as to their understanding of the instructions, what they had tried to do at
various points in the experiment, and what they thought they really had done, especially as to the
speed of "falling" under various conditions. All reported that they had fallen at about the same
rate throughout, and none of them appeared to have the least comprehension of the rather
complex phenomena which they had been executing. In view of the nature of the experimental
outcome of the control, it is especially noteworthy that six of the eight subjects had understood
that the pretended trance was to terminate at the point where the individual opened his eyes at the
challenge of the experimenter, which means that these individuals, contrary to the original plan of
the experiment, were not really simulating hypnosis during the postural suggestion tests at all.
The main results of this investigation are shown in condensed form in Table 53 (not
available). A glance over this table shows a (317) picture surprisingly similar to that presented by
tables where genuine hypnosis was involved. Six of the eight subjects show a more prompt
response to suggestion when in the pseudotrance. The difference between the means of the two
conditions has the fairly satisfactory critical ratio of 3.77. The reduction in the time of postural
response in the pseudo-trance, 11.2 per cent, is noticeably less, however, than that usual for
groups where the genuine trance was induced. Thus the third deduction from the major
hypothesis seems to find confirmation.
It may be observed, incidentally, that these non-suggestible simulating subjects manifest
the phenomenon of practice effects quite as clearly as do genuinely suggestible individuals. This
appears whether the composite practice curve is plotted for the four "falls" each day or by whole
days, as may be seen in Figure 55.5 (not available) (318)
IS HETEROACTIVE HYPERSUGGESTIBILITY MUTUALLY OPERATIVE
AMONG WAKING SUGGESTIONS? Deduction 4. If hypnosis is only a special case of the
generalized facilitation of response to all prestige suggestions produced by positive response to
any closely preceding direct prestige suggestion, we should expect mutual heteroactive
hypersuggestibility to be manifested in the case of two postural waking suggestions.
In order to test this hypothesis, Hull, Patten, and Switzer carried out a special experiment (9). In
this investigation the subjects were not hypnotized. Instead there were employed two distinctly
different waking suggestions: (1) that the head should fall forward upon the chest and (2) that the
horizontally extended arm, supported by the swinging platform (Fig. 15 (not available)), should
sway forward. Clearly, if the above hypothesis is operative in a heteroactive sense, either of these
suggestions following positive response to the other should take place in less time than when
preceding. Thirteen college students previously found by trial to be positively responsive to
postural suggestion were employed as subjects. They were tested four times on each of fifteen
successive days, twice on each type of reaction. On seven of the days the two head suggestions
preceded, and on eight the arm suggestions preceded. One minute of quiet was interposed
between the end of one suggestion and the beginning of the next. The suggestions were given
entirely by means of phonograph records specially constructed for the purpose. In this way the
results were protected from any possible constant error due to variability in the nature of the
suggestions. The reactions of the subjects were traced on smoked paper by two recorders of the
usual type actuated by threads connected indirectly to the subject's body.
The results of the investigation for the head movements (319) are quite in agreement with
those for the arm, as might be expected. The data for the latter are summarized in Table 54. An
inspection of the pairs of means in this table shows that they are quite definitely in conflict with
the hypothesis that positive response to direct suggestion evokes a generalized
hypersuggestibility. The hypothesis demands that the first arm trial following head response
should be faster than the first arm trial preceding head response, and the same for the second arm
trials. Actually, the arm responses following the head reactions are a little longer. The negative
nature of these results is emphasized by the fact that the second trials of both arm and head are
distinctly more rapid than the first, which is the manifestation of a well-established form of
heightened (320) responsivenes to suggestion resulting from practice. The fact that the
experiment is sensitive enough to show this known form of homoactive hypersuggestibility at all
four possible points, yet fails completely to manifest any tendency to heteroactive
hypersuggestibility, makes the experiment peculiarly convincing and correspondingly fatal to the
hypothesis.
Do Waking Suggestions Manifest Homoactive H ypersuggestibility? Deduction 5. If
hypnosis is only a special case of the general tendency characteristic of all direct prestige
suggestion, then ordinary waking suggestion under suitable conditions should display the
phenomenon of homoactive hypersuggestibility; that is, it should display a net heightening of
responsiveness to suggestion of the same kind after the effect of all known non-suggestion
influences has been eliminated.
The experimental test of the above deduction is complicated by the fact that ordinary
practice effects also produce a facilitation of responsiveness at each successive trial. These
practice effects are not readily distinguishable from those of homoactive hypersuggestibility. It
might be supposed that the two tendencies could be differentiated on the basis of the greater
permanence of practice effects as contrasted with the transitory effects of hypersuggestibility.
Unfortunately, this approach is unsatisfactory because in certain kinds of learning the effects of
practice are known on the basis of experiment also to be exceedingly transitory. Ebbinghaus, for
example, found in his classical study of forgetting that the amount of material retained after
twenty-four hours was only about a fifth of the amount known by the subject immediately at the
close of practice. However, an escape from this difficulty would seem to be offered by the
expedient of testing some form of suggestion behavior for homoactive hypersuggestibility after
the activity in question had been practised until it had ceased to show permanent facilitation.
Such a procedure should reveal unambiguously any hypersuggestibility which is genuinely
distinct from practice effects. (321)
A special experiment was instituted by Patten, Switzer, and Hull (I4) with a view to the
testing of the above deduction. The basic plan of the project was to test a waking suggestion for
homoactive hypersuggestibility when it had reached the stage where practice no longer produced
a facilitation with a twenty-four-hour permanency. Three standing postural suggestions in the
waking state were given to each of sixteen susceptible subjects for twenty successive days. After
a no-practice interval of seventy days, practice was resumed for six more consecutive days. On
anyone day a period of one minute elapsed between the termination of each suggestion and the
beginning of the next. The score, as usual by this technique, was the number of seconds of
continuous suggestion required to produce a "fall".
The mean experimental values in terms if suggestion time of the sixteen subjects on each
of the twenty-six days of practice are shown for each of the three successive trials per day in
Table 55. These values pooled in groups of four days are shown graphically in Figure 56. It may
be seen from an inspection of this graph that, contrary to the hope of the investigators, the limit of
practice had not been reached when practice was interrupted at the end of twenty days. The
experiment accordingly fails to yield a direct test of the truth of the deduction.
It is conceivable, however, that indirect indication of what would take place at the limit of
practice may be obtained by appropriate examination of the course of the three practice curves
shown in Figure 56 (not available). Our present hypothesis assumes, it will be remembered, that
at the limit of practice these curves must converge in such a way as to coincide as a single
horizontal line if no hypersuggestibility is mingled with the obviously existent practice effects.
On the other hand, if there is mingled with the practice effects a certain amount of genuine
hypersuggestibility, then the three curves at the limit of practice should become horizontal but be
separated by some finite (322) distance, the magnitude of the separation constituting in some
sense an index of the amount of hypersuggestibility present. It is clear that under the latter
conditions the amount of permanent gain from each day's practice will be zero, but the temporary
gain will be a positive value. If, therefore, the amount of permanent gain should be divided by the
amount of (323) temporary gain, the quotient would of necessity also be zero. Our present
experimental results, on the other hand, show that at the outset of practice the amount of
permanent gain divided by the amount of immediate gain yields a quotient with a value
approximately 0.4. It follows, therefore, that between the beginning and the termination of a
complete practice sequence the ratios of the permanent gain to the immediate gain must decrease.
Since the diminution in the ratio under consideration is assumed to be dependent in the main
upon the diminution in the amount of the permanent gain, which is known from experiment to
shift very gradually, it is reasonable to expect that the ratios would diminish progressively
throughout the period of practice. Consequently, if the ratios in question should be (324) found to
diminish rapidly and consistently during the early stages of practice, this would indicate as
possible an ultimate zero value and so a certain amount of genuine hypersuggestibility; whereas,
if the ratios should remain constant or should grow larger, the indication would be definitely
against such a possibility. Accordingly, quotients of the kind just described were calculated from
the values presented in Table 55. Since these ratios proved somewhat erratic, doubtless because
of limitation in the size of the sample, the values for the twenty days of consecutive practice were
grouped into four convenient periods. Similar ratios for the six days of resumed practice after the
seventy-day retention interval were also computed and pooled. The five mean ratios thus obtained
are as follows:
[First period 36% Second period 36% Third period 31.5% Fourth period 21%' Recovery
period
35%]
An inspection of the above pooled ratios shows that the secondary criterion also yields a
somewhat ambiguous answer to the question implicit in the deduction with which we are
immediately concerned. The first and second practice periods, together with the recovery period,
if considered by themselves, would indicate that no hypersuggestibility exists in this form of
suggestion. However, the third and fourth periods, taken in conjunction with the second, present a
fairly marked and progressive diminution which points in the direction of a possible
hypersuggestibility. However, even if we leave out of account the recovery period, which has a
somewhat uncertain status in this connection, there is a serious question whether an extrapolation
of the falling tendency manifested by the ratios of the period of genuine practice would ever
arrive at zero. Upon the whole we must conclude, then, that so far as the ratios just considered are
concerned the indications of homoactive hypersuggestibility in waking suggestion are slight.
Moreover, the marked tendency of the three curves in Figure 56 to converge at the end of the
practice period indicates rather clearly that if any hypersuggestibility whatever exists in this form
of suggestion it must be very small in amount. (325)
CAN HETEROACTIVE HYPERSUGGESTIBILITY BE INDUCED IN WAKING
SUGGESTIONS BY MEANS OF A SPECIAL SUPPLEMENTARY SUGGESTION TO
THAT EFFECT? Deduction 6. The apparent failure of waking suggestions of arm and head
movements to show any trace of mutual heteroactive hypersuggestibility (p. 320) led to an
attempt to test Hypothesis 3 as a possible alternative. It will be recalled (p. 310) that this
hypothesis assumes a special belief on the part of most subjects that once their lids have been
closed by means of waking suggestion they immediately will become very much more
suggestible than previously, and that this state of heightened suggestibility will persist as long as
the subject considers himself to be under hypnosis. From this hypothesis two consequences
follow. First, if subjects could be found who do not hold the view assumed by the hypothesis,
these individuals ought not to show hypnotic hypersuggestibility. Secondly, if a strong suggestion
of a nature analogous to that assumed to be operative in the case of hypnotic hypersuggestibility
could be implanted in the minds of a group of subjects in respect to a suggested reaction which
does not normally show this phenomenon, hypersuggestibility should then be manifested.
The writer knows of no direct evidence bearing on the first deduction. It is true that a
certain percentage of subjects probably do not manifest hypersuggestibility by the methods at
present employed to detect it. Possibly these individuals do not happen to have acquired this
belief. If this particular angle of the problem had been appreciated earlier, it would have been a
relatively simple matter to question each of the thirtyfour individuals (p. 298) who were tested for
hypnotic hypersuggestibility at the termination of his participation in the experiment and thus
determine whether there was any connection in this respect between belief and performance.
A special experiment to test the second deduction was performed by Patten (13). Ten
college students who showed definite responsiveness to waking suggestion were first given the
indirect suggestion contained in the following letter to Professor Patten which purported to come
from a psychologist favorably known to the subjects by reputation: (326)
"The following interesting psychological phenomenon has recently been discovered by
experimental psychologists: When a normal human being is sitting blindfolded and hears another
person tell him that his arm is swinging forward, he will react to this suggestion by moving his
arm (when his arm is supported on a movable cradle). Then if he is told, a few seconds later, that
his head is falling forward on his chest, he will also respond by moving his head. The most
interesting thing about this behavior in normal subjects is that the second response occurs much
more rapidly than the first response no matter which suggestion happens to be given first. In other
words, if the head-movement suggestion comes first, it will increase the rapidity of the arm
movement which comes afterward, and if the arm-movement suggestion comes first, it will
increase the rapidity of the following head movement. The above fact is extremely significant and
seems to have been well established by careful scientific investigators at several universities. The
only question remaining unanswered is the extent of the influence of the first suggestion upon the
second. That is, what is the speed of the normal subject's response to head movement following
arm movement, and vice versa? This is the problem which it is hoped you and your students at
Miami will be able to solve. The subjects in the experiment should be informed as to the nature of
the problem before the experiment takes place, in order that they may understand its
psychological significance. However, they should be carefully instructed not to let this knowledge
of the problem influence their behavior in any way."
This letter was read to the class in psychology from which the subjects came, and later
was shown to them individually. Every effort was made to produce a strong belief and
expectation on the part of the subjects.
Otherwise the experiment was substantially like the one reported by Hull, Patten, and
Switzer, except that the project continued through eight consecutive days, two cycles of four days
each. The subjects were divided into two equal groups, the cycles of one group being of the A B
B A type and those of the other being of the B A A B type. On the A days two arm suggestions
were followed by two head suggestions; on the B days two head suggestions were followed by
two arm suggestions. According to the present hypothesis, this experiment would be expected to
show a faster reaction of (327) both arm and head responses when each follows the other than
when preceding, even though no such difference should normally appear in suggested reactions of
this type.
The condensed results of this portion of Patten's investigation are shown in Table 56 (not
available). In general the outcome of this experiment is indistinguishable from that by Hull,
Patten, and Switzer (p. 320): the head responses show a slight mean facilitation when following
the arm responses, and the arm responses are slightly retarded when following the head
responses. Neither of these tendencies has a satisfactory statistical reliability. The mean of the
two tendencies is slightly in the direction of heteroactive interference rather than of facilitation,
as demanded by this hypothesis. (328)
A comparative examination of the results of the two studies involving sequences of arm
and head responses to suggestion shows that in both cases the arm moved more slowly after the
head responses. This suggests that there may be some active perseverative interference involved.
If this were true, it is conceivable that such a tendency might be strong enough to override a
genuine but weak tendency to hypersuggestibility and thus mask its existence. In searching for a
possible cause for such an effect it was recalled that at the end of each head and each arm
response in the experiments under consideration the movement was reversed, i.e., the member
was replaced by the experimenter in its original position as if in preparation for the next response.
In anticipation of this, moreover, several of the subjects would restore the member to its normal
position voluntarily. In hypnosis, on the other hand, there is ordinarily no such reversal; the lids
normally remain closed throughout the period of subsequent suggestions. In other words, to carry
out the analogy of the hypnotic lid reaction in an experiment involving a possible perseverational
heteroactive suggestion influence of an antecedent head movement on a closely following arm
movement would require that the head should remain forward on the chest throughout the
responses to arm suggestions . With a view to satisfying this demand, Patten continued the
experiment just described with the same subjects for eight more days, but with this change in the
technique: he made new phonograph records in which the suggestion was explicitly given that
once a suggested response was performed the part of the body involved must remain in position
unless the experimenter himself returned it to its original place.
An examination of Patten's table presenting these latter results shows that the change in
technique produced no detectable difference in the mean results. Once more we find, exactly as in
the two preceding studies involving this general technique, that the head shows a slight
heteroactive facilitation effect and the arm shows a somewhat greater heteroactive inhibiting
effect.
In order to determine so far as possible whether the individual participants had really
received the suggestion, a mimeographed questionnaire was prepared to be filled out by each
subject at the conclusion of the experiment. In various (329) ways this questionnaire attempted to
elicit indications from the subjects as to what was supposed to be the purpose of the experiment.
All of the subjects responded in such a way as to indicate that they had fully comprehended the
suggestion. This fact tends to make still more convincing the negative nature of the present
outcome. It is difficult to say what these results mean, but it seems fairly clear that they furnish
no support for belief in the truth of Hypothesis 3.
Another interesting and possibly significant fact was also revealed by this questionnaire:
everyone of the ten subjects reported feeling drowsy during the experiment, despite the fact that
nothing whatever concerning either sleep or hypnosis was contained in the suggestions. This
touches indirectly Hypothesis 2, since if incipient sleep tendencies were somehow elicited by the
technique these probably would involve relaxation. Since true sleep, as Bass's graphs show, is a
gradual and progressive change, it is reasonable to suppose that the drowsiness and relaxation
would be greater at the second pair of suggestion reactions than at the first pair. This, according
to Hypothesis 2, should produce a relatively greater facilitation of the response to the second pair
of suggestions. Since this did not take place, Patten's experiment tends somewhat to discredit the
hypothesis that relaxation facilitates response to direct prestige suggestions. The fact that
incipient sleep was probably also involved requires that the conclusion be held very tentatively.
Indeed, if it could be shown that any considerable degree of true sleep was involved, this might
possibly explain the failure of the experiment to yield positive results, since sleep retards
responses to suggestion and may wholly inhibit them.
SUMMARY AND CONCLUSIONS: We have seen that subjects as a rule show a fairly
marked tendency to both homoactive and heteroactive hypnotic hypersuggestibility. These two
tendencies are related to each other very much like the immediate effects of training and transfer
effects of training respectively. Whether or not this hypothetical relationship is the true one, at
least the two forms of hypersuggestibility conform to it in so far as relative strength is concerned:
homoactive hypersuggestibility is considerably (330) stronger than the heteroactive variety, just
as the effects of training are stronger on the activity trained than on some different activity. The
evidence at present available indicates that homoactive hypnotic hypersuggestibility is somewhat
more than twice as strong as suggestibility in the normal state, whereas the heteroactive
manifestation is somewhat less than twice as strong.
 It is particularly to be noted, however, that the hypnotic hypersuggestibility revealed by
these experiments has not an absolute, but only a relative, significance. While most persons
actually are more suggestible when in the hypnotic trance than when they are in the non-trance
condition, it does not follow by any means that all persons in hypnosis are more suggestible than
any person in the normal condition. This point may be illustrated concretely by the results shown
in Table 47. Subject No. 4 is nearly twice as suggestible in the trance as in the non-trance state.
Even so, this subject when in the trance state has a mean postural suggestion time of 42.5 sec.,
and no other subject in the group has a mean waking suggestion time greater than 31 sec.
Despite the limitation in the meaning of hypnotic hypersuggestibility just noted, it has
been regarded as presenting some possibilities of being a basic and essential characteristic of
hypnosis. For this reason the nature and origin of hypersuggestibility becomes of considerable
theoretical significance. A plausible hypothesis which would make hypnosis but a special case of
direct prestige suggestion is that whenever a direct prestige suggestion is reacted to positively
there is generated within the reacting organism a heightened susceptibility to react positively to
all other direct prestige suggestions. It has been possible to subject this hypothesis to a number of
experimental tests. These experimental results show that waking suggestions seem to facilitate
closely subsequent trance inductions as the hypothesis demands. Moreover, the hypothesis would
seem to hold also for purely voluntary simulated suggestion responses.
On the other hand, the small amount of experimental data available fail to show a
progressive weakening of hypersuggestibility during an inactive trance as the hypothesis would
lead us to expect. Moreover, a careful experiment has failed (331) completely to show any
indication of heteroactive hypersuggestibility in the case of waking postural suggestion. Both of
these tendencies are definitely demanded by the hypothesis. Also, a last experimental attempt to
produce hypersuggestibility in acts of waking suggestion by means of a special indirect
suggestion that this should take place at a certain point in the experimental procedure likewise
failed completely.
Finally, it may be pointed out that in the case of Hypothesis 2 we have probably witnessed
the early death of an exceedingly attractive conjecture. If the experimental test of the hypothesis
on the level of purely waking (head and arm) suggestions had been indefinitely delayed, the
hypothesis might conceivably have enjoyed a long and honored existence, found a respectable
place among the so-called theories of hypnosis, and so have contributed materially to the
continued confusion in this unfortunate subject. Meanwhile the fundamental question of hypnotic
hypersuggestibility still lies in an extremely unsettled condition. Rich returns are likely to follow
vigorous experimental pursuit of the numerous clues now available. (332)

20. CH 12: HYPNOSIS REGARDED AS HABIT: CLARK HULL: HYPNOSIS AND

SUGGESTIBILITY: APPLETON-CENTURY-CROFTS: NY: COPYRIGHT 1933: 1961
IT HAS long been known that successive repetitions of the hypnotic trance facilitate the
induction of the trance on later occasions. Husson's committee, appointed by the French Academy
of Medicine, stated in 1831, "When once a person has been thrown into the magnetic sleep, it is
not always necessary to have recourse to contact and passes in order to magnetize him afresh."
James Braid wrote in 1843, "It is important to remark, that the oftener patients are hypnotized,
from association of ideas and habit, the more susceptible they become." Forty-five years later we
find Binet and Fere writing from a different point of view, yet agreeing in this. They state, "The
hypnotic sleep, which is produced with so much difficulty and delay in fresh subjects, occurs with
alarming rapidity in those who have been long under treatment." Bernheim, referring to hypnotic
"sleep," says, "The habit of sleep is very soon acquired by the organism." Moll remarks,
"Training not only makes the hypnosis deeper, but makes it appear more quickly." Forel states,
"That one increases the suggestibility of a person by repeated hypnotizing is an assured fact."
Bramwell writes, "Hypnotic suggestion tends to gain strength with repetition." Pavlov
characteristically concludes, "We can, therefore, regard 'suggestion' as the most simple form of
the typical conditioned reflex in man." On no phenomenon from the whole field of hypnosis,
perhaps, is there more complete agreement by the writers on the subject. Clearly, hypnotic
susceptibility is greatly facilitated by practice.
OBJECTIVE CHARACTERISTICS OF HABITUATION: It is a remarkable fact, and
an interesting commentary on the tardy scientific development of hypnotism, that this striking
and well-known phenomenon aroused the experimental curiosity (334) of no one for at least a
hundred years. But, with the advent of the modern experimental approach to human behavior, it
was inevitable that the phenomenon should give rise to numerous queries which could be
answered only by appropriate experiment. These queries naturally center around the question of
whether hypnosis is in fact a habit phenomenon. However, the fact of facilitation by practice is
but one of a number of features characteristic of habituation. A second characteristic is that the
rate of facilitation in the practice activity is usually greatest at the beginning of practice, and
always grows less and less in the later stages of practice. This yields what is known as a practice
curve of negative acceleration. A third characteristic is that after a relatively prolonged period of
no practice there is observed a partial loss in the facility previously acquired. A fourth
characteristic is that, other things being equal, the loss of facilitation following disuse is less
where the original repetitions were widely spaced than where closely spaced. A fifth
characteristic is that with the resumption of practice this loss is recovered, the new practice curve
showing a picture of negative acceleration. A sixth characteristic usually encountered is that the
amount of loss resulting from disuse is recovered with less practice than was required for its
original acquisition. If hypnosis should be found to conform in these numerous detailed respects
to the known characteristics of habituation, the similarity could hardly be attributed to
coincidence. There will accordingly be established, in addition to the matters of fact, which are of
interest and scientific value in themselves, a certain presumption that hypnosis is essentially a
habit phenomenon and that ultimately it may be best understood and explained from this point of
view.
TECHNIQUE OF MEASURING PROGRESS IN HYPNOTIC HABITUATION:
With a process which presents such numerous phases as hypnosis it becomes difficult to settle a
question of this kind by means of a single experiment. Perhaps an ideal program would
investigate every possible phase by means of a separate (335) experiment. If a random sampling
of the phases should all prove to conform to the criteria of recognized habits, the results would be
convincing. By virtue of the same principle, if one of the more important of these aspects is
investigated and is found to possess the characteristics of habit, some presumption in the same
direction will have been established. Such an investigation is reported by Krueger.
Of the numerous phenomena which might have been employed as indicators of the
onset of the trance, the closure of the eyelid resulting from suggestions to sleep was chosen. This
was a wise choice. Because it is an overt movement, it not only may be observed without
difficulty but it may be made to trace a precise and permanent record of its occurrence in
response to suggestion. The normal eyelid, while apparently very delicate and often moving at
high speed, has abundant energy to activate the delicate recording device employed by Krueger.
A diagram of the apparatus is shown in Figure 57 (not available).
"A fine capillary glass tube (C) 16 cm long and 30 mm in diameter, one end of which was
fastened to the timer with a (336) few drops of cement, served as a recording stylus for the lid
movements. A small piece of tin foil (T) about 5 mm square was attached to the subject's upper
eyelid with a drop of cement. A fine silk thread fastened to the tin foil rose vertically to the glass
ring (H) passed through the ring (H'), the guide ring (F) and thence to the glass marker. A
downward movement of the lid consequently produced an upward movement of the stylus. Glass
was used for the rings in order to minimize friction. A universal clamp interposed between rods
(G) and (1) permitted (G) to be placed in any desired position. In order to prevent breaking the
glass marker by too great an excursion, a pin (E) was fastened to the thread below the guide (F).
In case of a gross downward head movement, the movement of the thread is interrupted by the
pin striking the guide, causing the tin foil to come off the subject's lid."
The subject rested in a half-reclining position in a comfortable Morris chair, his head
securely supported by the back cushion. Because of the subject's position the thread attached to
the eyelid could extend vertically to point H without touching the eyebrow. Obviously the thread
could not be seen by the subject. The glass stylus was so delicate that after the foil T had been
attached one or two minutes the process of sensory adaptation rendered the subject entirely
oblivious to its presence. Indeed, subjects would frequently inquire at the end of the experiment
whether the thread had not become detached.
The tracings of the lid behavior while the subject is going into the trance have a special
interest of their own. In some sense they may be considered graphic representations of the
progress of hypnotization as a function of time-a kind of curve of the trance-induction process.
Subjects differ more or less in the records which they produce. Some stare at the hypnotic crystal
without blinking. These subjects show a gradual lowering of the upper lid until the two lids come
together as in sleep. Others blink normally at first, but as the suggestions continue the blinking
becomes so rapid that it should be described very properly as a flutter. Presently this gives place
to complete and permanent closure, though a certain tremulous movement persists. Such a record
is reproduced as Figure 58 (not available). (337)
HABITUATION EFFECTS WHEN HYPNOSIS IS REPEATED AT SHORT
INTERVALS: Krueger investigated two phases of the problem-that of the increase in rate of
lid-closure when the successive trances were in close succession, and that when they were
separated from each other by longer intervals. In the first investigation seven male subjects were
employed. Because of the rather intensive nature of this experiment, only subjects who showed
fairly prompt lid-closure were chosen. A small, bright fixation object was used, and suggestions
of lid-closure and sleep were given. In general these subjects gave evidence of being in a
moderately deep trance. All were specifically tested on their ability to open their eyes when
challenged to do so after having received the suggestion that they could not. All subjects failed.
(338) to do this, but gave evidence by facial expression and otherwise that they were making a
genuine attempt. The trance was, induced on each of these subjects ten times in succession on one
day, and again five times in succession twenty-four hours later. The subjects were allowed to
remain in the trance two minutes. They were wakened by the experimenter's counting to three and
then snapping his fingers. Suggestions for each succeeding trance were begun five minutes after
the subject was aroused from the preceding one. The same general procedure was followed for
the five trances on the second day. The time required to induce the trance was read from the time
line between the signal marks A and B (Fig. 58 (not available)).
The experimental results from this part of the investigation aTe shown in Table 57 (not
available). There it may be seen that, while the (339) subjects vary considerably in their speed of
trance induction, all without exception show very marked decreases in the tranceinduction time
with successive repetitions. With a view to securing a kind of composite picture of the process
which would be largely free from the distortions of accidental factors, thus enabling the more
refined criteria of habituation to manifest themselves if present, the mean time of lid-closure for
all seven
subjects has been computed for each of the fifteen trances. This is shown in the last column of the
table. From this series of means has been plotted Figure 59 (not available) .
From an examination of Table 57 (not available) and Figure 59 (not available) several
conclusions may be drawn.
1. Repetition markedly facilitates the closure of the lid in trance induction.
2. The curve falls most rapidly at first, more slowly later.
3. A period of disuse produces a marked loss in the effect of the previous repetitions. (340)
4 Resumption of practice restores the lost facilitation.
5 The recovery of the facilitation probably takes place at a more rapid rate than did its original
acquisition.
The truth of this last principle may be seen from an inspection of the two curves shown in
Figure 59 (not available). It comes out even more clearly if we divide the mean induction time of
trances 14 and 15 by that of II and 12, and compare this ratio with a similar one obtained from
dividing the mean value from trances 9 and 10 by that from 6 and 7. The former yields a ratio of
0.58, showing a comparatively steep fall, whereas the latter yields a ratio of 0.77, indicating a
distinctly slower rate of fall. This experiment accordingly shows lid-closure as conforming to all
the quantitative characteristics of true habituation.
HABITUATION EFFECTS WHEN HYPNOSIS IS REPEATED AT INTERVALS
OF TWENTY-FOUR HOURS: Krueger's second experiment on habituation to hypnosis was
substantially like the one already described, except that the interval between the successive
trances was at least twenty-four hours instead of five minutes, and the retention period (341) was
approximately three months instead of twenty-four hours. Six subjects were used. Their
trance-induction (lid-closure) times for the first ten experimental periods are shown in Table 58
(not available). Average values for the group as a whole are shown in the last column. These
values when plotted yield an exceedingly smooth and perfect practice curve.
Krueger's retention results for this experiment are complicated by the fact that his subjects
varied somewhat in the number of periods devoted to the original practice series. A substantially
accurate picture of the situation was secured, however, by tabulating the results in such a way that
it would be possible to plot a composite curve of a large section of the original training series
which immediately preceded the period of no-practice, to the end that the relative amount of loss
and the comparative rate of recovery might be seen. These results are shown graphically in Figure
60 (not available). (342)
An inspection of Figure 60 (not available) shows for the most part the same general
picture as that of Figure 59 (not available). Figure 60 presents a remarkably smooth and
consistent practice curve with the characteristic negative acceleration. It also shows the
anticipated loss following the period of disuse. It is to be noted that it is strikingly small in extent
as compared with the relatively great amount of loss from only twenty-four hours of no practice,
as shown in Figure 59 (not available). Actually the average loss from the three months of
no-practice was a shift from 93 sec. to I07, or only 14 sec., which is slightly less than half the loss
resulting from a twenty-four hour retention interval with the first group. This is all in conformity
with parallel phenomena from the field of laboratory experiments in learning. The rate of
recovery in this second experiment, however, does not conform to the typical learning formula.
The rate of progress on the resumption of practice, instead of being more rapid than the original
acquisition, is actually slower. Nevertheless, the quantitative picture as revealed by Krueger's
investigation as a whole is decidedly one of habituation by practice.
PREMIMINARY STUDIES SHOWING HABITUATION IN DIRECT WAKING
SUGGESTION: We must now raise the question of whether direct waking suggestion shows
habituation effects parallel to those shown by Krueger to be characteristic for the process of
hypnotization. In 1929 Hull and Huse found clear evidence that suggestions to postural
movement given alternately in the normal And the hypnotic condition progressively reduces the
time required for the evocation of the maximal response. Their published graph also shows a
tendency to negative acceleration. The following year, in an experiment of a somewhat similar
nature, but so organized that it was possible to plot a more extended practice curve, Williams
found unmistakable evidence of facilitation of the maximal postural response through practice
and also of the negative acceleration shown by Krueger for the eyelid in hypnosis and
characteristic of learning curves in general.
While valuable as corroborative evidence, both of the above studies were complicated by
the fact that half of the practice in (343) the postural suggestion was given when the subjects were
in the hypnotic state. A more serious limitation lies in the fact that neither of them throws any
light on the retention and recovery behavior of the process. Fortunately, a subsequent study by
Patten, Switzer, and Hull makes good both defects and enables us to obtain a fairly accurate
comparison with the parallel results of Krueger on hypnotic lid-closure.
A STUDY OF HABITUATION AND RETENTION IN DIRECT WAKING
PRESTIGE SUGGESTION: The investigation by Patten, Switzer, and Hull (I2) used the
ordinary postural suggestion technique in the normal waking state on sixteen university students
who had previously been found to give maximal responses, i.e., who would fall completely
forward under the influence of verbal suggestion (344)
Three "falls" were given on each of twenty consecutive days. The suggestions for each
"fall" were begun one minute after the preceding fall. After a seventy-day period of no-practice
these subjects were put through an additional six-day period of practice to determine whether
there would be a loss of the facilitation and, if so, the mode of its recovery.
The average of the time required for the first falls of each day by the sixteen subjects on
each of the twenty days of original training and on each of the six days of resumed practice after
an interval of no-practice of seventy days is shown in the first column of Table 55 (not
available). These data are shown graphically in Figure 61. This composite graph shows in general
a very clear practice effect with a decided negative acceleration, quite in harmony with the known
law of habituation and with the results on hypnotic lid-closure. There is also a very clear loss in
facility of response after the seventy-day interval of no-practice. Lastly, the six-day period of
retraining shows indications of a steeper slope than did the last six days of the original training
period.
PRACTICE EFFECTS ON THE INDUCTION OF TONIC IMMOBILITY IN A
FOWL: By way of contrast to the above, there may be mentioned in closing this account some
observations reported by Mowrer concerning the effect of practice on the induction and duration
of animal "hypnosis" (tonic immobility) in the domestic fowl. While a number of chickens were
being subjected repeatedly to intense light, one rooster was observed consistently to fall into a
state of immobility which sometimes lasted as much as three minutes if the bird was left
undisturbed. Following this observation it was accidentally discovered that the same condition
could be induced by stroking the rooster's comb with the hand and rubbing downward on the
eyelids. This led to systematic observations on the time required to induce the state and the
duration of the "trance" thus brought about. The number of minutes elapsing from the beginning
of the stimulation until the lids had remained closed thirty seconds was regarded as the cataleptic
induction time. The duration of the "trance" was the number of minutes (345) elapsing from the
termination of the stimulation until the bird got to his feet. The detailed scores for the nineteen
days during which the experiment was continued are given in Table 59 (not available)
A perusal of these results reveals the following facts: There is a very marked and
consistent tendency for the rooster to fall into its catalepsy more rapidly at each successive trial
on any given day. This parallels exactly the tendency of human subjects to respond more
promptly to suggestions at closely successive trials on a given occasion On the (346) other hand,
there is an equally remarkable lack of positive practice or habituation effects in the total time
required for the three cataleptic inductions from day to day. Indeed, there are distinct hints of a
negative practice tendency near the end of the experiment, where the catalepsy induction time has
increased from seven or eight minutes to eleven or twelve. Lastly, in the total duration of the
catalepsy thus induced (last column, Table 59 (not available) ) these hints of a negative practice
effect become gross and unmistakable. These zero or negative practice effects are in sharp
contrast to the positive effects found in true human hypnosis and waking suggestibility and point
to a decided and significant differentiation of the two states. These results definitely confirm the
view of Tromner, as reported by Schilder and Kauders, who offers as one of six criteria of
distinction between animal catalepsy and human hypnosis the fact that they are affected by
repetition in an opposite manner.
SUMMARY AND CONCLUSIONS: We have seen that the available experimental
results as a whole show that the process of hypnosis manifests the following characteristics:
Practice in the act facilitates its performance; the rate of gain is more rapid early in the practice
than later; a period of disuse is followed by a partial loss of the facilitation resulting from
practice; the amount of loss from disuse is greater where the practice intervals are closely spaced;
a resumption of practice produces a recovery of the lost facility; the curve of the recovery is one
of negative acceleration; and the general rate of recovery of facility is faster than was its original
acquisition at the point in question. Such a remarkable and detailed conformity of the phenomena
of hypnosis to the known experimental characteristics of ordinary habituation can hardly be
accidental and without significance. The indication would seem to be that, whatever else hypnosis
may be, it is-to a considerable extent, at least-a habit phenomenon and that quite possibly this
hypothesis may furnish the basis for an ultimate understanding and explanation of its hitherto
largely inexplicable characteristics.
While not quite so complete experimentally, the parallel (347) results on direct waking
suggestion show exactly the same characteristics as those of hypnosis, just enumerated. This not
only serves to identify waking suggestion as in some sense a habit phenomenon, but to strengthen
still further the kinship between hypnosis and direct suggestibility already indicated by the
evidence of correlational studies, as well as general a priori considerations. These positive
habituation phenomena also serve a useful purpose in differentiating hypnosis and direct waking
suggestion from animal catalepsy on the one hand and from indirect or non-prestige suggestion
on the other.
Finally, it should be noted that all of the experiments summarized in the present chapter
have been based on the homoactive principle, i.e., the criterion of habituation in all cases has
been the influence of repetition upon the act repeated. Hypnosis should also show a complete set
of heteroactive practice phenomena in the form of facilitated responsiveness to miscellaneous
non-practiced suggestions in successive trances. The experimental determination of this matter is
logically the next step in the development of this branch of the subject. The findings may be very
significant as to the nature of hypnosis. (448)

21. CH 13: HYPNOSIS AND NON-PRESTIGE SUGGESTION: CLARK HULL: HYPNOSIS

AND SUGGESTIBILITY: APPLETON-CENTURY-CROFTS: NY: COPYRIGHT 1933: 1961
FREQUENTLY throughout the preceding pages, in connection with discussions of
various phases of hypnotic or prestige suggestion, we have had occasion to mention other forms
of suggestion the detailed consideration of which has been reserved for the present chapter.
Because of the general lack of the prestige characteristic among these forms, we shall follow the
practice of A veiling and Hargreaves by calling them non-prestige suggestions.
BINET'S PROGRESSIVE-INCREMENT OR PERSEVERATIONAL
SUGGESTIONS: It may be recalled that Binet was closely associated with Charcot's hypnotic
activities and that in collaboration with Fere he wrote (I888) a book on the subject. Some years
later, his interests having shifted to the development of tests for the determination of individual
differences, but still regarding suggestibility as important psychologically, he cast about for a
convenient method of measuring it. He judged it inadvisable to use the actual induction of the
hypnotic trance or any similar procedure as a test, because of widespread popular fears regarding
hypnotism. Moreover, it was forbidden to use hypnosis in the schools, where his researches were
mainly to be conducted. He therefore devised the progressive-lines test and the somewhat similar
progressiveweights test in order "to judge the suggestibility of a person without resorting to
hypnotism or similar procedures". These tests have been used rather widely, though usually with
some modifications, since 1900, when Binet's major work describing them was published. Of
special importance for our purposes was a modified procedure developed in 1916 by Brown (350)
The latter, with still further modifications, was extensively employed by Williams (I930) in an
attempt to determine the relationship of some typical non-prestige suggestions to hypnosis.
The apparatus employed by Williams, and later by Hull and Forster, for the administration of
the progressive-lines suggestion test (I7) is shown diagrammatically in Figure 62. The lines were
presented in succession at the window, A. This aperture, 2 X I1.5 em. in size, was placed in the
center of the screen, which was 40 em. high and 56 em. wide. The lines to be presented were
drawn in India ink on a disk of very stiff white cardboard which was attached at its center to the
back of the screen in the position indicated by the dotted circle shown in Figure 62. The
approximate positions and relative lengths of the lines are indicated by the broken radial lines
within the circle. The actual lines were, of course, unbroken, like the shortest one, B, which is
represented as being seen through the window. The disk would be turned by means of the handle
E in such a way as to expose the lines at suitable time (351) intervals. The lengths of the lines in
the order of presentation to the subject were:
No. mm.
 I, 12
2, 24
3, 36
4, 48
5, 60
6, 60
7, 60
8, 60
9, 60
10, 60
11, 60
12, 60
13, 6')
14, 60
15, 60
As may be seen in Figure 62 (not available), the lateral positions of the lines as they
appear in the window vary more or less at random, in order that the subjects should not form
habits of judging the lengths of the lines by noting the distance from one side or the other of the
aperture. It is well known, however, that the position of a line within an aperture has an important
influence on its apparent length. To prevent constant errors from this source from systematically
distorting the reactions, Hull and Forster used nine different disks which so varied the position of
the stimulus lines that each line of the series occupied each lateral position an equal number of
times.
One of the verbal formulas employed by Hull and Forster in the progressive-lines
experiment was as follows: "I am going to show you some lines and I want you to tell me if the
second line that you see is longer or shorter or just the same length as the first line that you saw.
Then I'll show you another line, and you tell me whether that one is longer or shorter or just the
same length as the line you saw just before it."
The subject will normally judge lines 2 to 5 longer because they are obviously so in fact;
but many subjects continue more or less to judge the later lines of the series longer than their
predecessors also. In this lies the indication of suggestibility. (352)
The apparatus for the progressive-weights suggestion test as employed by Williams, and
later by Hull and Forster (IO), consisted of fifteen white cardboard boxes each 41 x 41 x 18 mm.
in size. On the top of each box, as used by Williams, was clearly marked the number indicating
the position of the box in the series. The boxes were placed on a table before the subject, as
shown in Figure 63 (not available). The weights of the respective boxes were as follows:
No. gm.
I, 20
2, 40
3, 60
4, 80
5, 100
6, l00
7, 100
8, l00
 9, 100
10, 100
II, 100
12, 100
13, 100
14, 100
15, 100
The instructions given to the subject in this test have been either printed or oral, though
there is no reason why they might not at present be given phonographically if an experiment
should demand the uniformity of direction thus obtainable. The wording of the instrnctions will
vary somewhat with the age and maturity of the subjects. The procedure employed by Hull and
Forster with boys between ten and eleven years (353) of age combined the oral method with
demonstration and preliminary practice. The verbal formula used by them was as follows: "I want
you to lift this weight [pointing to the first weight] and then this weight [pointing to the second
weight] and tell me whether this one here is heavier or lighter or just the same weight as that one
[pointing again to the first weight]. Then lift this one [pointing to the third weight] and tell me
whether this one is heavier or lighter or just the same weight as that one [pointing to the second
weight] and then do the same for the rest of these."
Under these conditions, subjects normally judge each successive weight from the second
to the fifth heavier because they obviously are so. But there is a strong tendency to continue to
call the succeeding weights heavier also, which they are not. In this, of course, lies the evidence
of suggestibility.
OTHER FORMS OF PERSEVERATIONAL SUGGESTION: A third form of
experimental suggestion which appears to be more or less of the perseverational variety is based
upon the perception of a faint warmth stimulus. It results, when successful, in a normal
hallucination of temperature. This experiment seems to have originated with Seashore, in 1895. It
has been used by a number of investigators since that time, and the apparatus has taken various
forms. In the method employed by Williams, the experimenter sits at the right side of a table, the
subject at the left. A coil of German silver wire occupies a conspicuous place between them on
the table. This may be heated gradually by leading through it a two-volt current from a battery.
So far as the subject can see, the current is turned on in the coil by closing a simple knife switch.
Actually the current does not go through the coil unless, in addition to the closing of the switch, a
button is also pressed. However, the button and its wiring are wholly concealed from the subject's
view. The former is placed on the side of the table leg and is operated by pressure from the
experimenter's knee.
The subject was given typewritten instructions which were adapted from Brown. They
were as follows: "The purpose of this experiment is to determine the smallest amount of heat that
you can feel with your finger. When the experimenter says 'ready' you are to place your finger on
the coil and hold it there until you can just barely feel the warmth. At the instant you feel any
warmth whatsoever remove your finger sharply from the coil. Then wait until the experimenter
tells you to replace your finger on the coil."
Continuing with Williams's report, "At the word 'ready' the experimenter closed the
knife-blade switch. For the first four stimulations the push-button switch was also closed, thus
closing the circuit so that the current heated the coil. The following six trials were made by
closing the knife-blade switch alone so that no warmth could be generated in the coil. The subject
continued to touch the coil for thirty seconds if he had not already responded to the suggestion of
warmth. If no response was made within thirty seconds the experimenter said, 'All right, you did
not feel it this time,' and the subject removed his finger. After each trial when heat was generated
in the coil, the experimenter felt the coil and continued to do so until the coil was back to normal
roam temperature. Exactly the same routine was followed in the cases when no heat was
generated and a pause was made equivalent to that necessary for the coil to cool after being
heated. This routine was followed irrespective of whether the subject responded to the suggestion
or not." 1 It will be seen that this test was conducted in a manner such as to render the results
obtained from it as comparable as possible with those from the lines and the weights.
A fourth perseverational suggestion experiment is based on the faint sense of touch. When
successful, it also, results in a simple normal hallucination. The apparatus of this experiment as
used by Williams consisted of a screen of black sateen cloth suspended from a 6o-cm. rod which,
in turn, was supported by two laboratory tripod standards. Six small cork weights used for
purposes of stimulation were suspended by threads from as many small rods.
The instructions which Williams used were an adaptation from those employed by Brown.
They were handed to the subject in typewritten form: (355) "You see this set of small cork
weights. Some of them are so light that they cannot be felt at all. It is the purpose of this
experiment to find the lightest of this set of weights which you can just barely feel on the tip of
your middle finger. When the experimenter says 'ready' you are to see whether you can feel the
weight or not and report to the experimenter immediately by saying 'yes' or 'no.' "
After the subject had read the instructions he was told to close his eyes, and one of the
heavier weights was placed on the tip of his second finger. He was asked whether he felt this, and
he invariably said that he did. This was then repeated with the lightest weight, which was not
heavy enough to be felt. The subject was next told to open his eyes and see the weight resting on
his finger, following which he placed his hand behind the screen for the tests. Gradually
diminishing weights were used for the first four stimulations, thus making the fourth barely
perceptible. There was no actual stimulation for the remaining six tests, but the rods were moved
about by the experimenter behind the screen in such a way as to produce a sound similar to that
which occurred while changing the weights when the real stimulations were made. After an actual
stimulation was made, a pause of three or four seconds intervened between the word "ready" and
the placing of the weight on the finger, in order to give the subject a period of anticipation and
thus prevent him from responding immediately after the word "ready" when there was no
stimulation. The noise of moving the rods and the pause were both intended to increase the factor
of suggestion in the tests.
Many other forms of experiment believed by one writer or another to test suggestibility
have been reported. Perhaps the greatest collection of these is that assembled by Brown, which
comprises twenty-one different forms, including the four described above. Depending somewhat
upon how they are administered, tests of this kind, while remaining indirect in their nature, tend
to acquire more or less extensive prestige elements. An example of the latter would be a situation
in which the subject has the task of making a judgment of length, size, weight, or some other
characteristic and at the conclusion (356) of the instructions there is appended a note stating that
most people tend to judge whatever it is as too small or too large, too heavy or too light, etc. If a
subject's reactions are influenced in the direction of a compensation against the supposed
tendency to error, this is considered an indication of suggestibility. Such indirect prestige
elements are effective with children and, according to Brown, adults are by no means immune to
them.
PERSEVERATIONAL NATURE OF THE PROGRESSIVE-LINES SUGGESTION
TEST: To immediate inspection, few processes could appear to differ more widely from the
direct prestige suggestion of the posture suggestion test than that of Binet's two forms of
progressive-increment suggestion. On the surface, at least, the distortion of judgment generally
accepted in the latter as evidence for suggestibility appears somehow to result from the
perseveration of the influence of a kind of habit or "set" acquired by the subject in reacting to the
first five members of the series, where it properly applies, inta the remainder of the series, where
it does not apply. Some objective evidence bearing on the nature of these forms of suggestion has
been reported by Hull and Forster.
Ten boys from ten to eleven years of age were given the progressive-lines test described
above for ten successive days. The lines were presented by means of the apparatus shown in
Figure 62 (not available). The boys were instructed in this case as follows: "I am going to show
you some lines and I want you to draw a line here [pointing to a ruled line on the record sheet]
which is just as long as the line you see."
Careful measurements were made of the length of the 1,500 marks thus obtained, and the
average judged length of each of the fifteen stimulus lines was computed. These averages are
represented graphically in Figure 64 (not available), in relation to the actual lengths of the
stimulus lines. There it may be seen that though the subjects as a group tend to underestimate the
lines presented, their drawings increase in length with remarkable (357) uniformity up to the
fifth. At this point, if there had been no influence of suggestion, the curve presumably should
have become horizontal, i.e., it should have followed the course of the broken line marked
"No-suggestion level." In general the tendency to increase the length of the lines clearly persists
throughout the entire series of ten test reactions, though, as might have been expected of a
perseverational tendency, the amount of gain grows progressively weaker with the increased
distance from its source. The remarkable smoothness and (358) uniformity of this graph is
distinctly convincing as to the genuineness of the tendency and its point of origin.
It is possible also to plot perseverational tendencies directly from the pooled experimental
results in the case of the heatsuggestion test and that of touch described above. Williams
publishes two perseverational graphs of the former, and one of the latter. While somewhat erratic
and not entirely consistent, these results are probably in fair agreement with the tendency
indicated in Figure 64 (not available).
DETERMINATION OF THE AMOUNT OF SUGGESTIBILITY AND ITS
PERSEVERATION WITH BINET'S WEIGHTS: The perseverational nature of the
progressive-weights suggestion test cannot be arrived at so directly as in the case of the three
forms of suggestion just considered. The reason is that the replies likely to be made by the
subjects are of three kinds, viz., that a given weight feels heavier than, lighter than, or the same
as, the one preceding it. Binet employed the somewhat clumsy method of representing separately
the scores of all three types of responses. Whipple recommends the "admittedly somewhat crude"
method of considering only the score of "heavier" judgments. This method of scoring seems to
have been taken over uncritically by Brown and by Williams.
It is quite possible that the scoring procedure recommended by Whipple was adequate to
the needs of a rough-and-ready method of classifying or "testing" groups of children. It is by no
means adequate for the more exacting requirements of a scientific investigation seeking to
determine fundamental principles regarding suggestible behavior. This becomes evident at once
when it is realized that a subject may have a very appreciable score of "heavier" judgments and
yet show no tendency to suggestion whatever. He might, for example, judge 33 per cent of the
weights "heavier," 33 per cent "lighter," and 34 per cent "same." In such a case there is just as
much reason to infer a negative suggestion effect, i.e., one in the direction of "lighter," as one in
the direction of "heavier" which the uncritical Whipple technique would do. Actually, such a
score constitutes no evidence of suggestion of any kind. The moral (359) is that in cases of this
sort the amount and nature of suggestibility can be determined only by a joint and simultaneous
consideration of all three types of scores. It is only as the scores of "same" judgments are shifted
as a whole to one side or the other of the strictly. central position assumed in the above example
that any evidence of suggestion exists. By the same token, it is evident that the amount of the
tendency to suggestibility is measured by the magnitude of this shift.
The method of determining the amount of distortion in the judgments of pairs of weights,
lines, etc., supposedly due to suggestion, may be illustrated by an example. The first three entries
in one of Hull and Forster's published tables represent respectively the mean number of "heavier,"
"same," and "lighter" judgments made by fourteen boys when tested three times each on the first
of a ten-day series. The mean values are: 76 per cent "heavier" judgments, 14 per cent "same"
judgments, and 10 per cent "lighter" judgments. How shall these three scores be converted into a
single score of suggestibility?
The method proceeds upon the assumption, commonly made in operating with data of this
kind, that all of the judgments taken together are distributed according to the "normal curve (360)
of error." Detailed tables of this function have been worked out by mathematicians; a simple and
convenient one appears in Thorndike's Mental and Social Measurements. Now, it will be seen by
an inspection of Figure 65 (which has been plotted according to Thorndike's table) that the
normal curve is bilaterally symmetrical; accordingly, 50 per cent of the judgments will fall in
each wing. It follows, then, that the 76 per cent of "heavier" judgments will occupy the entire
right wing of the distribution and also the inner 26 per cent of the left wing. Reference to the
table shows that in this region of the distribution 26 per cent of the judgments must occupy a
space extending from the middle of the total judgment distribution to a position 0.71 0 distant.
Next to this in the distribution must be placed the 14 per cent of "same" judgments; this makes a
total of 26 + 14, or 40 per cent on the inner portion of this wing. A second reference to the table
show5 that 40 per cent occupies a space extending from the center of the distribution to a point
1.28 0 distant from the middle. With one margin of the "same" judgments at 0.71 0 and the other
at 1.28 0, it is obvious that its mid-point falls at a point slightly in excess of 0.99 0 from the
middle of the distribution. But since the "normal" position of the mid-point of the "same"
judgments is in the exact middle of the entire distribution, it is evident that the displacement or
distortion of judgment in the present case must be 0.99 o.
By means of the above procedure, Hull and Forster (IO) were able to secure, from the
specially pooled lifted-weights scores of the fourteen boys mentioned above not only quantitative
values of the strength of the tendency to overestimate each successive weight throughout the
purely suggestion portion of the series (weights 6 to 15) but, for purposes of comparison, values
for the strength of the tendency to call the successive weights heavier from No. 2 to NO.5, where
they really were heavier. The results of this procedure are shown in Figure 66. In order to
facilitate comparison, it may be noted that the values in this graph correspond to the differences
between the successive estimated values for lines as shown in Figure 64. Despite the radical
difference in the method by which these two sets of values were obtained, there is (361)
substantial agreement, though the procedure with the weights appears somewhat more prone to
sampling influences than is the method of direct measurement which may be used as an
alternative procedure with the lines. The first suggestion judgment on the (362) weights series
(weight 6, Figure 63 (not available)) shows slightly less than half as great a tendency to the same
reaction as the reaction immediately preceding it, where there was a genuine difference in the
weights judged. This tendency gradually grows weaker as the series continues, exactly like that
with the lines but less consistently and at a less rapid rate. It may be that the apparently slower
weakening of this perseveration effect is due to the possibility that a portion of what appears to be
suggestion here, in reality is due to a well-known tendency to judge heavier the second of a pair
of weights, apart from any suggestion whatever.
DISTRIBUTION OF NON-PRESTIGE SUGGESTIBILITY: It is evident that
practically all of the questions which have been asked throughout the preceding chapters
concerning prestige suggestibility, hypnosis, and their relationships, may also be asked
concerning non-prestige suggestion. In many cases no experimental evidence is yet available
where such questions logically arise, though we are fortunate in several instances in possessing it.
The remainder of the present chapter will be devoted to the presentation of such evidence. It is
hoped that a perusal of this will aid the reader in forming an intelligent judgment as to whether
the non-prestige suggestions are essentially related to hypnosis or whether the fact that they are
called suggestion is a mere accident of nomenclature.
The first question which we shall consider concerns the distribution of susceptibility to
non-prestige suggestion. It may be recalled that we considered above at some length whether
direct suggestion is distributed in the ordinary population according to the so-called "normal
curve of error." The same question also arises here. Fortunately it may be disposed of rather
easily. Aveling and Hargreaves, who believe that the prestige suggestions show a U-shaped
distribution, (363) agree that the non-prestige forms involving progressive lines and progressive
weights yield a distribution substantially normal in its general shape. Brown publishes
twenty-two distributions of test scores of suggestibility for a great variety of suggestions which,
for the most part, are of the non-prestige variety. Of these all but one show a clear tendency to the
normal form, and the single exception probably is due to an obviously non-linear scale which is
used as the basis for plotting. As an example of the "normal" shape of many of Brown's
distributioll8 there may be instanced the one shown in Figure (364) 67 (not available). This has
been constructed from a table computed from Brown's published data.
In addition to the shape of the distribution of non-prestige suggestibility, there arises, in
the comparison of prestige and non-prestige suggestibility, the question of whether the latter
manifests at one extreme of the distribution a certain percentage of genuine negative
suggestibility. This is something more than a simple failure to be influenced by suggestion. It is
comparable rather, to the backward movements of certain subjects in the posture test of direct
verbal suggestibility when "forward" suggestions are given. The analogy in the progressive-lines
and the progressive-weights experiment would be the consistent underestimation of lines and
weights beyond the fifth in the series. It is to be noted that forms of suggestion like that of
warmth and touch described above cannot show such a tendency. The possible presence of such a
negativism in progressive-weights data is not manifest upon the surface, however, as may be
realized by a glance at Figure 67 (not available). Brown himself, with these particular data in
mind, remarks,
"These experiments are successful in so far as they afford what has the appearance of
being a measure of the suggestibility of nearly every person tested. Only two individuals among
the 140 escaped wholly from the suggestion of Progressive Weights ... " The two exceptional
individuals referred to are, of course, the two with zero scores represented in the distribution
shown in Figure 67 (not available).
When we recall the principle put forward above in connection with Figure 65 and the
method of computing the degree of suggestibility from such scores (p. 36Iff.), it becomes
apparent that we must be extremely cautious about accepting crude "heavier" judgments alone as
evidences of suggestibility. If we possessed the number of "same" judgments of each of the 140
persons, it would be possible by the methods already described to calculate a definite suggestion
index in terms of the a of the judgment distribution for each subject. However, there is available a
statement that the average woman called the weights equal 3.5 times out of ten trials, whereas
(365) the average man called them equal 3.7 times in the ten trials. If all subjects showed a
uniform tendency to call 36 per cent (35 + 37over 2) of the weights equal, this would leave 64 per
cent of the weights to be called either heavier or lighter. If these latter were equally divided
between heavier and lighter, there would, of course, be no evidence of suggestibility in either
direction, i.e., a subject would normally give 64 -;- 2, or 32 per cent of "heavier" scores without
showing any indication of suggestibility whatever. By referring to Figure 67 it will be seen that
fifty-two individuals gave scores of 30 per cent "heavier" judgments (three out of a possible ten)
or less. On the face of things, therefore, something like fifty individuals of the total of 140 appear
not only not to be positively suggestible but, rather, to be negatively suggestible. Strictly
speaking, individuals showing 36 per cent "same" judgments and 30 per cent "heavier" judgments
must show 34 per cent "lighter" judgments, which corresponds to a suggestibility of - 0.06 o. By
similar reasoning, a score of 2 (20 pcr cent) on the distribution shown in Figure 67 corresponds to
a suggestion value of - 0.40 0, a score of 1 to a suggestion value of - 0.59 0, and one of 0 to a
theoretical suggestion value of something like - 1.45 o.
Such data, however, are not adequate to demonstrate with certainty the existence of
genuine negative non-prestige suggestibility. It must be remembered that only ten reactions were
taken from each subject. This means that the element of chance in such scores must be large. It is
accordingly possible that the fifty individuals with scores of 3 or less "heavier" judgments may
have reacted thus largely, or even entirely, through chance. On this assumption, moreover, it is
likely that approximately an equal number of subjects gave equally large scores in the opposite
direction also through chance alone, i.e., scores greater than 3.2 of "heavier" judgments. These
two chance groups would, of course, make up a total of something like 100 persons out of the
140 who wholly escaped from the suggestion of the progressive weights, leaving, perhaps, no
more than forty or fifty who were really susceptible to this form of non-prestige suggestion. (366)
We conclude, then, that this form of behavior does not depart fundamentally from the
"normal" or bell-shaped distribution. In this respect, it probably resembles susceptibility to
hypnosis and direct waking suggestion as well as most other forms of human behavior. It is
possible that non-prestige suggestion differs from the hypnotic forms in not showing at one
extreme of its distribution the phenomenon of negativism, though as yet there appears to be no
satisfactory experimental evidence on the question. Data from the progressive weights and the
progressive lines, when treated by an adequate statistical technique, should show this if it exists.
Much larger samples of each subject's behavior would need to be taken than in the investigation
just considered which, unfortunately, would introduce complications due to practice effects.
CORRELATION OF NON-PRESTIGE SUGGESTIBILITIES: We pass at this point,
in our quest for evidence bearing on the relationships of the non-prestige suggestions, to the
consideration of investigations which approach the problem from the point of view of correlation.
The problem before us was seen over thirty years ago by Binet, who wrote: (367) "We shall have
to find out, for example, whether people who are easily hypnotizable are more sensitive to our
tests than are people who are very refractory to hypnotism."
We have already seen (p. n:ff.) that hypnotic susceptibility correlates positively and rather
high with susceptibility to direct waking suggestibility as revealed by the postural technique. We
must now raise the question: Do the non-prestige forms correlate in a similar manner with
hypnotic susceptibility? And how do the non-prestige suggestibilities correlate with waking
suggestibility of the postural type and with each other?
So far as the present writer has been able to discover, there has never been an attempt,
despite the age of the problem, to determine the correlation of hypnotic susceptibility with
susceptibility to any of the non-prestige forms of suggestion. This is particularly to be regretted,
since clean-cut results from this point of view would be of special value in defining the functional
relationship, if any, of the non-prestige suggestions to hypnosis. We are fortunate, however, to
possess in the study of A veling and Hargreaves a number of correlational results showing
relationships between certain non-prestige suggestions and two forms of prestige suggestion:
hand rigidity and hand levitation. These correlation values are shown in Table 60 (not available).
The entries in this table are averages of rank coefficients from somewhat miscellaneom: sources.
The subjects were English school-children of both sexes in the neighborhood of twelve years of
age. The number of subjects represented in the several entries of the table varies from 121 to 59.
An inspection of this table shows only one moderately high value, that of + 0.55, between
the two forms of prestige suggestion. The correlations between these and the four nonprestige and
semi-prestige forms, the center of our chief interest, is strikingly lower. For the most part,
however, they are positive. So far as these data go, they indicate a certain very slender but
positive kinship between the two. Lastly, there may be mentioned the correlations among the
non-prestige suggestions themselves, which are even lower than the values between the
contrasted groups.
There may also be mentioned in this connection an elaborate series of correlation values
published by Brown. Brown's published coefficients taken individually are of little significance
(368) because they are in practically all cases averages from rank coefficients which have been
computed from small groups at varying sizes. Moreover, there are no reliability coefficients
available, so that it is impossible to secure any notion as to the extent to which the values have
been "attenuated" by inadequacies of measurement.4 Even so, these data constitute the greatest
mass of experimental evidence available in this particular field and undoubtedly indicate a
significant tendency. Dtspite the fact that some possibilities of a sex difference in correlation may
exist, we have ventured in the interests of simplicity of exposition to combine the correlation
coefficients of the two sexes for the first nine of the twenty tests investigated by Brown, since
they have the appearance of being more definitely tests of non-prestige suggestibility. These data
are presented in Table 61 (not available). The occasional negative values are probably due to
chance factors involved in the sample of the subjects employed, and pretty certainly do not
represent negative relationships. An examination of this table shows a remarkably uniform lack
of correlation among these forms of suggestibility. While it is undoubtedly significant that the
average correlation of each form with all the others is positive, the uniformly low values of these
means is equally striking. It is evident, after all due allowance has been made for the effects of
attenuation, that none of these forms of suggestion, at least with adult subjects, has any
considerable amount in common with the others. Such facts as these emphasize the
appropriateness of speaking of the non-prestige suggestions in the plural rather than in the
singular.
We may say, then, that direct evidence concerning the correlational similarity of hypnosis
and the various non-prestige suggestions is as yet lacking. However, there are some indirect
indications that, if there is any relationship whatever between 'them, it is probably far less close
than that between hypnosis and prestige postural suggestion. (370)
SEX AND SUSCEPTIBILITY TO THE NON-PRESTIGE SUGGESTIONS: We are
fortunate in the matter of the relative susceptibility of the sexes to non-prestige suggestion in
possessing two major investigations, one based on a large number of children of various ages and
the other on a considerable population of young adults.
Messerschmidt (12) tested approximately twenty-five children, of each sex and at each
age from six to sixteen years, on eleven different forms of indirect suggestion. By an ingenious
statistical procedure she was able to convert into comparable values the somewhat diverse raw
data from the several forms. These converted values or indices have been averaged for each sex
and age and the results arranged in Table 62 (not available). There it (371) may seem that, upon
the whole, the girls are more susceptible to non-prestige suggestions than the boys; the boys
average more suggestibly than the girls at only three of the eleven years investigated. The mean
difference, while not great, has a critical ratio of 2.7 which means that a difference of this size
would be obtained by change only once in about thirty times.
The second major study, that of Brown to which (372) reference has already been made,
was based on university students. The number of subjects of each sex tested varied more or less
from one form of suggestion to another, but the round number employed (sixty men and eighty
women) in the progressive-weights test already considered is fairly typical.
The various forms of suggestion employed, together with the outcome with respect to sex
difference as judged by Brown, are given in the first two columns of Table 63. In the last three
columns of this table are placed the results of certain statistical treatments of Brown's published
data relating to the magnitude of the sex differences obtained and their critical ratios. An
irtspection of this table shows that, of the twenty-one forms of indirect suggestion which showed
some indications of a sex difference, eighteen, or 85 per cent, are in the direction of the women's
being more suggestible. The first nine of these which were capable of satisfactory statistical
treatment all yield a more or less statistically reliable difference, indicating that the women are
more suggestible. The difference in all cases, however, is slight. It averages only 0.27 a of the
male distribution, which is only 14 per cent as great a difference as that in the mean standing
height of the two sexes (1.92 a). This, it may be recalled (p. 83), is substantially the same
outcome, both as to sex and amount, as has been found in the case of hypnosis and the other
prestige forms of suggestion.
To these major investigations there may be added the results of several less extensive
studies. Of these, Gilbert, Seashore, Dressler, Yung, and Wolfe all report women or girls more
susceptible than men or boys on one or another form of indirect suggestion. Only one author,
Otis, reports boys very slightly more suggestible than girls, though no claim is made for
conclusiveness of these results. Hurlock, however, using Otis's tests in a study of both white and
Negro subjects, reports the boys of both races (373) as more resistant to suggestion than girls,
though the statistical reliabilities of the differences are not very high.
 The experimental results on sex differences in this field may be summed up as indicating
females fairly consistently, but only very slightly, more suggestible than males. In this the
non-prestige suggestions appear to resemble hypnosis and the other forms of direct prestige
suggestion.
AGE AND SUSCEPTIBILITY TO THE NON-PRESTIGE SUGGESTIBILITIES:
Perhaps the earliest systematic investigation of the relation of a non-prestige suggestion to age
was reported by Guidi. He tested a small number of children at each age from six to fifteen for
susceptibility to the illusion of heat produced by his "pseudo-stove" apparatus. This device, while
(374) mechanically very different from that employed by Williams (p. 354), apparently involved
a somewhat similar psychology. Guidi's results are shown graphically as Figure 68. These data
indicate rather strongly a rise in suggestibility during the early years, after which there is a
gradual fall beginning at about nine or ten years and continuing well into adolescence.
The most extensive experimental results in this field are those reported by Messerschmidt
and summarized in Table 62. These results are shown graphically in Figure 69. Each age point of
this curve represents approximately 450 measurements divided about equally between the two
sexes and somewhat unequally among eleven different forms of suggestion. The same general
picture is presented as that reported by Guidi, (375) though the consistency of the results from
age to age is naturally greater because of the relatively slight distortion resulting from the effects
of chance upon the choice of the samples of children tested. The rnost obvious characteristic
revealed by this curve is the rapid diminution of suggestibility with advancing age, which
confirms Guidi's earlier results. Children of sixteen years are only about one-sixth as suggestible
as those of seven. It is also noteworthy that both the boys and the girls average less suggestible at
six years than at seven, which also agreeS very well with Guidi's study. An examination of
Messerschrnidt's original tables does not show this for all forms of suggestion, however. Instead,
some forms show the maximum at eight years instead of seven. It is impossible to say how much
of this irregularity is due to chance, and how much to differences in the characteristics of the
forms of suggestion in questioll' It is reasonable to suppose, however, that genuine differences of
the latter sort may exist.
A comparison of the above results with those on direct prestige suggestion presents a
rather striking similarity, particularlY in respect to the umque reversal in the direction of the
growth curve in the neighborhood of seven or eight years, after which resistance tends to
increase. The indication would seem to be that both forms of suggestion are second-order habit
phenomena, the stimulus releasing the secondary or resistance component in the case of prestige
suggestion being primarily social and that releasing the correctional component of nonprestige
suggestion being primarily non-social.
HABITUATION AND RETENTION CHARACTERISTICS OF NOW PRESTIGE
SUGGESTION: With all insight very characteristic of his excellent work La Suggestibilite,
Binet foresaw over thirty years ago the question now Defore us, as well as the general nature of
its answer.
He says: "Finally it was to be noted that a repeated suggestion is less effective the second tlrne
than the first. This weakening is undoubtedly particular to those indirect suggestions in the
waking state which do not, properly speaking, involve any constraint upon the intelhgence of the
mdlvldual. In hypnotic (376) experiments, on the other hand, the suggestibility of the hypnotized
individual increases with the number of hypnotizations." It has remained, however, for later
investigators to supply the quantitative experimental proof.
 Williams, incidental to the investigation of a different problem, publishes graphs
purporting to be practice curves for all four of the non-prestige suggestibilities presented above.
While probably indicating a general tendency in the direction of a negative practice effect, the
results with the lines and weights suffer from the same inadequacies of scoring as do those of
Brown. Moreover, it is probable that Williams's subjects, being rather sophisticated adults, like
those of Brown, did not show enough suggestion effects to make negative practice effects easily
determinable.6 At all events, Williams's results do not indicate any positive facilitation of the
non-prestige suggestions such as we have seen to be so characteristic of hypnosis and related
forms of suggestion.
In an attempt to determine the nature and extent of practice effects in the progressive-lines
experiment, Hull and Forster tested ten ten- to eleven-year-old boys three times each day for ten
days, having the subjects reproduce the stimulus lines with a pencil. The 3,000 lines thus
produced, when measured in millimeters and averaged by days, yielded the following ten values:
54.1,54.5,56.0,56.7,55.1,54.6,53.8,55.1,55.3, 54,9. It would be difficult to imagine a set of
empirical data. which would show a more complete absence of practice effects.
In a second (unpublished) attempt to secure evidence of practice effects on
progressive-lines suggestibility, Hull and Forster tested fifteen ten- to eleven-year-old boys three
times per day for ten successive days on the progressive-lines procedure, using the oral-response
technique described above. The composite (377) results were plotted as a practice curve, which is
shown in Figure 70 (not available). An inspection of this graph reveals some indications of a
negative practice effect, since falling values by this system of determination indicate a weakening
tendency. From these two studies with independent groups of subjects and by distinct
procedures, it seems probable that practice effects on progressive-lines suggestion are slight,
though such as exist appear to be in the negative direction.
Hull and Forster, in a third experiment (unpublished) attempted to study the habituation,
perseveration, and retention characteristics of the progressive-weights suggestion experiment by a
procedure which paralleled in its major outlines the one (378) performed from a similar point of
view by Patten, Switzer, and Hull, and reported above (p. 323ff.). Twelve boys between ten and
eleven years of age were tested on the progressiveweights experiment three times each day for ten
consecutive days. This was followed by a no-practice period of 120 days, after which practice
was resumed for four more days. The mean scores of these subjects on "heavier," "same," and
"lighter" judgments are shown by days in Table 64. These data, when subjected to the statistical
treatment described above (p. 361), yielded the series of suggestibility values which appears in
Figure 71 (not available). A glance at this graph shows at once that the lifted-weights suggestion,
like waking and hypnotic (379) suggestion (Figs. 60 and 61), manifests a marked and typical
practice effect. A resumption of practice after four months also produces a further fall in the
curve, just as in hypnosis and direct waking suggestion. Somewhat paradoxically, the loss of a
noticeable proportion of the practice effects from a considerable period of no-practice, which is
so conspicuous in the prestige forms of suggestion, seems here to be largely lacking. Lastly, and
perhaps of central significance, it must be noted that this practice curves differs absolutely from
those based on prestige suggestion in being negative; the more practice given in the activity, the
weaker the tendency to suggestion becomes.
It is clear from this brief survey of the influence of repetition on non-prestige suggestions
that much work remains to be done (381) before we can speak with complete assurance
concerning the group as a whole. The evidence so far available indicates that such suggestions
may differ rather widely among themselves, as their low intercorrelations might lead us to expect.
The work of Hull and Forster on progressive lines and progressive weights indicates that both
(probably) are habit phenomena, in which respect they resemble hypnosis. They differ radically
from hypnosis, however, in that their practice effects are either zero or negative.
IS NON-PRESTIGE SUGGESTIBILITY INCREASED DURING HYPNOSIS? As
we have so frequently had occasion to note in connection with other aspects of non-prestige
suggestion, Binet envisaged the questions at issue more clearly perhaps than any writer since. In
1900 he wrote:
"We shall have to see, also, whether in somnambulistic states, which induce a notable
increase in suggestibility, people become more sensitive to our tests than they are in their waking
state."
Binet himself, however, seems never to have investigated the question experimentally.
The only experimental attack on the problem known to the writer was made by Williams (I7) in
1930. Williams, in (381) conjunction with an intensive investigation of the comparative
susceptibility to direct prestige suggestion in the trance and the waking state, made rather
elaborate tests of the same kind for evidences of heightened suggestibility in progressive lines,
progressive weights, temperature, and touch. He employed as subjects in this rather laborious
investigation one group of eight university students and a second group of six. With the second
group, tests for five diff'erent kinds of suggestion were administered in fairly close succession on
each experimental occasion. This experiment, possibly through the confusion of perseverational
influences among the various forms of suggestion employed, was probably defective in
technique. An additional reason for suspecting this is that it yielded rather decisively atypical
results on the relative degree of direct prestige (postural) suggestion, which has been sufficiently
investigated to enable us to judge a normal performance with some confidence. With the first
group, on the other hand, each type of suggestion was investigated on a separate occasion.
Because of this circumstance it may perhaps be assumed that these latter results are somewhat
more significant. The results from the two groups are summarized in Table 65 (not available). An
inspection of this table shows very little consistency, either for the same form of suggestion with
the different groups or for the different forms of suggestion with the same group. The only two
agreements in the first, or supposedly better, group are those of the lines and the weights.
Unfortunately, as we have noted above, a seriously defective form of scoring was employed with
these particular types of suggestion.
It is clear that if any of the mean values appearing in Table 65 (not available) represents
a genuine tendency, it must be very small in amount-probably much less than the tendency to
heightened suggestibility to direct waking suggestions. This, in turn, may be due to the
exceedingly weak tendency of at least two of these forms of suggestion among sophisticated
university men. At all events, the chaotic and inconsistent nature of the results appearing in this
table makes it impossible to answer with any confidence the question propounded so long ago by
Binet and to-day crying louder than ever for an answer. If we might hazard a guess based largely
on a combination of a priori expectation and experimentalist's judgment, it would be that (382) in
the alert trance condition as distinguished from the lethargic states in which the subject falls more
or less into a physiological sleep, hypnosis will be found ultimately to have no substantial
influence on the non-prestige suggestibilities.
In addition to the experimental evidence concerning nonprestige suggestion which has
been presented in detail, there may be merely mentioned two further relationships. The
relationship of non-prestige suggestion to intelligence has been considered in an earlier chapter
(p. 86ff.). A second, the relationship to perseverational tendencies, may be dismissed with the
statement that non-prestige suggestion either does not manifest this phenomenon or that, if it
does, the tendency is exceedingly slight and in a positive direction. A number of aspects such as
have been reviewed in Chapter IV have not been considered in the present chapter because, so far
as the writer has been able to discover, no pertinent experimental evidence exists. Among these
neglected aspects may be mentioned: the relation of non-prestige suggestibility to character; to
general mental morbidity; to special morbid mental conditions such as hysteria, catatonia, etc.; to
delinquency; to various drugs such as scopolamine; and to dissociation as such. The mere
mention of this formidable list of unexplored regions, to say nothing of the innumerable
inadequacies of the available evidence in the portions of the field considered in the foregoing
pages, emphasizes the great need of further research. Meanwhile, let us survey briefly the
evidence at present available, particularly in its relation to hypnosis.
SUMMARY: We have seen that the external appearance of the non-prestige suggestions
is very different from that of hypnosis. We find, moreover, that the a priori conclusion that the
progressiveincrement tests of suggestibility are perseverational phenomena is confirmed by
experiment. In all, we have found evidence bearing on seven of the possible criteria of kinship of
non- prestige suggestion and hypnosis. While all of these leave something to be desired in the
way of conclusiveness, some possess experimental evidence which is tolerably convincing. Three
of the seven, (1) the shape of the distribution of the trait in the general population, (2) its relation
to age, and (3) its relation to sex, show a resemblance to prestige suggestion. There is, of course,
nothing remarkable in the first, since it is common to a very large number of behavior forms
which otherwise are known not to be specially related to each other. The extent of the agreement
as to sex susceptibility is very slight and may be nothing more than a coincidence. But the nature
of the age agreement, especially the apparent existence of the reversal of the growth curve in
both, may very well prove to be a more significant agreement. Such a reversal is so unusual in
growth curves that the chance of its being a coincidence would seem to be relatively small.
The fact remains, however, that all agreements of whatever sort may be coincidences, and
it IS difficult to obtain convincing proof of essential similarity from such evidence. The situation
is more or less reversed in the case of differences. It may require only one difference, if it is an
essential one, to establish a significant distinction. The difficulty in this case becomes, of course,
that of determining what is an essential difference.
Passing in review our seven criteria of kinship from the point of view of differences, then,
it may be said with a fair degree of confidence that non-prestige suggestibility differs from
hypnosis in at least four respects: (1) its correlation with susceptibility to hypnosis, (2) its relation
to scholastic intelligence, (3) the direction of the habituation effects, and (4) the tendency of the
hypnotic state to heighten susceptibility. Unfortunately the experimental evidence on the first two
of these is definitely weak; that upon the third is more conclusive. The fourth requires special
discussion.
Of the seven criteria of kinship, the one which comes closest to being an essential
difference is the heightening of suggestibility in the hypnotic state. Among the supposed
characteristics essential to hypnosis, such as amnesia, catalepsy, dissociation, rapport, sleep, etc.,
this only appears to remain as essential. If the non-prestige suggestibilities are not heightened
(384) in the hypnotic trance, then they must be different phenomena from the direct
prestige-suggestion forms which are known to be so heightened. This question, in the opinion of
the writer, is the central one in the present discussion. It is much to be regretted that the
experiments so far directed to its solution have, for some reason, failed to establish the usual
amount of certainty. Meanwhile, the writer is proceeding on the tentative assumption that the
final verdict will be negative, i.e., that non-prestige suggestion is essentially different
from-indeed, quite without significant relation to-hypnosis and direct prestige suggestion in
general. If this hypothesis could be firmly established it would. aid greatly in clarifying the
present lamentable confusion as to the fundamental nature of both hypnosis and suggestibility.
(385)

22. CH 14: INTERPRETATIONS: CLARK HULL: HYPNOSIS AND SUGGESTIBILITY:

APPLETON-CENTURY-CROFTS: NY: COPYRIGHT 1933: 1961
THE preceding pages have presented a considerable body of evidence concerning various
phases of hype nosis and suggestibility. Necessarily, much of this has been of a technical nature.
No doubt more than once the reader has grown impatient over the meticulous weighing of
evidence and the endless discussions concerning the adequacy of controls, the number of subjects
employed, the number of measurements made, the size of probable errors, and the scientific
reliability of the evidence in general. He has wished to know the gist of the thing without further
ado; he has desired a systematic framework upon which might be projected the otherwise largely
disconnected facts. This is not only a natural, but a legitimate, desire, and in this brief concluding
chapter the writer will attempt to satisfy it. The reader should understand, however, that
conciseness in such a procedure can be attained only by oversimplification, by ignoring
exceptions to prevailing tendencies, and frequently by dogmatizing where the evidence is too
complex for brief summary and the reasoning behind the conclusions is too intricate for brief
presentation.
It is obvious that the theory of human behavior is much too scantily developed, and that
the phenomena of suggestibility are too imperfectly known at the present time, for a complete and
ultimate theory of hypnosis to be possible. The evidence is sufficiently complete in certain
portions of the subject, however, to permit of limited generalizations. It is also sufficiently
complete on a number of controverted points to enable us to state with some confidence that the
supposed phenomena really are not characteristic of the hypnotic trance, and therefore need not
be embraced in a general theory of the nature of hypnosis as such. By this removal of
pseudodifficulties we may at least be opening the way to a possible theoretical development
which may be expected in the future. (387)
PSEUDO- DIFFICULTIES IN HYPNOTIC THEORY: One pseudo-difficulty which
has been a stumbling block in the way of certain writers is the belief that rapport is an inherent
and essential characteristic of hypnosis. Our experience with this phenomenon is wholly in
agreement with that of Braid, Bramwell, and Young, among others, in indicating that the
phenomenon called rapport is never encountered unless suggested to the subject either directly or
indirectly. This does not, of course, abolish the ultimate necessity of accounting for the
phenomenon, but it definitely simplifies the theoretical problem of the nature of hypnosis itself.
An adequate explanation of hypnotic anesthesia should also serve to account for the phenomenon
of rapport, since rapport appears to be nothing but a special form of selective anaesthesia.
A second pseudo-difficulty which has been the cause of considerable futile theoretical
effort is the belief that catalepsy is an essential and inherent characteristic of hypnosis. This error
has been particularly pernicious in its indirect effect of falsely linking hypnosis with the tonic
immobility which may be induced in many animals such as hens, rabbits, toads, lizards, etc., by
means of violent stimulation or of physical restraints. There is some reason to believe that the two
phenomena are effectively differentiated by their differences in response to practice: hypnosis and
prestige suggestion in general show markedly positive practice effects, whereas animal catalepsy
seems to show negative practice effects. Moreover, hypnosis and prestige suggestion appear to
originate in habits which must previously have been acquired, whereas the tendency of animals to
pass into a state of tonic immobility seems to be innate and unlearned. But probably the most
convincing evidence indicating differentiation between human hypnosis and the tonic immobility
of animals is the well-known fact that suggested catalepsy can be induced about as readily in the
waking condition as in the trance state. Thus catalepsy appears to be anything but peculiar to
hypnosis.
A third pseudo-difficulty which has presented a major obstacle to many hypnotic theorists
is the belief that hypnosis is a form of true sleep. Bass's experiment, in the opinion of the present
writer, has proved conclusively that hypnosis is not real physiological sleep and is in no
significant way related to it. Indeed, the fact that subjects during hypnosis (as indicated by
post-hypnotic amnesia, say) can perform the most complex processes of learning, computation,
and problemsolving should have prevented any such view from ever having been widely held.
People do not do such things when really asleep. IncidentaIly, the sleep experiments enable us to
clear up the mystery of the old distinction between the alert state of hypnosis and the so-called
lethargic state. The alert state is the true hypnotic condition, whereas the lethargic state is real
sleep of mild degree brought on by the suggested immobility, closed lids, etc., which are
favorable for the appearance of sleep under any circumstances.
A fourth pseudo-difficulty which has agitated writers in hypnotic theory (5, 86; 6, 89) is
the belief that subjects under hypnosis show a greatly heightened sensitivity to faint stimuli. We
have seen, on the contrary (p. 24Sff.), that the evidence for hypnotic hypersensitivity is negligible
but that the evidence for normal or even subnormal sensitivity is good. Similarly, much effort has
also been wasted in seeking to explain why hypnotic subjects judge time more accurately when
under hypnosis than when in the normal state. The simple facts of the case appear to be (p. 248)
that persons in the hypnotic state really do not judge time any better than when in the normal
condition. The outstanding positive result surviving from the experiments just aIluded to is the
fact that subjects when in the hypnotic state usually show all signs of believing implicitly that
they possess the heightened sensitivities which have been suggested to them. This, however, is a
phenomenon of quite a different order from that of hyperesthesia, and one which must find its
explanation along with other phenomena of hypnotic hypersuggestibility.
A fifth pseudo-difficulty which has complicated the problem of hypnotic theory has been
the belief that hypnosis is a pathological condition, presumably aIlied to hysteria. It must be
granted at the outset that hysteria, with its amnesias, anaesthesia, etc., probably is dependent to a
certain extent upon the same mechanisms as enable the corresponding phenomena to (389)
be evoked artificially in the hypnotic trance. It is even likely that a comparative study of the two
phenomena by modern experimental procedures would yield important light regarding the
psychoneuroses. However, the correlational studies which have been performed reinforce the
consensus of observational reports that persons susceptible to hypnosis are quite as normal and
quite as intelligent as are persons who are resistant to suggestion.
A sixth pseudo-difficulty in the development of a theory of hypnosis has been the belief
that it is a state of dissociation. We have fairly good experimental evidenceindicating that
hypnosis is a state of dissociation neither in the sense that persons in that state can carry on two
independent menta! processes more effectively than when in the non-trance state, nor in the sense
that persons susceptible to hypnosis can in the normal state carryon two independent mental
processes relatively more readily than can persons who are not susceptible to hypnosis. It is true,
however, that specific suggestions to that effect are able in hypnosis to make certain memories
inaccessible to voluntary recall, and to reduce to varying degrees the responsiveness of the
organism to painful stimuli. These are, in a certain real sense, dissociation phenomena, but by no
means such in the sense of a dissociation into two independent "minds," one conscious and the
other subconscious. These latter notions are probably survivals of an outworn metaphysics which
conceived the mind as a kind of disembodied spirit which was associated with a living body only
by reason of a kind of metaphysical coincidence.1 It seems fairly clear, however, that the
dissociations observed are not inherent and essential to the hypnotic state, but are always the
result either of direct or indirect suggestion. Accordingly, they must find their explanation in the
general theory of suggestion, rather than in the theory of hypnosis as such. (390)
HYPNOSIS IS A STATE OF RELATIVELY HEIGHTENED SUSCEPTIBILITY
TO PRESTIGE SUGGESTION: So much for what hypnosis is not. We must now face the
question as to what it is. The answer is simple: The only thing which seems to characterize
hypnosis as such and which gives any justification for the practice of calling it a "state" is its
generalized hypersuggestibility. The difference between the hypnotic state and the normal is,
therefore, a quantitative rather than a qualitative one. Despite the widespread and longstanding
belief to the contrary, the author is convinced that no phenomenon whatever can be produced in
hypnosis that cannot be produced to lesser degrees by suggestions given in the normal waking
condition.
This raises the further question of how great an increase in suggestibility takes place upon
entering the hypnotic trance. Theoretically an answer to this question is susceptible of
determination in connection with every process which is responsive to direct prestige suggestion
in any measurable way. No doubt the degree of hypersuggestibility will differ to a certain extent
from one reaction to another. The evidence at present available is based largely upon the duration
of continuous suggestion required to evoke a given amplitude of a variety of simple movements.
The general indication from this evidence is that suggestibility in the hypnotic trance averages
something like twice as great as in the normal condition. As further evidence accumulates from
measurements based on other processes, this figure quite certainly will be revised. As a
preliminary indication in this direction may be mentioned Williams's results on muscular fatigue,
where hypnotic suggestion yielded an amount of work only r6 per cent in excess of that produced
by waking suggestion. On the other hand, it is likely that amnesias and anaesthesias may be
produced with far greater readiness in hypnosis than in the waking state.
THE MERE SUSCEPTIBILITY TO PRESTIGE SUGGESTION IS NOT
HYPNOSIS: It is further to be noted that this hypnotic hypersuggestibility has a relative and not
an absolute significance. The essence of (391) hypnosis thus lies in the fact of change in
suggestibility. The evidence at present available indicates that of two people who show a distinct
heightening of suggestibility following the hypnotic procedure, one individual may be
considerably more responsive to suggestion before going into the trance than the other is
afterwards. Not only this, but persons differ greatly in the amount of hypersuggestibility induced
by a given technique. It is indeed probable that a certain small percentage of persons who appear
to go into the trance and who display such phenomena as post-hypnotic amnesia do not show any
overt motor hypersuggestibility whatever. It is even possible that occasional individuals will
show, when apparently in the trance, a diminution in suggestibility, at least for certain activities.
If a subject after submitting to the hypnotic procedure shows no genuine increase in
susceptibility to any suggestions whatever, there seems no point in calling him hypnotized,
regardless of how fully and readily he may respond to suggestions of lid-closure and other
superficial sleeping behavior. This amounts, of course, substantially to an arbitrary definition of
hypnosis. It is believed that the general adherence to such a definition might aid materially in the
clarification of certain ambiguities which still cling to the subject. This is not to say that other
influences such as fatigue, certain drugs, etc., may not also produce an increased susceptibility to
suggestion. It is to say, however, that if the subject has responded positively to the explicit
suggestions of the hypnotist the hypnotic technique usually does result in a state of generalized
hypersuggestibility (which itself has not been specifically suggested) as a kind of accessory or
excess effect. It is believed that this phenomenon is of sufficient significance to be made the basis
of at least a working definition.
Accordingly, the mere susceptibility to prestige suggestion, no matter in what degree, is
not hypnosis. Its essence lies in the experimental fact of a quantitative shift in the upward
direction which may result from the hypnotic procedure. So far as the writer can see, this
quantitative phenomenon alone remains of the once imposing aggregate known by the name of
hypnosis. But this undoubted fact is quite sufficient to give significance and value to the term.
The dual level of prestige suggestibility thereby indicated should be sufficient to justify the
retention of the term hypnosis for the upper level, and to prevent its application to the lower or
normal level as proposed by Jenness and, possibly, by Wells.
But why should the particular stimulations used in inducing the hypnotic trance produce
this remarkable effect? Certain experiments, such as that of Hull and Huse and, particularly, those
of Williams and that of Jenness, have seemed to indicate that hypersuggestibility is a kind of
spontaneous excess effect resulting from positive response to any direct suggestion, something
analogous to transfer of training. But the two Miami experiments clearly contradict this by
showing that heteroactive waking hypersuggestibility is not generated by responses to ordinary
waking suggestions. It is difficult to reconcile the results of the two groups of investigations. The
problem involved is of such central importance that the experiments should be repeated with
judicious variations to make certain whether or not some hidden defect of technique may not have
produced this seeming inconsistency, particularly between the study of Jenness and that of Hull,
Patten, and Switzer.
The study by Patten seems to have negated the hypothesis that the hypersuggestibility is
due to a special belief held by most subjects that a heightened suggestibility will supervene as
soon as the hypnotic technique has produced Iidcatalepsy. The only further untested hypothesis
which has presented itself in this connection is that the hypersuggestibility characteristic of the
hypnotic trance may be due to the general relaxation of muscular tension which ordinarily is
suggested both directly and indirectly by the hypnotic technique. Meanwhile we have no
alternative but to reserve judgment regarding this fundamental matter until the laboratory
produces more adequate evidence. No question in the whole subject of hypnosis and
suggestibility is in such urgent need of critical experimentation.
(393)
PRESTIGE SUGGESTION CONCEIVED AS A HABIT PHENOMENON: There
remains the still more basic problem of the nature of suggestion itself. In the consideration of this
question it must be observed, in the first place, that we are not here concerned with any forms of
behavior not essentially related to hypnosis. On this ground we discard from further consideration
the nonprestige forms of suggestion such as Binet's progressive weights and progressive lines. A
careful survey of the technical phenomenology of these forms of suggestion has convinced the
author that a definition of suggestion from the point of view of hypnosis would err in attempting
at the same time to envisage either these or any other forms of non-prestige suggestion. This
greatly simplifies the problem of arriving at the essential nature of the form of suggestion which
is actually heightened during hypnosis, and which therefore is relevant to our present problem.
The experimental evidence indicates that whatever else prestige suggestion may be, it at
least is a habit phenomenon. That both hypnosis and waking suggestion manifest the classical
behavior of habituation in remarkable detail has been shown by Krueger, and by Patten, Switzer,
and Hull. This general thesis is supported indirectly by Messerschmidt's results, which show that
children in early childhood are not as suggestible as they are at seven or eight years of age. This
latter fact is interpreted as meaning that learning the ordinary habitual responses to language
stimuli is an essential component of acquiring the tendency called suggestibility.
It may be noted that the present view is closely akin to that held by Bernheim, when a
moderate allowance is made for the difference between the prevailing psychologies of that day
and the present. Bernheim reduced suggestion substantially to the association of ideas; the view
here presented reduces it to the strictly physical basis of the associations between stimuli and
responses, ideas becoming purely physical symbolic acts. (394)
SUGGESTION DIFFERENTIATED FROM VOLITION ON THE HABIT
HYPOTHESIS: From the introspective point of view, one of the most characteristic differences
between actions performed through the influence of suggestion and ordinary acts is that the latter
are usually felt somehow to be willed, whereas the former acts are felt not to be willed. This
being the case, it is obvious that no wholly satisfactory account of suggestion can be given until
there is available an adequate psychology of volition. Since no satisfactory systematic view of
voluntary action has yet been formulated, the following is offered as a skeleton working
hypothesis upon which to project a systematic account of hypnosis and suggestibility.
The basic assumption of the present hypothesis is that there exist two fairly distinct levels
of habitual reactions, an upper or symbolic level, and a lower, non-symbolic or instrumental
level. Symbolic acts may perhaps best be described as purestimulus acts (IO, II), acts which
function purely as stimuli to evoke other acts. The most common form of symbolic or
pure-stimulus act is speech.
The distinction between symbolic and instrumental acts may be illustrated by the
following example. Suppose a ma.n is standing where one needs to pass. Theoretically he may be
removed either by an instrumental or a symbolic act on our part. The instrumental procedure
would involve lifting him bodily out of the way just as if he were some inanimate object. The
symbolic procedure would be to execute a few speech movements, such as saying, "May I pass,
please?" and the man, if he has previously acquired the usual language and related habits, will
ordinarily step aside at once. But speech need not necessarily be symbolic. A non-symbolic or
instrumental use of speech would be exemplified by a mechanic speaking the word "See!"
repeatedly into a laboratory voice key to test the accuracy of its adjustment. Most of the ordinary
actions of everyday life, such as chopping wood, repairing ma-: chines, stoking furnaces, cleaning
streets, etc., are instrumental or non-symbolic activities. So much for the basic habituation
hypothesis.
 The symbolic act (speech) may be executed by one person (395) and its instrumental
sequel may be executed by another; this is believed to be the essential mechanism mediating
heterosuggestion. In such cases the communication usually takes place through the sense of
hearing or vision. On the other hand, the symbolic act and its instrumental sequel may be
executed in a continuous sequence by the same individual; this is believed to be the essential
mechanism mediating acts ordinarily called voluntary. In this latter case the symbolism is much
more varied and the processes involved may be extremely complex. For our present purposes,
however, we may consider only the form known as subvocal speech, which is believed to be the
chief mechanism of volition. Here the symbolic act stimulates the proprioceptive sense-organs
within the muscles of the symbolic reaction system and the neural impulses thus released evoke
the consequent instrumental or voluntary act.
Suggestions Differentiated from Commands: Responses to commands are conceived as
being either voluntary or involuntary, though it would seem that frequently the two mechanisms
may be simultaneously active. In the case of an instant response to a sharp command such as
might be given by an officer to a soldier, the reaction would presumably be non-voluntary-a
simple habitual reaction without involvement of the soldier's symbolic processes. A truly
voluntary response to a command, on the other hand, would be a case in which the command first
evoked an intraorganic symbolic sequence on the part of the subject, which finally led to a
subvocal symbolic reaction substantially equivalent to the command emanating from the other
person, which, in turn, evoked the reaction commanded. It is not inconceivable that suggestioni:
might also set going a symbolic sequence in the subject similar to that caused by the command
just considered, in which event we should probably have a case of simulation.
Considerable color is given to the general hypothesis here put forward by an observation
made by Williams, while administering continuous forward postural suggestions to strongly
negativistic catatonic dementia precox patients. Under these conditions he often observed the
subjects (396) whispering to themselves continuously the words, "Backward, backward,
backward," etc. We seem to have in this remarkable observation a case of voluntary action where,
probably because of the distortions of disease, the mechar..ism of volition has been laid bare. It is
significant that under such circumstances the patients, instead of swaying forward in response to
Williams's suggestions, actually swayed backward presumably in response to their own (counter)
suggestions. According to the present hypothesis, those autosuggestions were actually
exaggerated volitions, and the resulting backward movements were truly voluntary.
In contrast to the above, a true suggestion response is one in which the subject's own
symbolic processes, instead of becoming active either in facilitating or in resisting the tendency to
action naturally arising from the experimenter's words, remain passive so far as the particular act
suggested is concerned. It seems reasonable to assume that this passivity is facilitated by the
suggestions usually given for the subject to relax and not to think of anything but sleep. This
withdrawal of the subject's symbolic activities would naturally leave his muscles relatively
susceptible to the symbolic stimulation emanating continuously from the experimenter.
Apparently some individuals are unable to withdraw to an appreciable extent the influence of
their own symbolic processes; these individuals would be classed as insusceptible to
heterosuggestion. At the beginning it is rather difficult for most subjects to do this, so that the
first responses to suggestion are both slow in taking place and slight in degree. As practice
continues, however, skill in this inhibition of symbolic interference is gradually acquired, which
is conceived as being the major factor in the practice effects so characteristic both of hypnosis
and of waking suggestion.
As corroborative evidence leading to the same conclusion, it may be pointed out that the
drowsiness reported by the subjects of Baumgartner and by those of Hull, Patten, and Switzer
following waking suggestions containing no hint of hypnosis or sleep was quite possibly due to
this extensive lapsing of the symbolic activities. In this same connection it may also be recalled
that Scott's subjects in the trance upon (397) the whole acquired the conditioned reflex more
readily than those who were conditioned in the non-trance state. The theoretical relevancy of this
last observation lies in the assumption that conditioning is interfered with by the activity of
related symbolic processes which, according to the present hypothesis, is reduced to a minimum
in the hypnotic state. As further evidence pointing in the same direction it may be noted that
several investigators of the conditioned reflex have sought to facilitate conditioning by otherwise
employing the symbolic processes through having the subjects engage in reading, adding, and
other irrelevant symbolic activity.
IDEOMOTOR ACTION AND MONOIDEISM CONCEIVED AS HABIT
PHENOMENA: According to Bramwell, Braid in his later years held that the various
phenomena of hypnosis resulted entirely from dominant ideas which "reacted on the body and
produced their physical equivalent." A number of other writers in this field, notably Bernheim,
have held a substantially similar view. That the concrete facts of normal psychology support this
hypothesis has been shown in considerable detail in the preceding pages (p. 41 ff.). A continuous
stimulation by words associated with a particular act will bring about the act, whether these
words are those of the subject himself or of some other person. Perhaps the most critical evidence
of all indicating the reality of ideomotor action is the fact (p. 44) that when a subject merely
observes an action, he tends automatically to execute it.
The present hypothesis recognizes very fully the role played by ideomotor action in the
field of hypnosis and suggestibility. It disagrees sharply with the view held by Braid and
Bernheim, however, who believed with most psychologists of their respective periods that an idea
is some non-physical entity which is evoked in the mind of the subject by the experimenter's
words, and which somehow in the case of ideomotor action is able to muster the physical energy
required to evoke the action suggested. According to the present hypothesis, the physical
substance of an idea is a symbolic or pure-stimulus act. The proprioceptive stimuli arising from
such acts, usually spoken words, are assumed when combined in certain patterns to have (398)
acquired during the previous history of the subject, through the process of association or
conditioning, the capacity to evoke the reactions of which they are the names.
In much of the literature on the subject of monoideism the concept of ideomotor action is
associated with that of attention. Indeed, the expression "dominant ideas" contains the essence of
the concept of monoideism. According to the present hy~ pothesis, monoideism means that the
(proprioceptive) stimulus emanating from a single idea (pure-stimulus act) plays continuously
upon the neuro-muscular equipment of the organism; this stimulation evokes the act of which it is
the "mental equivalent." Attention in such a situation means merely that the organism reacts to a
single stimulus without substantial change (disturbance from the intrusion of stimuli associated
with conflicting action patterns) for an appreciable length of time. In the case of heterosuggestion
and hypnosis, the conditions of attention or monoideism would be fulfilled when, either through
the quiescence or the withdrawal of the stimulation arising from the subject's own symbolic
processes (ideas), proprioceptive intrusions would be reduced. This removal of sensory
competition would give the continuous stimulation emanating from the symbolic processes
(ideas) of the experimenter a kind of right of way to the control of the subject's movements.
Thus we find in essential functional agreement the points of view of ideomotor action,
monoideism, and attention when they are viewed as habit phenomena.
SUGGESTIVE CONTROL OVER CERTAIN N ONVOLUNTARY PROCESSES:
One of the phenomena of hypnosis which has impressed workers profoundly is the fact that
suggestion, particularly in the hypnotic trance, is sometimes able to evoke phenomena which the
subject is presumed not to be able to call forth voluntarily (p. 272 ff.). The seemingly paradoxical
nature of such results lies largely in their conflict with the usual but unwarranted assumption that
voluntary control of physical processes is the maximum control possible. As a matter of fact,
large numbers of physiological processes which take place all the time are normally quite beyond
voluntary control. Familiar (399) exemplifications of this principle are the processes of glandular
secretion, water metabolism and excretion, the galvanic skin reaction, and pupillary dilation and
constriction. The paradox is still further attenuated when it is recalled that such nonvoluntary
processes readily become conditioned so that stimuli previously unable to evoke them will later
do so with ease. In a word, conditioned stimuli regularly evoke reactions not possible to
voluntary effort.
The method by which the present habituation hypothesis explains the long-standing
paradox mentioned above is not difficult to understand. In the process of suggestion a series of
verbal stimuli (ideas) from another person are assumed to have become conditioned to the
reaction in question during the previous life of the subject. He mayor may not possess the
necessary conditioned (proprioceptive) excitatory tendency to evoke the same reactions by means
of his own symbolic processes. It is likely that in most cases he really does possess them, but
having no occasion for the self-evocation of such reactions as gastric secretion, pupillary
constriction, etc., he has never made the attempt and therefore does not know of their existence.
There is strong reason to believe, however, even in case the chance experiences of life have not
provided the subject with such proprioceptive excitatory tendencies, that they could easily be
acquired. This is to say that all reactions susceptible to control by suggestion are ultimately
capable of being controlled voluntarily by suitable conditioning to or association with
proprioceptive stimuli arising from one's own symbolic activities.
This whole matter is illuminated by a brilliant experiment recently reported by Hudgins.
The experimenter repeatedly pronounced the word "Contract [" shortly before a strong light was
thrown into the subject's eyes. The light of course produced a marked contraction of the pupil.
After a good many such paired stimulations it was found that the experimenter could cause a
contraction in the subject's pupil by merely speaking the word "Contract." The reaction thus
produced duplicates experimentally the characteristics of a reaction which is susceptible of
heterosuggestion yet incapable of voluntary control.
But the sequel to this experiment is still more illuminating. (400) The procedure described
above was followed by a number of paired stimulations in which the subject himself spoke the
word "Contract!" in conjunction with the strong light being flashed into his eyes. After a number
of such paired stimulations the subject was found to have acquired a control over his own iris
movements such that he could now produce the contraction either by speaking the word aloud or
by merely "thinking" the word. This latter fact, according to the present hypothesis, would mean
that "thinking" the word involves a subvocal speech movement which provided adequate
proprioceptive stimulation for the evocation of the act without the aid of the ordinary auditory
component.
 HYPNOTIC HALLUCINATIONS INTERPRETED AS HABIT PHENOMENA:
Hypnotic hallucinations also find a ready interpruation when regarded as habit phenomena. This
principle may be illustrated by an observation reported by Scott. It will be remembered that Scott
conditioned subjects in the trance state by pairing a buzzer stimulus with a shock applied to the
tips of the subject's fingers. Several of these (totally amnesic) subjects, when placed in the
apparatus in the normal condition, made vigorous conditioned retraction movements when
stimulated by the buzzer alone, and later reported the feeling of having received an electric shock
on the fingers. At least one of his subjects insisted that he had received a bona fide shock, though
the experimenter was positive, from testing the apparatus, that no shock could have been
received. The interpretation of these striking observations is that experiences in essence are really
actions. In most cases the activity is believed to be so slight that it is not ordinarily observable by
another person, and is rarely observed by the subject himself as movement.
Let us suppose that stimulus SA naturally evokes the reaction RA and stimulus SB
naturally evokes reaction RD' Suppose, further, that stimulus SD has acquired by the conditioning
process the power to evoke the reaction RA, previously evoked by SA alone. According to the
principles just laid down, the subject would be likely, when reaction RA took place, to report
(401) that he had received stimulus SA, even though the reaction may in reality have been evoked
by SB' Such a report on the part of a subject would of course satisfy the conditions ordinarily
accepted as evidence of an hallucination. That such hallucinations are not more common is
probably due to the fact that conditioned reactions as a rule differ more or less radically from
unconditioned reactions, both qualitatively and with respect to their intensity. Moreover, such
conditioned reactions would normally be accompanied by the reaction RB, which should serve as
a sign to the subject enabling him ultimately to differentiate the two types of origin of RA.
On the above hypothesis, it would seem that in the hypnotic condition, owing to the
relative inactivity of the subject's own symbolic processes, symbolic stimuli from another person
would be able to arouse far more complete duplications of reactions to unconditioned stimuli than
would normally take place. That this is a fact seems to be borne out by experimental
observations. The perfection of the duplication of the behavioral pattern characteristic of the real
stimulus should favor the error which constitutes the hallucination. On the other hand, it is also
possible that in such cases the symbolic stimulation from the experimenter may sometimes
actually get control not only of the subject's non-symbolic activities but of his symbolic processes
as well. In such an event the subject would report the hallucination as reality, not as a reaction to
the proprioceptive stimulations resulting from the gross reaction to the suggestions but as the
direct result of the suggestion itself.
CONCLUSION: Thus ends an attempt at a scientific account of the major phenomena of
hypnosis and suggestibility. Many similar attempts have been made in the past, and many more
will be made in the future. Too many of the works on the subject in the past have fallen short of
the scientific ideal. Doubtless many things have contributed to this weakness, but surely a major
factor must be the inherent difficulty of the problems involved, the fundamental elusiveness of
the phenomena, and the consequent subtlety necessary in the experimental controls. These
difficulties are so great that to enter seriously on a program of investigation in this field is a little
like tempting fate; it is almost to court scientific disaster. Small wonder that orthodox scientists
have usually avoided the subject [ Yet each generation may be expected in the future, as in the
past, to produce a very few rash souls who will not only risk the dangers of making scientific
errors but will also have the courage to brave the semi-superstitious fears of the general public
and the uneasy suspicions of their orthodox scientific brethren. It is to them that the present work
is really addressed.
We have done our best to see the problems with a fresh eye, to avoid the omnipresent
experimental pitfalls, to devise really adequate experimental controls, and to be docile in the face
of facts. The history of the subject teaches us to have no illusions in regard to the success of such
efforts. No worker can wholly escape the ideology of his time. Many of the things here gravely
put down as securely established will be rejected with reason by investigators of the future, just as
we have found reason to reject more or less of what many earlier workers have regarded as
certain. This is both inevitable and proper, for science is to a certain extent a trial-and-error
process. (403)