"—o. . “O---"Wr—v— W

. ' W — -—————————— ______
f ,e '7 'w.-m-'°0w-”qu~~M-¢~wowx-j-ygflu— -- ~.——..¢-a—"W—ar ., , H- v - H- «.‘*o.~qva¢ ‘ «to .c. -¢H~-o-o ‘ -0. _- o ‘ . V ' . ‘
- . ‘7 a
o
I

 

-
.
.

I - CRIIICAL FACTORS IN THE
I ARTIFICIAL INDUCTION OF CONFLICTS:
I ~ A HYPNOTIC PARADIGM FOR REPRESSION

 

Thesis for the Degree Of M. A..
MICHIGAN STATE UNIVERSITY
ALICE MADELINE WOLFE. —

 

 

2' {111 Q -. 'L'c
O

fie"; U Alix-{15‘}; } 7

(I

. "I "* “View”, m-

 

 

 

ABSTRACT

CRITICAL FACTORS IN THE ARTIFICIAL INDUCTION OF CONFLICTS:
A HYPNOTIC PARADIGM FOR BEPBESSION

By

Alice Madeline Wolfe

Previous research into the connection between repression
and psychOpathology (Reyher, 1958, 1962, 1963, 1967; Perkins,
1965; Sommershield, 1969) has suggested that an artificial
conflict induced under hypnosis can activate an individual%
defensive structure with the resulting appearance of char-
acteristic symptomatology. This study was designed to
examine the contribution of suggested amnesia and a destruc-
tive impulse to the symptomatic outcome. Since a previous
study (veenstra, 1969) which had excluded the destructive
impulse from its design failed to produce symptomatology, it
was hypothesized that the destructive impulse was the
critical variable for activation of psychopathology. The
amnesia variable was added because an earlier study (Bobbitt,
1958) had made its contribution to the activation of the
conflict unclear.

Each of sixteen female S's served in two of the four
experimental conditions. While the presence of the destruc-
tive impulse was varied for each 8, 8'8 were randomly placed
in either an amnesia or no amnesia condition. Two E's of

opposite sex selected their S's at a group hypnosis session

and then saw each 8 for three individual sessions. After
the first hypnotic induction at each of these sessions 8
was asked to free associate to each of fifteen words on a
word list. Under hypnosis at each of the two experimental
sessions, 8 experienced the anger-inducing paramnesia, was
told that certain words (1/3 of the word list) would revive
the anger posthypnotically, and then was given instructions
for the amnesia and/or destructive impulse corresponding to
the experimental condition. During the 35 second free
association period, G.S.R. measurements were taken. Later
the 6.8.3. data was analyzed and the free associations were
rated using Beyher's (1967) symptom scale.

This procedure failed to produce the type of florid
symptomatology or G.S.B. activation which had been observed
in previous studies. Neither of the experimental variables
had an effect significantly greater than chance. There was,
however, a significant experimenter effect even before the
experimental procedures were administered and a significant
difference between the pre- and post-experiment data.
Analysis of these outcomes suggests that the absence of
significant symptomatology in this and Veenstra's (1969)
study is the outcome of the use of a paramnesia that arouses
anger without simultaneously stimulating guilt. The fact
that S's were not distracted from the focus of E's interest
may have also prevented the occurrence of florid reactions.

Finally, the significant experimenter effect points out the

previously unexplored, but highly important, area of
experimenter differences in expectations, sex, and person-

ality characteristics.

CRITICAL FACTORS IN THE ARTIFICIAL INDUCTION OF CONFLICTS:
A HYPNOTIC PARADIGM FOR REPRESSION

BY

Alice Madeline Wolfe

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF ARTS

Department of Psychology

To:
Dr. Thomas Leider
Whose faith and patience made this and all my future work
Possible.

ii

ACKNOWLEDGMENTS

I wish to thank my thesis chairman, Dr. Joseph Beyher,
for his ideas, assistance, and encouragement throughout the
planning, execution, and writing of this study. To Aaron
Karnilow, my co-experimenter, I shall be forever grateful
for the work, cOOperation, and tolerance which made the study
possible. I also would like to express gratitude to
Dr. Paul Bakan, Dr. Mary Leichty, and Dr. Albert Rabin for
serving on my committee.

I want to give special thanks to Karen Iverson for her
work in rating the protocols. Finally, I would like to
thank Robert Gurney, who not only develOped our analysis of

variance but also endured my many questions on its application.

111

TABLE OF

List of Tables . . . . . .
List of Appendices . . . .
Introduction . . . . . . .
Method . . . . . . . . . .
Subjects . . . . . .
Materials . . . . . .
Procedure . . . . . .
Measures . . . . . .
Results . . . . . . . . .
Data Description . .
Data Analysis . . .
Discussion . . . . . . . .
Previous Findings . .
Procedural Elements .
Experimenter Effects
Word Type Differences
References . . . . . . . .

Appendices . . . . . . . .

iv

CONTENTS

Page

.............13
. . . . . . . . . . . . . l3
. . . . . . . . . . . . . 13
. . . . . . . . . . . . . 15
. . . . . . . . . . . . . 23
. . . . . . . . . . . . . 2h
. . . . . . . . . . . . . 24
. . . . . . . . . . . . . 26
. . . . . . . . . . . . . 38
. . . . . . . . . . . . . 39
. . . . . . . . . . . . . b5
. . . . . . . . . . . . . #9

oooooeooooo-..52

Table
l.

G.S.R. data - 2 nested in 2 X 2 X 3

LIST OF TABLES

analysis of variance

Verbal data - 2 nested in

analysis of variance

Verbal data - baserate vs
2 X 2 X 3 analysis of variance

Verbal data - baserate 2 x 3 analysis

of variance . . .

R (repression) as computed using Reyher's

formula .

2 X 2 X 3

condition I

Page

28

29

32

33

36

LIST OF APPENDICES

Appendix Page
A Repression computation formula . . . . . . . 57
B Reyher's symptom classification scale . . . . 59

vi

INTRODUCTION

Earl; Research on Posthypnotic Conflict

Although the connection between repression and psycho-
pathology, particularly psychosomatic symptoms, is well
accepted and documented by clinicians, little research has
been done along these lines.

Luria (1932) was a pioneer in the use of hypnosis to
simulate unconscious conflicts in normal subjects. Working
in Pavlov's laboratory, he had his subjects depress a lever
each time they reSponded to a word from a word association
list. Luria hypothesized that the pressing response would
become conditioned to any associations to individual words
on the list, whether spoken or unapoken. While his subjects
were in a hypnotic state, Luria suggested a reproachable act
to them involving a serious misdeed on their part and a fear
of its discovery. In reaction to the words connected with
the induced experience, these subjects exhibited disturb-
ances in both their associations and pressing behavior.
Huston, Shakow, and Erickson (1934) replicated Luria's
original research, reporting the same type of minor dis-
turbances in sleep, appetite, and affect on the night and

day following implantation of the suggestions.

Counts and Mensh (1950) tested the effects of posthypnotic
anger aroused through the implantation of a paramnesia (an
artificial experience) and directed towards E1, who
administered the Rorschach before and after the induction of
the paramnesia. The S's were interviewed by another experi-
menter, E2, before and after the implantation of the par-
amnesia with its associated posthypnotic anger towards El.
Although the interviews with E2 revealed S's struggle with
anger, the S's Rorschach scores remained essentially unchanged.
The single protocol presented in the literature reveals
symptomatic reactions by S.

Bobbitt (1958) examined the Freudian theory of repres-
sion by equating degree of suggested amnesia with degree of
repression. Degree of suggested amnesia for an induced
paramnesia and the feelings it arouses was varied in the
different conditions, with each S serving under all condi-
tions. The results indicated that suggested complete amnesia
for the paramnesia and its associated feelings or suggested
complete awareness produced little or no disturbance.
Significant disturbances of affect did, however, occur when
the S's were only partially aware of the induced conflict
and its associated feelings. Simulating S's were not
included in the design.

Wolberg (l9#7) observed psychopathological reactions in
two S's as a result of a posthypnotic conflict. He gave
these S's the posthypnotic suggestion of an "irresistable "
urge to eat a candy bar that was on his desk. Concurrently

he told them that they would know that it is very wrong and

impolite to eat candy that does not belong to oneself. One
8 avoided looking at the chocolate, felt thirsty, dizzy, and
faint, broke out into a cold sweat, started to tremble, and
experienced a negative hallucination for the candy bar.

The other S ate the candy but then developed nausea, stomach
pains, and finally vomited up the candy. These informal
observations suggest that the apparent pathogenecity of the
posthypnotic conflict is related to impulses which become
anxiety-producing when activated posthypnotically.

Eisenbud (1937) was able to exacerbate a patient's psycho-
pathology by implanting conflicts parallel to the patient's
natural conflicts and then discussing the conflicts and the
symptoms they produced in psychotherapy.

Conclusions relevant to psychOpathology drawn from the
preceding studies are restricted by two limitations in
experimental design: (1) a method of assessing the demand
characteristics of the experimental design is unknown
(Orne, 1959; Reyher, 1967); (2) there is no way to separate
the effects of direct suggestion from the spontaneous
behavior produced by the hypnotically induced process
(Reyher, 1962, 1963, 1967). All the paramnesias included
suggestions as to how S should react to the paramnesia
or hypnotically induced process.

To assess the demand characteristics of the experi-
mental design Orne (1962) recommends using simulating,

insusceptible S's with E blind as to whether S is susceptible

or insusceptible. Reyher, however, advises that using
susceptible S's eliminates the possibility that distinguish-
ing personality characteristics of insusceptible, simulating
S's are correlated with the dependent variable.

Reyher recommends that no suggestions which specify S's
behavior should be included; the effects of direct suggestion
can then more easily be sorted out from the spontaneous
effects of the induced process. If experiments using par-
amnesias are to parallel normal (non-hypnotic) functioning,
the induced process must spontaneously produce S's behavior.
Reyher (1962, 1963) has included both of these features in
his paradigm for determining the clinical relevance of
hypnotically induced psychopathology:

"First, the induced process must in no way include

cues as to how the eXperimenter expects the

subject to respond in any other respect . . .

Second, the induced process must produce other

processes and behavior; that is, it must be

response-producing.

Third, some of these responses must satisfy the

defining criteria for inclusion in some system of

psychOpathology.

Finally , . . . some of the subjects must be asked

by a co-experimenter, unknown to the experimenter,

to fake hypnosis in order to determine the demand

characteristics of the research." (1962, 3&8)

Using a paramnesia generating intense posthypnotic
anger associated with a destructive impulse, Reyher (1963,
1967) observed that many S's spontaneously repressed the

impulses but develOped a wide variety of symptoms instead.

These symptoms, which occurred in 87% (N=23) of ”normal" S's,

included headaches, skin disturbances, nausea, tics, tremors,
anxiety, paranoid reactions, and disturbances of affect.
Simulating S's neither exhibited symptoms nor showed the same
pronounced G.S.R. activation. Under hypnosis the 3's were
told that certain words relating to the paramnesia were to
serve as cues for a sudden experience of rage and the
execution of the destructive impulse after they awakened.
When these words were later presented tachistoscopically,
a posthypnotic conflict was produced. An index of repression
based upon S's verbalized awareness of the impulses was
(1) inversely related to the frequency of symptoms, and
(2) directly related to the proportion of somatic symptoms.
Reyher hypothesized that the sudden presentation of words
associated with the conflict produces a temporary disequili-
brium between repression and the anxiety-eliciting impulses.
As a result the individual's own idiosyncratic defenses are
brought into play, resulting in the manifestation of symp-
toms and pronounced G.S.R. activation until either the
destructive impulse is acted upon or repression is restored.
Subsequent research from Reyher's laboratory has shed
light on the four variables that he feels are critical.
These are: (a) sudden surge of overwhelming rage, (b)
towards an authority figure, (c) coupled with an overwhelming
impulse to destroy property belonging to said authority
figure, (d) activated suddenly by a posthypnotic cue
(conflict word).

Moore (196#) reduced "a" to ”almost overwhelming" instead
of "overwhelming" anger and aggression and substituted the
interpersonal relationship with E (an interview) as the
posthypnotic cue for anger instead of word recognition (d).
He found that high achievers experienced more repression of
the anger-aggression than did low achievers. Although the
incidence of psychOpathology was relatively less than what
Reyher had reported, it was still substantial (28%).

Perkins (1965) replicated Reyher's procedures and
predicted that the TAT protocols of good repressers (S's
who did not act upon the destructive impulse) would differ
from poor repressers (S's who acted upon the destructive
impulse). Pine's scoring system was used to measure (1)
degree of drive representation, (2) integration of drive, and
(3) socialization of drive. The two groups differed sig-
nificantly on all three criterion measures. Essentially
the same results as Reyher in regard to incidence of psycho-
pathology and its relationship to degree of repression were
also reported. A control group of simulating susceptible
S's was relatively free of symptoms and showed markedly
different patterns of G.S.R. activity.

Basch (1968) using a replicated design confirmed the
relationship between degree of repression and frequency of
symptoms utilizing non-hypnotic methods (the Byrne Represser-
Sensitizer Scale). She rated personality inventory items

in terms of degree of drive representation and correlated

this with frequency of reported physical symptoms on a
physical health questionnaire. Both of her correlations of
.43 and .57, reapectively, were significant at the .01 level.
Sommershield (1969) replicated Reyher's procedures,
including a sexual paramnesia with oedipal characteristics
in addition to the anger paramnesia. Reyher's previous
findings were confirmed. Furthermore, the sexual-oedipal
paramnesia is apparently more pathogenic than the anger
paramnesia: (1) it produced greater spontaneous repression
of the posthypnotic impulses as demonstrated by fewer ver-
balizations of impulse awareness, and (2) it produced more
symptoms. Once again a separate group of simulating sus-
ceptible subjects differed in terms of symptoms and G.S.R.
Veenstra (1969) has identified 13 variables in Reyher's
procedures: (1) Hypnosis, (2) age regression, (3) paramnesia,
(#) hypnotic arousal of anger, (5) hypnotic arousal of
conflicting emotion, (6) anger associated under hypnosis
with critical words, (7) anger directed towards an authority
figure in the experimental situation, (8) suggested loss of
control on presentation of conflict words. (9) posthypnotic
arousal of anger, (10) posthypnotic suggestion of a destruc-
tive impulse, (11) induced amnesia, (12) tachistoscopic or
sudden presentation of conflict words, and (13) accepting,
non-directive experimenter attitude. Reyher believes that
anger directed towards an authority figure in the experimental

situation (7), posthypnotic arousal of anger (9), posthypnotic

suggestion of a destructive impulse (10), and tachistoscopic
or sudden presentation of conflict words (12) are the
variables critical to the production of psychopathology.

In his own research Veenstra (1969) eliminated age
regression (2), hypnotic arousal of conflicting emotion (5),
anger directed towards an authority figure in the experi-
mental situation (7), and posthypnotic suggestion of a
destructive impulse (10). This reduced design produced
virtually no symptomatology as measured by Reyher's symptom
checklist (see Appendix B). Unlike the subjects in
Perkins', Reyher's, and Sommershield's experiments, all but
one of these S's acknowledged feelings of anger. The only
significant disturbance found by Veenstra was in response
latency, and he furthermore discovered that direct suggestion
was just as effective as the paramnesia in producing this
change. It seems likely, therefore, that the critical
variable or variables were among those excluded.

Although Veenstra's reduced design, which included the
suggested amnesia, produced no psychopathology, Bobbitt's
(1958) study indicates that suggested amnesia may also be
a critical variable. Bobbitt's design differed from Reyher's
in that her paramnesia involved guilt and anxiety rather
than anger and in that she examined physiological measures
and a conditioned light response rather than verbalizations
of impulse awareness and of symptoms. She also used varying

levels of suggested amnesia for the paramnesia and its

associated feelings. Reyher, it should be remembered, sug-
gested amnesia only for the paramnesia, not for the emotions
attached to it. Nonetheless, Bobbitt's study yielded a strong
curvilinear relationship between the extent of suggested
amnesia (total, partial, and no amnesia) and the amount of
disturbance, with the greatest disturbance occuring when there
was partial amnesia.

A recent study by Sheehan (1969) sought to assess Reyher's
(1961) findings about posthypnotic conflict using simulating
insusceptible subjects as controls instead of using simulating
susceptible subjects as controls as Reyher did. Unlike
Reyher, Sheehan gave his control S's Specific instructions as
to how S's should respond to the paramnesia. Since differences
in hypnotic susceptibility may be the result of differences
in personality, the results of an experiment comparing the
effect of an independent variable on susceptible and insuscept-
ible subjects may be confounded by the differences in person-
alities. Sheehan's controls were inferior to Reyher's in
this reSpect. Sheehan's experiment also did not include what
Reyher considers the critical variables for the production of
psychopathology: (a) a sudden surge of overwhelming rage,

(b) towards an authority figure, (c) coupled with an over-
whelming impulse to destroy property belonging to said
authority figure. Instead, a reproachable act was used for
the paramnesia. In view of these differences from Reyher's
design, it is not surprizing that Sheehan found no significant

differences between the experimental and control groups.

10

In Reyher's procedures an amnesia for the paramnesia
is suggested to guard against the breakdown of the post-
hypnotic conflict. However, it is important to keep in mind
that the suggested amnesia for the paramnesia (11) does not
include amnesia for the posthypnotic anger (9) and destructive
impulse (10). To the contrary, the anger and destructive
impulse are strongly suggested to occur posthypnotically.

Sheehan (1969) and Veenstra (1969) present the first
evidence that an amnesia for the paramnesia may not be a
critical variable in the production of pathogenic post-
hypnotic conflict. Veenstra's research is of particular
interest because no symptoms were reported or observed when
anger was not tied to a destructive impulse. Apparently anger
is less anxiety-producing for S when it is not tied to a
destructive impulse. To act upon the destructive impulse
would increase anxiety to a prohibitive degree because
undesirable intrapsychic (anxiety and guilt) and inter-
personal consequences (rejection and retaliation) would
result.

The destructive impulse (lO) and amnesia (11) were the
focus of the present investigation. Veenstra's research is of
particular systematic importance because all but one 8
verbalized being angry. There apparently was relatively
little repression of the posthypnotic anger, and 8's did not
manifest psychopathology as measured by the symptom rating
scale (see Appendix B). These results are in marked contrast
to previous work on Reyher's procedures and strongly suggest

that the destructive impulse is the critical variable in

11

producing repression and psychopathology. Anger alone is not
sufficient, although it might be necessary.

The suggested destructive impulses correSpond closely

to the libidinal impulses in the psychoanalytic construct

of repression. According to psychoanalytic theory the impulse
should activate the individual's defenses and induce a spon-
taneous repression of the impulse without the help of the
suggested amnesia. The significance of the suggested amnesia
is not clear at this point. Bobbitt found that the extent of
suggested amnesia significantly influenced the extent of
disturbance, but Veenstra observed that the amnesia did not
differentially affect the production of psychopathology.
Because of this contradiction, we decided to further investi-
gate the effect of suggested amnesia.

Four conditions exhaust the ways these two variables

can be manipulated: (1) both the amnesia for the paramnesia
and the destructive impulse are suggested, (2) no amnesia is
suggested but the destructive impulse is suggested, (3)
amnesia for the paramnesia is suggested but the destructive
impulse is not, and (4) neither the amnesia for the paramnesia
nor the destructive impulse are suggested. The conditions

all include posthypnotic suggestion of anger.

The hypotheses to be tested were:

1. The rated frequency of symptoms for the conditions
including the destructive impulse differ significantly
from the frequency of symptoms for the conditions
without the destructive impulse.

2. The central nervous system activation as measured

12

by galvanic skin reSponse for the conditions
including the destructive impulse differs signifi-
cantly from the C.N.S. activation for the conditions

without the destructive impulse.

METHOD

Subjects
The subjects were 16 female college freshmen at M.S.U.

who were enrolled in an introductory psychology course and
had volunteered for the experiment. Each 8 had scored 9 or
above on the self-scoring version of the Stanford Test of
Hypnotic Susceptibility at a group hypnosis session and was
free from obvious psychopathology. In addition, each S was
capable of achieving posthypnotic amnesia as tested by an
additional individual session with the experimenter.

To screen potential subjects, several initial group
hypnotic sessions were held. The session was explained to
the students as being simply a demonstration of hypnosis.
The possibility that attendance might later lead to partici-
pation in an eXperiment was mentioned, but it was made clear
that participation in this initial session in no way incurred
an obligation to participate in the eXperiment. At this
session the self-scoring form of the S.H.S.S. was admin-
istered.
Materials

The experiment was conducted in a sound-proof, ventilated
room furnished with two comfortable chairs, a bed, two tables,

and the G.S.R. apparatus. The subject sat in one chair with

13

14

the apparatus on a table to the right of him and the experi-
menter sat to her right beside the apparatus. A manuscript
reading "Very Important, Don‘t Touch!" lay on the table next
to the reclining chair in which the subject sat. A Grass
Model #5 Polygraph was used to record the G.S.R., and finger
electrodes manufactured by the Yellow Springs Equipment Co.
were attached to the right hand, which had been immobilized
through a posthypnotic suggestion.

A list of 15 words was used to stimulate free association.
For each condition one third of the words (C words) served
as cues for posthypnotic arousal of anger and were matched
with five non-conflict words (C'words) which were used as
conflict words in the other condition, and with five neutral
words (N words, never used as conflict words). All three
word lists were matched for frequency in the English language
according to the Thorndike-Large word count. The two groups
of five conflict words served as N words in one condition and
C words in the other condition. Each word was typed in pica
type on a plain 3 X 5 index card and was presented singly

in random order. The list was as follows:

C or C C or 5 Neutral
penny pickle triangle
nickle cookie rectangle
dime sandwich square
quarter milk oblong

dollar bread sphere

15

Procedure

A multiple factor nested design was used. Each S served
in one experimental condition and then returned for an addi-
tional session which included the destructive impulse suggestion
if it had been excluded in the previous session or excluded
the destructive impulse if it had previously been included.
Assignment of S's to experimental conditions was random. The
four experimental conditions were: (1) suggestion of the
amnesia for the paramnesia and of the destructive impulse,

(2) suggestion of the amnesia but not of the destructive
impulse, (3) suggestion of the destructive impulse but not of
the amnesia, and (4) neither the amnesia nor the destructive
impulse suggested.

Female S's scoring 9 or above were contacted by E and
asked if they would be interested in participating in an
eXperiment. The S's were told that the purpose of the experi-
ment was to assess the effect of a hypnotically induced
paralysis on the G.S.R. Those S's who agreed to participate
were asked to come to an individual hypnosis session. During
this session E hypnotized S, gave deepening suggestions, and
tested a posthypnotic amnesia. To obtain baseline data, an
initial presentation of the word list was made after S was
awakened. S was asked to free associate to each word for 30
seconds. E refrained from Speaking during this and subsequent
free association periods.

The instructions given prior to the presentation of the

word association list were as follows:

16

"Now I am going to show you some words. From the time
I show you each word until the time I tell you to stop I
want you to tell me everything that comes to mind; everything
that comes to your attention, no matter what it is. Do
you have any questions? All right, let's begin.”

At the following session, one week later, S was again
hypnotized. While S was in a hypnotic state the paramnesia
was implanted using these instructions:

”Now as you continue to rest in a deep, sleep-like state,
I'm going to recall to your mind an event which occurred
not too long ago. As I recount this event to you, you will
recall fully and completely everything that happened. As I
recall this experience, you will remember each and every
detail fully. Now bear in mind that while I repeat what I
know of this event, you will recall fully and completely
everything just as it happened, and more than that, you will
remember the emotions which you had at this time, and you
will feel as you did while this occurrence was taking place.
Nod your head if you understand.

"Now the particular event of which I am going to tell
you happened at the bookstore. The bookstore was very busy,
crowded with people. As soon as you see the bookstore
crowded with people, let me know by nodding your head. Nod
your head to let me know when you see the large number of
peOple that were milling around. Ybu were carrying a full

armload of books and had your money, the bills and the coins

17

in your hand. On top of the armload of books you had a
lunch bag and a hard-bound book that you had really been
looking forward to buying. Nod your head when you see that
bag lunch and that special book. Having picked up all the
books you needed, you walked to the end of the checkout line.
Nod your head when you reach the end of the checkout line.
"The line was long. It inched forward so slowly; so
slowly that you wondered if it was moving at all. Ybu grew
tired of waiting; you were tired and impatient. Nod your
head when you feel the impatience you felt then. The line
moved so slowly that you grew tired of holding your books.
The books became heavier and heavier. Your arms ached from
holding them. Nod your head when you feel the aching in
your arms. Other people had fallen in line behind you as you
waited. Ibu waited and waited and grew more and more im-
patient and tired; your arms ached more and more. The person
behind you shoved into you; this irritated you. Nod your
head when you recall the shove. It irritated you that people
were so inconsiderate and rude. Out of the corner of your
eye you saw a man walking towards you. He pushed his way
through the line right in front of you, bumping you and
almost making you drop that special book you had been looking
forward to reading. Nod your head when you see that man
cutting through the line and bumping you. That really made
you mad. Ybu were thinking that you'd had just about enough.
It would take just one more thing like that to make you boil.

18

over with anger. You turned your head to see how many people
were lined up behind you when someone tapped you on the
shoulder. You turned your head back, and you saw that it

was the instructor in one of the courses you had enrolled in.
He asked you to step back. Assuming that he wanted to go
through, you stepped back, bumping into the person behind
you, who snapped at you crossly, 'Watch it!’ To your
astonishment, your instructor stepped into the line, taking
your place. That did it! Anger surged up within you. Ybu
thought to yourself, 'Who does he think he is?‘

“Just then a man who was with him started to step into
line, and you stepped forward to close the gap, but he bumped
into you, Spilling all your books, your money and your bag
lunch to the floor. There were books, money, and the con-
tents of your lunch bag all over the place. What a mess! Nod
your head if you see it. As you stooped to pick up the
books, money, and lunch, your instructor's friend stepped
into your place. That made you furious! YOu picked up a
book and saw that the binding was broken; the pages were
crumpled and smudged with dirt. It was your favorite book,
the special one you had been looking forward to reading.

Nod your head if you see the broken book. YOu were just

boiling with anger inside; you were fuming and seething with
anger. As you knelt down to pick up your money, books, and
lunch, your instructor commented sarcastically, 'You dropped
something.' Those words really burned you; you were infuri-

ated.

19

"As though that was not enough, he turned to his friend
and snickered, 'You want to see something funny, look.‘ Then
you heard them both laughing; laughing and laughing; really
enjoying your predicament. With every laugh your anger
surged up more uncontrollably. Ybu were swept up in over-
whelming feelings of anger and rage. You hated these men.
Once again you find yourself feeling those powerful emotions.
You feel them right now with your whole body.

”Now you can feel your anger fading away with each
breath. Your feelings of anger are draining away as you
gradually feel more and more calm and relaxed. The memory
of the bookstore and your anger are fading away. Nod your
head when your anger is gone.

“Now listen carefully. After you have awakened, anything
that comes into your mind that is associated with (food or
money - alternating) will stir overwhelming feelings of
rage."

Following implantation of the paramnesia and while the
S remained in a deep state of hypnosis, suggestions were
given for amnesia and/or the destructive impulse, depending
upon the eXperimental condition S was to serve in. S's
serving in a condition including the destructive impulse
were given the following instructions:

'When these feelings of hate and anger boil up you will
realize that it is the person who wrote the manuscript on

the table next to you that you hate, and you will have an

20

overwhelming urge to tear them up. Nod your head if you
understand."

S's serving in a condition including the amnesia were
given these instructions:

”After you awaken, you will not be able to remember
anything about your experience in the bookstore. It will be
just like a dream that you had while you were asleep that
you don't remember after awakening. Just like a word on the
tip of your tongue that you just can't remember even if you
should try. Nod your head if you understand."

After implantation of the paramnesia and of the experi-
mental condition instructions was completed, S was awakened.
Electrodes for the G.S.R. were then placed upon S's fingers,
and the equipment was prepared for use. When preparations
were completed, 8 was presented with the words in the word
list one by one and was asked to free associate to each
word for 30 seconds. A tape recording was made of the associ-
ations, and G.S.R. measurements were taken throughout the
free association period. Once the word list was gone
through, S was rehypnotized, and all suggestions, including
the induced paramnesia, were removed.

One week later S returned for the second session. She
was again hypnotized and presented with the paramnesia. If
the destructive impulse had been suggested at the previous
session, it was excluded at this one. If the destructive
impulse had not been included in the previous session, it

was suggested at this session. Absence or presence of amnesia

21

was held constant across the two sessions. 3 was again
awakened and presented with the word list. At the conclusion
of the free association period, S was rehypnotized and given
a posthypnotic cue for recalling and recounting the par-
amnesia. This procedure provides data indicating whether
S accepted the paramnesia as a real experience and provides
B an Opportunity to note spontaneous distortions and omis-
sions. Special efforts were made to remove all traces of the
paramnesia before S was dismissed.

The experiment could be described chronologically as
follows:

(1) Group hypnosis session

(2) Individual hypnosis session
Word associations - baseline data

Sessions 3 and 4 varied depending upon the experimental con-

dition S was assigned to. The four patterns were:

Condition: Amnesia and destructive impulse

(3) Hypnosis
Paramnesia - posthypnotic anger
Amnesia and destructive impulse suggested posthypnotically
S awakened
Word associations and G.S.R.
Hypnosis - treatment removed

(4) Hypnosis
Paramnesia - posthypnotic anger
Amnesia suggested posthypnotically; no destructive
impulse
S awakened
Word associations and G.S.R.
Hypnosis - treatment removed
S awakened
Discussion of experiment

22

Condition: Amnesia and no destructive impulse

(3) Hypnosis
Paramnesia - posthypnotic anger
Amnesia suggested posthypnotically; no destructive
impulse
S awakened
Word associations and G.S.R.
Hypnosis - treatment removed

(4) Hypnosis
Paramnesia - posthypnotic anger
Amnesia and destructive impulse suggested post-
hypnotically
S awakened
Word associations and G.S.R.
Hypnosis - treatment removed
8 awakened
Discussion of experiment

Condition: Destructive impulse and no amnesia

(3) Hypnosis
Paramnesia - posthypnotic anger
Destructive impulse suggested posthypnotically;
no amnesia
S awakened
Word associations and G.S.R.
Hypnosis - treatment removed

(4) Hypnosis
Paramnesia - posthypnotic anger
No amnesia; no destructive impulse
S awakened
Word associations and G.S.R.
Hypnosis - treatment removed
Discussion of experiment

Condition: No amnesia and no destructive impulse

(3) Hypnosis
Paramnesia - posthypnotic anger
No amnesia; no destructive impulse
S awakened
Word associations and G.S.R.
Hypnosis - treatment removed

(4) Hypnosis
Paramnesia - posthypnotic anger
Destructive impulse suggested posthypnotically;
no amnesia
S awakened
Word associations and G.S.R.
Hypnosis - treatment removed
Discussion of experiment

S's were assigned to these four conditions randomly.

23

Measures

The data in this experiment falls into two categories.
One type of data consists of S's verbal responses to the word
list. G.S.R. measurements are the other form of data. Only
G.S.R.'s of 1,000 ohms. or greater were counted during the
35 seconds following stimulus presentation. S's verbal
responses were classified in terms of Reyher's categories of
psychopathology (see Appendix B). Category eight, ”Distur-
bance or distortion in perception of the tachistoscopic
stimulus,” was eliminated in this instance because the words
were not presented on a tachistoscope. Category 11,
"Delayed awareness of one or both aSpects of the conflict"
was replaced by "Obsessional statements or ruminations."
S's score was the total number of reactions in categories 1
through 11. The rating was done by one individual (an
undergraduate psychology major) who was ignorant of the

hypotheses under investigation.

RESULTS

Data Descripgigg

Statistical analysis can serve only as a summary of the
salient characteristics of the experimental data. Of
necessity the coding system used for this study excluded
many potential observations from examination. For example,
the personal attitude of S towards E was not examined,
although this E observed wide variation. Some S's expressed
a personal interest in E and seemed to regard her as a
potential counselor or model for identification. Other S's,
on the other hand, were neutral in their attitude towards
E and mainly interested in acquiring extra credit for their
introductory psychology course.

Another statistically ignored variable was response
style. Some S's responded with strings of loosely connected
words such as in Protocal 44: "Dollar - bill, green, cash,
capitalism, America." Other S's Spoke about previous eXperi-
ences or momentary mental images.

Protocal 20 - “Dollar - I see these rich men with

lots of bills in their pockets. They're handing

out money to these handicapped children. I see

beautiful houses with . . . You can tell that

they cost a lot of money. I see a store with, a
furniture store."

24

25

Each 3 exhibited a relatively consistant response style.
A subject who recited strings of words during the pre—
experimental session continued to do so throughout the ex-
periment. With very few exceptions S's who reported images,
experiences, or associations also remained consistant in
their response style. More S's in the author's group produced
images, experiences, and associations than those in the group
run by the other experimenter.

The obvious symptomatology which appeared in these
protocals was minimal. There were a few reports of head-
aches, of coming up with "blanks," but no subject reported
feeling anxious or depressed. The majority of the rated
symptomatology fell into the category of "conscious correlates."
For example, the subject who produced Protocal 9 gave this
response to a C word, "Quarter - Fat people, Spend money all
the time and buy fat foods. No excuse for it. Loneliness,
sadness, peOple fighting, wars."

No S tore up the papers, but three verbalized a special
interest in the papers' owner. Only one S appeared to be
completely aware of the paramnesia, but even she denied any
memory of the instructions on questioning. However, the
progression of her associations reveals considerable memory
for the paramnesia and awareness of the destructive impulse.

". . . Spilled, angry, ticket. Who wrote this? . . .

angry, tear, rip, destroy, rage, store, rather

hold back, . . . Spilled, angry, tear, person,

bookstore, mad, scream. . . spilled, whoever wrote

that, angry, mad, tear . . . angry, tear, rip,
bookstore, mess, mad."

26

It would appear that this study failed to produce the
florid symptomatology reported by Reyher (1958, 1967),

Perkins (1965), and Sommershield (1969).

Data Analysis

The data for statistical analysis consisted of G.S.R.
responses which had been recorded on paper tape and verbal-
ized free associations which were tape recorded and then
transcribed. For the G.S.R. data the maximum deflection in
millimeters of the recording pen occuring during the 35
seconds following presentation of the individual word
stimulus was used as the basic datum. The free association
ratings (using a modified version of Reyher's system; see
Appendix B) for each word presented in a particular session
to a S were summed across categories 1 through 11, with
this sum serving as the basic datum. A separate 2 nested in
2 X 2 X 3 analysis of variance was computed for each type of
data from the two eXperimental conditions. The independent
variables tested were: Amnesia (or no amnesia), destruc-
tive impulse ( or no destructive impulse), word type (C words,
5 words, and N words), and experimenter (Aaron Karnilow or
Alice Wolfe). The linear model for this analysis can be

described as listed below:

27

A,

+4 Wadi 4% Icy/e + m + 777%
*C/ {7% *4 W ”VS/fl ”(A + A 7727,
‘* 0955/77 ”(5% “7092/ + 5x: +77%?
+0957 + 092 4/3/4 Egg/”7’7

This model was generated by Robert Gurney from the principles

MW =74 + RAM/W + 0% VIII/a

set forth by Myers (1966) and Lee (1961).

The results of the analysis of variance for the G.S.R.
data are shown in Table 1. Using a two-tailed F test with
a minimum acceptable alpha level of .05, only two contri-
butions to the total variance are significantly greater than
chance - differences in word type (C words, C'words, and
N words) and the interaction between eXperimenter and amnesia.
(Both were significant beyond the .01 level.) Neither the
amnesia nor the destructive impulse, the two experimental
variables, had a greater than chance effect.

An identical 2 nested in 2 X 2 X 3 analysis of variance
was performed on the verbal data with the results as listed
in Table 2. This time three contributors to the total
variance had a significant effect: 1. EXperimenter dif-
ferences, 2. differences in word type, and 3. the inter-
action between experimenter and word type. Comparison of

the G.S.R. data to the verbal data reveals one consistantly

TABLE 1

28

G.S.R. Data - 2 Nested in 2 X 2 X 3 Analysis of Variance*

 

Degrees of

 

 

Component F Freedom p) . 05 Etaa
A .4664 1/12 no .0062
B 3.8495 1/12 no .0516
C 12.9733 2/24 .01 .1277
D 100879 1/12 no .0115
AB 14.9368 1/12 .01 .2002
A .3576 2/24 no .0035
AD .1930 1/12 no .0020
BC 2.9540 2/24 no .0290
CD 1.7237 2/24 no .0140
ABC .6401 2/24 no .0063
ABD .0123 1/12 no .0001
BCD .9506 2/24 no .0077
ACD 1.0460 2/24 no .0084
ABCD 2.4961 2/24 no .0202
A = amnesia
B = experimenter
C = word type
D = destructive impulse

*

2 Tailed Test of Significance

TABLE 2

Verbal Data - 2 Nested in 2 X 2 X 3 Analysis of variance*

 

Degrees of

 

a

Component 2*, Freedom p12.05, Eta
A .2017 1/12 no .0045

B 6.7476 1/12 .05 .1517

C 5.6619 2/24 .01 .0827

D 1.9074 1/12 no .0137
AB .2197 1/12 no .0049
AC .6875 2/24 no .0100
AD .8690 1/12 no .0062
BC 4.3844 2/24 .05 .0640
RD 2.0039 1/12 no .0144
CD 1.2580 2/24 no .0070
ABC 1.2449 2/24 no .0181
ABD .8058 1/12 no . 0058
BCD 1.5461 2/24 no .0087
ACD .5097 2/24 no .0028
ABCD 1.0111 2/24 no .0056

 

A.= amnesia

BI: eXperimenter

C

D

* 2 Tailed Test of Significance

word type

destructive impulse

30

significant effect -- differences in word type. This result
suggests that the three categories of words were either
initially different in their response eliciting qualities or
that the words in each category were affected in a different
fashion by the experiment.

Etaawas computed for all sources of variance generated
by both types of data (Tables 1 and 2). Subject differences
accounted for 27% of the variance of the verbal data and
16% of the variance of the G.S.R. data. When the inter-
actions of subject differences with other experimental
variables were added, 60% of the verbal data variance and
50% of the G.S.R. data variance was accounted for. The great
contribution of subject differences to the total variance is
an expected result in this research, since individual
differences in personality type and degree of pathology
should differentially affect reSponse style.

To determine whether the significant experimenter and
word type effects were obscuring small but significant main
effects, separate 2 X 2 analyses of variance were performed
on both types of data. For these analyses only the two
critical variables (amnesia and destructive impulse) were
taken into consideration. Out of these 12 analyses of
variance, there was only one significant F ratio, the effect
of amnesia on 5 words with the author as experimenter. Since
there were 36 F ratios altogether, the occurance of one

significant F would occur with approximately .97 Probability.

31

Because the two critical variables tested (amnesia and
destructive impulse) had no significant effect in the 2
nested in 2 X 2 X 3 analysis of variance for either type of
data, a 2 nested in 2 X 3 analysis of variance was computed
for the verbal data, comparing the baserate data to that
produced by the first eXperimental session. Again a two
tailed test of significance with a minimum alpha level of
.05 was used. Surprisingly, every independent variable and
all their interactions had significant F ratios (See Table 3).
Thus it would appear that not only was there a significant
(.05) difference between the baserate and Condition I data,
but there was also a significant (.01) amount of variation
in word type and eXperimenter. Since the significant word
type and experimenter differences appeared in the larger
analysis of variance, only the condition difference (base-
rate vs Condition I) provides a new bit of information.
Apparently something in the experimental procedure was
associated with a change over baserate levels of rated
symptomatology.

Since both experimenter and word type variables produced
significant F ratios in the baserate vs Condition I analysis
of variance as well as in the Condition I vs Condition II
analysis of variance, an additional 2 X 3 analysis of
variance was performed on the baserate data. (See Table 4.)
There was a significant (.05) experimenter effect even before

the experimental treatments were administered. However,

32

TABLE

Verbal Data - Baserate vs Condition I,

2 X 2 X 3 Analysis of variance*

 

 

Degrees of
Component L Freedom 2 Z , 05
B 14.0815 1/14 .01
C 8.1455 2/28 .01
E 7.2123 1/14 .05
BC 8.7648 2/28 .01
BE 5. 0085 1/14 . 05
CE 7.4774 2/28 . 01
BCE 5.7466 2/28 .01

 

B. =- experimenter
C a word type
,3 - condition.(baserate vs Condition I)

* 2 Tailed Test of Significance

TABLE 4

Verbal Data - Baserate 2 X An sis of variance*

 

 

Degrees of
Component F Freedom 2 :05
B 7. 0422 1/15 . 05
C .0279 2/28 no
BC 3.6328 2/28 .05

 

B a experimenter

C - word type

* Two Tailed Test of Significance

34

there were no meaningful differences between the three
types of words before experimental treatment was adminis-
tered.

The chief statistical findings as measured by a variety
of analyses of variance can be thus summarized: Even
before the experiment began there was a significant differ-
ence either in the subjects chosen by the two experimenters
or in the way the experimenters affected the subjects. On
the other hand, the three categories of words were approxi-
mately equal in their symptomatology-eliciting effect before
the experiment was begun.

As reflected in the verbal data, there was a signifi-
cant difference in rated symptomatology between the pre-
experimental and post-experimental conditions. Therefore,
something in the experimental procedure was associated with
a change in rated symptomatology. Furthermore, the addition
of the experimental procedure also produced a significant
difference in the rated symptomatology elicited by different
word types (C words, C'words, N words).

When the experimental variables (amnesia and destructive
impulse) were analyzed in a 2 nested in 2 X 2 X 3 analysis
of variance for the verbal data, they had no significant
effect. Even when this data was broken down by experimenter
and word type, the experimental variables still had no effect
greater than that attributable to chance. An identical
analysis of variance performed on the G.S.R. data similarly

35

indicated that the experimental variables had no significant
effect on the G.S.R.

As suggested in the proposal, R (repression) was com-
puted using Reyher's formula (see Appendix A) for each 8 in
each of the baserate and experimental conditions. The
results of this computation are listed in Table 5. It should
be noted that there is actually an.inverse relationship
between repression and the computing formula, which evalu-
ates the amount of conscious expression of the repressed
material. This data was not, however, recomputed, since
neither the differences between the groups nor the variation
within the groups would change with the conversion. A
T-test was computed on the baserate repression data compar-
ing S's run by the two different experimenters. Not only
was the difference between the means significant (p a .025),
but there was also a large difference in the amount of
variation within the two groups. variance within the group
run by the author was 20.00, while variance within the group
run by the other experimenter was 1.19.

One hypothesis might account for such a difference in
variance. Subjects may be categorized according to response
style or personality type as reflected in the repression
score. Perhaps some error in experimental procedure placed
nearly all S's of one type in Mr. Karnilow's group but mixed
them with other types of S's in the author's group. If this
was the case, the experimental variables might still have a

R

Re ression As Com uted Usi

TABLE

 

 

Baserate
" 0133“ .4002
- .9338 - .8004
- 08671 10 0000
- .6003 .8004
- .3335 .4002
- e 7337 3.2001
- .2668 .2001
" e 9338 .4002
" 08671 " 0800“
" e 9338 -10 0000
" e 9338 ‘10 0000
- .8004 -l.0000
-1.0000 - .8004
- 0800“ ‘l' .4002
- .8004 -1.0000
" e 9338 .4002

Re her's Formula

 

3.0000

0.0000

1.0000

- .2001
1.0000

3.6003

0.0000

.8004

" e 800“
- .6003
“100000
- .8004
-1.0000
-l.0000

0.0000
- .8004

 

36

Condition I Condition I;

 

IIIIIIII

........

........

........

IIIIIIII

37

significant effect when only one group of 8's was considered.
To test this hypothesis, all S's with baserate repression
scores ranging between —.86 and -1.0 were arbitrarily placed
in one group.

A 2 X 2 analysis of variance was computed for this
group of S's, using the C word verbal data only. By coin-
cidence, the cell frequencies were equal. Neither of the
experimental variables produced significant P's. In fact the
F ratios for both the amnesia and the destructive impulse
were approximately zero. While the F ratio for their inter-
action was larger (3.85), nevertheless, it was not signifi-
cant at the .05 level of probability.

IA similar 2 X 2 analysis of variance was performed on
the seven S's with the highest R scores. This time the cell
frequencies were unequal. Once again, none of the P ratios
were significant.

It would seem, therefore, that no matter how the data
is rearranged, the experimental variables have no signifi-

cant effect.

DISCUSSION

The results of this study do not confirm the tested
hypotheses. None of the analyses of variance, ranging from
the largest, which took experimenter and word types into
consideration, down to the smallest, which looked separately
at each destructive impulse - amnesia interaction for each
experimenter and word type, revealed any significant effect
of the destructive impulse or its interactions. Even when
an attempt was made to regroup the subjects by repression
scores, the destructive impulse appeared to exert virtually
no effect. This lack of significance was equally true for
either G.S.R. or verbal data.

When we examine the other experimental variable,
suggested amnesia, we again discover an absence of signifi-
cance. By itself, suggested amnesia had a significant effect
only in one of the 12 separate 2 X 2 analyses of variance.
Since this was the only significant F ratio out of a group
of 36 F ratios, it may very well be a chance effect. In
the 2 nested in 2 X 2 X 3 analysis of variance for G.S.R.
data, the interaction between suggested amnesia and experi-
menter was significant at the .01 level. This effect did
not appear in the verbal data analysis.

38

39

The significant amnesia-experimenter interaction is,
perhaps, part of the larger effect of experimenter differ-
ences. Even before the experiment was begun, there was a
significant difference between the experimenters' subjects
in terms of mean repression score. variance around these
means was 1.19 for one experimenter and 20.00 for the other.
Taking only the verbal data into consideration, there was a
significant experimenter effect throughout the experiment
which showed up in all the analyses of variance. For the
G.S.R. data in the 2 nested in 2 X 2 X 3 analysis of
variance, the experimenter effect by itself did not have a
significant F ratio, but its interaction with suggested
amnesia was significant at the .01 level.

The consideration of experimenter differences provides
an incidental, but very important, experimental variable.
None of the studies to be reviewed below used more than one
experimenter. Yet, using virtually the same procedure
some experimenters (Reyher, Perkins, Sommershield) derived
significant results, while others (veenstra, Larison) had

no significant findings.

ggevious Findipgs

This entire line of research is an outgrowth of
Reyher's (1958) doctoral dissertation, 'Hypnotically Induced
Conflict in Relation to Subception, Repression, Antisocial
Behavior and Psychosomatic Reactions." Although the

40

experiment was designed to test perceptual defense as a
function of an induced conflict, Reyher found that his
subjects experienced various symptomatic reactions which were
reflected in the side comments they made while reporting
what they saw on the tachistoscope screen.

In a post-doctoral study (1967) Reyher replicated his
original research with only minor modifications in the
procedure. His purpose was to further examine the psycho-
pathological reactions displayed by his subjects in what had
originally been designed as an experiment on perceptual
defense. This second experiment differed from the first
both in the inclusion of simulating, control subjects and in
the variation of the strength of the anxiety-arousing
impulses posthypnotically over a scale from 1 to 4. Once‘
again there were vivid symptomatic reactions, and the
intensity of these symptoms seemed to increase as the post-
hypnotic signals for increasing the strength of the induced
impulses were given.

There are a number of procedural differences between
Reyher's research and the present study. To begin with, the
S's attention was focused on the cognitive task of recog-
nizing words flashed briefly on the tachistoscope screen.
They were not aware that incidental comments about their
subjective reactions were of interest. In the present study
S's were alloted no task other than reporting their subjec-

tive impressions (free associations). Whereas in Reyher's

41

studies S's reSponses were either correct or incorrect, S's
were told to report anything which came to their minds in
this research. It seems most likely that the former pro-
cedure would produce a higher level of anxiety than the
later, which is less evaluative.

While Reyher varied the intensity of impulses during
each session in his replication study, the instructions to
8'8 were uniform throughout this study. The emphasis placed
on intensifying various aspects of the conflict may have
made it seem more real to the subjects and thus be more
pathogenic.

Perkins' (1965) study "Repression, Psychopathology, and
Drive Representation: An Experimental Hypnotic Investi-
gation of the Management of Impulse Inhibition” further
investigated Reyher's (1958, 1967) findings. The major
innovations in Perkins' study included the division of S's
into good and bad repressers on the basis of their aware-
ness of the conflict and the addition of a control group of
simulating S's who were themselves capable of achieving a
deep trance state. S's were also given TAT's following the
one experimental session. S's again focused their atten-
tion on looking at the tachistoscope screen rather than
monitoring their personal reactions. The intensity of the
conflict and its various elements was varied during the

experiment.

 

 

 

ll! '1 ‘1' l l lulIl. l‘. I'

42

The results of Perkins' experiment revealed that the
behavior of the hypnotized S's was qualitatively different
from that of the simulating S's. As the intensity of the
conflict was increased, there was a progressive rise in the
rated symptomatology of the good repressers. Poor repres-
sers were initially at a high level of rated symptomatology
which first increased and then decreased to a point below
that of the good repressers. It would appear, therefore,
that the suggested intensification of the conflict had some
influence on the experimental results.

These results were confirmed in a study by Sommershield
(1969), 'Posthypnotic Conflict, Repression and Psychopath-
ology.“ Sommershield's procedure was similar to that of the
present study in that the stimulus words were presented on
index cards. S's were told to pronounce each of 57 words and
then report to E on 'how they were doing'. However, in
addition to this task S's were given the TAT prior to the
first experimental session and a symptom questionnaire
following the last session. Furthermore, a different
paramnesia was used in each of the two experimental sessions
to test the effectiveness of a sexual conflict as apposed
to a hostile-aggressive conflict. Finally, all of Sommer-
shield's S's were males.

The symptomatic reactions observed by Reyher and
Perkins were also observed by Sommershield. He found that

the sexual and aggressive paramnesias were approximately

 

 

I ll lilllll I II II [If]

43

equal in their pathogenicity as measured by G.S.R. and rated
symptomatology. Using the TAT as a measure of degree of
repression, Sommershield found a significant negative cor-
relation between degree of repression and frequency of
symptoms.

.A fourth attempt to replicate Reyher's results failed
to produce significant symptomatology. The focus of Veenstra's
(1969) study, “The Effectiveness of Posthypnotically Aroused
Anger in Producing Psychopathology," was to examine the
effective pathogenicity of direct suggestion as opposed to
the paramnesia. While the study revealed that these two
methods of conflict induction were equally effective, the
lack of significant symptomatology in this experiment puts
this conclusion in question. Veenstra attributed this lack
of symptomatology to the absence of the destructive impulse
in his experimental design.

veenstra's procedure was nearly identical to that of
the present study. He used the same paramnesia and virtu-
ally identical instructions. Each 8 was asked to free
associate to each word on the word list for 20 seconds and
then report his emotions and feelings afterwards. Prior to
the experiment, S's were given practice in free association,
which may have biased the type of response offered. An
additional similarity to the present study lay in the
addition of a no amnesia condition. All but one of the
subjects in veenstra's experiment acknowledged angry feelings

and displayed little or no symptomatic reaction.

 

 

44

Though not yet written up, a followup to the present
study was conducted by Grey Larison to determine whether the
specific wording of the instructions given to S on how to
report his impressions was responsible for the lack of
results in the present study. Larison's procedure differed
from that o: the present study only in that he varied the
free association instructions with the "pump priming”
instructions used by Sommershield, Perkins and Reyher. In
both conditions there remained an absence of significant
pathology, thus suggesting that the instructions were not
reSponsible for the lack of results.

Another, as yet incomplete, study conducted by Bruce
Burns has produced florid symptomatology like that which
appeared in the earlier studies of this series. Burns used
Sommershield's oedipal paramnesia to create a conflict
stimulated by the posthypnotically anticipated appearance of
the woman in the paramnesia. Rather than have his S's
pronounce a list of words or identify words diaplayed briefly
on a tachistoscOpe screen, Burns asked his subjects to close
their eyes and describe any visual images or emotions which
came to awareness.

Burns' experiment differs from the previous line of
research in its use of free imagry (Reyher, 1963; Reyher and
Smeltzer, 1968; Reyher and Morishige, 1969) to stimulate
images, sensations, and symptomatology. The stimuli were

spontaneous visual images (primary process material) rather

“5

than words (cognitive, secondary process material) presented
by the experimenter. It is to be expected that the images,
which tap a primitive level of integration, would stimulate
even more primary process elaboration than the word stimuli
used in the present research. Although this experimental
procedure differs significantly from that used in the
previous line of research, the fact that symptomatic

behavior was associated with materials reflecting the induced
conflict once again confirms the effectiveness of the

hypnotic paradigm in producing symptomatology.

Erocedural Elements
Out of the 13 variables in the hypnotic paradigm which

veenstra (1969) identified, he reported eliminating five:
Age regression, hypnotic arousal of conflicting emotion,
anger directed towards an .Iuthority figure in the experi-
mental situation, and posthypnotic suggestion of a destruc-
tive impulse. A.major idea behind the present study was
that the omission of the destructive impulse was responsible
for the absence of symptomatology from Veenstra's study.
However, the variation of the destructive impulse did not
significantly affect the results in the present study.

On the other hand, both this study and that of Veenstra
excluded the age regression variable. Excluding age regres-
sion was a function of the particular paramnesia used. While

the bookstore paramnesia used in this study takes place in

46

the immediate past, the paramnesia used by Perkins (1965)
takes place during S's childhood, when he was in sixth grade.
It is difficult to imagine why the age of S in the paramnesia
should make great difference in the production of symptoms,
except that childhood incidents frequently have a greater
impact than those which occur during adulthood. On the other
hand, the intensity of the conflict may very well contribute
to the pathogenicity of the paramnesia.

In Sommershield's anger paramnesia 8 commits a repre-
hensible act (breaking an art object) and then is insulted
(made to feel guilty) by the authority figure in the conflict.
Similarly, Sommershield's sexual paramnesia taps guilt by
having 3 imagine making love to an older woman and then
realize his inadequacy to satisfy her, thus reactivating
the oedipal conflict. Guilt is also activated by Reyher's
(1958) anger paramnesia. Here 8 steals some money and then
is blackmailed by a fellow student who serves later as the
authority figure in the induced conflict. This same par-
amnesia was used successfully by Perkins (1965).

In all these paramnesias S has already performed a
reprehensible act before he is asked to act out posthypnoti-
cally. In the paramnesia used in the present study 8 has
no reason to feel guilty if she does not tear up the
”important" paper unless anger is wholly inacceptable to
her superego. Even this act might be considered Justifiable

because the professor taunts 8 without provokation. On the

47

other hand, with the other paramnesias S has reason to feel
guilty even before he tears up the papers; in fact S's

anger serves to defend against acknowledging his guilt.
Therefore, to tear up the papers would involve not only
expressing anger in an antisocial destructive fashion but
would also mean that S was offering the authority figure suf-
ficient reason to again point out S's guilt over having per-
formed the initial reprehensible act.

Prom a psychodynamic orientation, the paramnesias used
by Reyher, Perkins, Sommershield, and Burns should be far
more pathogenic (anxiety arousing) than the paramnesia used
by Veenstra and in the present study. In the former par-
amnesias S's superego is actively involved in disapproving
of 8's having acted on libidinal impulses. Either S feels
guilty, or his ego defends against the guilt. The introduc-
tion of the conflict-inducing instructions not only introduces
a new libidinal impulse (aggressive acting out) to be defended
against but also threatens to reactivate the initial super-
ego reaction. In this instance ego is threatened by a strong
superego reaction from two sides and is thus highly likely
to activate its characteristic modes of defense. On the
other hand, in the veenstra paramnesia S risks a superego
reaction only if he tears up the papers, and even then, his
mechanisms of defense are called into play only if aggressive
acting out under duress is inacceptable to his superego. It

is easy to understand why no 3 in this study either tore up

.il),,|‘ I‘ r ,‘lll‘l Ill-l [I'lllul‘ !
«I. III III! III! (I. (I). .

48

the papers or exhibited outstanding symptomatology. If
aggressive impulses are acceptable to S when they are not
acted upon, then S can entirely avoid the conflict and the
need for defense by failing to tear up the papers.

In the present study and that of Veenstra, S was
instructed only once during each experimental session to
tear up the papers. However, Reyher (1967), Perkins, and
Sommershield all had their S's experience the antisocial
impulse at several different intensities. In the procedure
used by Perkins and Sommershield, S was advised by E during
hypnosis that 8 would experience the destructive impulse at
each of three (as apposed to Reyher's four) different inten-
sities as triggered by a posthypnotic signal. Very possibly
an impulse suggested under hypnosis and reinstituted several
times by a posthypnotic signal is more actively pathogenic
than an impulse suggested only once in the trance state.
However, the validity of this conclusion is questionable since
significant symptomatology appeared without the posthypnotic
intensification of the impulses in Reyher's original study
(1958) and in the Burns (1971) study.

Another possible contributor to the positive outcome of
the early studies and the negative results of the present
study is the focus of 8's attention during the data collecting
period. Reyher and Perkins both had their S's identify
words presented briefly on a tachistoscOpe screen. This task

in itself was anxiety arousing because S's answer was either

49

correct or incorrect. The verbal comments S made about his
reactions were seemingly not the focus of E's investigation.
Therefore, it seems likely that 3 would not put much effort
into censoring what he said.

Sommershield's study, on the other hand, used much the
same mode of stimulus presentation as the present study. S's
were asked to pronounce each of 27 words, free associate to
them, and then comment on their feelings during the free
association period. This procedure differs from that of the
present study only in the addition of the comment period
following free association. However, Sommershield used a
different list of 2? words for each of his two experimental
sessions, while the same list of 15 words was presented to
each 8 three times during the present study. In Sommershield's
study 8 may have accepted pronouncing the words as a task
with a correct or incorrect response. It was, however, clear
to S in the present study that E was interested in his free
associations. The likelihood that S would censor his verbali-
zations was probably greater in the present study than in any

of the earlier, more successful research.

Experimegter Effects

As reported in the section on results, there were sig-
nificant experimenter differences in repression scores and
rated symptomatology even before the experimental instruc-
tions were presented. This finding introduces a new dimen-

sion to be considered as contributing to the eXperimental

50

outcome. At least three factors may be responsible for the
experimenter differences: 1. Subject selection, 2. vari-
ations in procedure, and 3. Demand characteristics.

Although the same procedure for subject selection was
used by both experimenters in this study, subjects were not
randomized across experimenters to control for selection
differences. Certainly this control would have been advis-
able, but it would have lengthened the time required to
complete the experiment beyond reasonable limits since far
more potential subjects would have necessarily been rejected.
Studies by Shor & Schatz (1960) and Dorcus (1963) indicate
that hypnosis is a phenomenon specific to a particular
subject and a particular hypnotist; a subject capable of
achieving a deep trance state with one hypnotist may not
necessarily reach the same depth of hypnosis with another
hypnotist.

Selection factors were probably Operating in this experi-
ment. S's first volunteered for the experiment knowing they
would be hypnotized. Although extra credit towards their
grade in introductory psychology served as the external moti-
vation for subjects to volunteer, there must have been
special reasons why they volunteered for this particular
experiment. Further, a considerable number of S's who wrote
their names on the signup sheet did not come to the group
hypnosis session. Since E contacted S by phone to inform

her of the date and location of the group, it is possible that

51

certain volunteers chose not to come because they did not

feel comfortable with the idea of being hypnotized by a mem-
ber of the same or opposite sex. Thirdly, at the group
hypnosis session S met E, was hypnotized, and then was asked
to report on the depth of the trance achieved. At this

point 8 probably made some evaluation of E in terms of
competence and trustworthiness, and this evaluation might

well have been unequal for the two experimenters. In the
observation of the author, some S's who appeared to be in a
deep state of hypnosis reported achieving only a light trance
on the questionnaire. Was this report made out of an accurate
subjective impression, or did 8 deny the depth of trance
because she was made too uncomfortable by the idea of relin-
quishing control to a female peer? It is impossible to

state the answer definitively, but this is certainly a fertile
area for future research.

Finally, the significant experimenter effect may have
been the outcome of the different demand characteristics
created by the two experimenters. The fact that one experi-
menter was male and the other female could have altered S's
perceptions of expected behaviors. For example, the greater
symptomatology found in the data acquired by the female
experimenter might have resulted from a greater willingness
on the part of the female S's to express their emotions to
another woman. To admit these feelings to a male could have

been perceived by S as an invitation to a more intimate

52

(sexual) relationship with the male E. On the other hand,
some female S's may have unconsciously interpreted the
hypnotic induction by the female E as a homosexual assault.
As a result, greater anxiety would be aroused and thus
ultimately more symptomatology would appear.

The two experimenters may also have unconsciously affected
the data differentially through nonverbal expressions of
their expectations. Direct verbalization of anger towards E
may have been less acceptable to one experimenter than the
other. Again, one experimenter may have expected S to
reapond in a different fashion than the other experimenter.
This hypothesis seems likely in view of the fact that one
experimenter elicited virtually nothing but stringsaof words
from his S's, while the other E also had S's who spoke in
phrases and sentences, reporting images and experiences.
Finally, one experimenter may have expected more pathology

to appear than the other expected.

Word Type Differences
Although the eXperimental variables had no significant

effect, the experimental procedure was effective in producing
a significant change in differences in rated symptomatology
for the three groups of words (C words, E'words, and N words).
Before the experiment the three groups of words were approxi-
mately equal in their rated symptomatology. After the first

experimental session, however, there were significant

53

differences between the three word types on rated symptoma-
tology. Furthermore, the destructive impulse - no destruc-
tive impulse conditions differentially affected the three
word types.

In essence these results confirm that a single sug-
gestion was effective, the suggestion that mention of a
certain class of words would stir up overwhelming feelings
of anger. The change in the class of words which produced
the greatest amount of rated symptomatology can be simply
accounted for by the fact that the class of words mentioned
in the posthypnotic suggestion was shifted either from food
to money or vice versa across the two experimental sessions.
There is no need to assume that the inclusion or exclusion of
the destructive impulse had any influence on the differences
in rated symptomatology between the three classes of words.

On the other hand, the existance of these significant
effects suggests that the experimental procedure did have
some effect. It would seem unlikely in view of the success
of this suggestion that the experimenters somehow presented
the instructions in such a fashion that the 8's rejected them.
This finding further supports the hypothesis that the inef-
fectiveness of the paramnesia is reSponsible for the absence
of significant pathology in the experimental findings. If
the experimental variables coupled with the paramnesia had
been as powerful stimulators of impulses as the direct
suggestion of anger, the results of this study would probably

have been far more significant.

‘.II .Illv

1.

2.

3.

7.

9.

10.

11.

W

Basch, Judith. Degree of repression and frequency of
psychosomatic symptoms. Unpublished master's thesis,
Michigan State University, 1968.

Bobbitt, Ruth A. The repression hypothesis studied in a
situation of hypnotically induced conflict. Journal
of Abnormal and Social Ps cholo , 1958, 56, 205-21 .

Counts, Robert M. and Mensh, Ivan N. Personality char-
acteristics in hypnotically-induced hostility.

Journal of Clinical Psychology, 1950, 6, 325-330.

Dorcus, R. M. Fallacies in predictions of susceptibilities
to hypnosis based upon personality characteristics.

American Journal of Clinical Hypnosis, 1963, 5(3),
1 3‘1700

Eisenbud, Jule. The psychology of headache: a case

studied experimentally. Esychiatric Quarterly, 1937.
11 9 592’6190

Erickson, Milton. The method employed to formulate a
complex story for the induction of an experimental

neurosis in a hypnotic subject. Journal of General
PS ChOIO , 191+”, 31’ 67-840

Huston, Paul; Shakow, David; and Erickson, Milton. A
study of hypnotically induced complexes by means of

the luria technique. Journal of General Psychology,
193“, 11’ 65‘970

Lee, Wayne. Experimental design symbolization and model
derivation. Ps chometria, 1966, 31(3), 397-412.

Luria, A. R. The nature of human conf icts, W. Horsley
-Gantt (trans.). New Ecrk: Liveright, Inc., 1932.

Moore, U. P. The effects of posthypnotic stimulation of
hostility upon motivation. American Journal of

Clinical Hypnosis, 196k, 7, 130-13 .

Myers, Jerome. Fundamentals of experimental des;g .
Boston: Allyn and Bacon, 19 .

54

12.

13.

14.

15.

16.

1?.

18.

19.

20.

21.

22.

23.

55

Orne, M. T. The nature of hypnosis: artifact and

essence. Journal of Abnormal and Social Psychology,
1959. 58. 277-299.

Orne, M. T. On the social psychology of the psychological
experiment with particular reference to demand
characteristics and their implications. American

Psychologist, 1962, 17, 776-783.

Perkins, Kenneth, Repression, psychopathology, and drive
representation: an experimental hypnotic investi-
gation of the management of impulse inhibition.
Unpublished doctoral dissertation, Michigan State
University, 1965.

Reyher, Joseph. A paradigm for determining the clinical
relevance of hypnotically induced psychopathology.

Psychological Bulletin, 1962, 59, 344-352.

Reyher, Joseph. Posthypnotic stimulation of hypnotically
induced conflict in relation to psychosomatic
reactions and psychopathology. Psychosomatic
Medicine, 1963, 23, 384-391.

Reyher, Joseph. Free imagry: an uncovering procedure.

Journal of Clinical gsychology, 1963, 19(4), 454-459.
Reyher, Joseph. Hypnosis in research on psychopathology,

In: Handboog of clinical and egperimental hypnosis,

Jo Gordon 1 e 0 New York: M30m1118n, InCe, 19 7e
Reyher, Joseph and Smeltzer, William. Uncovering preper-

ties of visual imagery and verbal association: a

comparative study. Jouypal 9f Abnormal gsychology,
1968, 73(3), 218-222. '

Reyher, Joseph and Morishige, Howard. Electroencephalo-
gram and rapid eye movements during free imagry and

dream recall. Journal of Abnormal Psychology, 1969,
74(5). 576-582-

Sheehan, P. W. Artificial induction of posthypnotic
conflict. Journal of Abnormal Psychology, 74, 16-25.

Shor, Ronald and Schatz, Joel. A critical note on barber's

case-study on subject j. Journal of Psychology, 1960,
50. 253-256-

Veenstra, Glenn, The effectiveness of posthypnotically-
aroused anger in producing psychOpathology. Unpub-
lighed master's thesis, Michigan State University,
19 9.

56

24. Wolberg, Lewis. Hypnotic experiments in psychosomatic
medicine. Psychosomatic Medicine, 1947, 9, 337-340.

was

”If repression is assumed to be the apposite of verbali-
zed awareness, then an index of repression can be devised
simply by counting instances of verbalized awareness:

B. FA +ZZT1;A + cc - CR
where R is the average degree of repression over the experi-
mental sessions: FA, full awareness upon c-word recognition:
PA, partial awareness upon c-word recognition (awareness of
either the hostility or destructive impulse); CC, conscious
correlates of the unconscious hostility upon c-word recog-
nition, such as feelings of annoyance with the task of recog-
nizing words; CR, complete repression upon c-word recognition
(no instances of symptoms or awareness): and Tc, the total
number of conflict words recognized.

In view of the omnipresent possibility of suppressiOn of
the verbalizations of experienced affect upon the recognition
of c-words, instances of full awareness were considered to be
more important in assessing the average degree of repression
than were partial awareness . . . Conscious correlates are
assumed to indicate the mislabeling of the conscious repre-
sentations of the relevant dynamics. The S is aware of some

affect, but it is weak enough so that his cognitive appraisal

57

58

of it may be distorted. Instances of c-word recognition with
no verbalized awareness and no symptoms were assumed to be
the outcome of complete repression.“

(Reyher, 1967, 130-131)

APPEND X B

Reyher's Symptom Classification Scale

1.

2.

3.

Symptoms characterized by the dominance of autonomic

system enervation such as feelings of nausea, gastric

distress, headache, tiredness, sleepiness, tachycardia,

pressure in head, sweating, skin disturbances, flushing,

organ dysfunctions, heaviness, temperature alterations,

and such feelings as ”queasy“ and 'antsy.“

Symptoms dominated by enervation of the somatic or mus-

cular nervous system such as stiffness, aches, pains,

tension, tics, tremors, physical discomfort, and so on.

Disturbances of affect

a.

b.

0.

d.

Flattening: lack of feeling, apathy, and the like
upon the recognition of a c-word, when symptoms
usually attend e-word recognition.

Superego reactions: feelings of being alone, aban-
doned, ashamed, depressed, disgusted, guilty,
worried, and so on. '

Inversion: definite feelings of well-being on the
recognition of a c-word.

Alienation: feelings that seem weird, strange, odd,

unreal, unnatural, foreign, and so on.

59

5.

7.

8.

60

UnSpecified distress that cannot be clearly categorized as

either physical or emotional in nature, in S's frame of

reference, and are eXpressed in such conventional terms

as being upset, fidgety, jittery, nervous, on edge,

restless, and bothered.

States of emotional agitation that reflect the reaction of

the ego to the threat of complete breakdown of repres-

sion, such as feelings of anxiety, fear, apprehension, and
terror.

States of confusion, doubt, and disorientation that

include statements that one's thoughts are being pushed or

pulled and that the content of thought cannot be specified.

Dissociative reactions

a. Somatic and ideational delusions such as limbs feeling
detached, "crazy” thoughts and intruding paranoid
ideas.

b. Strong compulsive urges not carried out in behavior,
such as wanting to move hands around, scratch at
something, and so on . . . .

c. Compulsive destructive urge acted out in behavior
without awareness of the relevant hostile or des-
tructive impulse, such as hitting or picking at the
papers without knowing why, including destructive
acts not directed at the papers.

Disturbance or distortion in perception of the tachisto-

scopic stimulus.

9.

10.

11.
12.
13.

61

Derivatives of the induced conflict. These are symbols
of the induced experience and/or the repressive forces
themselves.
Conscious correlates of the unconscious hostility, such
as feelings of irritation, annoyance, and frustration.
Delayed awareness of one or both aSpects of the conflict.
Immediate awareness of one aSpect of the conflict.
Immediate and complete awareness of both aSpects of the
conflict.

(Taken from Reyher, 1967)

aunts m

HICHIGRN STRTE UNIV. LIBRARIES

31293100200868