‘lulllflllllllll lillllll ‘

5649

 

 

 

 

 

LIBRARY

Michigan State
University

 

 

 

This is to certify that the

thesis entitled

The Role of Annihilation Anxiety

in Schizophrenia
presented by

Mark E. Heim

has been accepted towards fulfillment
of the requirements for

Masters—degree in W

Major professor

Bertram P. Karon, Ph.D.
Date October 30, 1996

 

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

PLACE Ii RETURN BOX to rornovo thb checkout from your roconi.
TO AVOID FINES return on or boioro am duo.

DATE DUE DATE DUE DATE DUE

NOV :2 r
I L .LI 7:“
i
, if“. ,1 7
.4 ill? gr." I I |

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

:l
1 || |

 

 

 

 

 

MSU ioAnAfiinnotivo AetioniEm-i Opportunity institution
Wan-no.1

 

THE ROLE OF ANNIHILATION ANXIETY IN SCHIZOPHRENIA

BY
Mark E. Heim

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
MASTER OF ARTS
Department of Psychology

1996

ABSTRACT
THE ROLE OF ANNIHILATION ANXIETY IN SCHIZOPHRENIA
BY
Mark E. Heim

Annihilation anxiety (AA) is defined as a fear of ego
disintegration or a loss of self which may include the fear
of non-existence, and has been speculated to be a central
component in psychotic disorders and relevant
symptomatology. Hurvich (see Hurvich, Benveniste, Howard, &
Coonerty, 1993) developed the Rorschach Content Scale (RCS)
to measure AA from projective tests. One goal of this study
was to check the reliability and ease of use of the ROS. The
second goal of this study was to use the RC8 to compare
Rorschach protocols from 20 college students to 20 diagnosed
schizophrenics. Results revealed that the RC8 is a reliable
measure that can be used with minimal training.
Additionally, as predicted, it was shown that the
schizophrenic group had a significantly higher level of AA
than the college student group. Results suggest that AA
should be a recognized component in the diagnosis and
treatment of psychotic disorders.

Hurvich, M., Benveniste, P., Howard, J., & Coonerty, S.

(1993). Assessment of annihilation anxiety from

tests. Perceptual and Motor sgills, 77, 387-401.

ACKNOWLEDGMENTS

I would like to express my sincere appreciation to
Bertram Karon for his mentorship and support throughout this
project--Thank you very much.

To Michael Teixeira, thank you for your support and
encouragement as well as fostering the ideas that were built
upon in this thesis.

For your nurturance that helped me persevere, my
gratitude is extended to Abby Gleason, Stephen Jefferson,

and my parents.

iii

List of Tables .

Table of Contents

List of Appendices .

Chapter
I.

II.

III.

Introduction.

A.

Theoretical Discussion

of Annihilation Anxiety .

B. Annihilation Anxiety in Schizophrenia .

C. Symptoms of Schizophrenia Consistent
with Annihilation Anxiety.
1. Catatonia .
2. Nightmares.
3. Hallucinations.
4. Suicide .

D. The Quality of Object Relations Measured
in Rorschach Protocols.

E. The Assessment of Annihilation Anxiety in
Rorschach Protocols: The Development of
the Rorschach Content Scale .

F. Study Overview.

Method.

A. Participants.

B. Materials .

C. Procedure .

Results .

A. Interrater Reliability.

8 Group Differences .

C. Effects of Demographic Variables.

D Frequency of Themes of Annihilation Anxiety

iv

Page
vi

vii

. 14

. 19
. 19
. 20
. 21
. 22

. 24

. 26
. 3O
. 31
. 31
. 33
. 33
. 35
. 35
. 36
. 37
. 38

Table of Contents (cont’d) Page

IV. Discussion . . . . . . . . . . . . . . . . . . . . 38
A. The Psychometric Properties
of the Rorschach Content Scale. . . . . . . . 38

8. Theoretical and Clinically Relevant

Findings Regarding Annihilation Anxiety . . . 41
References . . . . . . . . . . . . . . . . . . . . . . . . . . 44
Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Appendices . . . . . . . . . . . . . . . . . . . . . . . . . . 54

Table

List of Tables

Significant Mean Differences of

Participant Characteristics .

Summary of Mean Group Differences and

Partial Correlations on Different AA Scores .
Demographic Predictor Variables of the percent of AA

within Groups .

Summary of R08 Categories Chosen Overall and by Groups.

Summary of the ROS Responses on the Rorschach Cards.

vi

Page

. 49

. 50

. 51
. 52

53

Appendix
A. The Hurvich Experience Inventory.
B. Experiential Themes of Annihilation Anxiety .
C. Discussion and Rational for Controlling for
Rorschach Protocol Length .
D. Informed Consent Form .
E. Training Procedure for the Rorschach Content Scale.
F. Calculation of Socio-Economic Status.

List of Appendices

vii

Page

. 54
. 55

. 56
. 58
. 59
. 61

 

 

The Role of Annihilation Anxiety in Schizophrenia
Introduction and Overview of the Present Studv

Annihilation anxiety has been discussed by several authors
for many years in the literature (e.g., Chadwick, 1929;
Fairbairn, 1952/1994; Guntrip, 1969/1989; Kohut, 1977;
Winnicott, 1965). However there has been little agreement of
its precise meaning or its etiology. Hurvich (1989; 1991)
surveyed the literature and compiled the major definitions of
what constitutes annihilation anxiety. From his work,
annihilation anxiety can generally be defined as a fear of ego
disintegration, or a loss of the self. That is, annihilation
anxiety is an experience that may be equated to feelings of non-
existence, impending danger or dread. Annihilation anxiety may
be aroused in response to either conscious or unconscious
malevolence from caretakers or noxious environmental stimuli
that threaten or impinge upon the infant. This fear may then
persist throughout life and detrimentally impact object
relations. Further, annihilation anxiety is believed to be the
earliest, most basic, form of anxiety that people experience.
It originates in early infancy, even predating what Freud
(1926/1959) discussed as the earliest experienced anxiety, that
is, the fear of the loss of the object.

It has been postulated that when a person has been
repeatedly exposed, over an extended length of time, to an
environment that would likely produce annihilation anxiety, he
or she may be more likely to develop a subsequent psychotic
disorder, especially schizophrenia. In fact, schizophrenia and

its symptom constellation may best be understood as a response

2
to, or way of coping with, annihilation anxiety (e.g.,
Bettelheim, 1967; Karon & Vandenbos, 1981).

This view is in contrast to the commonly held belief that
schizophrenia is primarily a biological disease(i.e., brain
dysfunction[neurotransmitter disorder) (e.g., Tsuang, Faraone, &
Day, 1988). From this viewpoint, researchers, propelled by a
medical model, may ignore or downplay the role of annihilation
anxiety. For example, in the Diagnostic and Statistical Manual
of Mental Disorders (1994) fourth edition, there is no mention
of any sort of anxiety or fear listed as a criteria for
recognizing and classifying schizophrenia, or any of the
psychotic disorders. Further, it is often believed that
schizophrenics have no affect (or have flat or blunted affect).

The following thesis shall begin by discussing the
theoretical background of annihilation anxiety as the first,
most basic, fear of life. This view will be contrasted with
Freud’s (1926/1959) revised theory of traumatic and signal
anxiety. Next, annihilation anxiety will be discussed as it
relates to schizophrenia, as well as the symptoms commonly seen
in schizophrenia. The focus of this study is to measure
annihilation anxiety from Rorschach protocols gathered from both
college students and schizophrenics using the Rorschach Content
Scale (RCS) devised by Hurvich (see Hurvich, Benveniste, Howard,
& Coonerty, 1993).

One goal of this thesis is to determine the usability of
the ROS by college undergraduates as well as to perform checks
on reliability. Second, assuming acceptable reliability of the
(ROS, this study will attempt to demonstrate that the

schizophrenic group will have higher scores reflecting

3

annihilation anxiety on the ROS than a reference college student
group. Thus, the present study is an attempt to support the
view that the schizophrenic is terrified of his or her world.
This fear will be discussed in terms of the driving force behind
the commonly seen symptoms in schizophrenia
Theoretical Discussion of Annihilatibn Anxietv

Freud’s (1926/1959) monograph Inhibitions. Symptoms and
Anxiety, which contains his revised theory of anxiety, posited
that there are two forms of anxiety: traumatic anxiety, and
signal anxiety. This revised theory of anxiety is the most
familiar in psychoanalytic circles and is often accepted and
incorporated into treatment modalities by therapists (e.g.,
Eagle, 1984; Paolino, 1981). As such, it will be used as a
comparison theory to contrast against the theory of annihilation
anxiety. As described by Freud, traumatic anxiety happens
automatically in response to a perceived experience of
helplessness or overwhelming on the part of the ego. These
traumata may happen often in early infancy, especially in an
undeveloped ego that lacks the organization necessary to
discharge or regulate an influx of stimuli. Over time, as the
ego develops, it is able to “signal,“ or predict, the possible
advent of a traumatic situation before it occurs full-blown.

Thus, anxiety can serve an adaptive, self-protective role,
in that it can initiate a fight-or-flight reaction, or trigger
appropriate defenses. The advantages of being alerted by such a
signal seem clear. Freud (1926/1959) called the situation in
which a person is signaled by anxiety a "danger-situation." He

.stated ”The signal announces: ’I am expecting a situation of

4
helplessness to set in,’ or: ’The present situation reminds me
of one of the traumatic experiences I have had before.
Therefore I will anticipate the trauma and behave as though it
had already come, while there is yet time to turn it
aside"(p.102).

Freud (1926/1959) specified that there are four basic
danger-situations common to early life: fears of loss of the
object, loss of love of the object, castration fears, and fear
of superego disapproval. When an infant (or toddler/young
person--as the case may be) experiences a perceived threat of
one of the four danger-situations, he or she is signaled to
anticipate (and hopefully avert) an experience of an earlier,
remembered traumatic situation of helplessness during which the
immature ego was overwhelmed.

Hurvich (1989) took issue with Freud’s four basic dangers
and contended that the fear of annihilation (e.g., overwhelming
and mortal danger) should constitute a basic danger situation.
Hurvich stated, “Based on the assumption that overwhelmed
helplessness can at some point be anticipated, I believe it
qualifies as a basic danger; indeed, as the first basic danger,
of which the later dangers, beginning with the loss of the
object, may be derivatives, and partial transformations”
(p.313).

To support his contention, Hurvich (1989) pointed out that
Freud discussed the fear of loss of the object (the first basic
danger) as a displacement from the growing tension due to need
onto the mother. Therefore, what subsequently becomes feared is

. the mother’s absence (because her presence was associated with

5
reducing tension). Thus, although the mother is recognized as a
protector against danger, Hurvich (1989) pointed out that "Freud
does not see the fear of the mother’s loss as the first danger.
The fear of being overwhelmed or annihilated precedes, and is
present during, the development of the fear of object loss."
(p.314).

Teixeira (1995) observed that Freud, in discussing the so-
called "wolf-man case,“ commented that oral phase anxiety is
manifested as a fear of death. Thus Teixeira pointed out that
Freud’s commentary suggested "that the fear of death, like
separation anxiety, is one of the earliest, and therefore, most
overwhelming of human anxieties" (pp. 4-5).

Other psychoanalytic authors after Freud (see below) have
also discussed experiences of fear and anxiety that occur before
the danger situation of the loss of the object as delineated by
Freud. This early, most primitive anxiety is often referred to
as annihilation anxiety in the literature, and its felt
experience in most likely akin to what Freud (1926/1959)
described as an overwhelming of the ego, or state of utter
helplessness.

According to Winnicott (1965), the infant requires the ego-
support from maternal care in order to survive. The maternal
care supplements the infant ego during the dependent state until
it becomes strong enough that the infant can detach and
differentiate into a separate personal self.

During the dependent state, Winnicott (1965) described the
infant as having an "inherited potential" of growth given the

‘ necessary maternal care. The earliest necessary maternal care

6
involves creating a holding environment which minimizes adverse
impingements upon the infant. If the holding environment is not
adequate, the infant will experience feelings of annihilation to
his personal being. Winnicott (1965) stated:

As a result of success in maternal care there is built up

in the infant a continuity of being which is the basis of

ego-strength; whereas the result of each failure in
maternal care is that the continuity of being is
interrupted by reactions to the consequences of that
failure, with resultant ego-weakening. Such interruptions
constitute annihilation, and are evidently associated with

pain of psychotic quality and intensity. (p.52)

Additionally, Winnicott (1965) described the dependent
infant as "living on the brink of unthinkable anxiety,“ and it
is the mother’s (or primary care-givers) responsibility, via the
holding environment, to prevent lapses into anxiety. Winnicott
provided further descriptions of unthinkable anxiety as going to
pieces, falling forever, having no relationship to the body, and
having no orientation.

Kohut (1977) made a distinction between two different
classes of anxiety experiences. He described the first class of
anxiety as being experienced by persons with a relatively
cohesive self, and the anxiety is in response to danger
situations as specified by Freud (see above). Kohut emphasized
the point that the experience of this first class of anxiety is
related to the danger situation and not to the state of the
self.

The second class of anxiety, according to Kohut (1977), is

7
"experienced by a person who is becoming aware that his or her
self is beginning to disintegrate; whatever the trigger that
ushered in or reinforced the progressive dissolution of the
self, the emphasis of the experience lies in essence on the
precarious state of the self and not on the factors that may
have set the process of disintegration into motion" (p. 102).
This description of the disintegration of the self is consistent
with what has been defined as annihilation anxiety.

Fairbairn (1952/1994) posited that when a child expresses
libidinal needs to a ridiculing or rejecting mother, the result
will be a devastating affective experience--one of discharging
libido into an emotional vacuum. When this happens to an older
child there will be feelings of intense humiliation over the
deprecation of his love. When the child is younger, Fairbairn
stated:

At a somewhat deeper level (or at an earlier stage) the

experience is one of shame over the display of needs which

are disregarded or belittled. . . His sense of his own

value is threatened. . . At a still deeper level (or at a

still earlier stage) the child’s experience is one of, so

to speak, exploding ineffectively and being completely
emptied of libido. It is thus an experience of

disintegration and of imminent psychical death. (p. 113)

Fairbairn (1952/1994) also discussed that separation-
anxiety is the earliest form of anxiety experienced by the
infant, and further, that this separation-anxiety is based on
the prototype of birth trauma--the first, original separation

_experience. Although Fairbairn denied any conscious memory of

8
the birth experience, he likened post-natal trauma, and
subsequent separation-anxieties, to a deep mental reactivation
of acute anxiety that was first modeled by the birth trauma.

It is interesting to note that Fairbairn (1952/1994)
described the period of separation as tumultuous. As the young
child transits between states of identification versus
separation with the primary object, both experiences may be
fraught with anxiety. In the state of identification, the child
may have a feeling of retention and bursting; whereas in the
state of separation there may be an experience of being emptied
or drained. From these descriptions of separation and
identification, the anxiety experienced seems to fit within the
definition of annihilation anxiety.

According to Guntrip (1969/1989), the core of psychological
distress is fear. When a weak infantile ego cannot cope with an
inadequate or traumatic environment, it will experience a
regressive longing and a state of profound infantile dependence.
Further, Guntrip proposed that although the regression may be an
attempt at security, in another sense it may be felt as
annihilation in that the regressive wish to return to the womb
may be felt as the wish to die.

Guntrip (1969/1989) delineated three situations that may
result in intense anxiety and a subsequent withdrawal (i.e., a
regression, or a schizoid state) from the outside world. First,
the primary care-giver may tantalizingly refuse to satisfy the
infants libidinal needs. Second, the care-giver may impinge on
the infant and overwhelm him or her. Third, the infant may be
‘ rejected and neglected. Given this third situation Guntrip

9
stated "In this case the danger of ego-loss and
depersonalization is at its maximum. . . [this situation] leads
to the most profound regression of all, which the patient can
experience as dying and death" (p. 75).

Melanie Klein also believed that infants only a few months
old are capable of experiencing fear so intense that it may
amount to a fear of death and a sense of absolute annihilation.
From the infant’s perspective, it’s ego maintains a relationship
with two part-objects--an ideal breast and a persecutory one.
Whereas the good breast (i.e., the nurturing mother) provides
love and nourishment and keeps persecution at bay, the infant
may also experience more than a mere lack of gratification; it
may experience a threat of persecution which is akin to
annihilation. (e.g., Guntrip, 1969/1989; Segal, 1964). Thus,
both Guntrip and Klein theorized about infantile fear and
anxiety in a period of life that would antedate Freud’s first
basic danger of loss of the object.

From a less theoretical perspective, there is evidence that
a child’s idea of being destroyed or annihilated may be
permeated in our culture in various manifestations. For
example, in discussing childhood punishments, Chadwick (1929)
stated:

Present-day parents may still in their anger make use of

threats and punishments that show a more or less direct

reference to the same idea [i.e., killing the child], which
call forth the cry from the child, "Then I will go away and
kill myself instead.” The parent who hears this will often

say that he or she cannot remember the words which provoked

10

the outcry of the child, which is an interesting example of

convenient forgetting, but one may surely assume that it

was something pertinent, and in the symbolic punishments
still in use in the modern nursery we may find the similar
idea of banishment representing death, in the sending out
of the room, away from sight, into the corner, where the
child cannot see others and its face is hidden, which seems
regarded by the child as a serious punishment as well by
its elders. One will remember in this connection that in
the Bible death is frequently referred to as hiding the
face, or being no more seen, sometimes as the result of God

hiding His face from those who had done wrong. (p. 326)

Another manifestation of annihilation anxiety may be
present in fairy tales. Chadwick (1929) discussed the role of
fairy tales as being both entertainment as well as being
instructional and serving admonitions to children. That is, in
most fairy tales good children are rewarded whereas bad children
are punished (e.g., stuffed in an oven, or eaten), often with
death as a consequence. Thus the child must work to please and
win the affection of the parents. If this struggle is lost,
then life itself may seem uncertain.

Bloch (1978, 1985) has proposed that infants have a
universal potential fantasy of being killed or annihilated in
some manner. These fantasies can act as an outlet to channel
feelings of rage, fear of abandonment, and other infantile
terrors. To wit, in the child’s fantasy, the fear of being
killed by his or her parents is displaced onto monsters and

_ imaginary creatures, thereby preserving the idealized image of

11
the parent, and to some extent, ensuring his or her own
survival.

Thus, as discussed above, several authors (mostly object-
relations theorists) have discussed an experience of anxiety in
infantile life that predates anxiety related to fear of the loss
of the object--Freud’s conception of the earliest fear--that may
be best described as annihilation anxiety.

Eagle (1984) made the observation that both Freud and the
object relation theorists (i.e., Winnicott, Kohut, Fairbairn,
and Guntrip) describe the ultimate anxiety as an experience of
annihilation via a threat to the integrity of the self.

Although the two classes of theorists do not disagree about the
result of intense feelings of anxiety, they do disagree about
the source of where these feelings originate. Regarding the
source of the threat, Eagle stated:

[For Freud] it is an excessive degree of excitation

resulting from, to put it as simply as possible,

accumulating undischarged instinctual drive tensions. For

[object relation theorists] the threat to the self results

from sensed and actual defects'and weakness in the

structure of the self interacting with situations in which
one experiences oneself either isolated from or merged with
the needed object. . . there is no "libidinal danger" to be
contrasted with danger to the self. There is only danger
to the self and different conceptions regarding the source
of this danger. (p. 46)
McCarthy (1981) also discussed that the fear of death is

equated with experienced threats to the self or to the fear of

12
the loss of the self. Like Eagle (1984), McCarthy also
distinguished classic psychoanalysts from contemporary
Freudians. To wit, the classic analysts attribute the fear of
death to guilt, the death instinct, castration anxiety, and
Oedipal guilt; whereas the contemporary Freudians (i.e., object-
relationists) are more likelyto stress a fear of maternal
destructiveness.

As witnessed above, annihilation anxiety has been discussed
by various psychoanalytic authors, both new and old, without a
unified conceptualization or definition of the concept. Hurvich
(1989), after a review of several works in the literature on
annihilation anxiety summarized that:

The major meanings of annihilation anxiety in the writings

of these authors are the fears of ego disintegration, the

loss of the self, the loss of identity and of personal
characteristics, the loss of the object world, breakdown of
self and object representations, the loss of control over
ego functions, the disintegration of the self, and the

perception of deficit. (p. 317)

Further, in a later paper, Hurvich (1991), addressed the

fact that there are several considerations that need to be taken
into account when attempting to define annihilation anxiety.
One consideration includes the level or organization and
developmental period of a person. In the worst case scenario,
Hurvich described annihilation anxiety as:

a virtually intolerable anxiety experience, felt and

believed to be over psychic survival and accompanied by

13

fears of imminent death or psychological destruction.

There is a sense of helplessness in the face of an utterly

frightening danger experienced as having no foreseeable

end. It is a danger against which the person feels he or
she can take no constructive action and that threatens to
overwhelm and disorganize him or her. It may lead to
panic, paralysis, and other maladaptive responses rather
than to effective or adaptive behavior; in some cases,
primitive fight-or-flight reactions occur, such as blind
assaultiveness, running headlong into oncoming traffic, or

jumping out of windows. (p.139)

In sum, the discussion has primarily focused on an
experience of anxiety that is tantamount to a feeling of
annihilation. It is acknowledged that the different authors do
not share the same precise meaning of annihilation anxiety
(Hurvich, 1991), though they do seem to capture the same general
phenomena. Thus, it was discussed that annihilation anxiety is
rooted in infancy in response to: an experience of impingement
from care-takers (e.g., Winnicott); the display of needs to a
rejecting care-taker (e.g., Fairbairn, Guntrip); or to the
anticipation of the dissolution of the self (e.g., Kohut). As
will be discussed later, these differing conceptions of
annihilation anxiety are used by Hurvich et a1. (1993) to
develop the Rorschach Content Scoring manual for annihilation
anxiety, which is described in greater detail below.

At this point it should be noted that this early anxiety
experience may persist throughout life, although in muted,
.disguised, and defended against forms. Annihilation anxiety may

14
exist in differing degrees in each individual and the extent to
which it persists throughout life may also run along a continuum
(Hurvich, 1991). Freud (1926/1959) similarly discussed that the
danger-situations and the resultant anxiety can work throughout
life at periods beyond the prescribed danger-situation, and
further, the danger-situations can exist along side one another
and compound the anxiety. Also, Freud discussed that the root
of the anxiety (i.e., which danger-situation is being
experienced) may be played out in the form that an ensuing
neurosis may take.

In addition to individual differences, certain groups of
people may have experienced differing degrees of annihilation
anxiety early in life. Central to the present thesis is the
role of annihilation anxiety in schizophrenia, and how an
understanding of annihilation anxiety may help make sense of the
constellation of symptoms commonly seen in schizophrenia.
Increased Annihilation Anxietv in Schizophrenia

Having established a general description and theoretical
background of annihilation anxiety, the focus will now be
shifted to a discussion of annihilation anxiety as it applies to
schizophrenia. Hurvich (1989) hypothesized that annihilation
anxiety is dominant in the psychotic disorders, whereas anxiety
over object loss and loss of love tends to be central in
borderline conditions, and castration and superego anxieties and
fears of loss of love are characteristic of neurotic disorders.

Several clinicians and researchers (e.g., Bettelheim, 1967;
Bloch, 1985; McCarthy, 1981; Teixeira, 1984) have observed that

lthe fear of annihilation (expresSed as a fear of death, a fear

15
of overwhelming, etc.) is a central component underlying, and
antedating, schizophrenia and psychotic disorders. Also, there
is some evidence which indicates that annihilation anxiety may
often immediately precede a psychotic decompensation in
schizophrenia (e.g., Karon & Vandenbos, 1981; Teixeira, 1984).

Winnicott (1965) discussed the role of parental impact on
the development of psychotic disorders as follows:

The mental health of the individual, in the sense of

freedom from psychosis or liability to psychosis

(schizophrenia), is laid down by [adequate] maternal care.

[Thus] schizophrenia or infantile psychosis or a

liability to psychosis at a later date is related to a

failure of environmental provision. (pp. 49-50)

Further focus on the role of maternal care as it relates to
schizophrenia is provided by Karon and Vandenbos (1981), who
stated that:

Schizophrenic pathology is usually the result of a pattern

of unconsciously malevolent parenting from the earliest

days of infancy onward. It is not merely the result of
isolated traumatic experiences, but of a pattern of
pressures that continues throughout childhood in somewhat
changing form. The basic problems that begin in infancy
are strengthened rather than reduced by the continuing
interaction between the preschizophrenic child and his or
her parents, particularly the mother. The child is the
victim of a series of subtle and unsubtle rejections, the
end effect of which is to make him or her feel worthless,

unlovable. But to be literally unlovable means that mother

16

will not love you, that she will abandon you and, to the

infant, this means pain and death. This is the infantile

terror that lurks behind the schizophrenic symptoms. The
schizophrenic individual’s_whole life is organized around

the need to defend psychologically against this danger. (p.

74) '

The data relevant to the hypothesis of malevolent parenting
as a precursor to schizophrenia are summarized by Karon and
Widener (1994). After reviewing a series of studies, they
reported that there was a consistent finding in families of
schizophrenic patients: there was an unconscious dynamic or
defense that was labeled "pathogenesis.“ Pathogenesis occurs
when the needs of the parent and the needs of the child are in
conflict. More specifically, the parent of a preschizophrenic
child, more often than other parents, acts in terms of the his
or her needs without regard to the potentially conflicting needs
of the child.

In their own study, Karon & Widener (1994) obtained a score
of pathogenesis as follows. Stories on the TAT were marked
pathogenic if there was an interaction between a dominant and a
dependent character with the potential for conflict, and the
dominate character does not take the dependent character’s needs
into account. Results revealed that mothers of normals average
a score of 35 percent pathogenic; whereas mothers of
schizophrenics average a score of 65 percent pathogenic.

Whereas Karon and Vandenbos (1981) discussed the effects of
unconscious parental malevolence, Kahr (1993) used

_psychohistorical research on diagnosed schizophrenics and found

17
that many of his patients had experienced profound death threats
and attempts on their lives in childhood and adolescence. He
likened this phenomena to "psychological infanticide" and
considered this to be a modern day version of past (though not
so ancient) “actual infanticide.‘ That is, many parents today
may still have the same "desire" to annihilate their infant, but
due to increased impulse control, different social standards, or
legal repercussions, their desire has been tempered and
transformed.

Teixeira (1984), working from an interpersonal perspective,
discussed how the effects of pathogenic experiences between
parent and child may be felt as annihilation anxiety to the
child and thus act as a harbinger to schizophrenia. He stated:

Characteristic and repeated pathogenic developmental

experiences (involving separation, deprivation, neglect,

rejection, hostility, conflict, criticism, and
intrusiveness) in the pre-schizophrenic's significant early
childhood relationships are internalized with traumatic
affect and the experience of “victimization" and

"helplessness" to form pathological experiential models of

the world, the significant others, and the self. . . . In

relation to the ontogenetic interpersonal experiences of
trauma and frustration, the pre-schizophrenic individual
has been conditioned to expect, and anticipates, criticism,
hostility, intrusiveness, rejection, and helplessness in
relation to others. These conditioned expectations in turn

arouse schizophrenic annihilation anxieties of "being

18

killed” in relation to others. (pp. 378-379)

From an empirical standpoint, further evidence has also
been published that demonstrated that fears and anxieties of
annihilation are a correlate of schizophrenia. For example,
Khanna, Khanna and Sharma (1988) found that on Templer’s death
anxiety scale, the trend was that schizophrenics rated higher on
death anxiety than manic-depressives or normals. Further, when
the scale was broken down into subcomponents, schizophrenics had
significantly higher scores than normals on “fears of personal
death,” and “concern about suffering and lingering death.“

Planansky and Johnston (1977) selected 80 schizophrenic men
out of 205 men that were admitted to a mental hospital over an
18 month period that explicitly expressed a fear of death or of
being killed. A semi-structured interview and daily
observations were used to determine fears and preoccupations
concerning death. Although the men were considered to be a
heterogeneous group (with the exception of having served in the
Armed Forces), the authors found that the fear of death
displayed a continuity throughout the whole course of a
psychoses. Further, Planansky and Johnston (1977) reported
that:

[The 80 men] were all tormented by the same primitive,

diffuse fear, or revealed the same constricted choice of

delusional constructs and projections, and employed the
same imagery for the description of their anguish. This
uniformity in their expressions of the fear of death could
have been brought about only by the force of psychotic

changes, independent of personal, cultural, or formative

19
influences. . . Since the dissolution of personality is the
basic event in all modes of active schizophrenia, the fear
of death as an expression of the psychotic existential
anguish can be regarded as a core component of
schizophrenic psychopathology. (p.196)

Symptoms of Schizophrenia Consistent with Annihilation Anxiety
In addition to discussing annihilation anxiety as a

 

precursor and a correlate to schizophrenia, it is also
interesting to consider that many of the symptoms of
schizophrenia are congruent with fears of annihilation and
resulting anxieties (Bettelheim, 1967; McCarthy, 1981).
Catatonia

It is well known that when animals are in a life-
threatening situation (i.e., a fear of annihilation) they may
resort to playing dead, which increases their chances of being
left alone and not eaten by the predator. In humans, Chadwick
(1929) discussed catatonia as a means of escape from a hostile
enemy. By becoming rigid and acting as if dead, the person, if
only in fantasy, can pretend that he or she has control over the
situation and thereby is spared the anguish and humiliation of
acknowledging a power greater than their own. Without resorting
to catatonia as a means of escape, the person would have to
succumb to the greater power and would be filled with
intolerable anxiety (i.e., annihilation anxiety) at the thought
of being controlled or impinged upon, which is menacing to the
ego and may be equated with death.

Catatonia has been frequently likened to a state of
. paralyzed fear in schizophrenia. Kahr (1993) observed that in

20

the schizophrenic patients he encountered, they often sought
refuge in a catatonic stupor for days or weeks at a time. He
speculates that the catatonia serves two functions: first, it is
a means of withdrawing from a terrifying world of interpersonal
relationships, and second, the catatonia serves as a way of
communicating an internal deadness to the surrounding world.

Karon and Vandenbos (1981) discussed the catatonic stupor
as the last line of defense in both animals and humans. Animal
research has revealed that nearly 70 percent of animals can
survive an attack by resorting to an immobile state, as if dead.
Subsequently, the predatory animal may disengage the attack and
save its "kill“ for later. Evolutionary advantages of the
catatonic defense can easily be seen. This emergency reaction,
commonly seen in schizophrenics, is the last stage of defense
when faced with an inescapable threat of violent death.
Nightmares

Nightmares are another symptom, or phenomena, that may be
indicative of annihilation anxiety and that are common in pre-
schizophrenics and schizophrenics alike. Levin and Hurvich
(1995) reviewed the literature on nightmares and reported that
nightmares can be so severe as to be likened to brief psychotic
episodes. Frequent nightmares (i.e., once per month with
chronicity) may be a sign of underlying psychological distress,
and further, Levin and Hurvich stated that investigators have
speculated that persons with schizophrenic disorders display a
continuity between frequent nightmares and waking
psychopathology. I

Levin and Hurvich (1995) also stated that it has been

21
speculated that childhood nightmares may predict adult
schizophrenia. Underlying these nightmares may be a sense of
overwhelming anxiety, panic, feelings of fragmentation, and a
sense of annihilation. It has been theorized that nightmares
may develop as a result of the internalization of early parental
objects who have been inconsistent and\or malignant. That is, a
child with these destructive internalizations may have weak or
disjointed ego boundaries which may leave him or her feeling
unsafe and possibly fragmented. As a result of this process the
child may experience manifest anxiety, general fearfulness, and
a heightened fear of death that is subsequently expressed in
nightmares.

To empirically validate the theories and speculation about
nightmares being an expression of underlying annihilation
anxiety, Levin and Hurvich (1995) gathered data on over 1300
college students. They measured the frequency, vividness,
severity, and perceived distress of the participant’s
nightmares, and they also obtained a level of annihilation
anxiety via the Hurvich Experiences Inventory. Results in this
study clearly demonstrated that people prone to frequent
nightmares also experienced heightened levels of annihilation
anxiety as compared to participants that did not report frequent
nightmares. '

Hallucinations

Another symptom of schizophrenia similar to nightmares is
hallucinations. Hallucinations have been likened to waking
dreams or nightmares, except that the motivation behind the

_hallucination is stronger (Karon & Vandenbos, 1981).

22
Hallucinations may be rooted in infancy or very early childhood
when the infant or child was ignored, under which circumstances
he or she withdrew from all interpersonal relationships into a
state of autism. Subsequently, auditory hallucinations replaced
real human contact. Karon and Vandenbos pointed out that even
malevolent voices (i.e., auditory hallucinations) are better
than being alone. .

Similarly, Teixeira (1984) reported that schizophrenics may
withdraw as a defense against oVer-stimulation (e.g., maternal
impingement), and that this process may lead to hallucinations.
He stated:

Although withdrawal may reduce annihilation anxieties

aroused in relation to others, withdrawal resulting in

isolation and increased autism can only intensify
hallucinations of condemning, terrifying, malevolent
internalized figures and fantasies based on the distortion
of actual traumatic interpersonal experiences and patterns
of relationship, leading to exacerbation of annihilation

anxieties. (p.379)

Suicide .

Suicide, although not a symptom, may be considered a common
sequel in schizophrenia that has some connection to (or roots
in) annihilation anxiety. Statistically, there has been a
consistent finding that suicide is excessively high in
schizophrenia (Allebeck, Varla, & Wistedt, 1986). In fact,
suicide is the leading cause of premature death among
schizophrenics (Caldwell & Gottesman, 1992). Allebeck (1989)

_ reported that in one ten-year follow-up study of 1,190

23
schizophrenic patients, the suicide rate for schizophrenics was
more than ten times higher than that for the general population.

These exceedingly high rates of suicide may be understood
as the schizophrenic’s attempt at escaping unbearable, chronic
states of anxiety, terror, panic, and torment. Thus, the
schizophrenic individual may turn to suicide as an escape from a
more overwhelming fear of living with its conjoint annihilation
anxieties (Teixeira, 1995).

It has been observed that immediately prior to a suicide
attempt, the schizophrenic individual tends to be highly
agitated and anxiety prone (Drake, Gates, Cotton, & Whitaker,
1984; Planansky & Johnston, 1971). It could be hypothesized
that an underlying state of annihilation anxiety (although
perhaps in a transformed manner) is being experienced by the
schizophrenic individual from which he or she is seeking escape.

In fact, Planansky and Johnston (1971) stated that some of
the suicidal schizophrenic patients used in their study
spontaneously reported unbearable feelings of autistic
separation and alienation as the reason underlying their
suicidal urges. There may be a connection between these
reported feelings of alienation as compared with transformed
feelings of an annihilated non-existence.

Similarly, McCarthy (1981) also stated that suicide and
self-mutilation among schizophrenics can be understood as a
defensive attempt to blot out feelings of depersonalization and
inhibit further disintegration. He stated:

Their cutting themselves and attempting suicide may serve

to combat the fear of the loss of the self by acting out

24

the experience of the disintegration of the self.

Ironically, schizophrenic or borderline individuals who

attempt suicide under these circumstances could be

characterized as trying to kill themselves in order to stay

alive. (p. 23)

In summary, theory and research has focused on the early
parent-child interaction and how, under unfortunate
circumstances, this interaction (i.e., malevolent parenting and
resultant annihilation anxiety) may lay the groundwork for later
schizophrenia. This process is perhaps most clearly evidenced
in that the role that annihilation anxiety appears to be
underlying and contributing to some of the most common symptoms
seen in schizophrenics. Further evidence of disturbed object-
relationships in schizophrenia has been found in studies using
the Rorschach.

The Quality of Obiect Relations Maasured in Rorschach Protocola

Lerner (1986) discussed how a knowledge of object relations
within different forms of psychopathology help in understanding
the pathology as well as in determining a treatment process. He
further suggested that the Rorschach is an ideal tool to assess
an individual’s representational world. That is, the Rorschach
can be thought of as a microcosm of the larger world and is thus
useful in determining how a person perceives himself, others,
and their world.

Mayman (1967) focused on the use of content scoring on the
Rorschach to assess the quality and type of object-relations a

person may have. He summarized a study that was done at the

25
Menninger Clinic in which an attempt was made to see how well
ratings of psychopathology could be established based on content
scores (focusing on self and other representations) on the
Rorschach as compared to other clinical ratings. To keep the
focus on content, references to color, shading, location, and
other cues were deleted before-the records were scored. The
content scores were transposed to a scaled score using the
Lubcrsky Health-Sickness Rating Scale. Subsequently, the data
from the Rorschach correlated well (most correlations were
between .60 and .70) with eight other patient variables (e.g.,
severity of symptoms, ego strength). Thus, Mayman made the
claim that a person's level of pathology could be inferred from
Rorschach content (see also Krohn & Mayman, 1974, for an
additional study that validates the use of projective measures
to determine the quality of object relations).

Blatt, Brenneis, Schimek, and Glick (1976) conducted a
study that compared control participants to adolescent and young
adult psychotic inpatients on their perceptions of human objects
in Rorschach protocols. As compared to the controls, the
psychotic inpatients showed a differentiation in the objects
they reported. First, when they had reported realistic human
perceptions (as denoted by an F+ score), the objects were
perceived as more malevolent and destructive, as well as being
active-passive or active-reactive in their interactions.

Second, those human perceptions that were perceived as
unrealistic (scored F-) were seen as more benevolent and kind.
With these findings, Blatt et a1. suggested that for psychotics,
_ their interactions with reality are painful and threatening,

26
whereas their escape to fantasy provides them with a world where
their interactions are perceived to be safer and provide comfort
and peace. A replication of this study by Ritzler, Zambianco,
Harder, and Kaskey (1980), found nearly identical results.
The Assessment of Annihilation Anxietv in Rorschach Protocols:
The Development of the Rorschach Content Scala

Relevant to the present study, Hurvich, Benveniste, Howard,
and Coonerty (1993) devised a measure, the Rorschach Content
Scale (ROS), that assesses annihilation anxiety from projective
tests. Because the ROS grew out of a theoretical background, as
described above, and previous research on annihilation anxiety,
it is necessary to consider the history of this line of work.

Before developing the R08, Hurvich et al. (1993) devised
another measure to assess annihilation anxiety, the Hurvich
Experience Inventory (HEI, see Appendix A for a copy of this
measure). The HEI is a paper and pencil test that uses a 25-
item scale in Likert format on which participants indicate to
what extent certain descriptive events are consistent with their
experiences. The questions were derived from therapy sessions
with patients that would theoretically be expected to have
annihilation anxiety.

In several different samples of participants Hurvich et al.
(1993) reported that the HEI was found to have alpha
coefficients consistently above .80, which is indicative of good
internal consistency. Also, when the researchers compared HEI
scores across different groups of participants, the mean group
scores reflected the expected differences between the patients.

. That is, a college student group received a lower mean score on

 

 

27
the HEI (indicative of less annihilation anxiety), as compared
to psychiatric patients (who received the 2nd highest score) and
phobics (the highest scoring group).

Next, when scores from the HEI have been compared to other
established measures of anxiety, the correlations have been
respectable. For example, college students’ scores (n=205) on
the HEI yielded a product-moment r of .69 with trait anxiety and
.56 with state anxiety on the Spielberger State-Trait Anxiety
Inventory. A comparison 218 participants scores on the HEI
revealed a correlation of .63 with the Taylor Manifest Anxiety
Scale. Thus, Hurvich et al. (1993) stated that the HEI seems to
be measuring what is generally agreed upon in clinical
psychology as anxiety.

Further studies were carried out with the HEI on more
unique groups of participants (e.g., blood phobics, abuse
victims, agoraphobics, frequent nightmare sufferers) to
determine if the measure is getting at general anxiety, or a
more distinct anxiety akin to annihilation anxiety. This series
of studies is reported in Hurvich et al. (1993), after which
they conclude that the HEI
appears to have good internal consistency, some construct
validity, and that it specifically measures annihilation
anxiety.

Encouraged by the results with the HEI, Hurvich et al.
(1993) decided to construct a second instrument, the R03, to
assess annihilation anxiety from projective measures. The R08
is presented in manual form and consists of nine interrelated

_ categories that Hurvich et al. determined were indicative of

28
varying perspectives of annihilation anxiety from an extensive
review of the literature.

These nine categories include: (1) the fear of being
overwhelmed or engulfed, (2) fear of merger, (3) fear of
disintegration, (4) fear of impingement, (5) fear of loss of
needed support, (6) fear of inability to cope, (7) fear of loss
of self-cohesion, (8) concern over survival, and (9)
catastrophic mentality. Hurvich et al. (1993) provided an
exemplary account of the content within the domains of the nine
categories which is reproduced in Appendix 8. Additionally, a
copy of the R06 is attached to this thesis.

The nine separate but interrelated categories listed above
are what constitute the ROS. Each of these categories are
further broken down into subcategories and specific items. As
an example, the first category, “The Fear of Being Overwhelmed,"
has six subcategories that further describe the category
concept. If a Rorschach response fits the description, a score
of 1.0 is marked for annihilation anxiety for Category I. Each
response is subsequently scored as 0.0 or 1.0 for annihilation
anxiety, and each positive answer is equally weighted.
Additionally, each positive answer can subsequently be scored
from 1 to 3 to indicate the severity of the annihilation
response, with 3 being the most severe. The front page of the
ROS manual provides the instructions and examples for
classifying the severity.

Construct validity for the R08 was supported by comparing
scores on the R08 with scores obtained on the HEI. In this
(initial study, Hurvich et al. (1993) used twenty participants

29
from a college psychology clinic and outpatient hospital
facilities. The Rorschach, theThematic Apperception Test, as
well as the HEI were administered to the participants. After
matching the zero points on the ROS and the HEI, the correlation
coefficient between the two measures was .56 (p < .01) for one
rater and .47 (p < .05) for a second rater. The interrater
reliability for the two raters was .94 when the presence or
absence of annihilation anxiety was assessed. The interrater
reliability dropped to 82 percent agreement when the raters had
to determine which of the nine categories (described above) best
represented the participants responses.

Further psychometric integrity of the ROS was supported by
a study carried out by Benveniste, Papouchis, Allen, and Hurvich
(1995). In this study, the ROS, as well as several other
measures of anxiety, ego functioning, and reality testing were
used to assess annihilation anxiety in groups of college
students, in inpatients with diagnosed Borderline Personality
Disorder (BPD), and in inpatients with diagnosed Schizophrenia
(N=75, n=25 in each group). As predicted, their results showed
that both the BPD and schizophrenia groups scored significantly
higher for annihilation anxiety as measured on the ROS than the
college student group.

When Benveniste et al. (1995) did three checks on
interrater reliability for two raters, the results showed
Pearson correlations above .95 for the total number of
annihilation responses in a record, for the presence or absence
of annihilation anxiety in each response, as well as for the

agreement on which category specific annihilation responses

30
belonged in (all correlations significant at the p < .001
level). Additionally, the researchers organized ten
psychologists and had them rate the RCS’s nine categories for
content validity. Overall there was an 81 percent agreement
that the nine 808 categories were content valid. The range of
agreement ran from 6 of 10 to 10 of 10 judges in agreement on
the ROS categories.
The Present Studv

The present study had two goals. To begin with, this was
be the first attempt to determine if the Rorschach Content Scale
(RCS) recently devised by Hurvich et al. (1993) is able to be
used outside the originator’s research group. To provide an
indicator of the RCS’s ease of use, this study trained 2
psychology undergraduates to use the instrument, which will also
provided a measure of reliability.

The second goal of the study, if the RC8 was found to be
usable with sufficient reliability (above 80 percent agreement),
was to determine if there was a difference in the amount of
annihilation anxiety between a group of schizophrenics as
compared to a college student control group. It was predicted
that the schizophrenic group would have a higher mean score for
the percent of annihilation anxiety in their protocols, as
measured by the ROS, when compared to the college student group.

A percentage score of AA was used because it was
anticipated that the schizophrenic group would give
significantly briefer protocols than the college student group
based on previous research (e.g., Benveniste et al., 1995; Karon

& Vandenbos, 1981). To help control for this difference, a

31

group mean was calculated based on the percent of responses for
each participant’s protocol scored positive for AA on the ROS.

A second method of controlling for differences in protocol
length (R) is to use a partial correlation that controls for the
number of Rorschach responses. This method is consonant with
suggestions in the literature regarding how and when to
normalize for different protocol lengths (Appendix 0 contains a
discussion of the literature on the rational for controlling
differences in Rorschach protocol length). When holding R
constant by partial correlation, it was hypothesized that being
in the schizophrenic group would be a significant predictor of
both a higher total score of AA, and a higher score of AA when
including the severity scores_(i.e., the 1 to 3 ratings).

The hypotheses stated above are in line with recent
findings from Benveniste et al.’s (1995) unpublished study.
However, the present study used different statistical
methodology (i.e., partial correlation as well as percentage
scores) and individuals from a different geographical location,
thus extending the validity of the RC6. The present study also
explored the effect of socio-economic status on AA.

Method
Earticipants
There were two groups of participants: a college student
group, and a group of diagnosed schizophrenics. The first group
was made up of twenty Michigan State University undergraduate
students that were volunteers from the psychology department
subject-pool. All students received extra-credit for their

participation, and for some it was a course requirement.

32

By putting restrictions on the volunteer sign-up sheets, an
attempt was made to match the college student group with members
in the schizophrenic group (discussed below) on the demographic
variables of age, race, and gender. In addition, this study
used mostly freshmen students to reduce the discrepancy in the
education level.

The second group was made up of a subgroup, chosen for
their age, of twenty out of a possible 36 participants that had
participated in the Michigan State Psychotherapy Research
Project (see Karon & Vandenbos, 1981). The project originally
consisted of 36 participants selected from patients admitted to
the Detroit Psychiatric Institute in the 1960’s. As part of the
project, the participants were assigned to different treatment
groups and their progress was monitored over eighteen months.
They were given a battery of measures throughout the project and
the present study used the Rorschach protocols that were
administered by advanced psychology graduate students before
treatment began.

The participants in the project were selected on the basis
of being: (1) unquestionably schizophrenic; (2) they had an
onset of blatant psychotic symptoms within 3 months prior to
admission; (3) this was their first admission; (4) they had no
ECT or insulin shock treatment, no organic brain damage, and no
history of alcohol or drug abuse (Karon & Vandenbos, 1981).

They were all selected within a four-month period and were
primarily poor, inner-city, and African-American. 0f the 33
participants used in the project (three of the original

participants were dropped due to drug or medical complications),

33
20 were selected for use in the present study to allow for
optimal matching in age to the college student group (Karon &
Vandenbos, 1981, report complete participant characteristics on
pp. 392-393).
Materials

The Rorschach Ink Blot Test was administered to the college
student group. As mentioned above, the Rorschach Content Scale
for annihilation anxiety was used on the Rorschach protocols
from both the college student group and the schizophrenic group
to obtain quantitative scores of annihilation anxiety.
Procedure

For the college student group, the Rorschach Ink Blot Test
was administered by the author to the college students. These
students were contacted by phone to set a meeting time and were
informed that they would be participating in research that
required them to take the "ink blot" test. They were not
informed of the study hypothesis or that their protocols would
be scored for annihilation anxiety.

When a participant arrived for their meeting, they were
engaged in a brief rapport-building session and basic
demographic factors were recorded (i.e., age, education, parents
education and occupational status). Additionally, the
participants were informed that their participation was
voluntary and that were able to terminate participation at any
point. This information was also included on the informed
consent form, contained in Appendix D, which each participant

was required to sign.

34

The Rorschach was administered in accordance with the
instructions discussed in Beck, Beck, Levitt, and Molish (1961).
These were the same instructions that were used to administer
the Rorschach to the schizophrenic participants. The
instructions are essentially the same as the more commonly used
method of administration discussed in Exner (1993).

The administration of the Rorschach took approximately 1.5
hours, after which the participant was thanked for their help
and general questions about the Rorschach were answered if the
participant was curious. '

After collecting the Rorschach protocols from the college
student group, those from both participant groups were typed and
printed in an identical format and each participant was randomly
assigned a subject number. The protocols were then intermixed
and scored blindly using the ROS. Two senior-level psychology
undergraduate students were enlisted to participate as volunteer
research assistants (one assistant offered his help while the
researcher was acting in the capacity of a teaching assistant,
the other was a co-worker of the researcher). After a one-month
training period in the use of the ROS (see Appendix E for a
complete description of the training procedure), each assistant
scored a random sample of 30 protocols as follows: Working
independently, the raters both scored 20 identical protocols to
provide a check of interrater reliability; additionally each
rater was assigned a random sample of 10 protocols that was not
given to the other rater. The assistants were blind to the
study hypotheses as well as to the source of the Rorschach

protocols.

35
Results
Intergater Reliability

The RCS was found to be developed enough to be used outside
the research group that developed the instrument, and to have
adequate interrater reliability: Two senior level undergraduate
psychology majors used the Rorschach Content Scale (R08) to
score a random selection of 20 out of the 40 Rorschach protocols
used in this study. A Pearson product-moment correlation was
used to check interrater reliability on two different scores on
the ROS. The first reliability check was performed on the two
rater’s scores of the total number of responses scored positive
for annihilation anxiety (AA) in each protocol. The two raters’
scores yielded a significant Pearson 2(17) = .89, g < .001. (An
alpha level of .05, used for all statistical tests in this
study).

The second reliability check was performed on a total AA
score for each protocol that included not just the sum of the
positive AA scores, but the sum of severity scores of AA when it
was present (each positive AA score was ranked from 1 to 3, with
3 indicating the most severe score of AA). The interrater
reliability of these scores yielded a significant Pearson g(17)
= .93, p < .001.

There was also an 83% agreement between the two scorers
regarding the presence or absence of AA for each Rorschach
response (i.e., the raters agreed that AA was present or absent
on 387 out of 465 responses that were included in the 20
protocols used for the reliability check). Further, there was a

78% agreement between the two scorers when they had to select

36

which of the nine categories best represented the type of AA in
the response (see Appendix B for a review of the nine different
categories). In subsequent statistics, for the 20 protocols
scored by both assistants, the 2 scores were averaged and then
the unified scores were used in conjunction with the other 20
protocols that were scored by only 1 rater.
Group Differences

Despite restrictions placed on the college student sign-up
sheets, there were significant differences between the 2 groups.
As shown in Table 1, the schizophrenic group had a slightly
lower education level and a much lower socio-economic status
(SES) (Appendix F contains a discussion of how SES was
calculated), and a higher mean age than the college student
group. Both groups were comprised of 15 African-Americans and 5
caucasians. The schizophrenic group contained 10 males and 10
females, the college student group had 9 males and 11 females.

As anticipated, there was a significant group difference in
the Rorschach protocol length (R), with the schizophrenic group
having a lower mean number of responses (M = 18.15, SQ = 15.02)
than the college student group (M = 28.60, _Q = 7.27), 3(38) =
2.80, p = .008, two-tailed (shown in Table 1). Additionally, R
was strongly and significantly correlated to both the total AA
score (Pearson £(37) = .77, p < .001) and the AA severity score
(Pearson 5(37) = .65, p < .001). Given these 2 conditions, as
discussed in Appendix C, R was normalized by using a group mean
percentage score of AA, as well as using partial correlations

that control for R, in the following analyses.

. 37

As shown in Table 2, the schizophrenic group, as
hypothesized, had a significantly higher group mean (M = 53.28,
SD = 25.50) for the percentage of responses scored positive for
AA as compared to the college student group (M = 38.29, §Q =
16.43), t (38) = -2.21, p < .033, two-tailed. Additionally, as
predicted, when the number of Rorschach responses was partialed
out, a subsequent Pearson correlation revealed that being in the
schizophrenic group was a significant predictor for having both
a higher score of AA (3(37) = .38, p = .017), as well as a
higher severity AA score (£(37) = .42, p = .008) (i.e., the AA
Total and AA Severity scores, respectively, as described in
Table 2).

Also shown in Table 2, when R was not taken into account,
there was not a significant difference in the total or severity
scores of AA between the groups.

The Contributing Effects of Demographic Vagiables on the
Percentage of AA Responses

Within the college student group, none of the variables in
this study (i.e., age, SES, race, education, gender) were
significant predictors for the percent of responses scored
positive for AA. Table 3 contains the data.

Within the schizophrenic group, the only variable that was
a significant predictor of the percent of responses scored
positive for AA was race (see Table 3). A significant Pearson
correlation 3(17) = .49, p = .027 indicated that Caucasian
participants (a = 5) were more likely to score higher for the
percent of responses reflecting AA than the African-Americans (fl

= 15) in the sample. Even though the groups were matched for

38
race, a subsequent partial correlation was performed on the data
from both groups that controlled for race which revealed that
being in the schizophrenic group was still a significant
predictor of having a higher percentage score of AA (Pearson
5(37) = .34, p = .034).
The Frequencv of the Themes of Annihilation Anxietv Chosen

For both groups of participants, the most frequently scored
categories of AA were 1) The fear of being overwhelmed or
engulfed; 2) The fear of disintegration, and; 3) The fear of
impingement. These three categories contained 75 percent of all
the AA responses given for both groups. Table 4 contains the
full breakdown of the frequency of responses in each of the 9
categories.

Discussion
The Psvchometric Properties of the Rorschach Content Scala

The Rorschach Content Scale (RCS) was found to be a
reliable instrument that could be used with minimal training
time by persons with undergraduate training in psychology.
Scores of interrater reliability and agreement of which category
represented the AA were very similar to the scores reported in 2
previous studies that used the R06 (Benveniste et a1. 1995;
Hurvich et al. 1993).

Additionally, Hurvich et a1. (1993) reported that the ROS
categories most frequently endorsed by the raters were the fears
of being overwhelmed, of disintegration, and of impingement,
which all together accounted for 72 percent of the Rorschach AA
scores. The present study found the same three categories, in

the same order of frequency, accounted for 75 percent of the

39
Rorschach AA scores. Similarly, Benveniste et al. (1995)
reported that the fear of being overwhelmed and of
disintegration were the two highest scoring categories (in that
order) for their schizophrenic group, and the same two
categories (in reverse order) were the highest endorsed for
their college student group.

The three lowest scoring categories in this study,
catastrophic mentality (not endorsed at all), inability to cope,
and the fear of loss of needed support, were also the least
endorsed in the Hurvich et al. (1993) study (similar results
were reported again in the Benveniste et al. 1995 study). As
mentioned in the Introduction section of this paper, Benveniste
et al. had 10 independent ratings from psychologists to assess
the ROS categories for content validity. It is noteworthy that
the 3 most used categories were amongst the highest rated for
content validity, and the 3 lowest used categories were amongst
the lowest rated for content validity.

In sum, the fears of being overwhelmed, of disintegration,
and of impingement make up the core of AA responses as they are
reflected on the Rorschach stimuli. The lesser used categories
of catastrophic mentality, inability to cope, and loss of
support may either reflect a lack of content validity for AA, or
the Rorschach stimuli may be inadequate to evoke responses that
reflect these categories. Benveniste et al. (1995) discussed
these low scoring categories as being representative of AA from
a theoretical or clinical perspective (i.e., from therapy
session and interview experience) that may not be reflected in

Rorschach responses.

40

Also of note, the college student group and the
schizophrenic group tended to report the same types of AA as
represented in the different categories. Thus it is not
possible to distinguish the groups based on the “type“ of AA as
reported on the Rorschach. As suggested by Benveniste et al.
(1995), the ROS may benefit from refining the scoring system
within the categories. For example, both “a human hand“ and
“rotting flesh“ would be scored positive for AA as a fear of
disintegration when it seems clear that the second response
reflects much stronger AA than the first.

On the Rorschach, cards IX and X accounted for the highest
number of 808 AA responses (frequencies = 81 and 80
respectively; see Table 5). These two cards are also the most
colorful, and often cause the most difficulty for people to
respond to (e.g., Exner, 1993). Hurvich et al. (1993), and
Benveniste et al. (1995) reported the same results in their
study and Hurvich et al. suggested that the affect-stimulating
color and less structured stimuli may result in responses that
reflect underlying AA. Rorschach card V was the least likely to
provoke AA responses in the present study as well as in the two
previous studies mentioned above. Card V is considered the
least ambiguous card and is monochromatic, and these factors
most likely account for its low "pull“ of AA responses.

Overall, the results of this study using the R08 are very
similar to, and in many cases identical to, the two previous
studies that used the ROS (i.e., Benveniste et al., 1995;
Hurvich et al., 1993). The result were similar in respect to

the magnitude of difference of the percent of AA responses

41
between college students and schizophrenics; in respect to the
most frequent and least frequently chosen categories; and in
respect to which Rorschach cards accounted for the most and the
least number of positive AA responses. These are very
encouraging results for a rather new measure that Hurvich et al.
considers to be in preliminary form.
Theoretical_and Clinically Relevant Findings Reoaggiflg
Annihilation Anxiety

The present study supports the theory that people with
schizophrenia live in a world of fear. As hypothesized, when
presented with ambiguous stimuli (i.e., Rorschach cards),
schizophrenics were more likely than a reference group of
college students to impose meaning on the stimuli that
corresponded to themes of annihilation anxiety (AA). This is
consistent with the theory that schizophrenics interact with a
negative transference to the world around them (e.g.,
Bettelheim, 1967; Karon & Vandenbos, 1981).

The results of this study are similar to the results found
by Benveniste et al. (1995). In their study, a college student
reference group was found to have 29% (versus 38% in this study)
of their Rorschach responses scored positively for AA and a
schizophrenic group scored 41% positive (versus 53% in this
study) for AA. Thus the Benveniste et al. study revealed that
12% more of the Rorschach responses-were scored positive for AA
in the schizophrenic group, versus 15% in this study. Although
the present study tended to score higher overall on the
percentage of responses positive for AA than Benveniste et al.,

there is a significant and reliable difference which reveals

42
that schizophrenics are more likely than the reference group to
interpret their world through lenses of fear and annihilation.

Additionally, in the Benveniste et al. (1995) study, the
college student group and the schizophrenic group did not differ
significantly in education. This lends support to the finding
in the present study that education is a non-significant
contributor to AA.

It is important to recognize that the schizophrenic
population has a significant amount of AA. This is often
overlooked by researchers and clinicians that hold to a medical
model of pathology. The oversight of AA may be best highlighted
by reviewing the Q§M;;M criteria.for schizophrenia. 0f the five
"characteristic symptoms“ of schizophrenia, there is no mention
of increased fear, anxiety, or AA. In contrast, the Q§M;;M does
list “affective flattening“ or “avolition” as a characteristic,
as well as disorganized or catatonic behavior. This study
provides support for the contention that what appears to be
affective flattening, as well as a catatonic state, may be
better understood as responses to extreme fear and AA.

In addition to rethinking the criteria for diagnosing
schizophrenia, understanding that AA may be a central component
in schizophrenia may also lead to a different attitude in the
treatment of this disorder. By conceptualizing a
schizophrenic’s symptomatology as reactions to intense fear, it
may be possible for the clinician to focus interventions on the
underlying fear. Thus, the clinician may be able to strengthen
the therapeutic alliance by addressing the negative

transference, and subsequently relieve related symptomatology.

43

It has also been accepted that schizophrenia results in
impaired information processing and cognition (e.g., Spohn et
al., 1985). Zilboorg (1943) points out that if the fear of
death is constantly on a person’s mind (i.e., constantly
conscious), they would be unable to function normally. Thus, if
the clinician addresses an underlying fear of AA when working
with schizophrenics it may result in cognitive improvement and
less disorganization.

On a cautionary note, care must be taken not to approach
therapy with a schizophrenic person in hopes of indiscriminately
removing their symptomatology. McCarthy (1981) discussed that
the loss of a symptom may signify a loss of the self and evoke
feelings of death. In the case of schizophrenia this can be
especially problematic given their heightened fear of death and
annihilation to begin with. Thus it may be useful when working
with a schizophrenic to discuss with them that you will not take
away anything from them that they believe they need--over time,
as the alliance strengthens, the patient may relinquish their
symptoms.

The idea that annihilation anxiety has been a central
component in schizophrenia has long been held by many clinicians
based on clinical experience and observation (e.g., Bettelheim,
1967; Karon & VandenBos, 1981; Teixeira, 1984). The present
study extended empirical support to these long-believed

contentions.

L IST OF REFERENCES

44

References

Allebeck, P., Varla, A., & Wistedt, B. (1986). Suicide and
violent death among patients with schizophrenia. AQLQ
Esychiatrv Scandinavia, 74, 43-49.

Allebeck, P. (1989). Schizophrenia: A life-shortening disease.
Schizophrenia Bulletin. 15.1, 81-89.

American Psychiatric Association. (1994). gmagnostic and
statistical manual of mental disorders (4th ed.).
Washington, 00: Author.

Beck, S.J., Beck, A.G., Levitt, E., & Molish, H. (1961).
Rorschagh’s test: Basic processes (3rd ed.). New York:
Grune & Stratton.

Benveniste, P., Papouchis, N., Allen, R., & Hurvich, M. (1995).
Rorschachyassessment of annihilation anxiety and ego
functioning. Unpublished manuscript, Long Island
University, Brooklyn.

Bettelheim, B. (1967). The empty fortress: Infantile autism and
the birth of the self. New York: The Free Press.

Blatt, S., Brenneis, C., Schimek, J., & Glick, M. (1976).
Normal development and psychopathological impairment of the
concept of the object on the Rorschach. Journal of
Apnormal Psvchologv.,85.4, 364-373.

Bloch, D. (1978). "So the witch won’t ea;_ma:. Boston:
Houghton Mifflin.

45

Bloch, D. (1985). The child’s fear of infanticide and the
primary motive force of defense. Psychoanalytic Review,
za.4. (Winter), 573-588.

Caldwell, 0., & Gottesman, I. (1992). Schizophrenia-A high-
risk factor for suicide: Clues to risk reduction. Suicide
and Life-Threatening Behavior. 22.4. (Winteri, 479-493.

Chadwick, M. (1929). Notes upon the fear of death.
International Journal of Psvchoanalysis, 10, 321-334.

Cronbach, L.J. (1949). Statistical methods applied to Rorschach
scores: A review. Psychological Bulletin.46, 393-429.

Drake, R., Gates, 0., Cotton, P., & Whitaker, A. (1984). The

 

Mgurnal of Nervous and MentaliDisease. 17gy10, 613-618.

Eagle, M. (1984). Recent developments in psychoamalvsis: A
critical evaluation. Cambridge, MA: Harvard University
Press.

Exner, J.E., Jr. (1992). R in Rorschach research: A ghost
revisited. Journal of Personality Assessment, 58, 245-251.

Exner, J.E., Jr. (1993). The Rorschach: A comprehensive system
(vol.1, 3rd ed.). New York: John Wiley & Sons, Inc.

Fairbairn, W. (1994). Psychoanalytic studies of the
personality. New York: Roultedge.

Fiske, 0., & Baughman, E. (1953). Relationships between
Rorschach scoring categories and the total number of
responses. The Journal of Apnormal and Social Psychology,
58, 25-32.

Freud, S. (1959). Inhibitions. symptoms_and amxietv (A.
Strachey, Trans.; J. Strachey, Ed.). New York: W.W Norton.

46

Guntrip, H. (1989). Schizoid phenomama. object-relations amp_
the self. Madison, CT: International Universities Press.

Hurvich, M. (1989). Traumatic moment, basic dangers and
annihilation anxiety. Psychoanalytic Psychology, 6.3, 309-
323.

Hurvich, M. (1991). Annihilation anxiety: An introduction. In
H. Siegel, L. Barbanel, I. Hirsch, J. Lasky, H. Silverman,
8 S. Warshaw (Eds.), Psychoanalytic Reflections on Current
Issaas (pp. 135-154). New York: University Press.

Hurvich, M., Benveniste, P., Howard, J., 8 Coonerty, S. (1993).
Assessment of annihilation anxiety from projective tests.
Perceptual and Motor Skills. 77, 387-401.

Kahr, B. (1993). Ancient infanticide and modern schizophrenia:
The clinical uses of psychohistorical research. Ina
Jounmal of Psychohistory. 20.3 (Winter), 267-273.

Karon, B., 8 Vandenbos, G. (1989). Psychotherapy of
Schizophrenia: The treatment of choice. Northvale, NJ:
Jason Aronson.

Karon, B., 8 Widener, A. (1994). Is there really a
schizophrenogenic parent? Psychoanalytic Psychology. 11.1,
47-61.

Khanna, N., Khanna, R., 8 Sharma, R. (1988). A study on death
anxiety. Journal of personality 8 Clinical Studies. 4.1,
47-51.

Kinder, B.N. (1992). The problems of R in clinical settings and
in research: Suggestions for the future. Journalyof

Personality Assessment. 58, 252-259.

47

Kohut, H. (1977). The restoration of the self. New York:
International Universities Press.

Krohn, A., 8 Mayman, M. (1974). Object representations in
dreams and projective tests. Bulletin of the Menninger
Clinic, 38, 445-466.

Lerner, H. (1986). Research perspectives on psychotherapy with
borderline patients. Psychotherapy. 23, (spring), 57-69).

Levin, R., 8 Hurvich, M. (1995). Nightmares and annihilation
anxiety. Psychoanalytic Psychology. 12.2, 247-258.

Mayman, M. (1967). Object-representations and object-
relationships in Rorschach responses. gournal of
Projective Techniques. 31, 17-24.

McCarthy, J. (1981). Psychoanalytic psychotherapy and the fear
of death. The Americam_aournal of Psvchoamalvsis. 41.1,
21-30.

Myers, J.K., 8 Bean, L.L. (1968). A decade later: A follow-ap_
of social class and mental illness. New York: John Wiley
and Sons, Inc.

Paolino, T. (1981). Psychoanalytic psychotherapy: Theory.i
technique. therapeutic relationship and treatability. New
York: Brunner/Mazel, Publishers.

Planansky, K., 8 Johnston, R. (1971). The occurrence and
characteristics of suicidal preoccupation and acts in
schizophrenia. ACTA Psvchiatricia Scandinavia, 47, 473-
483.

Planansky, K., 8 Johnston, R. (1977). Preoccupation with death
in schizophrenic men. Qiseases of the Nervous System;

38.3, 194-197.

48

Ritzler, 8., Zambianco, 0., Harder, 0., 8 Kaskey, M. (1980).
Psychotic patterns of the concept of the object on the
Rorschach test. Journal of apnormal Psychology, 89.1, 46-
55.

Segal, H. (1964). Introduction to the work of Melanie Klein.
New York: Basic Books.

Spohn, R., Coyne, L., Mittleman, F., Larson, J., Johnston, M.,
Spray, J., 8 Hayes, K. (1985). Effect of neuroleptic
treatments on attention, information processing, and
thought disorder. Psychopharmacology Bulletin. 21.3. 582-
587.

Teixeira, M. (1984). Annihilation anxiety in schizophrenia:
Metaphor or dynamic? Psychotherapy. 21.3 (Faill, 377-381.

Teixeira, M. (1995, May). Confrontation with suicidality.and
the dark side of love. Paper presented at the meeting of
the Michigan Psychoanalytic Council, Grand Rapids, MI.

Tsuang, M., Faraone, S., 8 Day, M. (1988). Schizophrenic
disorders. In A. M. Nicholi, Jr. (Ed.), The new harvard
guide to psychiatry (pp. 259-295). Cambridge, MA: Belknap
Press.

Winnicott, D. (1965). The maturational processesmand the
Iacilitating environment: Studies in the theory of
amotiomal development. London: Hogarth Press.

Zilboorg, G. (1943). Fear of death. The Psychoanalytic
Qparterlv. 12. 465-475.

49
Table 1
Significant Mean Differences of Participant Characteristics

 

 

College Students Schizophrenics
(D. = 20) - (n = 20)
Characteristic M SQ M §Q
SES‘ 38.10 11.71 61.65 8.23
Education 14.10 0.97 10.55 1.93
Age 19.80 1.77 22.15 3.41
Number of Responses 28.60 7.27 18.15 15.02

 

Note. There is a significant difference between the groups on the
variables of SES, Education, Age, and Number of Responses (p < .01).
'The higher the number the lower the socio-economic status. See

Appendix F for a detailed description of how SES was calculated.

50

Table 2
Summary of Mean Group Differences and Partial CorrelaMions on Different

 

 

 

 

AA Scores
College Students (9:20) Schizophrenics (p=20)
Type of AA Score M SQ M SQ
AA Percent' 38.29 16.43 53.28 25.50
AA Total 10.85 4.95 9.53 9.00
AA Severity 14.88 9.15 15.48 17.44
Partial Correlations (Controlling for the Number
of Responses) Between Group Status and AA Scores
Type of AA Score Pearson r (M = 40)
AA Total' .38
AA Severity' .42

 

MpgaaQ * indicates a significant difference, p < .05.

--AA Percent = the mean score of the percent of responses of each
protocol scored positive for AA in each group.

--AA Total = the mean score of the number of responses that were scored
positive for AA in each group.

--AA Severity = the mean score of the sum of severity scores (the 1 to
3 score for each response marked positive for AA) of each group.

--In the partial correlation, being in the schizophrenic was predictive

of having the higher AA scores.

51
Table 3
Demographic Predictor Variables of the Percent of AA within Groups

 

Percent of Annihilation Anxiety

College Students Schizophrenics

 

E E
(n = 20) (n = 20)
1. Age .02 .25
2. Race -.40 .49*
3. Gender .18 ' .18
4. Education .37 .10
5. SES .22 .24

 

Note. All correlations nonsignificant except *Race\Schizophrenic group

(p = .03).

52

Table 4

Summary of the ROS Categories Chosen Overall and by Groups

 

 

College Students Schizophrenics Overall

a = 20 m = 20 m = 40

Category Frequency Frequency Frequency
Overwhelmed 68 60 128
Merger 24 13 37
Disintegration 42 57 99
Impingement 36 38 74
Lost Support 7 4 11
Unable to Cope 2 O 2
Loss of Self 24 12 36
Survival 5 8 13
Catastrophe 0 O 0

 

Note. For the 20 protocols that were scored by 2 raters, the average
score was used and added to the remaining protocols. Also, the
category names are abbreviated--please see Appendix B for complete

description of the categories.

53
Table 5

Summary of the RC6 Responses On the Rorschach cards

 

College Students Schizophrenics Overall

 

m = 20 p s 20 m = 40
Card Frequency Frequency Frequency
I 28 2O 48
II 27 - 36 63
III 39 25 64
IV 50 19 69
V 22 13 35
VI 34 30 64
v1 I 35 ‘ 25 so
VIII 33 30 66
IX 35 46 81
X 41 39 80

 

Note. For the 20 protocols that were scored by 2 raters, both of their

scores were included in the frequencies.

APPENDICES

54

Appendix A
Hurvich Experience Inventory

After reading each statement, decide which of the following best
describes your experience. Then put the number beside the statement.

EXAMPLE.--1. Never 2. Not Very Often 3. Often 4. Very Often.

1. I feel I’m going to shatter or fall apart.

2. I am very afraid of fear.

3. I wonder who I really am.

4. I worry about my survival.

5. I feel like I am destroyed as a person.

6. I have trouble falling asleep.

7. I am afraid to get emotionally close to others.

8. I feel terror and panic.

9. My body feels like it doesn’t belong to me.

10. I think about the world coming to an end.

11. I had frightening nightmares as a child.

12. I feel the dread of dying at any moment.

13. I feel that I have more than one self.

14. I feel intruded on, mentally or physically.

15. I keep searching for an identity I don’t quite have.
16. Time seems to run very fast or almost stand still.
17. I need someone to reassure me when I become afraid.
18. I worry about my physical health.

19. I feel I can’t pull myself together.

20. I have nightmares.

21. I feel like my mind is falling into bits.

22. As a child I was afraid of dying.

23. I have a feeling of falling in space.

24. When something makes me nervous it’s hard for me to get over it.
25. I feel like I am being overwhelmed.

Note. From “Assessment of Annihilation Anxiety from Projective Tests,"
by M. Hurvich, P. Benveniste, J. Howard, and S. Coonerty, 1993,
Perceptual and Motor Skills. 77, p. 400.

Note.
by M.

55
Appendix B
Experiential Correlates (Themes) of Annihilation Anxieties

Fear of bein overwhelmed or engulfed

A. Fear 0 overstimulation

8. Fear of reexperiencin a terrifying situation

0. Fear of disturbing af ects: anxiety\depression, anger guilt
0. Fear of inner con licts and pressures

Fear of merger

A. Fear of being swept up or lost in another person

8. Fear of losing one’s separate sense of self

C. Fear of loss of body boundaries

Fear of disintegration
A. Fear of falling apart

8. Fear of dissolving
0. Fear of shattering into bits
0. Fear of exploding

Fear of impingement

Fear of being devoured

Fear of being smothered

Fear of being trapped

Fear of being intruded upon physically or mentally
Fear of being controlled

ear of loss of needed support

Fear of falling

Fear of abandonment

Fear of re’ection

Fear of si ence, aloneness, darkness

nability to cope (loss of ego function)

Fear of being unable to deal with people

Fear of being unable to generate an organized response
Fear of bein unable to think

Fear of para ysis

Fear of facing problems one is incapable of mastering
Fear of going insane ~ ‘

WINDOW>H DOW)?! MDOCD>

Fear of loss of self-cohesion (fragmentation, depletion, or
enfeeblement)

A. Fear of being destroyed as a person

8. Fear of not being sure of who one is

C. Fear of gaps in the continuity of self-experience

D. Fear of suffering shame or humiliation

Concern over survival

A. Fear of fatal disease

8. Fear of fatal accident

C. Fear of environmental catastrophe--exaggerated (fires,

floods, tornados, earthquakes, nuclear holocausts, warfare)

Catastrophic mentality
A. Tendency to anticipate calamities from ordinary events
8. Tendency to exaggerate ordinary stresses

From ”Assessment of Annihilation Anxiety from Projective Tests,“
Hurvich, P. Benveniste, J. Howard, and S. Coonerty, 1993,

Perceptual and Motor Skills. 77, p. 401.

56
Appendix C

Qiscusaionyand Rational for Controlling for Rorschach Protocol Length

A debate has been carried out in the literature since the 1940’s
over how to account or control for differences in Rorschach protocol
lengths (known as 'R'), especially when conducting research (e.g.,
Exner, 1992). According to Exner (1992) most authors on this topic
have cited back to Cronbach (1949)and\or Fiske and Baughman (1953),
who addressed this issue and gave recommendations in how to deal with
differences in R.

Cronbach (1949) stated that different methods of analyzing data
from the Rorschach will lead to different results and that the analysis
should not be carried out indiscriminately based on the experimenter’s
subjective judgement. He addressed the issue of R as problematic,
especially if groups differ in R, stating that it may require some sort
or normalizing. Fiske and Baughman (1953) followed suit with Cronbach,
and both articles recommend normalizing R, usually by placing subjects
in different groups that have a restricted range of R.

Exner (1992) agreed that normalizing for R may be appropriate
when there is extreme variation, as there often was several decades
ago. Currently, Exner believes that normalizing for R is not usually
necessary because Rorschach protocols are much more uniform in R. This
uniformity is evident in Exner’s (1993) reference samples, where 700
adult non-patients have a mean R of 22.67 (SQ = 4.23), and 320
inpatient schizophrenics have a mean R of 23.44 (SQ = 8.66). Exner
(1992) further stated that even if there is a significant difference in
R between groups, logic (versus probability values) should dictate if
the magnitude is sufficient to warrant some sort of normalizing for R.
Given a meaningfully significant difference in R, Exner (1992)
suggested that a stronger case could made for normalizing R if it is

correlated with the variables under investigation. He went on to state

57

that using a percentage score may be too simplistic and suggested the
use of a partialing technique.

Relevant to this study, both percentage and partialling
techniques are used because of the magnitude of difference in R between
the schizophrenic group and the college student group (R=29 versus 18,
respectively). And further, R is significantly correlated with the 2
different scores of annihilation anxiety (AA only 5(37) = .77; AA Total
3(37) = .65, p’s <.001) such that as R increases the AA scores
increase. Thus, this study fits the conditions described by Exner

(1992) and others (e.g., Kinder, 1992) that allow for normalizing of R.

7.

58
Appendix D

Informed Consent
I have freely chosen to participate in psychological research
being conducted by Mark Reid and supervised by Dr. Bertram Karon.
I can contact Dr. Karon in 108 Psychology Research Building (or
call 355-2159) if I have any questions regarding my participation
in this study. '

I have been informed that I will be taking the Rorschach Ink Blot
test and that this process will require approximately 1 hour of
my time.

It has been explained to me that my responses to the ink blots
will be written down and scored for certain psychological
concepts. I have also been informed that my responses will be
kept strictly confidential, and that a subject number instead of
a name will be used on all written material; only the primary
researcher will have access to the names and corresponding
numbers and he will keep this information under look.

I understand that I can discontinue my voluntary participation at
any time, and further, that there are other options besides
research participation to fulfill course requirements (e.g.,
writing short papers).

I understand that any known risks to my participation have been
explained to me and that I agree to participate in this research.
I also understand that this research does not guarantee any
beneficial results to me.

It has been explained to me that I can learn more about the
Rorschach Ink Blot after my participation in the study.
Additionally, I can receive information about the results of this
particular study at a later date (early summer) if I provide
optional mailing information below.

I have been provided a copy of this consent form, and I
understand that an identical copy will be kept on file.

Signed: .
Witness:

Date:

 

 

Optional Information (to obtain results)

Name:

 

Address:

 

 

59

Appendix E
Ipaining Procedure for the Use of the Rorschach Content Scalp

The primary investigator met a total of 5 times with both of the
research assistants for approximately 2 hours per meeting over a one
month span. The first meeting consisted of informing the assistants
that they would be required to learn the use of the Rorschach Content
Scale (ROS) and would subsequently have to score a total of 30
protocols--a task requiring 20 to 40 hours of their time after
training. Time was also spent informing the assistants about the
concept of annihilation anxiety (AA) and how the RC8 was developed, and
they were given relevant articles to read before the next meeting
(i.e., the articles by Hurvich 1989, 1991). The remaining time was
spent going through the ROS page by page to become familiar with the
instrument. The researcher and the assistants were then given the task
to score three protocols before the next meeting.

For the next 4 meetings, the researcher and the assistants went
through each Rorschach response and compared answers. When their was a
disagreement about whether or not 8 responses reflected AA, or which
category the response should be placed in, a discussion ensued until a
consensus was met. If a consensus could not be agreed upon, the
researcher made the final decision. This procedure worked very well
and additional examples were added to the R08 that helped reduce
discrepancies at each subsequent meeting.

Almost 300 responses were eventually compared and by the last
meeting there was approximately a 90 percent agreement between the
three raters. Additionally, it became clear by the end of the third
meeting that certain responses resulted in confusion as to which of the
nine categories best represented the type of AA (e.g., a response that

included a fight or conflict). Subsequently, the researcher made an

60

addendum to the R08 to clarify some differences and help insure
consistency in the scores. The addendum is added below.

RORSCHACH SCORING MANUAL FOR ANNIHILATION ANXIETY

ADDENDUM

Liam
Score arguing under category I C (Fear of Being Overwhelmed--Disturbing
affects or conflicts)
Score actual fights under category I 8 (Fear of Being Overwhelmed--
Overstimulation) unless the fight includes a stabbing, gunshot, or some
other 'Impingement' then score under category IV.
Score a pending fight under category IV 8 (Fear of Impingement) if it
seems clear that the fight will lead to an impingement; if an impingement
is not evident or predictable, then score the pending fight under category
I B (Overwhelmed--out of control).
miscellaneous

Responses that don’t have a definite form quality (e.g., smoke,
clouds, bodies of water, heaven, hell) get scored under category II (Fear
of Merger). The exception to this would be 'disturbing affects“ (e.g.,
'I feel scared”, 'this gives me a
headache“) which would get scored under category I C (Overwhelming).

Also scored under category II (Merger) are “Pure C' responses--the
"C“ stands for color. An example here would be a response such as 'Oh,
green, blue, and yellow." (Note: the pure 0 must be the whole response--
not just a ”comment“ on the colors).

If you can’t figure out which category a response fits under, it may
help to back away from the details of the response and try to determine
more generally which category it 'would fit under by re-reading the
descriptions under the category titles. When this fails, list both
categories that you think are appropriate but please circle which one is

your main preference.

61
Appendix F

The Calculation of Socio-Ecgngmic Status

The participant’s socio-economic status (SES) was calculated using
Hollingshead’s Two Factor Index cited in Myers and Bean (1968).
Hollingshead originally used a three factor index to calculate SES which
included the occupation, education, and residential setting of the head
of household. The residential setting was dropped from the index because
it was determined that a detailed knowledge about the residential setting
within the city or suburb was necessary for this factor to be accurate.
Subsequently, the education (measured in years of completed schooling) and
occupation (divided into 7 categories to represent earning power) of the
head of household became the common factors used to assign an SES.

In this study, if the partiCipant was married and lived independent
from the parents, either the participant or the participant’s spouse,
whichever would result in the highest SES, was used in the calculation of
SES. If the participant lived with his or her parents or was dependent
upon them for support (as in the case of most college students), the
parent that placed the participant in the highest SES category was used
as the head of the household for calculations.

Both the education and occupation indexes are multiplied by
weighting factors then the two scores are summed. The range of scores run
along a continuum from 11 to 77, with the lowest score
equating to the highest SES. The range of scores are then subdivided into

five social class categories as follows:

Range of Computed Scores Social Class
11 - 17 I
18 - 27 II
28 - 43 III
44 - 60 IV
61 - 77 V

In this study the participant score (from 11 to 77) was used in
statistical calculations to eliminate the use of a categorical variable,

and subsequently improve statistidal inferences.