MSU

LIBRARIES
.—,—.

 

 

RETURNING MATERIALS:
Place in book drop to
remove this checkout from
your record. FINES wi11
be charged if book is
returned after the date
stamped below.

 

 

 

 

 

 

 

AN INVESTIGATION INTO THE RELATIONSHIP
BETWEEN ALEXITHYMIA, IMAGING CAPACITY

AND AFFECTIVE DIFFERENTIATION

BY

Terence L. Bradley

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements

for the degree of
DOCTOR OF PHILOSOPHY
Department of Counseling Psychology

1987

ACKNOWLEDGMENTS

I would like first to thank my parents, Scott and
Beatrice Bradley, for instilling in me the drive for
knowledge and truth that kindled the motivation for this
project. I would like also to express appreciation to my
committee members, Drs. Gershen Kaufman, Steven Raudenbush,
Frederick Lopez and, particularly, Dr. William Hinds as
advisor, committee chair and friend. Finally, I extend
heartfelt thanks to Geoff, Michelle, Carolin and Diane,
without who's patient assistance this project could not have

come to fruition.

ii

ABSTRACT

AN INVESTIGATION INTO THE RELATIONSHIP
BETWEEN ALEXITHYMIA, IMAGING CAPACITY
AND AFFECTIVE DIFFERENTIATION

by

Terence L. Bradley

The purpose of the study was to empirically examine the
relationship between alexithymia, as the dependent variable,
and affective differentiation and mental imagery as the
independent variables in a sample of health services
consumers. A secondary purpose was to examine how anxiety
and depression relate to alexithymia. The previous
literature reports clinically anecdotal validation of the
alexithymic concept with few attempts to empirically
establish it's psychological sequelae.

A total of 150 subjects drawn from two outpatient
medical clinics in southern California volunteered for full
participation in the study. The Schalling-Sifneos
Personality Scale was used as the measure of alexithymia.
The Beck Depression Inventory, State-Trait Anxiety
Inventory-Form Y, Differential Emotions Scale-IV, and the
Survey of Mental Imagery-Form A were administered as

measures of the independent variables. It was hypothesized

that affective differentiation and mental imagery would
demonstrate an inverse relationship with alexithymia while
anxiety and depression would not systematically covary with
alexithymia.

The study was of an observational nature using a
hierarchical multiple regression analysis to test the
research hypotheses. Independent variables were ordered
hierarchically prior to the regression analysis. F-tests
were performed in order to test the significance of
incremental variations. Bivariate analyses were performed
on the demographic variables by treating them as grouping
variables. A Bivariate analysis was also performed using a
new variable, alexithymia, which was created using a cutoff
score of 45 on the SSPS. The .05 level of significance was
used for all hypothesis tests.

The major findings of the study suggested that:

1) A significant and negative relationship exists between
alexithymia and scores on the SMI-A measure of mental
imagery.

2) A non-significant relationship exists between alexithymia
and scores on the DES-IV measure of affective
differentiation.

3) Non-significant relationships exist between alexithymia

and scores on the Trait and State measures of anxiety.

4) A significant and positive relationship may exist between
alexithymia and scores on the BDI measure of depression.

5) Non-significant relationships exist between alexithymia
and the demographic variables of age, sex, and race. A
significant and negative relationship may exist for
education.

6) The SKI-A was the single best predictor of SSPS scores in

the model tested.

TABLE OF CONTENTS

LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . .V
CHAPTER ONE - INTRODUCTION . . . . . . . . . . . . . . . .l

The Problem .1

Need. . . . . . . . . . . . . . . . . . . . . . . . .2
Theory. . . . . . . . . . . . . . . . . . . . . . . .5
Purpose . . . . . . . . . . . . . . . . . . . . . 17
Research Hypotheses . . . . . . . . . . . . . . . . 17
Overview. . . . . . . . . . . . . . . . . . . . . . 18

CHAPTER TWO - LITERATURE REVIEW. . . . . . . . . . . . . 19

Alexithymia . . 19

Affect Theory . . . . . . . . . . . . . . . . . . . 31
Mental Imagery. . . . . . . . . . . . . . . . . . . 41
summary 0 O O O O O O O O O O O O O O O O O O O O O 52

CHAPTER THREE - METHODOLOGY. . . . . . . . . . . . . . . 54

Sample. . . 55

Instruments . . . . . . . . . . . . . . . . . . . . 60
Procedures. . . . . . . . . . . . . . . . . . . . . 74
Hypotheses. . . . . . . . . . . . . . . . . . . . . 77
Research Design . . . . . . . . . . . . . . . . . . 78
Data Analysis . . . . . . . . . . . . . . . . . . . 79
Summary . . . . . . . . . . . . . . . . . . . . . . 85

CHAPTER FOUR - RESULTS . . . . . . . . . . . . . . . . . 87

Sample Descriptive Statistics . . . . . . . 87

Hierarchical Regression Analysis. . . . . . . . 89
Descriptive Statistics for Grouped Variables. . . . 93
Bivariate Analyses. . . . . . . . . . . . . . . . . 99
Hypothesis Tests. . . . . . . . . . . . . . . . . .102
All-Subsets Analysis. . . . . . . . . . . . . . . .108
Stepwise Regression Analysis. . . . . . . . . . . .108
Summary . . . . . . . . . . . . . . . . . . . . . .109

iii

TABLE OF CONTENTS (Continued)

CHAPTER FIVE - SUMMARY AND CONCLUSIONS . . . . . . . . .112

Summary . . . . . . . . . . . . . . . . . . . . . .112
Discussion. . . . . . . . . . . . . . . . . . . . .119
Limitations of the Study. . . . . . . . . . . . . .127
Implications for Future Research. . . . . . . . . .133
Summary . . . . . . . . . . . . . . . . . . . . . .135
REFERENCES . . . . . . . . . . . . . . . . . . . . . . .137

APPENDICES O O O C O C O C O O O O O O O O O O O O O O O 14 9

Appendix A (Schalling Sifneos Personality Scale). .150
Appendix B (Differential Emotions Scale IV) . . . .151
Appendix C (Survey of Mental Imagery-Form A). . . .154
Appendix D (State-Trait Anxiety Inventory-Form Y) .164
Appendix E (Beck Depression Inventory). . . . . . .166
Appendix F (Study Introduction) . . . . . . . . . .168
Appendix G (Consent Form) . . . . . . . . . . . . .169
Appendix H (Demographic Sheet). . . . . . . . . . .170
Appendix I (Means & Standard Deviations on

SMI-A Subscales x Alexithymia). . . . .171
Appendix J (T-tests on SMI-A Subscales

x Alexithymia). . . . . . . . . . . . .172
Appendix K (Means & Standard Deviations on

DES-IV Subscales x Alexithymia) . . . .173
Appendix L (T-tests on DES-IV Subscales

x Alexithymia). . . . . . . . . . . . .174

iv

Table

Table
Table
Table
Table

Table

Table

Table

Table

Table

Table

Table

Table
Table
Table

Table

3-5

4-1

4-2

4-3

4-4

4-6

4-7

4-8

4-9

4-10

4-11

LIST OF TABLES

Gender and SSPS Scores for
Partially-Participating Subjects . .

Age of Subjects. . . . . . . . . . .
Race of Subjects . . . . . . . . . .
Gender of Subjects . . . . . . . . .
Educational Level of Subjects. . . .

Sample Means and Standard Deviations
SSPS, BDI, and STAI-Y(State & Trait)

Sample Means and Standard Deviations
SMI -A o o o o o o o o o o o o o o o 0

Sample Means and Standard Deviations
DES-IV o o o o o o o o o o o o o o 0

Multiple Correlation Coefficients and
Variance Proportions for the Hierarchical

Regression Analysis. . . . . . . . .

Analysis of Variance for Multiple
Correlation Coefficients . . . . . .

Correlation Matrix of Predictor
Varialbes for Full Model . . . . . .

Regression Table for Hierarchical
Analysis . . . . . . . . . . . . . .

Descriptive Statistics for Age Variable.

Descriptive Statistics for Sex Variable.

Descriptive Statistics for Race Variable

Descriptive Statistics for Education
Variable . . . . . . . . . . . . . .

56

57

58

59

59

88

88

89

89

90

91

92

94

95

96

97

Table

Table
Table
Table

Table

Table

Table

LIST OF TABLES (Cont.)

Descriptive Statistics

Variable . . . . .
Bivariate Analyses
Bivariate Analyses
Bivariate Analyses

Bivariate Analyses
(five levels). . .

Bivariate Analyses
(two levels) . . .

Pairwise Contrasts

vi

for
for
for

for

for Alexithymia

Age (five levels)
Sex (two levels).
Race (two levels)

Education

Alexithymia

Education x SSPS.

. 98

.100

.100

.101

.101

.102

.108

CHAPTER I

INTRODUCTION

The Problem

It has long been recognized that a substantial number
of people constitute a significant drain on health care
resources by presenting physical conditions in which there
is either no underlying organic basis found or for which
standard medical treatments have little or no remedial
effect. Indeed, many of Freud's earliest psychoanalytic
cases revolved around baffling physical complications that
rendered the traditional medical treatments of the day
useless wastes of time and healthcare expenditure (Freud,
1895). Despite an anecdotally rich and extensive
literature that has purported a vast array of psychological
treatments since that time (adequately summarized by
Alexander, 1943; Wittkower, 1974; Wolff, 1968; and Kellner,
1975), the problem of health care management of the
so-called "psychosomatic" disorders continues largely
unabated to the present. Schreter (1978) has gone as far
as suggesting that up to 60% of the patients presenting in
health care settings today may present some form of
somatization disorder as described in the American
Psychiatric Association's Diagnostic and Statistical Manual
- 111. Our capacity to provide meaningful and cost
effective assistance has been far outstripped by an

increasingly overwhelming demand for health care services.

1

2
One approach to the problem has been developed by

Sifneos (1967), who collected observations of psychosomatic
patients under his care and launched a burgeoning
literature in the theoretical understanding and potential
treatment of these conditions. He coined the term
”alexithymia" to describe an unusually consistent cognitive
and affective patterning in his patients characterized by:
1) a relatively mechanistic thought pattern, dominated by
environmental detail and substantially devoid of fantasy
associated with feeling states; and 2) a relative inability
to experience, evoke and verbalize differentiated affective
states. Additionally, he noted that the patients tended to
be quite well adjusted to their roles in society but tended
to have rather superficial and shallow interpersonal
relationships. Since that earlier work, the disorder of
alexithymia has been found to overlap a number of
traditional diagnostic categories which include certain of
the psychosomatic disorders as well as substance abuse
disorders (Krystal, 1962 and 1970), severe post-traumatic
states (Krystal, 1971) as well as certain of the eating

disorders (Weiner, 1976).

Eggg
While the alexithymic concept has broadened our
understanding of the nature and function of many of the
psychosomatic disorders, clear explication of the

clinically observed cognitive and affective sequelae has

3
been difficult to isolate and identify (Krystal, 1979:

Leventhal and Tomarken, 1986; Nemiah, 1974 and 1977;
Sifneos, 1967 and 1977; Apfel, 1979; Blanchard, et. al.,
1981; Lolas, et. al., 1980; Martin, et. al., 1984; Shipko
and Noviello, 1984). This lack of clarity in the affective
and imaginal domains, as they relate to the clinical
manifestations of alexithymia and psychosomatic disorders
in general, has contributed to a clouded conceptual
articulation and a paucity of effective treatment
approaches (Sifneos, 1974, 1975, 1977, 1983; Stephanos,
1975, 1976, 1979; Shands, 1977, Schreter, 1978; Krystal,
1979; and Von Rad, 1979). Sifneos (1974 and 1975) has
summarized the difficulty with the use of traditional
psychodynamic therapies in the treatment of alexithymia as
these treatments uniformly rely heavily on the client's
ability to gain access to and experience a full range of
relatively well differentiated emotions, tolerate anxiety
that is inevitably generated in "uncovering" therapies, and
the ability to use fantasy in the forms of waking imagery
and/or dreams. The alexithymic person appears to come to
therapy deficient in each respect and is unable to
articulate these cognitive and affective deficits.

Krystal (1979) is more hopeful about positive
therapeutic intervention and recommends alterations in the

techniques used:

"The task is to help the patient to interpret,
organize and recognize his own feelings. One

4
often has to help patients.....to understand

what they are feeling and to give it a name." (pg.27).
and further:

"It is in this process of working with verbali-

zation of affects that one observes progress

toward affect differentiation and desomatiza-

tion." (pg. 29)

Clearly, advances in our theoretical and empirical under-
standing of the role and mechanisms of these interacting
affective and cognitive realms will lead to enhanced
therapeutic interventions and outcomes.

The relationship between, and relative
psychophysiological primacy of, emotion vs cognition has
been long debated in various theories of psychological and
physical well-being (Lazarus, 1984; Zajonc, 1984; Leventhal
and Tomarken, 1986; Izard and Schwartz, 1986; Tomkins,
1984; Plutchik, 1980 and 1984; Krystal, 1975a, 1975b and
1977). Additionally, there is a growing literature in the
psychophysiological mechanisms of mental imagery as it
relates to the mediation of a variety of acute and chronic
psychophysical conditions and psychoneuroimmunological
processes (Achterberg and Lawlis, 1984; Locke and
Hornig-Rohan, 1983; Schneider, et. al., 1983).

While there exists an abundance of theoretical
positions offered to explain isolated research findings,
there is a spate of unified theoretical understanding of
the function, development and interaction of these

affective and cognitive psychological domains. There

appears to exist a need to theoretically articulate and

5

empirically investigate the relationships between these
realms and their potential effect on the somatic and
psychic health and equilibrium of the individual. It is
the position set forth in the present study that such an
articulation can be fruitfully pursued through the
integration of Differential Affect Theory and Developmental
Object Relations Theory via the empirical investigation of

alexithymia and its psychological sequelae.

My

Although it is true that Freud never used the term
"psychosomatic" in his writing to describe the interface
between the psyche and the soma,his concept of
"conversion" dominated the early psychoanalytic literature
concerning the treatment of physical disturbances that did
not demonstrate medically verifiable causes. The
conversion model could be summarized as an entirely
intrapsychic process involving the repression of
unacceptable wishes or ideas out of the conscious awareness
of an individual. Once unconscious, the ideas are
"dissociated" from the emotional excitations originally
attached to them and rendered innocuous through the process
of "converting" the discomforting emotional excitations
into a bodily, and thus disguised, form of expression. It
was presumed that the physical "conversion" of the symptom
might certainly be disturbing in its own right but less so

than the distress it replaced. The conversion model was

6

clearly based on the conflict resolution theory of neurotic
symptom formation as outlined by Fenichel (1945).

There flourished, in the neo-Freudian era, a
substantial body of literature forwarding the conversion
model as the primary etiological factor involved in any
number of specific physical disturbances falling under
voluntary innervation that might share a sexualized
symbolic underpinning. The conversion and specificity
literature has been extensively summarized by Wolff (1968),
Wittkower (1974) and Alexander (1950) and held considerable
sway over psychosomatic research in Europe and the United
States until the early 1940’s.

Following the lead of Cannon (1920) and similar to
Selye (1946), Alexander (1943) extended the conceptual
broadening of the psychosomatic model by postulating a
second major group of psychophysiological disorders which
did not play a primarily symbolic psychological nor tension
reduction role. These disorders Alexander referred to as
"psychogenic vegitative" conditions which were not seen as
substitute expressions of repressed ideas and emotions but
rather as expected physiological concomitants of chronic
emotional states such as extended stimulation of the
adrenal system and subsequent elevations in blood pressure
as physical correlates of the emotional experience of rage.
He viewed these conditions similarly to the Flight/Fight
response of Cannon (1920) and the General Adaptation

Syndrome of Selye (1946).

7

The overall importance of this conceptual shift was
felt in at least two major ways; first, it offered a
theoretical base for the psychological research and
treatment of a greater number of physiological disturbances
and, secondly, it allowed for a shifting of etiological
focus to earlier developmental periods. As Shur (1955) and
Nemiah (1974) have pointed out, the conversion model has
been shrouded in conceptual mist and is open to further
criticism on the grounds of the presumption of a relatively
developmentally advanced personality structuring of the
self, achieving at least psychosexual development at the
oedipal level. Thus, the conversion model offered no
clarity in those clinical cases suggesting preverbal levels
of disturbance such as the severe personality disorders or
schizophrenia. As discussed by Engel and Schmale (1967):

”we are proposing that a (type of) conversion

mechanism at pregenital as well as phallic

levels (of psychosexual development) may....

play a role in the background of a variety

of somatic disorders not heretofore conceptual-

ized as involving a conversion process." (pg. 347)

In an attempt to further outline this theoretical
shift, Engel and Schmale go on to write:

"Conversion is a psychological concept, the

definition of which cannot be bounded by

neuroanatomy, even though the function and

structure of the nervous system may be

involved secondarily as a result of the

biological meaning of the conversion (in

the broader sense) to the organism. The

parts or systems of the body capable of

being involved in conversions (of this sort)

are determined not by the voluntary of

involuntary nature of their innervation but

by their capability to achieve mental repre-
sentation, a process involving innervations,

8
perceptions through distance receptors, and

fantasy. Important here are object relating

activities for when object relating involves

a body part in discharge, expression, or

communication, concomitant physiological as

well as pathophysiological processes may

also acquire mental representation, thereby not

only becoming capable of reactivation by

symbolic stimuli but also of participating in

primary symbolic expression." (pg. 346)

Stephanos, et. al. (1975, 1976, 1978 & 1979) have been
most active in extending the object relations theory of
psychosomatics. This group basically views the category of
psychosomatic disturbances (as differentiated from neurotic
forms of conversion reactions) as a clinically definable
condition growing out of developmental disturbances in
primary psychological identifications whereby the
internalization of the maternal object by the infant is
disrupted by tension filled interactions with her. This
disturbed primary identification process is viewed as the
precursor to the lack of capacity to develop a fantasy life
which normally acts as a transitional buffer between the
child and overwhelming somatopsychic frustration.
Substantial disturbances in this developmental flow render
the individual vulnerable to direct somatic expression of
these overwhelming frustrations with limited capacity for
adaptive use of higher level imaginal and cognitive
functions. The adult phenomenlogical manifestations have
been described as the psychosomatic person's
"characteristic lack", in the sense of his/her:

”absence of psychic fantasy life, his (/her)

inability to develop interpersonal relations,
and general psychic emptiness." (Stephanos,

1975, pg. 179)

A separate but related line of inquiry has been
conducted by Nemiah and Sifneos (Nemiah, 1977; Sifneos,
1967, 1977) whereby they have encountered and documented a
very similar clinical picture in their psychosomatic
patients. They have coined the term "alexithymia" to
describe an inability to experience and verbally express
affect together with an unusual lack of fantasy life
appropriate to the expression of feelings. Thinking tends
to be flat, mechanistic, dichotomous and dominated by
details of the environment.

The phenomenological similarities of these separately
described conditions are striking and akin to the types of
cognitive functioning outlined by Liebovich (1973) and
Schreter (1978) in their respective discussions of
borderline personality functioning in medical settings.
Interestingly, Ammon (1979) has described the developmental
and ego functioning similarities of borderlines and
psychosomatics, offering the major line of distinction as
being the mode of discharge of aggressive feelings. He
asserts that borderlines tend to externally discharge
aggression while psychosomatic individuals tend to
internalize and somatize these powerful affects.

Other clinical investigators have carved out more
specific theoretically pathogenic relationships such as
Levitan (1976), who postulates a correlation between early

interpersonal loss, rage, and ulcerative colitis; Musaph

10
(1974), who also points to the role of aggression and

object relating in the formation of psychosomatic
disturbances; and Mitscherlich (1977), who discusses the
importance of the "transitional object" as an important
psychic organizer around which early imaginal processes are
crystalized, a repeatedly noted deficit in the cognitive
functioning of psychosomatic individuals.

Thus, contemporary psychodynamic thinking has focused
less on a conflict or psychological trauma model while
closer attention has been paid to the normal and deviant
lines of psychological development.

The innate differentiability of emotional states is
another important aspect of the present study. In a recent
review of the dominant theories of emotion, Leventhal and
Tomarken (1986) have pointed out a major distinction
between competing theories as revolving around the question
of differentiability of feeling states, physiologically as
well as cognitively.

A central proponent of the Cognition-Arousal Theory,
Schacter (1962) asserts that emotional states are grounded
in nondifferentiated physiological arousal and the labeling
of different affects relies solely on the cognitive
appraisals of social ones as guides. Following an
extensive review of the misattribution of arousal to
emotional and neutral cues, excitation transfer and bogus
feedback literatures, Leventhal and Tomarken (1986)

conclude, unenthusiastically,

11
"Overall, research testing predictions derived

from Cognition-Arousal theory has yielded

disappointing results." (pg. 574)

An alternative view of emotion, and one which is
consistent with the thrust of the present study, suggests
that emotion is biologically innate, complete with
differentiable and specific states of physiological arousal
and forms of expression. However, forms of behavior
appropriate to the expression of affect is not innate and
must be learned. Darwin (1904) noted the evolutionary
significance of affect in man and Ekman (1982) has
demonstrated the cultural universality of facial expression
attached to specific affects. Several authors have
suggested integrated developmental models of affective
differentiation (Schmale, 1964; Tomkins, 1962 and 1982;
Izard, 1971; Krystal, 1975) and Plutchik (1980) has
forwarded a complete psychoevolutionary synthesis.

Silvan Tomkins (1962, 1963, 1982, and 1984) has
theorized that the affects are not only biologically
innate, but, indeed, form the primary motivating system for
humans. He notes that drives such as hunger and sex must
be "fused" with, or "amplified" by affect before they
become motivators.

"The affect system is, therefore, the primary
motivational system because without its

amplification, nothing else matters, and with

its amplification, anything else can matter."
(Tomkins, 1984, pg. 164)

Tomkins further believes that affects are tied to

12
facial and vocal expressions, being triggered at

subcortical centers. He earlier specified facial
musculature as the feedback bridge but has recently shifted
to facial skin (1984). He first proposed eight primary
affects, each with its own specific facial concomitant. A
ninth has recently been added, further differentiating
dissmell from disgust (1984). The nine primary affects and

their associated facial expressions are as follows: (1984,

 

pg. 167)

Affect Facial Expression

1) Interest - Excitement 1) Eyebrows down, stare fixed
or tracking an object.

2) Enjoyment - Joy 2) Smiling response.

3) Surprise - Startle 3) Eyebrows raised, eyes
blinking.

4) Distress - Anguish 4) Crying response.

5) Fear - Terror 5) Eyes frozen open in
fixed stare or moving
away from object to the
side, skin pale, cold,
sweating, trembling, and
with hair erect.

6) Shame - Humiliation 6) Eyes and head lowered.

7) Dissmell 7) Upper lip raised in a
sneer.

8) Disgust 8) Lower lip lowered and
protruded.

9) Anger or Rage 9) Frown, clenched jaw and

red face.
The first three affects comprise the positive affects and
the last six are considered negative affects. Presumably

the list of negative affects is longer due to the

13

evolutionary function of emotion as a signaling system in
order to avoid danger. Apparently there were more dangers
to avoid than pleasures to approach.

Summarizing their review of current state of theory of
emotion, Leventhal and Tomarken (1986) reinforce the
conceptual bridge, through empirical study, between Tomkins
work and the object relations view of alexithymia outlined
above. They state:

"Our reading of the evidence on the universality
of expressive behavior and the lateralization of
emotional processing suggests an innate, central-
motor system for constructing expressions and
feelings with the critical function of cementing
interpersonal relationships. Further support
comes from the ability to detect these processes
during the first year and even the first days of
life. It is a system that thrives on social
contexts, able to react to social stimuli as well
as provoke them. Common sense and the data
showing links between emotion and memory suggest
that the emotion system undergoes elaborate
development over the individual life span. In
turn, the power of social cues in eliciting and
responding to emotional expression suggests that
such emotional development occurs primarily in
social interactions. It is in these same social
contexts that the infant becomes aware of the
distinction between self and other, and that the
child elaborates distinctive self-attributes and
develops a sense of self-esteem and effectance.
Emotion and self-concept are intimately bound in
memory." (pgs. 599 & 600)

Krystal (1977) has written extensively about early
affective development and differentiation and views the
evolving capacity for language, verbalization and
symbolization as key factors.

"....the process of putting emotions into

gestures, sounds and finally words is a

basic one. Language can be viewed as an
outgrowth or a manifestation of emotions...

(P9- 10)

14

As an individual grows up, his (/her)
affect becomes progressively verbalized
and desomatized...(pg. 11)

....the main advantage of verbalization is
that it manifests the child's ability to
experience specific emotions rather than a
generalized response. (pg. 11)

The development of language and symbolization
is the fundamental event in the development of
affects...failure to do so would lead to the
child being flooded with somatic and
undifferentiated affects..." (pg. 16)

Shifting the point of emphasis from the
affective/physiological domain to the cognitive/
physiological domain, Ahsen (1982) suggests that the
prevalent models of imagery have been focused too narrowly.
He asserts that previous debate over the nature and
function of imagery has been constricted to a discussion of
the cognitive aspects of the image and/or its semantic,
lexical meaning structures. He points out the importance,
from a theoretical and clinical perspective, of the
emotional and somatic linkage of the image to the cognitive
realm. He writes:

"Whether one believes that conscious experience
ensues upon an upward discharge from the
hypothalamus to the cerebral cortex as in the
Cannon-Bard theory...or upon return impulse from
the muscles as in the James-Lange theory...,the
simple fact remains that upon seeing an image (I),
(the sight of a bear in the woods), a specific
somatic or neurophysiological change (S) ensues as
either (a) a strong overt response observable by
another person or (b) an inner response which is
observable only by the subject himself. The S is
always of a specific type - skeletal,
proprioceptive, motor neural impulse, sensory
experienCe and so forth." (pg. 172)

Ahsen is certainly not alone in viewing the

imagery/emotion link as a powerful somatic mediator.

15
Achterberg (1984) points out that:

"the image serves both a reactive and a causative
role with regard to physiology. And, as part of
this reciprocal position, it can be consciously
accessed both to understand the physiologic
pattern and evoke change." (pg. 1)

The emerging field of psychoneuroimmunology provides
some key conceptual interfaces whereby the same
neuroanatomic structures involved in immunity are also
implicated in the transmission of the image as well as
emotional responses. Brain and central nervous system
structures, particularly the hypothalamus and the pituary
gland, as well as the limbic system and the reticular
formation, appear to be involved in the activation or
suppression of the immune system(Harrell, et. al., 1981;
Rogers, et. al., 1979; Spector and Korneva, 1981; Stein,
et. al., 1976). In fact, several authors have demonstrated
fairly specific interaction between imagery and various
aspects of immune function, even at the cellular level
(Achterberg and Lawlis, 1984; Schneider, et. al., 1983).

Sheikh and Kunzendorf (1984) have offered three
potential explanations of the psychosomatic effects of
imagery. The self-regulatory mechanism hypothesis assumes
that both the perceptually induced innervation and the
imaginally induced innervation of brain structures with
conscious qualities serve innately to regulate autonomic
responses. This self-regulatory mechanism would include

the immunological functions described above. A second

hypothesis, referred to as the associative-mechanism,

16

assumes that the imaginally induced innervation of brain
structures with sensory qualities elicits whatever
autonomic responses have been associatively connected with
those same brain structures during perceptual innervation.

A Third hypothesis, the structural-mechanism, assumes that

 

the innervation of brain structures possessing imaginal
qualities is neuroanatomically similar to the innervation
of brain structures controlling autonomic phenomena.
Again, a similar position as offered by Achterberg (1984)
and Ahsen (1982). Sheikh and Kunzendorf (1984) suggest
that these three functional interfaces need not be seen as
mutually exclusive or in competition as all are supported
by significant research literature.

The debate regarding the relative primacy of affect vs
imagery in brain/mind development can be assumed to
continue and is beyond the scope of the present discussion.
However, the emotion/imagery/physiology connections appear
manifest. The alexithymic concept would further appear to
be a useful organizing conceptual bridge for the
observation of relationships between and among these
domains from an empirical standpoint.

Thus, the present study can be viewed as an empirical
validation of the theoretical linkages between alexithymia,
affective differentiation and imaging capacity.
Additionally, the treatment of affect as independent from
cognition in this study provides the opportunity to

empirically investigate whether affect and affective

17

differentiation are embedded in cognition or not. Finding
a relationship between affective differentiation and
alexithymia, independent of mental imagery, would support
the theory of the innate psychobiological properties of
emotion as forwarded be Tomkins, Izard, Plutchik, and
Ekman. Conversely, failing to discover an independent
relationship would be consistent with the Cognition-Arousal

model of affective experience of Schacter and Singer.

Purpose

The purpose of the present study was to empirically
examine the relationship between alexithymia and its
primary psychological concomitants, affective
differentiation and imaging capacity in a sample of health
care consumers. A secondary purpose was to investigate the
role of anxiety and depression in the manifestation of

alexithymia.

Research Hypotheses
There were five fundamental research questions that
guided the present study. Stated as broad hypotheses, they
were:

1. Persons who manifest higher levels of alexithymia
will inversely manifest a reduced capacity to
engage in mental imagery contrasted with persons
who manifest lower levels of alexithymia.

2. Individuals who demonstrate higher levels of

18
alexithymia will inversely demonstrate a reduced

ability for affective differentiation compared to
individuals who demonstrate lower levels of
alexithymia.

3. Individuals who manifest higher levels of
alexithymia will show little or no differences in
levels of anxiety contrasted with individuals who
manifest lower levels of alexithymia.

4. Persons who demonstrate higher levels of
alexithymia will show little or no differences in
levels of depression compared with persons who
demonstrate lower levels of alexithymia.

5. Individuals who manifest higher levels of
alexithymia will demonstrate no differences in
the demographic variables of age, sex, race or

educational level.

Overview

The pertinent literature is reviewed in Chapter Two
under the following headings: Alexithymia, Affect Theory,
and Mental Imagery. In Chapter Three, the study sample,
instruments, procedures, hypotheses, research design and
data analysis are described. A summary of the results of
the data analysis is presented in Chapter Four. A
discussion of the results and limitations of the present
study, along with a discussion of the implications for

future research are presented in Chapter Five.

CHAPTER II

LITERATURE REVIEW

The three major variables under investigation in the
present study were alexithymia, emotional differentiation
and mental imagery. The pertinent literature will be
summarized in this chapter in regard to each major variable
under the headings of; Alexithymia, Affect Theory and

Mental Imagery.

Alexithymia

 

The summary of the literature in the following review
must be prefaced with one clear caveat; the overwhelming
bulk of the literature concerning alexithymia has been, to
date, theoretical, observational and exploratory. Theories
have proliferated regarding the manifestations and possible
etiology of alexithymia but the validation of the clinical
observations have lagged significantly behind. As Sifneos
(1977) pointed out in an address opening the 11th European
Conference on Psychosomatic Research devoted to
alexithymia, there are more questions than answers provided
by the current research literature. In the following
section, a number of the theoretical and empirical
perspectives on the condition will be summarized. They
span the biopsychosocial range of interactive domains but
with little connective conceptual or empirical

underpinning.

19

20
The present study can be viewed as an attempt to

empirically establish the linkages between the
observations, from various perspectives, of alexithymia,
affective differentiation and imaging capacity and does not
presuppose one single etiological factor. However,
affective and imaginal/cognitive development, as outlined
in the Object Relations and Affect Theory perspectives
summarized in Chapter One, is used as the overarching
conceptual framework.

The term alexithymia, as coined by Sifneos, derives
from the Greek ‘a' for lack, ‘1exis' for word, and ‘thymos’
for emotion (Sifneos, 1977). Considering his pivotal
position in the articulation of the alexithymic condition,
a summary of his earliest observations seems in order.
Those observations pertained to patients presenting at the
Beth Israel Hospital Psychiatric Clinic and were summarized
into 13 categories contrasting alexithymics and neurotics.

Those observations were as follows (Sifneos, 1977):

 

Alexithymic Neurotic

1. Presenting Complaints

a) Endless descriptions of a) Less emphasis on
physical symptoms, at times physical complaints

not related to an underlying
medical illness

Example: an ulcerative colitis b) Elaborate description
patient complaining more about of psychological dif-
aches and pains all over his ficulties (symptoms and/
body than about his bloody or interpersonal prob-
diarrhea. lems)

2. Other Complaints

Tension, irritability, frustra- a) Anxiety described in
tion, pain, boredom, void, terms of fantasies and

restlessness, agitation, thoughts rather than in

21
nervousness.

3. Thought Content
Striking absence of fantasies
and elaborate description of
trivial environmental details
(pensee operatoire)

4. Language
Marked difficulty in finding
words to describe feelings

5. Crying

a) Rare

b) At times they cry copiously

but crying seems not related to
an appropriate feeling such as

sadness or anger

6. Dreaming
Rare

7. Affect
Inappropriate

8. Activity

Tendency to take action
impulsively, action seems to
be a predominant way of life

9. Interpersonal Relations
Usually poor with a tendency
at marked dependency or pre-
ference for being alone,
avoiding people

10. Personality Make-Up
Narcissistic, withdrawn, pas-
sive-aggressive, or passive-
dependent, psychopathic

11. Posture
Rigid

physical sensations

b) Depression described
in terms of feelings of
worthlessness, guilt,
during sleepless nights,
etc.

a) Rich fantasy life

b) Marked ability to des-

cribe feelings in elo-
quent terms

Appropriate in descri-
bing feelings

a) Appropriate to speci-
fic feeling

Often

Appropriate

Appropriate to situation

Specific conflicts with
people but generally good
interpersonal relations

Flexible

Flexible

22
12. Countertransference

The interviewer or therapist Easy communication with

is usually bored by the patient patient whom the inter-

whom they find "dull" viewer or therapist finds
"interesting"

13. Relation to social, educa-
tional, economic, or cultural
background
None Considerable

It is clear from this original work that the
cognitive/affective difficulties demonstrated by the
alexithymic are often consistent with the manifestation of
psychosomatic disorders as well. However, this earliest
work points to no clear etiology of the alexithymic
condition or to any possible linkage with the psychosomatic
process. A

While Sifneos collected the first routine observations
of the alexithymic condition, Nemiah (1977) was
instrumental in beginning to articulate the different
theoretical perspectives that were starting to appear in
the literature regarding the etiology of alexithymia. He
initially dichotomized the theories (as if mutually
exclusive) into Psychological and Neuroanatomical groups of
theories with the former focusing primarily on the psychic
conflicts of the individual and the latter focusing
primarily on brain structure and function. An outline of
his organization of the theories is presented below:
I. Psychological Theories

A. Psychodynamic Models

1. Conversion Model (outlined in Chapter One)

23
2. Specificity Model (outlined in Chapter One)

3. Denial Model
This position broadly spans the Conversion and
Specificity Models. Nemiah suggests that
massive denial can be differentiated from
repression on the grounds that denial, as a
defense mechanism, may tend to be globally
applied to the totality of affective
phenomenon. Repression is here viewed as more
specifically applied to certain areas of
psychic conflict. He offers this view of the
denial process as augmenting the existing
psychodynamic models.
B. Deficit Model
This model is based on the tenets of Object
Relations Theory as outlined in Chapter One.
C. Developmental Models
As presented by Nemiah, the Developmental Models

are actually the developmental mechanisms

 

underlying the Conversion, Specificity and
Deficit Models outlined previously.
II. Neuroanatomical Theories
A. Structural Models
Nemiah here proposed that there is either an
absence of, or a deficit in, pathways between
neuronal centers underlying affect OR an absence

of, or a deficit in, the neuronal centers

24

themselves.

B. Developmental Models
Two developmental models are offered to account
for the structural anomolies presented above.
1. Defective post-natal development of neuronal
structures resulting from a lack of adequate
environmental stimulation and support.
2. Genetic abnormalities based on a inhereted

defect in anatomical structures.

After initially dichotomizing these various
perspectives, Nemiah concludes that only the Denial,
Deficit and Structural Models could account for the breadth
and scope of the alexithymic's lack of affect and fantasy
and calls for an open-minded integration of theories.

In an updated review of the alexithymic concept,
Lesser (1981) expands on the work of Nemiah, outlining in
greater detail the current neuroanatomical etiological
positions. He cites Nemiah's early work (Nemiah, 1975) and
points to a dysfunction of the striatum which inhibits
access to consciousness of neocortical sensory inputs as
well as affective input from the limbic system.

Lesser reports two studies which bear on the
split-brain phenomenon and possible alexithymic
characteristics. Hoppe (1977), studying patients who had
surgical commissurotomies for intractable epilepsy, noted

that these patients exhibited a paucity of dreams and

25
fantasies and had poor ability to symbolize. The

similarity to the "pensee operatoire" and alexithymic
behavior led him to introduce the concept of a "functional
commissurotomy" in patients with severe psychosomatic
disturbances. He theorized that a lack of interhemispheric
communication resulted in concern of the left hemisphere
with the translation of thing-presentations into
word-presentations, while the right hemisphere, unable to
verbalize, hypercathected bodily sensations.

Relatedly, Buchanan, et. al. (1980) present the case
of a man with a severed corpus callosum who manifested
severe alexithymic characteristics together with the
inability to report dream states. they conclude that one
facet of alexithymia may be this subtle cross-hemispheric
disturbance.

In a more recent study suggesting some
cross-hemispheric role in alexithymia, Rodenhauser, et. al.
(1986) looked at the relationship between handedness and
alexithymia in a sample of 100 college students. Their
results indicated a higher prevalence of alexithymia in
non-right hand dominant individuals raising questions,
again, about the role of right-hemispheric or anomalous
hemispheric dominance in fantasy life and awareness of
emotions. The most likely assumption would involve a
relative deficit in the development of left-hemispheric
dominance implying a residual deficit in language accessing

of the right-hemispheric productions.

26
Kaplan and Wogan (1976/77) developed an experimental

approach to cerebral laterality and alexithymia by using
subjective response to a painful stimuli and EEG monitoring
to assess subjects‘ ability to manage pain. They found
that left-hemisphere activation was associated with a
reported increase in pain, while right-hemisphere
activation was associated with a reported decrease in pain,
and with the apparent mobilization of fantasies following
experimental instructions. The investigators concluded
that the instruction to fantasize allowed subjects to
endure their pain better than under conditions where
fantasy was not prescribed. By analogy, alexithymic
individuals, unable to utilize fantasy, could experience
more prolonged and painful bodily symptoms. The
speculative nature of these conclusions needs to be
emphasized.

A possible genetic contribution to alexithymia was
reported by Heiberg & Heiberg (1978). With the use of the
Beth Israel Psychosomatic Questionnaire, monozygotic (M2)
and dyzygotic (DZ) twin pairs were compared. Intrapair
differences on the questionnaire were significantly
different, with the M2 pairs having more similar scores
than the D2 pairs. The conclusion was that there is a
strong hereditary component in alexithymia. However, this
needs to be interpreted with great care for several
reasons. First, only 33 pairs were studied. There were no

data on sociocultural variables, and, since the twins were

27

presumably reared together, they shared similar
environments where social learning and styles of
communication would influence their later behavior.
Similar kinds of studies have been forwarded to explain the
genetic basis for such disorders as schizophrenia with
similar, and well documented, methodological pitfalls.
Martin reports several studies on alexithymia
conducted with undergraduates from McGill University (1984,
1986). Martin theoretically extrapolated from Nemiah's
earlier work (1975) which suggested a certain dissociation
between alexithymic's physiological response mechanisms and
their subjective cognitive/affective response mechanisms
under stress situations. Martin (1984) selected 53
subjects divided into high and low alexithymia groups based
on the results of the Schalling-Sifneos Personality Scale
(SSPS). The subjects underwent a psychophysiological
stress-monitoring procedure and a dietary tryptophan
manipulation. The results showed that high alexithymics
manifested significantly lower levels of blood volume pulse
as well as a significantly greater percentage increase in
free tryptophan as compared to low alexithymics. The
author suggests that the results provide preliminary
supportive evidence for a dissociation between
physiological and subjective responses to stress under high
alexithymic conditions. However, the 1984 study included
only physiological measures and thus response

characteristics of a more psychological nature, such as

28
anxiety or depression, could not be directly determined.

In a follow-up study, Martin(1986) studied 30
subjects, again divided into high and low categories of
alexithymia based on SSPS scores. Frontal EMG, digital
blood volume pulse amplitude and heart rate constituted the
physiological measures. Measures of
repression-sensitization, social desireability and
state/trait anxiety comprised the psychological measures.
Interestingly, the results provided clear support for
earlier work that suggested that alexithymia is independent
of other psychological variables as there were no
significant correlations between SSPS scores and the
psychological measures. Secondly, high alexithymics
manifested higher levels of sympathetic activity during a
stress condition (in this case placing a hand in ice-water
until pain was reported) and during recovery periods. This
high level of sympathetic activity is conducive to the
development of a stress-related disorder either through
general systemic deterioration (Selye, 1946) or through
chronic demands on a genetically based specific organ
weakness (Sifneos, 1977). Thirdly, there was limited
support for the contention that alexithymics dissociate
their physiological responses from their ratings of
subjective levels of stress. These and the previous lines
of physiologically based research suggest that clearly
there are physiological concomitants to the alexithymic

condition as one would expect. However, to demonstrate

29

such a physiological connection is not to rule out
influence in the psychological realm as well. In fact,
literature will be cited in subsequent sections of this
review that will suggest that early learning experience can
alter brain structure and function, rendering this type of
nature/nurture question into a complex interactional
network of mutually reciprocal processes.

One final series of studies points to the subtle
psychological influences upon even the most tightly
controlled psychophysiological research. Embarking from
the linguistic/symbolic facet of the alexithymic
dysfunction, Von Rad & Lolas (1982) have extensively
studied the speech patterns and verbal productions of
alexithymics and psychoneurotics. Examining 80 subjects,
verbal production to TAT card 3BM, Rorschach fantasy
production, and Gottschalk & Gleser (1969) analyses of
speech content, were gathered. In the first research run,
all of the procedures took place in a lively interactive
context between interviewer and subject similar to the
context of patient and doctor interview in which
alexithymic characteristics were first, and primarily
continue to be, elicited and observed. The main results
were as follows: the alexithymic subjects (compared to
neurotics) produced less words on the TAT and open-ended
story, used the word ‘1' less and more frequently used the
word ‘one', and employed a smaller number of ‘affect-laden’

words. The Gottschalk-Gleser scores for the first 1,000

30

words of the interview indicated that alexithymic patients
expressed less guilt, shame, separation anxiety as well as
less ambivalent and inner-directed hostility. These
results are as might be expected. However, controls were
minimal and there was a lack of a normal control group.
However, on a second run with a similar research design
using 353 subjects, some interesting findings emerged.
Although the tasks were similar in design to the first
study, the subjects were either instructed to respond into
a tape recorder or interacted with an interviewer
instructed to provide no active feedback or interactional
stimuli. Under this situation no differences were apparent
either for pure groups (alexithymic vs neurotic) or for
mixed groups with "predominantly psychosomatic components"
or ”predominanly neurotic components". While there were no
physiological measures included in order to determine the
possible interactive effect of the interpersonal field with
the physiological response patterns, the authors still
tentatively conclude:
”The phenomenon of alexithymic behavior was first,
and continues to be, observed in the interactive
situation of the doctor—patient relationship. What
has been described as pensee operatoire, was an in-
teractional behavior in which the patient perceives
the other in a technical, stereotyped way without
clear recognition of his lively uniqueness, often
handling him like a thing. Only in the explicitly
dyadic situation of the conversation with a partner
who is alive and who intervenes and asks about
affects and is himself emotionally involved and
manifests his presence through reinforcement or
objection making the patient unsure about his
stereotypes and moving him, will alexithymic be-

havior become clear, making it possible to see the
described differences between psychosomatic and

31

psychoneurotic subjects."
and further:

"We seem justified in concluding, on the basis of

the so-far gathered experience, that alexithymia

should be defined and studied within the context

of dyadic verbal interaction; it is also probably

true that perhaps we are on the wrong track if we

expect it to be exclusively an ‘organismic’

attribute of patients.” (Von Rad & Lolas, 1982,

pgs. 99 a 100)

With this statement, the pendulum swings fully back to
the starting point first described by Sifneos (1977) in his
clinically oriented observations. Since the clinical
criteria and measurement of alexithymia have not been
clearly standardized, it seems somewhat premature to be
focusing exclusive attention on possible etiological
models. Rather there is need to establish empirical
validity for the concept itself. The present study is
viewed as just such an attempt at validating the
alexithymic concept without over-reliance on any single
etiological formulation. The literature summarized in the
remainder of this chapter sheds further light on the

possible interconnections between affect and imagery and

lay further groundwork for the present study.

Affect Theory
In this section the theoretical and research
underpinnings of the role of affect as viewed in the
present study will be outlined. Competing theories abound
regarding the nature and function of emotion in humans.

As Leventhal & Tomarken (1986) have suggested, there

32

have been, historically, four main theoretical approaches
to emotion articulated in the literature: 1)
Darwinian-Evolutionary Theory which assumes the existence
of multiple emotions and focuses on their form (expressive
response patterns), functions and evolutionary history
(Darwin, 1904; Izard, 1971; Plutchik, 1980); 2) Body
Reaction Theory which assumes that different emotions are a
product of different patterns of autonomic responses
(James, 1950; Schacter, 1957); 3) Central Neural Theories
which seek to find the neural structures involved in
emotional expression, feeling and behavior (Davidson, 1984;
Pribram, 1984); and 4) Cognition-Arousal Theory which
attempts to define emotion as the integration of thought
and/or perception with arousal (Schacter & Singer, 1962:
Lazarus, 1984).

Leventhal & Tomarken (1986) go on to suggest that much
of the current confusion surrounding the integration of
theoretical perspectives on emotion relies on a reluctance
to grant independent conceptual status to emotion as such.
They write:

”This "begrudging" attitude has three components:

(a) the behavioristic legacy and its suspicion of

subjective concepts; (b) the traditional cognitive

hold on our thinking in which emotion is a combi-
nation of arousal and cognition...; and (c) the
reluctance of cognitively oriented scientists to

view an emotion as anything more complex than a

"stop" or interrupt rule in a simulation of mental

operations....Admitting a richer concept of emotion

to the lexicon could generate major upheavals in
cognitive theory as emotions theory addresses the
growing theoretical and empirical knowledge in

neuroscience and molecular biology." (Leventhal &
Tomarken, 1986, pg. 566).

33
The innate psychobiological differentiability of

affect is a key conceptual assumption underlying the
present study. As such, it relies jointly on an
integration of aspects of the Evolutionary, Body Reaction
and Central Neural theories briefly outlined above and it
stands in stark contrast to the cognitively based
Cognition-Arousal theory as forwarded by Schacter (1957,
1962) and expanded by Richard Lazarus (1984).

The Cognition-Arousal theory of emotion has dominated
the social-psychological research literature for the past
20 years. As it’s major proponent, Schacter (1957, 1962)
asserts that emotion is basically reduceable to an
interaction between non-differentiated physiological
arousal (as characterized by a hightened sympathetic
activation) and a cognition or thought about the nature or
cause of the arousal. Since this arousal is perceived to
be emotionally non-specific, it determines only the
intensity of emotional states, while thoughts (cognitions)
determine their quality. One of the core assumptions

requires that both arousal and cognitions are necessarily

 

components for emotional experience. As Lazarus (1984)
states it:

"the quality and intensity of an emotional
reaction are determined by cognitive appraisal
processes, that is, the person’s continually
reevaluated judgments about the significance
of demands and constraints in ongoing trans-
actions with the environment and about the
options for meeting them. Second, cognitive
appraisal processes underlie coping activities
which, in turn, continually shape the emotional
reaction by altering in various ways the meanings

34

of ongoing relationships between the person and
the environment, that is, by affecting appraisal
itself via reappraisal.” (Scherer & Ekman, 1984,

pg. 222)

In their extensive review of the empirical literatures
testing the Cognition-Arousal theory, Leventhal & Tomarken
(1986) point out that investigations appear to fall into
one of two categories: those that clearly fail to support
the theory and those that provide only limited support,
with some, but not all, findings consistent with
predictions. Summarizing the literature on the
misattribution of arousal to emotional cues, they conclude
that the evidence in favor of the theory is quite weak and
that the findings which have leant partial support to the
theory have been based on such highly restrictive
experimental conditions as to render them nearly
meaningless in respect to generalizability beyond the
laboratory.

In summarizing their review of over 40 studies in the
misattribution of emotional as well as neutral cues,
excitation transfer, and bogus feedback literatures,
Leventhal & Tomarken unenthusiastically conclude:

"Overall, research testing predictions derived

from Cognition-Arousal theory has yielded

disappointing results." (pg, 574).

The alternative view of emotion, which is consistent
with the orientation of the present study, suggests that
emotion is psychobiologically innate, complete with
differentiable and specific states of physiological

arousal, experiential phenomenology and forms of

35

neuromuscular expression. However, forms of behavioral
expression appropriate to differentiated affect is not
innate and must be learned in developmentally sequential
interactions with the social and non-social environment.
Zajonc (1984) provides a powerful response to the
Cognition-Arousal school ala Schacter and Lazarus. He
asserts that the necessity of cognition in the experience

of emotion has been established by definition rather than

 

by empirical investigation. His argument is substantial
and will be quoted at some length.

"Assuming that cognitive appraisal is always a
necessary precondition of emotion, preempts
research on the matter. It is my preference
to leave the question of cognitive appraisal
open for empirical research, postponing the
task of precise and extensive definitions of
both processes until we know more about them.
Solving problems by definition is not an in-
centive for further study. It is a useful
temporizing maneuver that allows us to proceed
with our work for a while, pretending that one
aspect of our problem had already been solved.
But we can pretend just so long. At some
point of theoretical development, we must look
to the empirical side of the problem and con-
front each element of our definition with
empirical reality and theoretical consistency.
This point of theoretical development has now
been reached, I believe. Of course, the question
that is contested here cannot be fully resolved
unless we have a full understanding of conscious-
ness. Such an understanding is at the moment
beyond our reach. But we have learned just
about enough about cognition and emotion to move
beyond definitional disputes. There are suf-
ficient number of conflicting results which I
pointed out in my paper (Zajonc, 1980), and a
sufficient number of suggestive experimental re-
sults that need to be integrated. Questions
about the independence and primacy of affect can
now be seriously asked at the empirical level.

I offered the notion of affective independence and
primacy as an hypothesis to be empirically veri-
fied--not as a definition to be disputed. Above

36
all, however, defining affect as heavily dependent

on cognition, should make it rather clumsy to

study the interaction of cognition and emotion,

and especially those forms of emotion where the

latter influences cognition (in such phenomena,

for example, as phobia and prejudice)." (Scherer &

Ekman, 1984, pg. 260).

In support of this contention, Zajonc (1984) goes on
to summarize the empirical literature on the independence
of affect and cognition crystalized around five major
clusters of findings. A detailed review of this literature
would be beyond the scope of the present study and the
reader is referred to Zajonc for an in-depth analysis. The
five clusters of findings can be simply put forward as:

l) Affective reactions show phylogenetic and ontogenetic
primacy; 2) Separate neuroanatomical structures can be
identified for affect and cognition; 3) Appraisal and
affect are often uncorrelated and disjoint; 4) New
affective reactions can be established without an apparent
participation of appraisal; and 5) Affective states can be
induced by drugs, hormones or electrical stimulation of the
brain.

Darwin (1904) first outlined the evolutionary and
adaptive significance of emotion in animals and humans,
noting species-wide modes of expression and cross-cultural
adaptations. Lorenz (1965) and Izard (1971) have
independently demonstrated the cross-cultural similarities
in emotional expressiveness with Izard's work,

particularly, paralleling a significant series of

cross-cultural studies carried out by Ekman and Friesen

37
(1969, 1975, 1980). Ekman and Friesen displayed pictures

of facial expressions across five culturally diverse
populations and then presented a list of potential
emotions. Observers gave the same interpretations to each
face.

Other investigators have contrarily suggested that
facial expression is culturally specific and functions more
like a culture-bound language system than a biologically
innate expressive mechanism (Leach, 1972; Mead, 1975,
LaBarre, 1947). If this position were correct it would
undermine the theory of innate differentiability of affect.
In an elegant study, Ekman and Friesen (1969) tested the
notion of display rules as a means of reconciling their
findings with those of Mead, Leach and LaBarre. They
essentially posited that affective expression is innate and
cross-cultural but that each culture develops social rules
that govern who can express what emotion under what social
contexts. For example, in Western cultures males should
not cry, females should not show anger, losers should not
cry in public and winners should not be too happy about
winning, etc. Ekman and Friesen (1969) successfully
demonstrated how affective expression is 995g universal (in
private with no display rules operating) and culture
specific (in public with display rules in operation).

Ekman (1983) has provided further evidence in support
of the innate psychobiological differentiation of emotion

by demonstrating emotion-specific activity in the autonomic

38

nervous system. In assessing the autonomic correlates
(primarily heart rate and temperature change) of six
distinct emotions (anger, fear, sadness, happiness,
surprise and disgust), the null hypothesis of no main
effect for emotion was heartily rejected F(25,317) =
2.51, P .001).

Numerous authors have suggested integrated
psychobiological developmental models of affective
experience and differentiation (Schmale, 1964; Tomkins,
1962 & 1982; Izard, 1971; Krystal, 1975). Plutchik (1980)
has offered a complete psychoevolutionary synthesis which
organizes affect and cognition around a series of
adaptational tasks that the organism faces in interaction
with the physical and social environments. He asserts
that, in an evolutionary sense, emotion serves to organize
experience and mobilize behavior that fosters survival and
adaptation in eight broad functional classes. Thus, he
suggests that there are three languages that may be used to
describe emotional states; a subjective language, a
behavioral language and a functional language.
Correspondences between these three language systems

results in the following table (Plutchik, 1984, pg. 200).

 

 

 

Subjective Behavioral Functional

Fear, Terror Withdrawing, Protection
Escap1ng

Anger, Rage Attacking, Biting Destruction

Joy, Ecstasy Mating, Possessing Reproduction

Sadness, Grief Crying for Help Reintegration

39

Acceptance, Trust Pair Bonding, Incorporation
Grooming or Affiliation

Disgust, Loathing Vomiting, Rejection
Defecating

Expectancy, Examining, Mapping Exploration

Anticipation

Surprise, Stopping, Freezing Orientation

Astonishment

Silvan Tomkins (1962, 1963, 1982, 1984) has theorized
that the affects are not only biologically innate, but,
indeed, form the primary motivating system for humans. He
notes that drives such as hunger and sex must be "fused"
with, or "amplified" by affect before they become
motivators to behavioral action.

"The affect system is, therefore, the primary
motivational system because without its ampli-
fication, nothing else matters, and with its
amplification, anything else gag matter."

(Tomkins, 1984, pg. 164)

Tomkins further believes that affects are tied to
facial and vocal expressions, being triggered at
subcortical centers, identical in his position to that
offered by Ekman. Tomkins earlier suspected that facial
musculature was the primary feedback bridge but has
recently shifted his focus on to the facial skin (1984).
He first proposed that there were eight primary affects,
each with its own specific facial concomitant. The
similarities to Plutchik are manifest. A ninth affect has
recently been added by Tomkins, further differentiating
dissmell from disgust (1984).

In the summary of their review of the current state of

4O
theory of emotion, Leventhal & Tomarken (1986) have

reinforced the conceptual bridge, through empirical study,
between Tomkins work and the object relations View of
alexithymia. Their words bear repeating here:

"Our reading of the evidence on the universality
of expressive behavior and the lateralization of
emotional processing suggests an innate, central
motor system for constructing expressions and
feelings with the critical function of cementing
interpersonal relationships. Further support
comes from the ability to detect these processes
during the first year and even the first days of
life. It is a system that thrives on social
contexts, able to react to social stimuli as well
as provoke them. Common sense and the data
showing links between emotion and memory suggest
that the emotion system undergoes elaborate
development over the individual life span. In
turn, the power of social cues in eliciting and
responding to emotional expression suggests that
such emotional development occurs primarily in
social interactions. It is in these same social
contexts that the infant becomes aware of the
distinction between self and other, and that the
child elaborates distinctive self-attributes and
develops a sense of self-esteem and effectance.
Emotion and self-concept are intimately bound in
memory." (pgs. 599 & 600)

Krystal (1977) has written extensively about early
affective development and differentiation and views the
evolving capacity for language, verbalization and
symbolization as key factors.

"....the process of putting emotions into
gestures, sounds and finally words is a
basic one. Language can be viewed as an
outgrowth or a manifestation of emotions..."

(P9: 10)

"As an individual grows up, his (/her)
affect becomes progressively verbalized
and desomatized..." (pg.11)

"....the main advantage of verbalization is
that it manifests the child’s ability to
experience specific emotions father than a

41

generalized response." (pg. 11)

"The development of language and symbolization

is the fundamental event in the development of

affects...failure to do so would lead to the

child being flooded with somatic and undif-

ferentiated affects..." (pg. 16)

In order to follow this progression in the direction
of cognition, language, and symbolization, the pertinent
literature regarding the role of imagery and affective

experience will be summarized.

Mental Imagery

 

As noted above, controversy rages over the issue of
the psychophysiological primacy of affect vs cognition (the
image). Within the narrower discussion of the nature of
imagery, debate is longstanding and intense regarding what
it is that actually constitutes an image and how it
functions in the psychic constitution of the individual.
While it is beyond the scope of the present discussion to
detail the theoretical and philosophical underpinnings of
the debate, a brief history will be presented for the
purpose of laying a foundation for the conceptualization of
imagery and its psychosomatic implications in the present
study. Following that historical brief, pertinent research
literature will be summarized.

As Kolers (1983) points out, in a recent review of
Perception and Representation, imagery has an old but
confused status in the study of the mind. For many it was

both the carrier of information from the environment to the

42
mind as well as the mind's expression of that information.

Richardson (1983) has noted that this duality in the nature
of imagery is in need of clarity by distinguishing the
nature of imagery (by contrasting it to perception) from
the function of imagery (by inquiring into the conditions
of its arousal and the information that it can make
available).

Holt (1964), in a broad historical discussion,
suggests that shifting philosophical paradigms from the
introspective idealism of James, with its emphasis on inner
subjectivity, to the behaviorally oriented
epiphenomenology, with its dualistic assertion that mental
events are caused by physical events but have no effect
themselves, has resulted in a scientific hiatus in the
study of imagery. Not until the mid 1960's did imagery as
a legitimate field of study re-emerge, cresting on the wave
of cultural preoccupation with the nature of consciousness
and subjective experience. This lingering debate over the
question of whether the image is a Egg; phenomenological
experience or whether the image exists separate from the
phenomenal field remains.

In further respect to the nature of imagery, Paivio
(1971) has summarized many previous works in the context of
a "Dual Code Theory" - suggesting we use two different but
redundant coding systems to represent the world, a pictoral
system and a linguistic system.

Pylyshyn's (1973) position contrasts pictorialists

43

from propositionalists - those who argue for the
fundamentally pictorial or spatial or ”analog"
representation of objects in the mind versus those who
argue that all mental events were best captured as a list
of statements something like those of a computer program.
The contrast is sometimes put, in fact, as between
pictorial mental representation and symbolic
representations, the latter identified with mathematical or
similarly language-like symbolization.

Holt (1964) suggests one form of conceptual bridge by
basing our thinking on a "double-aspect-coded-meaning"
position. This position tentatively accepts that
consciousness may make a considerable difference and
therefore suggests the term imggg when we are speaking
about a phenomenal content of a sensory or quasisensory
nature (pictorialist) and the term representation when we
are speaking about the lexical meaning structure
(propositionalist) as mediated by a brain process without
awareness. This sense of the nature of mental imagery as a
sensory or quasisensory phenomenally based experience in
the mind is the meaning basis for the present study.

Given the clear somatic linkages proposed in the
present study, it is necessary to extend the previous
discussion into the realm of the physiologic. Ahsen (1984)
states that:

" Paivio's Dual Code Model...implicitly involves a
statement against the status of physiological

correlates in imagery by virtue of omission. The
somatic (S) side is not given any representation

44

in the model, which posits a parallel relationship
between the image (I) and the meaning (M) as word or
language. The image represents the concrete side
and the verbal represents the abstract side of the
same experience. Obviously, if the somatic side

is given a due representation, it would change the
Dual Code Model into a Triple Code Model" (pg. 16).

As indicated in chapter one, and reiterated here, Ahsen

(1982) also states:

" Whether one believes that conscious experience
ensues upon an upward discharge from the hypo-
thalamus to the cerebral cortex ...or upon re-
turn impulse from the muscles..., the simple
fact remains that upon seeing an image (I), (the
sight of a bear in the woods), a specific somatic
or neurophysiological change (8) ensues as either
(a) a strong overt response observable by another
person or (b) an inner response which is observ-
able only by the subject himself. The S is always
of a specific type - skeletal, proprioceptive,
motor neural impulse, sensory experience and so
forth." (pg. 174).

This is a direct attack on the epiphenomenalist
position by suggesting that there is an interactive effect
between the physiology and the conscious experience of the
image. Kolers (1983) joins this attack by pointing to the
work of Greenough (1975 & 1978) and Freeman (1979) which

provides:

" reason to think that cognitive learning and (by
extension) imagery actually modifies biological
structure, making a distinction such as Pylyshyn
(1973) proposes very difficult to maintain, if
not vacuous." (pg. 148).

Plutchik (1984) takes a moderate position on the
epiphenomenalist position but links imagery directly to the
evolutionary properties of emotion in the coping and
survival of the organism. He suggests:

" that images are related to emotions in the following
ways: (1) Images are part of the mental maps we

45

make of the environment to help us adapt more

successfully; (2) Images reflect emotional states:

(3) Images can intensify emotional states; (4)

Images can habituate or reduce the intensity of

emotional states; (5) Images can sometimes help an

individual gain mastery over a situation; and

(6) Images have both a phylogenetic and ontogenetic

sequence of development." (pgs. 110 & 111).

Sheikh & Kunzendorf (1984) take a similar stance,
stating that the epiphenomenalist‘s division of mind and
body has warded off any attempt to define the image in its
physiological context and continues to influence the modern
medical community's attitude toward potential psychosomatic
effects. They go on to cite literally hundreds of research
studies in the areas of; imagery and heart rate, imagery
and electrodermal activity, imagery and voluntary muscles,
imagery and blood flow, imagery and body chemistry, imagery
and ocular effects and, finally, imagery and the treatment
of disease as a counterpoint to the epiphenomenalist
position.

The domain of imagery and disease is of particular
import to the thrust of the present study. Sheikh &
Kunzendorf (1984) cite eleven studies in the treatment of
experimentally induced but disease related pain whereby
subjects instructed to imagine that a painful sensation
feels more numb or more pleasant increases both their
subjectively reported tolerance of pain and their
behaviorally measured tolerance. Further, subjects with
specific food allergies have been able to consume those

foods without becoming sick by imagining that other foods

were being eaten instead (Ikemi, 1965).

46
Achterberg (1984) extends the imagery/physiology

linkages by stating:
" the image serves both a reactive and causative

role with regard to physiology. And, as part

of this reciprocal position, it can be consciously

accessed both to understand the physiologic pattern

and evoke change" (pg. 1).

As mentioned in chapter one, the emerging field of
psychoneuroimmunology provides some of the key conceptual
interfaces whereby the same neuroanatomic structures
involved in immunity are also implicated in the
transmission of the image as well as the affective response
system. Brain and central nervous system structures,
particularly the hypothalamus and the pituary gland, as
well as the limbic system and the reticular formation,
appear to be involved in the activation or suppression of
the immune system (Harrell, et. al., 1981; Rogers, et. al.,
1979; Spector 8 Korneva, 1981; Stein, et. al., 1976).

In a series of studies conducted with patients
diagnosed with widely metastic, Stage IV cancer (Achterberg
8 Lawlis, 1978), three blood chemistry/hemotological
factors and two psychological factors were found to be
related to concurrent disease states. One of the
psychological factors measured denial and the other
psychological factor measured imagery dynamics of the
disease, with the imagery dynamics of disease having the
highest correlation. For the two-month follow-up, none of

blood chemistry/hematological factors were found to be
predictive. On the other hand, four psychological factors

47
were found to be predictive, including; denial, locus of
control, investment in the self and the imagery dynamics
factor. Again, the imagery factor was the most powerful
predictor of disease change.

The imagery analysis developed above has also been
extended to cover diabetes and low back pain (Achterberg 8
Lawlis, 1984). The diabetes assessment was found to
predict levels of blood glucose and the back pain imagery
has proven to be a useful diagnostic and prognostic tool in
determining several factors related to potential for
surgery and rehabilitation.

Schneider, et. al., (1983) trained subjects to image
specific behaviors of neutrophils (a type of white blood
cell which is not targeted for any special disease). In
several reiterations of their design, they demonstrated
that such imagery would reliably result in changes in both
the number of the cells and in the type of cell behavior
imaged.

Achterberg (1984) states:

” These and other findings describe a neuroana-
tomical bridge between the image and physiology

that is not only intact but quite ancient.

Further, the association is considerably more

primitive in terms of phylogeny and ontogeny

than is the link-up of either image or phys-

iology to the language system....this tie-in

between the location of the image and the area

associated with emotions (right brain) is

extremely important, since many of the autonomic

functions associated with health and disease are

emotionally triggered....these two sets of infor-
mation - the involvement of the central nervous
system in immune function and the image in

physiology - should be of critical concern to
the practice of medicine." (Pgs. 5 8 6).

48
The convergence of Achterberg’s position with that of

Plutchik's is evident. So too is it consistent with the
Triple Code Model of imagery offered by Ahsen. Achterberg
writes:
" The studies cited above support the Triple Code
Model (Ahsen, 1982) which considers the image
to be the primary phenomenon, followed by the

somatic responses and lastly by meaning or the
lexical and verbal aspects." (Achterberg, 1984,

pg, 10).

As briefly outlined in chapter one, Sheikh 8
Kunzendorf (1984) have forwarded three general hypotheses
explaining the mechanics through which imaging affects the
body; the Structural Mechanism hypothesis, the Associative
Mechanism hypothesis, and the Self-Regulatory Mechanism
hypothesis.

The Structural Mechanism hypothesis assumes that the
the innervation of brain structures possessing imaginal
qualities is neuroanatomically similar to the innervation
of brain structures controlling autonomic phenomena ala
Achterberg's discussion above.

There are three current positions held relative to the
Structural Mechanism hypothesis. One version of the
Structural Mechanism is based on the "image-hemisphere
identity" hypothesis: the hypothesis that neural states
with imaginal qualities are located in the right hemisphere
of a right-handed subject's brain. According to this
version of the hypothesis, imaginal control facilitates
autonomic control because both processes require

right-hemisphere activation, and better imagers should

49
exhibit better autonomic control because they utilize their

right hemispheres to a greater extent. To date the latter
claim has not been supported by the literature but the
general assertion that both imaging and autonomic
responding involve right-hemisphere activation has been
supported by all six studies cited to date (Sheikh 8
Kunzendorf, 1984).

A second version of the Structural Mechanism
hypothesis is based on the "image-ERG identity" hypothesis:
the hypothesis that visual images are the sensory qualities
of centrifugally innervated retinal states. If the
image-ERG identity hypothesis is correct, then imaginal
responses and autonomic responses are the results of
structurally similar processes which may or may not overlap
in terms of location. Two studies are reported in support
of this hypothesis (Sheikh 8 Kunzendorf, pg. 116).

Friedman (1978) has put forth a third version of the
Structural Mechanism hypothesis, based on evidence that the
ability to image rises and fall in 90-minute cycles
throughout the nocturnal portion of that day. Friedman has
suggested that, inasmuch as autonomic processes are
regulated by structurally similar biological rhythms, any
desynchronization of the imaging cycle may either cause or
reflect a disruption in the self-regulatory cycles of the
autonomic nervous system. It would appear that Price‘s
(1975) finding that better visualizers exhibit more regular

heart-beat intervals and more stable respiratory cycles is

50
consistent with Friedman's suggestion.

The second broad explanation is referred to as the
Associative Mechanism. An extensive, and contradictory
literature has accumulated suggesting that, within the
classical conditioning paradigm, mental images may play the
role of an unconditioned stimulus (UCS), a conditioned
response (CR) or a conditioned stimulus (CS). Following a
detailed summary of the associative literature (beyond the
scope of the present review) Sheikh 8 Kunzendorf conclude
that:

" The hypothesis that Pavlovian conditioning

mediates autonomic control through imagery

can be valid only if there is no validity

to King’s (1973, 1974) counterhypothesis,

that imagery mediates Pavlovian conditioning.

...In fact, autonomic conditioning in humans

did not appear to be imaginally mediated,

until Bell 8 Schwartz (1975) asked their

human subjects to describe how they controlled

heart rate: their human subjects reportedly

imaged UCSs. Unfortunately, it is not possible

to ask Pavlov's dogs whether (a) the bell CS

automatically elicited a salivary CR, as

Pavlov hypothesized, or whether (b) the bell CS

initiated retrieval of a food image, which
evoked a salivary UCR, as King hypothesizes."

(pg. 120.)

Finally, the Self-Regulatory Mechanism is offered as
the third possible explanation. The Self-Regulatory
Mechanism assumes that the neural processes behind
self-regulation of the body include not only unconscious
brain states, but also brain states with conscious
qualities. Accordingly, conscious images of heat are
likely to be concomitants of a self-regulatory process

which warms up the body. Moreover, any trauma-induced

51
blocking of those brain states that are identical with

conscious images of heat disrupts the self-regulation of
body heat and distorts conscious images of heat. However,
such disruption or imbalance in a self-regulated neural
process can supposedly be treated by generating the
traumatic image in question, and thereby inducing a
cathartic unblocking and rebalancing of the self-regulated
process. A total of 13 studies are reviewed that support
the Self-Regulatory Mechanism hypothesis ranging across
cases of chronic migraine, heart rate studies, skin
temperature and immune function (Sheikh 8 Kunzendorf,
1984).

Whether a self-regulatory mechanism, an associative
mechanism, or a structural mechanism best accounts for the
autonomic effects of imagery remains a matter for future
research to determine. Yet each of the three mechanisms
assumes that conscious images are qualities of specific
brain states which interact mechanically with autonomic
brain states, thus providing the necessary linkages to
physiology.

The theoretical significance of the interaction of
imagery and emotional access, differentiation and
regulation is the conceptual linchpin in the present study.
Ahsen’s (1982, 1984) discussion of the Triple Code Model of
imagery offers one interactive model of the processes
involved. In order to examine how the experience of affect

was influenced by individual differences in imagery

52
ability, Suler (1985) selected 30 ”high imagers" and 32

"low imagers" out of 400 male undergraduates who responded
to the Questionnaire Upon Mental Imagery (Sheehan, 1967)
and to the Vividness of Visual Imagery Questionnaire
(Marks, 1973). Affective reactance was measured by
self-report, primary process thinking and
psychophysiological changes. High visualizers manifested
greater levels of electrodermal activity associated with
affectively loaded cue words (punish, bloody, naked, seduce
and kill). The hypothesis that free association imagery
can enhance the experience of affect was further supported
by the result that all subjects using visual associations,
regardless of imaging ability, reported a higher personal
relevance of their associations as compared to subjects
using verbal associations.

Thus, the current literature on the nature and
function of mental imagery points decidedly in the
direction of significant interaction between imagery,

affect and physiology.

Summary
By way of summary, it appears clear that the attempt

of the present study, to empirically establish the
conceptual linkages between alexithymia, affective
differentiation and mental imagery, is well grounded in the
respective literatures reviewed. Regardless of the

theoretical stance on the etiology of alexithymia

53

(psychodynamic or neuroanatomical), there do appear to be

manifest psychobiological interfaces which are consistent

with the contemporary literatures regarding the nature and
function of emotion and imagery. In a sense, the present

study can be viewed as an empirical attempt to support the
construct validity of a clinically defined syndrome which,
heretofor, has been phenomenologically described and

presumptively accepted by definition.

CHAPTER III

METHODOLOGY

The purpose of the present study was to examine the
relationship between alexithymia and its primary
psychological concomitants, affective differentiation and
imaging capacity. A secondary purpose of the study is to
examine the relationship between alexithymia and anxiety,
depression and demographic information including age, sex,
race and educational level.

Subjects drawn from health care populations were
assessed using the Schalling-Sifneos Personality Scale
(SSPS), the Differential Emotions Scale-IV (DES-IV), the
Survey of Mental Imagery-Form A (SMI-A), the Beck
Depression Inventory (BDI) and the State-Trait Anxiety
Inventory-Form Y (STAI-Y). A multiple regression analysis
was conducted in order to establish the pattern of
relationships between the dependent variable, alexithymia,
and the sets of independent variables of affective
differentiation, imaging capacity, anxiety, and depression.

In the remainder of this chapter, the study sample,
instruments and procedures will be described. The research
hypotheses guiding the study as well as the design and data
analysis used in order to test the hypotheses will also be

outlined.

54

55
Sample

The population from which the sample for the present
study was drawn, consisted of health clinic users
presenting for medical services at the Isla Vista Medical
Clinic as well as the Montecito Medical Offices, located in
Isla Vista and Montecito, California, respectively.

Both clinics provide a full range of outpatient
medical services including diagnosis, treatment and
referral. The two clinics draw from all socio-economic
segments of the greater Santa Barbara area but their
proximity to the University of California, Santa Barbara
campus results in a relatively heavy student representation
in their clientele. As a result, the average age of the
clinic's patients is likely biased in the direction of the
18-24 year old range while the average level of education
is likely to be higher than that of the typical outpatient
medical clinic.

A total of 195 individuals initially volunteered for
participation in the study in the months of June and July
of 1987. Of those 195, 150 individuals volunteered to
participate completely by scheduling dates for the
administration of the total battery of questionnaires,
including the Beck Depression Inventory (BDI), State-Trait
Anxiety Inventory-Form Y (STAI-Y), Differential Emotions
Scale-IV (DES-IV), and the Survey of Mental Imagery-Form A
(SMI-A). The remaining 45 participants initially completed

only the Schalling-Sifneos Personality Scale (SSPS) but

56

were excluded from complete participation for one of three
possible reasons: 1) the person left no phone number by
which they could be contacted in order to schedule test
administrations; 2) the person wished to participate
further but was unable to schedule the necessary time; or
3) the person changed his/her mind between filling out the
SSPS and scheduling the remaining questionnaires.

Given that the primary demographic data was collected
at the time of the follow-up administration, the only
information gathered from the 45 people who did not fully
participate was gender and SSPS score. A Total of 11 males
with an average score of 56.0 on the SSPS did not
participate further in the study while a total of 34
females with an average score of 54.9 on the SSPS chose not
to participate fully. This information is summarized in

Table 3-1.

Table 3-1

Gender and SSPS Scores of Partially-Participating Subjects

 

 

SEX N % Mean(SSPS)p S.D.(SSPS)
Male 11 24.4 56.0 4.0
Female 34 75.6 54.9 3.4

 

The gender proportions in the partial-participation
group of 24.4% male and 75.6% female was very close to the
proportions in the full-participation group which consisted
of 25.3% male and 74.7% female, suggesting essential

equivalence of the two groups on this attribute.

57

A t-test was conducted on the mean scores on the SSPS
for the two groups in order to assure group equivalency on
the dependent variable. The results of the t-test affirmed
the null hypothesis of no significant difference between
the two groups on the dependent measure of alexithymia,
t=1.34, df=1, p< .05.

The demographic information for the 150 subjects who

participated fully in the study is summarized in Tables 3-2

 

 

through 3-5.
Table 3-2
Age of Subjects
AGE CATEGORY N %
15 - 19 9 6.00
20 - 24 59 39.34
25 - 29 26 17.33
30 - 34 13 8.67
35 - 39 17 11.33
40 - 44 8 5.33
45 - 49 5 3.33
50 - 54 4 2.67
55 - 59 9 6.00
60 - 64 0 0.00

 

As can be seen from Table 3-2, the age distribution of
the sample was unimodal and heavily biased in the direction
of younger subjects. 94 (62.67%) of the total sample fell
within the ages of 15 to 29 years with the 20-24 year olds
reflecting the modal category showing a total of 59

(39.34%) of the 150 total subjects.

58
The racial breakdown of the sample, reflected in Table

3-3, again demonstrates a unimodal distribution with a
predominantly white representation with 130 (86.66%)
falling in this category. Hispanics and Asians were nearly
equally represented with 7 (4.66%) and 6 (4.00%) in each
group. Blacks and Native Americans were least represented
with 3 (2.00%) and 3 (2.00%) in each category. This
overall ranking accurately reflects that of the greater
Santa Barbara area although, by percentage, the sample was
more predominantly white than is the greater metropolitan

area .

Table 3-3

Race of Subjects

 

 

RACE N %
Black 3 2.00
Hispanic 7 4.66
White 130 86.66
Asian 6 4.00
Native Amer. 3 2.00
Other 1 0.68

 

The gender composition of the sample, summarized below
in Table 3-4, was primarily female, 112 (74.40%). 38
(25.60%) males participated in the study. As noted, this
breakdown was nearly identical to that of the 45 partially

participating subjects.

59
Table 3-4

Gender of Subjects

 

 

 

GENDER N %
Male 38 25.30
Female 112 74.70

Table 3-5

Educational Level of Subjects

 

 

LEVEL OF EDUCATION N %

Less Than High School 8 5.33
High School Graduate 12 8.00
Part. Completed College 85 56.67
Comp. Col. Degree (4yr) 18 12.00
Grad. or Prof. School 27 18.00

 

The educational level of the sample is summarized in
Table 3-5 above. While each educational level is
represented, the sample is heavily biased in the direction
of partially or fully completed undergraduate or
graduate/professional degrees. 130 (86.67%) of the
participants have begun or completed their college or
professional degrees. 12 (8.00%) subjects graduated from
high school only and 8 (5.33%) individuals did not complete
a high school degree.

While the Isla Vista Medical Clinic and the Montecito
Medical Offices are not formally affiliated with the
University of California, Santa Barbara, the sample

demographics(particularly age and education) would suggest

60
that the clinics draw heavily from the student population.

It would seem that, resultantly, students from UCSB were

heavily represented in the study sample.

Instruments

The following section will consist of a description of
the instruments used in the present study.
Schalling-Sifneos Personality Scale

The instrument used to screen study participants for
the presence of alexithymia was the Schalling-Sifneos
Personality Scale (SSPS) (Apfel 8 Sifneos, 1979) (see
appendix A). The SSPS is a 20-item, self—report
questionnaire designed to elicit responses to statements
about the respondent on a four point scale ranging from 1
(does not apply at all) to 4 (applies completely). It was
originally developed by Apfel and Sifneos at the Beth
Israel Hospital Psychiatric Service, Boston, Mass. in order
to screen patients for the presence of alexithymia in a
more time effective manner. Previous assessment procedures
have involved time consuming structured clinical interviews
which have demonstrated inconsistent inter-rater
reliability and are subject to unintended
observer/investigator effects (Shipko 8 Noviello, 1984;
Lolas, et. al., 1980).

At this time, the SSPS appears to be the most widely
used of the available instruments and appears to have good

face validity. Additionally, evidence has been presented

61
indicating that the SSPS scores measure something distinct

from other psychological constructs such as anxiety or
depression (Blanchard, et. al., 1981; Martin, et. al.,
1984; Taylor, et. al., 1981). These findings provide
evidence for the discriminative validity of the SSPS.
Scores on the SSPS are derived by a simple tally across
items with a range of 20 to 80 for a total score. Items,
3,4,12,15, 8 20 are scored in the opposite direction of the
remaining items (Martin, 1987). A score of 50 or below is
considered demonstrative of the relative presence of
alexithymia (Blanchard, et. al., 1981).

Blanchard, Arena and Pallmeyer (1981) conducted a
series of four psychometric studies on the SSPS, providing
initial data on the prevalence of alexithymia in a normal
college population. In their sample of 230 undergraduates,
alexithymia was found to be distributed approximately
normally with a prevalence of 1.8% in the female group and
8.2% in the male group (pg. 64). Two factor analytic
studies were conducted revealing the following three
factors: I. Difficulty in expression of feelings; II. The
importance of feelings, especially about people; and III.
Daydreaming or introspection (pg. 67). Finally,
intercorrelations between scores on the SSPS and scores on
the Beck Depression Inventory, State-Trait Anxiety
Inventory, Rathus Assertiveness Scale and the Psychosomatic
Checklist were quite low, ranging from -.290 to -.067 (pg.

69). These low correlations suggest that the SSPS is

62

relatively orthogonal to measures of other dimensions of
psychological disturbance.

In an attempt to provide further descriptive data for
the SSPS and also to replicate the Blanchard, et. a1.
(1981) factor analytic studies, Martin, et. al., (1984)
administered the SSPS to 430 undergraduates at McGill
University. The overall prevalence rate for this sample
was 1.9% which was not further broken down by gender (pg.
147). The results of the factor analytic studies yielded
three similar factors to those found by Blanchard, et. al.
(1981) suggesting the factor structure of the SSPS is
stable and replicable across samples and is, therefore,
psychometrically robust.

Finally, in a further replication conducted on a
sample of 46 undergraduates at the University of
California, Irvine, Shipko 8 Noviello (1984) found a
test-retest reliability estimate of .76 and replicated
Blanchard’s three primary factors (Shipko 8 Noviello, 1984,
pg. 87.).

According to the factor loadings on individual items
in all three factor analytic studies, item 10 (I like
people to be precise and to describe details) and 19 (I
prefer to use my left hand) bore no significant
relationship to the primary factor structure. It is
reasonable to delete items 10 and 19 in order to increase
the instrument's overall sensitivity to the presence of

alexithymia. In fact, one author has gone as far as to

63

recommend that only nine of the original 20 items be
retained, only those most heavily loaded on the three
primary factors (Shipko 8 Noviello, 1984). Other authors
have not supported such an extreme alteration of the
instrument however. The revised scale resultantly
consisted of 18 items and was scored from 18 to 72, with a
score of 45 or below suggesting the presence of
alexithymia. Items 3,4,11,14 8 18 were scored in the
opposite direction in the revised scale. The cut point of
45 retained the proportional properties of the original
scale.
Differential Emotions Scale-IV

Affective differentiation was assessed using the
Differential Emotions Scale-IV (DES-IV) (Blumberg 8 Izard,
1985, in press) (see appendix B). The DES-IV is a 49 item,
paper-and-pencil self-report questionnaire which measures
the following emotions: Interest, Enjoyment, Surprise,
Sadness, Anger, Disgust, Contempt, Fear, Shame, Shyness,
Guilt and Inner-Directed Hostility. Respondents rate how
often over the past 30 day period they have experienced the
feelings expressed in phrases which comprise the items over
a five point scale from 0 (rarely or never) to 5 (very
often). Scoring simply consists of summing the items that
make up the 12 affect scales yielding a frequency score for
each emotion as well as a sum total for all emotional
experience by adding the individual emotion scale totals.

The parent version of this scale, the Differential

64
Emotions Scale, was originally developed to test

Differential Emotions Theory (Izard, 1971), which was based
on the initial work of Silvan Tomkins (1962 8 1963). The
original DES was designed as a state measure of subjective
emotional experience but was soon adapted to a trait
measure by shifting to a frequency-over-time response scale
similar to the one employed in the present study (Izard,
1977). Izard, (1977) reports a factor analytic study of
the DES yielding 10 factors: 1. Interest; 2. Enjoyment; 3.
Surprise; 4. Distress; 5. Anger; 6. Disgust; 7. Contempt;
8. Fear; 9. Shame/Shyness; and 10. Guilt. Test-retest
reliabilities for the trait instructions (DES-II) ranged
from .68 to .87 (Izard, 1977, pg. 126).

The DES-IV is identical to an earlier version of the
DES (DES III) with the exception of the addition of two
subscales hypothesized to target patterns of emotions
specific to depression, shame and guilt (Izard, 1972). The
DES-III was developed specifically for use with a college
and adolescent population by replacing emotional adjectives
with short phrases for the items. The DES-III has been
shown to possess reliability and validity in adolescents
and college students, and equivalency with the parent DES
and DES-II (Izard, et. al., 1974; Kotch, et. al., 1982).
The DES-IV, which was used in the present study, has been
shown to correlate significantly with the Children's
Depression Inventory in a sample of older children

(Blumberg 8 Izard, 1985, in press) and with the Depression

65
subscale of the Mental Health Inventory (Nelson, 1985).

Complex emotions such as anxiety and depression are
considered to be patterns of more finely differentiable
primary affects. The DES has been used to delineate these
patterns specific to complex emotions (Izard, 1972). In a
sample of college students, the following pattern of
primary affects were used by subjects to describe the
experience of depression (listed by order of frequency):
Distress; Hostility Directed Inwardly (a combination of
anger, disgust and contempt); Fear; Fatigue; Guilt;
Surprise; Shyness; and Joy (Izard, 1972). The instrument
offers clear distinction of the affective patterning.

Survey 92 Mental Imagery

 

Imaging capacity was determined using form A of the
Survey of Mental Imagery (SMI-A) (Switras, 1979) (see
appendix C). This instrument consists of two untimed,
paper-and-pencil, parallel-form surveys of 86 items each of
self reported mental imagery. For the purpose of the
present study, only form A was used as there was only one
administration required for each study participant. The
survey assesses two aspects of imaging capacity,
Controllability and Vividness. In addition, the survey
assesses imagery in seven sensory modalities; Visual,
Auditory, Olfactory, Gustatory, Tactile, Somesthetic and
Kinesthetic.

Scoring of the SMI-A requires simple addition.

Subtest scores exist for both controllability and Vividness

66

in each of the seven sensory modalities. For example,
there is both a visual controllability and a visual
Vividness score, an auditory controllability and an
auditory Vividness score, etc. Each subtest score is
obtained by adding the blackened numbers of the items
comprising that particular subtest. In addition to the 14
subtest scores, there is a total controllability score, a
total Vividness score, and a sum total score. The total
controllability scores are obtained by adding together the
seven controllability subtest scores, while the total
Vividness score is obtained by adding together the seven
Vividness scores. The total score is the sum of all the
items making up the test.

The SMI-A is a relatively recent instrument that was
standardized using 350 (129 male and 221 female)
undergraduate students from Iowa State University. The
author reports scale homogeneity data, using Chronbach’s
alpha, ranging from .68 to .97 (Switras, 1978, pg. 381).
Reliability estimates using correlations between
alternative forms ranged from between .74 and .91 (pg.
381). Analysis of scale convergence and discrimination
supported sensory modality and controllability-vividness
discrimination between the various subtests. The same
analysis supported same-sensory-modality and same-dimension
(controllability or Vividness) convergence among the
various subtests. A factor analysis performed on the 28

subtests rendered the following seven factors: I. Visual

6?
Imagery; II. Olfactory Imagery; III. Somesthetic Imagery:

IV. Kinesthetic-tactile Controllability; V. Gustatory
Imagery; VI. Kinesthetic-tactile Vividness: and VII.
Auditory Imagery.

Validity checks using correlations with several
existing measures of mental imagery were low, with
correlations ranging from .00 to .31 (Switras, 1979, pg.
34). However, the author points out that for several of
the sensory modalities there exist no other published
measure for comparison purposes and the measures that were
used for comparison comprise widely discrepant imaging
tasks. Further validity study is clearly called for.

In addition to the reliability and validity estimates
reported above, Hull and Render (1984), using the SMI,
forms A and B, to assess imaging controllability and
Vividness in 96 college students, conducted split-half
comparisons using the Spearman-Brown formula. Adjusted
reliability estimates ranged from .97 to .99 (pg. 73).
Parallel-form reliabilities were also estimated for each
subtest, the overall controllability and Vividness totals
and the sum total. All correlations were found to be
significant at or above the .05 level, ranging from .50 to
.95 (pg. 73).

Considering the above reliability and validity
estimates and the unique nature of the instrument, the

SMI-A was deemed appropriate for use in the present study.

68
State-Trait Anxiety Inventory-Form X

The State-Trait Anxiety Inventory—Form Y (STAI-Y) was
used as the measure of anxiety in the present study
(Spielberger, et. al., 1983) (see appendix D). The STAI-Y
has been used extensively in research and clinical practice
for over 20 years and yields independent measures of Trait
Anxiety as it manifests over time and State Anxiety as it
manifests phenomenologically in the moment. Thus, the
instrument is made up of two, 20-item, self-report scales;
the State inventory consisting of statements pertaining to

the respondent's experience ip the moment, with the Trait

 

inventory consisting of statements pertaining to how the
respondent generally feels over time. The State items are
scored on a four point scale, ranging from 1 (not at all)
to 4 (very much so). The Trait items are similarly scored
on a four point scale, ranging from 1 (almost never) to 4
(almost always). Nearly half of the items are scored in
reverse and, resultantly, scored from 4-1. The scores are
tallied across items, resulting in separate State and Trait
totals ranging from 20 to 80.

Over 2,000 studies have been conducted with the STAI
since 1970 and extensive normative data exists for working
adults, college students, high school students and military
recruits. Alpha coefficients of reliability are reported
by the test author between .86 and .96 with test-retest
reliability figures being somewhat lower, ranging from .73

to .86 (Spielberger, et. al., 1983). Given the transitory

69

nature of expressed anxiety, the measure of internal
consistency, alpha, is considered the more valid measure of
reliability, falling well within acceptable limits.

The main populations with which the STAI has been used
as a measure of anxiety includes high school students,
college students, working adults, military personnel, and
psychiatric, psychosomatic, medical, surgical and dental
patients. Application to the present study is clearly
within these sample parameters.

Construct validity of the STAI has been demonstrated
through the contrasted groups procedure whereby scores
achieved by normals were compared to scores achieved by
psychiatric groups in which anxiety was a major component
of the presenting complaint. All of the psychiatric groups
scored substantially higher on the T-scale than the normal
groups save the Character Disorder group, in which the
expectation of low reported levels of anxiety would be
predicted (Spielberger, et. al., 1983). Concurrent
validity of the STAI has been demonstrated through
correlation with other well known measures of anxiety,
including the IPAT Anxiety Scale (Cattell 8 Schier, 1963)
and the Taylor Manifest Anxiety Scale (TMAS, 1953). On
groups of college females, college males and psychiatric
patients, the correlations ranged from .73 to .85
(Spielberger, op. cit.).

Finally, two factor analytic studies conducted on

Form-Y of the STAI strongly supported the State-Trait

70

distinction as well as replicating the general factor
results of five previous factor analytic studies conducted
on the earlier, and psychometrically less robust, Form-x of
the STAI (Barker, et. al., 1977; Gaudry 8 Poole, 1975;
Gaudry, et. al., 1975; Kendall, et. al., 1976; 8
Spielberger, et. al., 1980). The first factor analytic
study on Form-Y was reported by Spielberger in 1980 using
the principle axis method of factor extraction with varimax
rotation using a population of high school students. Four
factors emerged: Factor I (State Anxiety Present), with
factor loadings ranging from .37 to .71; Factor II (State
Anxiety Absent), with factor loadings ranging from .53 to
.68; Factor III (Trait Anxiety Absent), with factor loading
from .39 to .66; and Factor IV (Trait Anxiety Present),
with loadings from .37 to .69. The second factor analytic
study was conducted by Vagg, et. al.(1980) with a sample of
1,728 Air Force recruits. The principle axis and varimax
rotation methods were again applied. Vagg's data yielded
both a two-factor and a four-factor solution. The
two-factor solution resulted in a State Anxiety Factor with
item loadings ranging from .42 to .68 and a Trait Anxiety
Factor with item loadings ranging from .32 to .72. The
four-factor solution yielded the same four factors as
reported by Spielberger, et. al., 1980.

Considering the wealth of clinical and research
applications of the STAI with samples similar in background

characteristics to the present study, it was deemed

71

appropriate for use.
Beck Depression Inventory

The Beck Depression Inventory (BDI) was used in order
to assess the presence of depression in the present study
(see appendix E). The BDI is a 21 item self-report
inventory consisting of a series of statements pertaining
to the respondent’s feelings. Responses to each item can
range from 0 - 3, resulting in a four point Likert-like
scale for each item. The range of responses varies in
severity for each individual item. For example, item one
ranges from "0 = I do not feel sad" to "3 = I am so sad or
unhappy that I can't stand it" (Beck, et. al., 1961; Beck
et. al., 1979). f

The BDI yields a single global score rating of the
severity of experienced depression ranging from 0 - 63.
Burns 8 Beck (1978) have provided the following
classification rating of resulting scores: 0-9 = no or
minimal depression; 10-14 = borderline depression; 15-20 =
mild depression; 21-30 = moderate depression; 31-40 =
severe depression; 41-63 = very severe depression. For the
purposes of the present study, these clinically derived
classifications will not be used. Rather, the global score
for each participant will be reported and used in the data
analysis.

According to Beck, et. al. (in press) the BDI has been
employed in over 1,000 research studies since its

development in 1961 and a summary of the psychometric

72
properties is currently in press. Citing from that review,

Beck, et. al., (in press) summarize 19 studies that have
addressed the internal consistency of the BDI as a measure
of reliability in both psychiatric and non-psychiatric
populations. For psychiatric populations, the nine
coefficient alphas ranged from .76 through .95, and the
mean coefficient alpha was .87. Similar values were found
for the 10 coefficient alphas that were reported for the
non-psychiatric samples. The range was from .73 to .92,
and the mean coefficient alpha was also .87.

Stability of the BDI is demonstrated by a summary of
nine studies reported by Beck, et. al., (in press)
providing test-retest correlations. Pre- and posttest
intervals ranged from hours to weeks. The range of the
Pearson product-moment correlation coefficients for
psychiatric patients ranged from .48 to .86, whereas the
coefficients for non-psychiatric subjects ranged from .62
to .83. It would appear from the lower boundary of the
non-psychiatric patient’s range that these subjects
displayed more stable BDI scores than the psychiatric
patients as would have been predicted. Furthermore, the
high correlations for the non-psychiatric groups suggest
that the BDI does demonstrate substantial stability over
time.

The test authors review five types of validity studies
conducted on the BDI, including; content, concurrent,

discriminant, construct and factorial.

73
The content validity of the BDI was originally based

on a clinical consensus of the symptomatology of depression
(Beck, et. al., 1961). Since that time the BDI has been
compared to the DSM-III criteria and found to accurately
reflect six out of nine DSM-III criteria well, two criteria
partially, and one criteria not at all (Moran 8 Lambert,
1983). The test author responds that these three criteria
(appetite gain, increased sleep and psychomotor aggitation)
reflect atypical depression and their inclusion in the
instrument would result in many false positives (Beck, et.
al., in press).

Concurrent validity of the BDI has been established
with clinical assessments and a variety of alternative
assessment instruments, including the MMPI, the Hamilton
Rating Scale for Depression, the Zung Self-Reported
Depression Scale and the Multiple Affect Adjective
Checklist (Beck, et. al., in press). While correlations
varied from study to study, the mean correlations within
studies conducted with the above instruments ranged between
.56 and .73 (pg. 17).

In regard to discriminant validity, the original
intention of the BDI was to reflect the depth or severity
of depression and not its subtype classification.
Resultantly, Beck. et. al. (in press) report several
studies that support the ability of the BDI to
differentiate between psychiatric and non-psychiatric

patients and two studies that demonstrate that the BDI

74

cannot differentiate between endogenous, involutional,
psychogenetic or mixed depressive diagnoses (Delay, et.
al., 1963; Schnurr, et. al., 1976).

In the same review, Beck, et. al. (in press)
demonstrates that the construct validity for the BDI is
strong with a number of hypothesized relationships between
physiological, behavioral and attitudinal variables
detected (pg. 20).

Factorially, 18 studies were reviewed with the BDI
appearing to be measuring a general second-order syndrome
of depression which may be decomposed into three highly
intercorrelated factors reflecting negative attitudes,
performance impairment and somatic disturbances.
Importantly, although the factor loadings of specific BDI
symptoms vary across diagnostic groups, the three
predominant factors still emerged. Specific factor
loadings were not reported by study (Beck, et. al., in

press).

Procedures
Subjects for the study were recruited from the
population of individuals presenting for medical services
at either the Isla Vista Medical Clinic or the Montecito
Medical Offices during June and July of 1987. The clinics
are located in Isla Vista and Montecito, California,
respectively. A box containing the Study Introduction (see

appendix F) with the SSPS attached was placed next to the

75

registration desk at each clinic. A 3x5 card was attached
to the box inviting clinic users to take and review the
Introduction, fill out the SSPS, and provide their name and
phone number. Completed Introductions and SSPS forms were
then returned to the respective receptionists who placed
the completed forms in manilla envelopes until collected by
the investigator at two to three day intervals.

After the questionnaires were collected, the
investigator telephoned the prospective subjects and
explained the purpose of the study in terms of
investigating the relationship between imagination and how
people experience their emotions. The subjects were also
informed about the nature of their further participation,
consisting of responding to several other paper-and-pencil
questionnaires. It was explained that these questionnaires
would ask about how they think and feel about themselves
and others and would take approximately 45 minutes to
complete. A total of 150 persons agreed to participate
completely in the study out of a total of 195 who
originally filled out the SSPS as discussed in the Sample
section above.

Next, depending on where the subject lived, small
group administrations of the remaining questionnaires,
including the BDI, STAI-Y, DES-IV, and the SMI-A, were
scheduled in quiet rooms of either the UCSB campus library,
the Goleta public library or the Santa Barbara public

library. Although not identical in size and decor, these

76

quiet rooms were similar in respect to providing a
comfortable, quiet and neutral environment for responding
to the questionnaires.

Subjects were grouped together based on time
availability considerations only with administrations
scheduled in the early evening hours, generally 5:00 to
8:00 p.m.. The number of subjects per administration
ranged from five to fifteen. The amount of time required
to respond to the questionnaires ranged from approximately
20 to 90 minutes, averaging approximately 45 minutes per
subject. All subjects who attended administrations
completed all questionnaires.

At the time of administration, another verbal
statement of the purpose of the study was provided along
with a review and signing of the Consent Form (see appendix
C) in accordance with the Ethical Principles ip the Conduct
pf Research with Human Participants (APA,1973) and the
standards of the University Committee on Research Involving
Human Subjects, Michigan State University. General
instructions for the questionnaires were provided at this
time and demographic information was gathered. Subjects
desiring a summary of the study results were invited to
provide mailing addresses (see appendix H).

Following the administration of the instruments, they
were coded and scored by the investigator. Introduction
and Consent forms containing the names of subjects were

gathered and secured for purposes of confidentiality.

77

Research Questions and Hypotheses

 

Several research questions guided the present study.
These questions and their attendant hypotheses are stated
below.

1. Will a relationship exist between the presence of
alexithymia and the capacity to engage in mental
imagery?

HYPOTHESIS I: Holding other variables constant, indivi-
duals demonstrating lower scores on the SSPS
measure of alexithymia will also demonstrate
lower scores on the SMI-A measure of imaging
capacity compared to individuals who score
higher on the SSPS.

2. Will a relationship exist between the presence of
alexithymia and the capacity for affective
differentiation?

HYPOTHESIS II: Holding other variables constant,
individuals demonstrating lower scores on
the SSPS measure of alexithymia will also
demonstrate lower scores on the DES-IV
measure of affective differentiation
contrasted to individuals who score higher
on the SSPS.

3. Will a relationship exist between the presence of
alexithymia and levels of self-reported anxiety?

HYPOTHESIS IIIa: Holding other variables constant, scores
on the SSPS measure of alexithymia will
not covary systematically with scores on
the STAI-Y measure of Trait Anxiety.

HYPOTHESIS IIIb: Holding other variables constant, scores
on the SSPS measure of alexithymia will
not systematically covary with scores on
the STAI-Y measure of State Anxiety.

4. Will a relationship exist between the presence of

alexithymia and levels of self-reported depression?

78

HYPOTHESIS IX: Holding other variables constant, scores on
the SSPS measure of alexithymia will not
co-vary systematically with scores on the
BDI measure of depression.

5. Will a relationship exist between the presence of
alexithymia and demographic variables including age,
sex, race, and education?

HYPOTHESIS Kg: Mean scores of the SSPS measure of alex-

ithymia will not differ between subjects
falling in different age categories.

HYPOTHESIS

I?

Mean scores on the SSPS measure of alex-
ithymia will not differ between males and
females.

 

HYPOTHESIS Mean scores on the SSPS measure of alex-
ithymia will not differ between subjects

falling in different racial groups.

'9

HYPOTHESIS

'5‘:

Mean scores on the SSPS measure of alex-
ithymia will not differ between subjects
differing in educational level.

Research Desiqp

The study was designed as an observational
investigation into the relationships among measures of
naturally occuring variables. As such, the study
represents a non-experimental design of a correlational
nature with single measures (Cohen 8 Cohen, 1983).

Random assignment to groups is not available as a form
of experimental control in this type of design. However,
potentially intervening variables including demographic
information, a measure of depression and a measure of
anxiety were built into the study for the purpose of
controlling the possibility of their confounding influence.

79
(Cohen 8 Cohen, 1983) Further controls to safeguard the

internal validity of the study were built into the
procedures of the study as outlined in the Procedures

section above.

Data Analysis

The data analysis was broken into two primary
sections. The initial purpose of the study was of a
confirmatory nature regarding the relationships between
alexithymia, affective differentiation and mental imagery
as previously suggested by the available literature. The
research hypotheses were laid out in respect to the
available theory. This confirmatory aspect of the study
was addressed by predicting a particular linear model and
then testing the hypotheses that derived from that model
using the hierarchical regression analysis described below.
An alternative test of hypotheses I through IV is provided
by the bivariate analyses conducted on the alexithymia
variable. Hypothesis V was tested by conducting bivariate
analyses on the grouped demographic variables of age, sex,
race and educational level.

A second aspect of the study was exploratory in
nature, given that the previous theory regarding the
relationship between the dependent and independent
variables has been criticized for it’s lack of empirical
validation. This exploratory aspect of the study was

addressed by conducting an all-subsets regression analysis

80
and a stepwise regression analysis in order to point out

some helpful directions for future research.

A full description of the data analysis used in the
study will be summarized below.
Hierarchical aggression Analysis

The data was initially analyzed using a hierarchical
multiple regression analysis. A hierarchical analysis
requires the researcher to sequentially order the
independent variables (IV's) as they appear in the
regression equation before the equation is run based on
predictions made by the theory. This ‘a priori'
arrangement of IV’s results in a more powerful statistical
test of the overall regression model and the research
hypotheses because it reduces the likelihood of
capitalizing on chance correlations as is possible if
conducting only a stepwise regression analysis with many
variables. In the present case, the variables of Mental
Imagery and Affective Differentiation were of primary
interest and were entered early in the equation. The Sex
variable was entered first, however, due to it's temporal
priority. One’s gender is determined before one develops
the capcity for mental imagery, affective differentiation,
anxiety or depression. The demographic variables of age,
race, and education were coded in a categorical manner
which prevented their inclusion in the regression analysis.
Treatment of these demographic variables is discussed

separately under the heading, Bivariate Analyses. The

81
variables of Anxiety and Depression were of secondary

interest and were consequently entered last (Cohen 8 Cohen,
1983). Thus, the variables were entered in the following
sequence for the initial regression analysis: 1) SEX; 2)
Mental Imagery (MI); 3) Affective Differentiation (AD); 4)
Anxiety (ANX); and 5) Depression (DEP), yielding the
following overall regression model for testing:
SSPS=C+SEX+MI+AD+ANX+DEP
The hierarchical analysis proceeds in a sequential manner,
entering the first variable first and calculating a
correlation coefficient. Subsequent variables are entered
one at a time with coefficients partialled from the
preceding variables such that the maginitude of the
incremental contribution of each successive variable to the
prediction of variation in the dependent variable can be
determined.

An analysis of variance was conducted to examine the
null hypothesis that the multiple correlation coefficients
for the population was zero. F-tests were conducted in
order to test the individual hypotheses by examining the
significance of the incremental contribution to variation
on the dependent variable for each individual independent
variable as it was entered into the regression equation.
The .05 level of significance was used for all F-tests.
Bivariate Analyses

As a test of hypothesis V, concerning the relationship

of demographic variables to alexithymia, bivariate analyses

82
were conducted by treating the demographic variables as

grouping variables and examining the relationships between
levels of the demographic variables and scores on the BDI,
STAI-Y, DES-IV, SMI-A, and SSPS. Where only two categories
exist on the demographic variables (for example sex),
t-tests were conducted for differences between means.

Where more than two categories exist (for example
education), analyses of variance (ANOVA's) were conducted.
Seatter plots were constructed and visually examined in
order to check for the possiblity of curvilinear
relationships between the variables. The variables of age
and race, originally consisting of nine and six categories
respectively, were collapsed into five and two categories
respectively due to limited cell frequencies in the
original sample.

Bivariate analyses were also conducted using the
cutoff score of 45 on the SSPS to designate scores as
falling into either alexithymic (45 or below) or
non-alexithymic (46 or above) groups. This results in the
creation of a new variable, alexithymia (ALEX). The
bivariate analyses conducted with ALEX provide an
additional test of hypotheses I through IV.

The .05 level of significance was used for all t- and
F-tests conducted to reject the null hypotheses of no

differences between mean scores on measured variables.

83
Assumptions Underlying the Analysis pf Variance

In an unbalanced, non-experimental design of this
nature, there are few methodological procedures available
with which to safeguard the assumptions underlying the
fixed-model analysis of variance. In a balanced,
experimental design, assumptions regarding the normality of
the distributions of variables can be addressed through
randomization procedures. The potential inequality of
group variances can be addressed by obtaining equal numbers
of subjects in contrasted groups.

In the present study, assumptions regarding the
normality of distributions cannot safely be made.

Therefor, descriptive statistics including the range,
skewness and kurtosis of the variable distributions are
reported. In cases where skewness or kurtosis reached
significance (BMDP, 1983), it was so indicated for tabled
values. The resulting t or F statistic should be
interpreted conservatively.

Due to the inequality of cell frequencies, Bartlett’s
Test for the homogeneity of variances was conducted prior
to each t-test or ANOVA. The Bartlett Chi-Square Test is a
powerful test for the homogeneity of variances and p< .01
was the alpha level used to reject the null of no
differences in group variances. Bartlett Chi-Squares are
reported and, where significant, are indicated with an *,
suggesting the t or F statistic should be viewed with

caution.

84
Pairwise Comparisons

All of the bivariate analyses consisted of post-hoc
contrasts between group means on various measures. Such
contrasts need not be orthogonal as in the case of planned
comparisons. Where significant overall results were found,
the Scheffe correction formula was applied in order to
examine the individual pairwise contrasts and control for
Type I Error.

All-Subsets Regression Analysis

Following the initial hierarchical regression
analysis, an all-subsets regression analysis was performed
post-hoc in order to determine the linear model that best
predicted outcomes on the dependent measure of alexithymia,
scores on the SSPS. The all-subsets analysis provides a
more clear view of the sample data set as it looks at all
possible combinations of variables entered into the
equation rather than being ‘forced' into a sequential
ordering by the experiementer. The all-subsets analysis
does not provide as powerful a test of the overall model or
the research hypotheses based on that model. It does,
however, provide a useful organization of the data set in
question in terms of generating future hypotheses. It
forms the basis of the exploratory aspect of the present
study.

Stepwise Rpgression Analysis
One rigorous stepwise regression analysis was

conducted post-hoc in order to determine the overall best

85

predictor in the model. The analysis was conducted using
the .01 level of significance to enter or remove variables.
Therefor, only those variables that contributed to
variation in the dependent measure beyond the .01 level
were generated for inclusion in this model. Again, this
analysis was conducted in order to provide exploratory

guidance and was not included in hypothesis testing.

Summagy

The purpose of the study was to examine the
relationship between alexithymia, as the dependent
variable, and mental imagery and affective differentiation
as the independent variables. 3A total of 150 subjects
drawn from the Isla Vista Medical Clinic and the Montecito
Medical Offices volunteered for full participation in the
study.

The Schalling-Sifneos Personality Scale was used as
the measure of alexithymia and was administered at the time
that subjects volunteered for the study. Administration of
the Beck Depression Inventory, the State-Trait Anxiety
Inventory-Form Y, the Differential Emotions Scale-IV, and
the Survey of Mental Imagery-Form A were scheduled and
conducted in the months of June and July, 1987. All 150
subjects completed all measures.

The study was of an observational, non-experimental
nature using a hierarchical multiple regression analysis to

test the first four research hypotheses. All independent

86

variables were ordered hierarchically prior to the
regression analysis. An analysis of variance was performed
in order to test the prediction of alexithymia by the
overall model. F-tests were performed on the incremental
contribution of each variable to the overall R squared in
order to test the individual hypotheses. An all-subsets
regression analysis was conducted in order to refine the
model and to provide guidance for future research. A
stepwise regression analysis was conducted in order to
isolate the best predictor in the model at the .01 level of
significance.

In order to test Hypotheses Va through Vd, bivariate
analyses were performed on the demographic variables of
age, sex, race, and education by treating them as grouping
variables and contrasting the mean scores on measured
variables by level of demographic information. A bivariate
analysis was also performed using a new variable,
alexithymia (ALEX), which was created using the cutoff
score of 45 on the SSPS. The bivariate analyses using the
ALEX variable provide an additional test of the major
research hypotheses. Bartlett’s Chi-Square Test for the
homogeneity of variances was performed prior to all
bivariate analyses and Scheffe's post-hoc correction was
calculated on the pairwise contrasts where appropriate.

The results of the data analyses are reported in
Chapter Four and a discussion and interpretation of the

results are presented in Chapter Five.

CHAPTER IV
RESULTS

The results of the data analysis are summarized in
Chapter Four. First, the descriptive statistics are
presented for the total sample with respect to scores on
the dependent and independent measures. Next, a summary of
the results of the hierarchical regression analysis is
presented. The descriptive statistics are then presented
for the grouped variables by demographic information and
alexithymia. Next is a reporting of the summary statistics
resulting from the bivariate analyses. Following is a
statement of the major research hypotheses and their
statistical tests. The results of the all-subsets and the
stepwise regression analyses are then presented. The

chapter is concluded with a summary of the major results.

Sample Descriptive Statistics

Table 4-1 contains a summary of the means and standard
deviations achieved by the overall sample on the SSPS, the
BDI, and the STAI-Y.

The means and standard deviations of the subscale
scores along with the total scores on the SMI-A are
presented in Table 4-2.

A summary of the means and standard deviations of the
subscale and total scale scores on the DES-IV is presented

in Table 4-3.

87

88

Table 4-1

Sample Means and Standard Deviations for
SSPS, BDI, and STAI-Y(State 8 Trait)

 

 

 

 

 

SSPS BDI STAI-Y(Sta) STAI-Y(Tra)
Mean 53.733 6.500 34.927 36.607
S.D. 6.588 6.509 12.110 10.856
Table 4-2

Sample Means and Standard Deviations for SMI-A

Scale Mean S.D.
Visual Controllability 44.047 6.307
Visual Vividness 60.087 14.604
Auditory Control. 34.920 4.947
Auditory Vivid. 46.680 11.695
Tactile Control. 27.207 4.309
Tactile Vivid. 36.500 10.107
Olfactory Control. 27.180 6.166
Olfactory Vivid. 34.020 11.469
Gustatory Control. 34.540 6.194
Gustatory Vivid. 44.773 13.283
Kinesthetic Control. 35.933 5.225
IKinesthetic Vivid. 49.120 11.929
Somesthetic Control. 25.447 4.408
Somesthetic Vivid. 33.187 8.890
Total Controllability 229.793 30.739
Total Vividness 304.433 68.328
Grand Total 534.227 95.692

 

89
Table 4-3

Sample Means and Standard Deviations for DES-IV

 

 

Scale Mean S.D.
Guilt 6.347 2.238
Shame 22.260 6.653
Inner-Directed Hostility 12.407 4.465
Anger 8.000 2.439
Fear 5.787 2.801
Sadness/Distress 9.720 3.577
Interest 11.367 2.420
Enjoyment 11.460 2.770
Surprise 7.893 2.583
Disgust 4.980 2.252
Contempt 6.220 2.252
Shyness 5.740 2.203
Total 112.300 22.189

 

Hierarchical Analysis

The initial series of hierarchical regressions

tested the linear model:

SSPS = C + SEX + SMI + DES + TRA + STA + BDI

The multiple correlations and the R squared values

for the hierarchical analysis are presented in Table

4-4.

Table 4-4

Multiple Correlation Coefficients and Variance Proportions
for the Hierarchical Regression Analysis

 

 

Variable R R Squared
SEX .113 .013
SMI .495 .245
DES .497 .247
TRA .497 .247
STA .501 .251
BDI .556 .309

90

As can be seen in Table 4-4, the overall multiple
correlation coefficient of .556 accounted for approximately
30% of the total variance in scores on the dependent
measure of alexithymia. An analysis of variance was
performed in order to reject the null hypothesis which
states that the true value of the multiple correlation
coefficient in the population for the model tested is zero.
The results of the analysis of variance is reported in
Table 4-5.

Table 4-5
Analysis of Variance for Multiple Correlation Coefficient

 

 

Source SS DF 4 MS F p
Regression 1880.998 6 313.500 10.365 .001
Residuals 4204.065 139 30.245

 

The multiple correlation coefficient of .556 was found
to be significant beyond the .001 level, F = 10.365
(6,139), p<.001. Thus, the null hypothesis is rejected.

It should be noted that the ANOVA was conducted on a
total N of 146 rather than 150 as it was found that four
cases produced undue leverage on the size of the mean
square error term. Those four cases were thus deleted from
the analysis. This reduces the degrees of freedom in the

residuals term from 143 to 139.

91
Table 4-6

Correlation Matrix of Predictor Variables

 

 

 

 

SEX SMI-A DES-IV TRA
SEX 1.000
SMI-A 0.007 1.000
DES-IV -0.093 0.181 1.000
TRA 0.018 -0.003 0.707 1.000
STA -0.013 0.079 0.592 0.714
BDI -0.126 -0.010 0.633 0.734
SSPS -0.126 0.466 0.137 -0.012
STA BDI SSPS
STA 1.000
BDI 0.690 1.000
SSPS -0.013 -0.141 1.000

 

Table 4-6 contains the correlation matrix of predictor
variables for the full model. An examination of Table 4-6
reveals that the correlations were low to moderately high,
ranging from .007 to .734. As can be seen, there were
fairly high and consistent correlations among all of the
affectively oriented measures including the BDI, the TRA
scale of the STAI-Y, the STA scale of the STAI-Y and the

DES-IV.

92

Table 4-7 contains the raw correlation coefficients,
standard error, standardized coefficients, t statistics
performed on the standardized coefficients and the
two-tailed alpha level associated with the t-tests. This t
statistic tests the null hypothesis of zero correlation

between the dependent variable and each variable in the

 

 

population.
Table 4-7
Regression Table for the
Hierarchical Regression Analysis

Variable Raw Coef. Std-Error Std-Coef. t p
CONSTANT 28.847 3.761 0.000 7.671 0.000
SEX -2.242 1.071 -0.151 -2.094 0.038
SMI-A 0.034 0.005 0.467 6.399 0.000
DES-IV 0.049 0.034 0.166 1.442 0.151
TRA 0.113 0.079 0.187 1.431 0.155
STA 0.003 0.058 0.006 0.057 0.954
BDI -0.413 0.121 -0.389 -3.408 0.001

 

Three of the variables reached statistical
significance at or beyond the .05 level. In decreasing
order of magnitude they were: SMI-A, t = 6.399, p< .001;
BDI, t = 3.408, p = .001; SEX, t = 2.094, p = .038. These
tests demonstrate that the respective correlations differ
from zero and that they should be retained in the equation.
However, they are not sufficient tests of the research
hypothses as they to not take the overall model into

account.

93

In the Systat statistical program, autocorrelations
are routinely conducted on the residuals. However, they
are not routinely reported unless found significant beyond
the .03 level. No autocorrelation was reported indicating

that it did not reach a significant level.

Descriptive Statistics for Groupgg Variables

The means, standard deviations, ranges, skewness and
kurtosis for the distributions of scores on the BDI,
STAI-Y, DES-IV, SMI-A and SSPS by grouped demographic
variable and the alexithymia variable are presented in
Tables 4-8 through 4-12. Significance of skewness and
kurtosis is reported for tabled values. Due to the limited
cell frequencies on the age variable, the original nine
categories were collapsed into five categories. The new
age categories consisted of:
1) ages 10-19; 2) ages 20-29; 3) ages 30-39; 4) ages 40-49;
and 5) ages 50-59. The race variable was similarly
collapsed from the original six categories to two new
categories consisting of White and Other due to limited
numbers of non-white participants in the overall sample.

Visual examination of scatter plots failed to reveal

curvilinearity in the relationships between variables.

94

Table 4-8

Descriptive Statistics for Age Variable

 

Age Category

 

 

Measure 1(n=9) 2(n=85) 3(n=30) 4(n=13) 5(n=13)

BDI

Mean 10.11 5.81 5.77 7.46 9.23
S.D. 7.93 4.84 8.15 7.33 9.07
Range 27.00 22.00 33.00 21.00 30.00
Skewness 1.58* 1.28* 2.44* .69 1.25*
Kurtosis 1.75 1.57* 5.42* -.78 .54
STAI(St)

Mean 41.57 33.86 34.07 36.77 37.46
S.D. 14.76 10.61 12.83 13.85 15.62
Range 39.00 41.00 50.00 47.00 54.00
Skewness .08 .57* 1.06* .99* 1.18*
Kurtosis -1.43 -.44 .43 .10 .72
STAI(Tr)

Mean 39.67 36.36 35.87 36.23 38.15
S.D. 14.46 10.09 12.02 9.64 12.52
Range 36.00 44.00 48.00 28.00 41.00
Skewness .08 .49* 1.01* .78 .65
Kurtosis -1.62 -.21 .57 -.77 -.63
DES-IV

Mean 115.78 113.91 107.40 110.54 112.46
S.D. 15.90 20.19 26.37 27.17 24.14
Range 43.00 93.00 130.00 85.00 69.00
Skewness -.25 .50* 1.76* 1.39* .42
Kurtosis -1.28 0.05 4.35* .69 -l.19
SMI-A

Mean 531.33 521.16 556.17 535.46 569.77
S.D. 96.41 96.87 94.12 95.50 86.58
Range 310.00 490.00 374.00 265.00 324.00
Skewness .02 -1.27* -.89* .17 -.93*
Kurtosis -.67 2.15* .54 -1.37 .66
SSPS

Mean 53.28 53.89 54.85 51.77 52.38
S.D. 7.95 6.54 6.66 6.37 6.34
Range 23.00 35.50 26.00 20.00 19.00
Skewness -.86 -.85* -.61* -.38 -.59
Kurtosis -.54 .70* -.42 -1.09 -.76

 

Note: * = significant

95

Table 4-9

Descriptive Statistics for Sex Variable

 

Sex Category

 

 

Measure Female(n=112) Male(n=38)
1.321
Mean 6.84 5.50
S.D. 7.09 4.28
Range 33.00 20.00
Skewness 1.79* 1.11*
Kurtosis -l.15* 1.83*
STAI(St)
Mean 35.02 34.66
S.D. 12.61 10.64
Range 56.00 47.00
Skewness .85* 1.01*
Kurtosis .12 1.11
STAI(Tr)
Mean 36.49 36.95
S.D. 11.26 9.70
Range 49.00 34.00
Skewness .74 .28
Kurtosis -.06 -.86
DES-IV
Mean 113.50 108.76
S.D. 22.14 22.23
Range 130.00 94.00
Skewness .96* .83*
Kurtosis 1.67* .35*
SMI-A
Mean 533.84 535.37
S.D. 93.55 103.04
Range 499.00 502.00
Skewness -.86* -1.22*
Kurtosis 1.00* 2.29*
SSPS
Mean 54.21 52.32
S.D. 6.58 6.50
Range 35.50 22.50
Skewness -.89* -.27
Kurtosis .67* -1.10

Note: * = significant

96

Table 4-10

Descriptive Statistics for Race Variable

 

Race Category

 

Measure White(n=130)p Other(n=20)
mu
Mean 6.32 7.06
S.D. 6.43 7.07
Range 33.00 29.00
Skewness 1.92* 1.60*
Kurtosis 4.83* 2.41*
STAI(St)
Mean 34.46 37.95
S.D. 11.46 13.73
Range 56.00 47.00
Skewness .89* .66*
Kurtosis .91* -1.17
STAI(Tr) .
Mean 36.40 37.95
S.D. 10.71 11.98
Range 49.00 37.00
Skewness .71* .22
Kurtosis .04 —.97
DES—IV
Mean 111.54 117.25
S.D. 22.14 22.45
Range 130.00 86.00
Skewness 1.04* .20
Kurtosis 1.91* -.45
SMI-A
Mean 536.32 520.60
S.D. 97.82 81.37
Range 499.00 263.00
Skewness -.87* -.23
Kurtosis 1.25* -.95
SSPS
Mean 54.04 51.75
S.D. 6.57 6.53
Range 35.00 23.00
Skewness -.81* -.78*
.31 -.24

Kurtosis

 

Note: * = significant

Descriptive Statistics for Education Variable

Table 4-11

 

Education Category

 

 

Measure 1(n=8) 2(n=12) 3(n=85) 4(n=18) 5(n=27),
BDI
Mean 9.75 10.25 6.83 4.28 4.30
S.D. 6.32 10.44 6.38 5.65 3.85
Range 18.00 32.00 33.00 25.00 13.00
Skewness .90 1.27* 1.70* 2.90* .85*
Kurtosis -.33 .34 3.54* 8.32* -.01
STAI(St)
Mean 39.87 34.50 36.23 33.06 30.78
S.D. 10.06 12.41 12.64 13.90 8.46
Range 30.00 37.00 50.00 56.00 28.00
Skewness -.54 .75 .71* 1.65* .78*
Kurtosis -.84 -.78 -.29 2.98* -.60
STAI(Tr)
Mean 41.87 39.42 37.28 32.11 34.67
S.D. 8.32 15.19 11.19 7.77 9.14
Range 21.00 41.00 49.00 29.00 34.00
Skewness -.06 .49 .60* .69* .35
Kurtosis -l.46 -1.21 -.20 -.09 -.94
DES-IV
Mean 123.00 112.17 116.13 104.56 102.30
S.D. 14.94 19.22 23.96 16.30 18.64
Range 39.00 67.00 130.00 64.00 64.00
Skewness -1.07 .68 .97* 1.17* .34
Kurtosis -.49 -.09 1.25* 1.28 -1.11
SMI-A
Mean 505.75 500.08 543.86 551.94 515.70
S.D. 108.82 114.12 91.24 94.71 96.67
Range 364.00 346.00 516.00 321.00 362.00
Skewness -1.10 -.26 -1.51* -.48 -.13
Kurtosis .59 -1.15 4.04* -.80 .63
SSPS
Mean 45.62 50.83 54.80 54.36 53.65
S.D. 7.96 8.14 5.27 7.37 7.09
Range 24.00 24.50 25.00 22.00 23.00
Skewness -.63 .40 -1.06* -.19 -.42
Kurtosis -.65 -.89 1.01* -1.15 -1.11

Note: * = significant

98

Table 4-12

Descriptive Statistics for Alexithymia Variable

 

Alexithymia Category

 

 

Measure Non-Alexithymic(n=120) Alexithymic(n=30)

BDI

Mean 6.21 7.67
S.D. 6.00 8.26
Range 33.00 32.00
Skewness 1.91* 1.58*
Kurtosis 4.48* 2.12*
STAI(St)

Mean 34.99 34.67
S.D. 12.34 11.32
Range 56.00 39.00
Skewness .99* .34
Kurtosis .53* -1.05
STAI(Tr)

Mean 36.77 35.97
S.D. 10.40 12.68
Range 49.00 43.00
Skewness .75* .45*
Kurtosis .01 -.71
DES-IV

Mean 114.88 101.97
S.D. 22.42 18.13
Range 130.00 75.00
Skewness .91 .79*
Kurtosis 1.38* .02
SMI-A

Mean 557.22 442.23
S.D. 77.74 106.33
Range 376.00 399.00
Skewness -.59* -.90*
Kurtosis .14 .38
SSPS
Mean 56.45 42.87
S.D. 3.87 2.97
Range 19.50 14.00
Skewness .16* -2.37*
Kurtosis -.34 6.63*
Note: * = significant

99

Bivariate Analyses

As tests of Hypotheses Va through Vd, and in order to
gain further insight into the data, bivariate analyses were
conducted on the grouped demographic variables, examining
the relationships between levels of the demographic
variables and scores on the BDI, STAI-Y, DES-IV, SMI, and
SSPS. Bartlett's Tests for the homogeneity of variances
were conducted in each case and are reported for tabled
values as Chi-Squares. The .05 level of significance was
used for all hypothesis tests. In cases where the F
statistic was found to be significant, Scheffe's post-hoc
correction was applied to pairwise contrasts.

A summary of the bivariate analyses are presented in
Tables 4-13 through 4-16. Table 4-13 demonstrates that no
significant differences were found between age categories
on any of the measures reported. Table 4-14 contains a
summary of bivariate analyses conducted using sex as a
grouping variable. No significant differences between
males and females were found on any of the variables
measured. T-tests conducted with race as a grouping
variable are presented in Table 4-15. The original six
categories of race (Black, Hispanic, White, Asian, Native
American, Other) were collapsed into two categories (White
and Other) due to limited cell frequencies. No significant
differences were found. A summary of the bivariate
analyses for Education is presented in Table-16.

Differences were found on the BDI, the DES-IV and the SSPS.

100

Table 4-13

Bivariate Analyses for Age (five levels)

 

 

ANOVA Chi-Square F p
Age x BDI 19.479* 1.866 .153
Age x STAI(State) 5.775 1.098 .360
Age x STAI(Trait) 3.831 .288 .885
Age x DES-IV 5.931 .546 .702
Age x SMI-A .257 1.249 .293
Age x SSPS .695 .658 .622

 

Note: * = significant at .01 level

Table 4-14

Bivariate Analyses for Sex (two levels)

 

 

t-test Chi-Square t p
Sex x BDI 11.655* 1.097 .275
Sex x STAI(State) 1.504 .158 .875
Sex x STAI(Trait) 1.159 .223 .824
Sex x DES-IV .001 1.138 .257
Sex x SMI-A .529 .085 .933
Sex x SSPS .008 1.542 .125

 

Note: * = significant at .01 level

101

Table 4-15

Bivariate Analyses for Race (two levels)

 

 

 

t-test Chi-Square t p
Race x BDI .302 .848 .398
Race x STAI(State) 3.816 1.201 .232
Race x STAI(Trait) .430 .593 .554
Race x DES-IV .006 1.072 .285
Race x SMI-A 1.006 .683 .496
Race x SSPS .001 1.452 .149

Table 4-16

Bivariate Analyses for Education (five levels)

 

 

 

ANOVA Chi-Square F p
Education x BDI 17.296** 3.002 .020*
Education x STAI(State) 6.603 1.506 .204
Education x STAI(Trait) 8.330 1.778 .136
Education x DES-IV 6.821 3.196 .015*
Education x SMI-A 1.369 1.187 .319
Education x SSPS 8.841 4.619 .002**
Note:

**

= significant at .05 level
= significant at .01 level

102
T-tests were also conducted using the cutoff score of

45 on the SSPS to designate scores as either alexithymic or
non-alexithymic. The research hypotheses would predict
that significant differences should be present in scores on
the DES-IV and SMI-A only. The results are summarized in

Table 4-17.

Table 4-17

Bivariate Analyses for Alexithymia (two levels)

 

 

 

t-test Chi-Square t p
Alex. x BDI 5.343 1.098 .274
Alex. x STAI(State) .333 .131 .896
Alex. x STAI(Trait) 1.943 .360 .719
Alex. x DES-IV 1.907 2.923 .004*
Alex. x SMI-A 5.089 6.698 .000**
Note: significant at .05 level

I-
II II

** significant at .001 level

Hypothesis Tests

The following section will contain a statement of the
research hypotheses together with their respective
statistical tests. F-tests were conducted on the
hierarchical analysis as tests of hypotheses I through IV.
Results of the t-tests conducted using the Alexithymia
variable will be reported as a secondary, and less
powerful, test of the major hypotheses. Where results from
the two analyses are in conflict, the F-test will be
considered the over-riding test. Hypotheses Va through Vd
will be tested using the bivariate analyses conducted on

the grouped demographic variables.

103
I. The first hypothesis was concerned with the

relationship between mental imagery and alexithymia. In
testable form it reads:

I Ha: Holding other variables constant, individuals
demonstrating lower socres on the SSPS measure
of alexithymia will also demonstrate lower
scores on the SMI-A measure of imaging
capacity compared to individuals who score
higher on the SSPS.

Partialling other variables, scores on the SMI-A made

a significant unique contribution to variation in SSPS
scores, F(1, 139) = 42.71, p< .001. The alexithymia by
SMI-A t-test was also found to be significant, t = 6.698,
df = 148, p< .001. Thus, the null hypothesis of no unique

contribution was rejected in favor of the alternative

hypothesis.

II. The second hypothesis was concerned with the
relationship between affective differentiation and
alexithymia. In testable form it reads:

II Ha: Holding other variables constant, individuals
demonstrating lower scores on the SSPS
measure of alexithymia will also demonstrate
lower scores on the DES-IV measure of
affective differentiation compared to
individuals who score higher on the SSPS.

Partialling other variables, scores on the DES-IV did

not make a unique contribution to variation in SSPS scores,
F(1, 139) = .37, p> .05. The alexithymia by DES-IV t-test
did demonstrate a significant difference between groups,

t = 2.923, df = 148, p = .004. This result is in conflict

with the F-test. However, the deviation of the

104
distribution of scores from normal and the wide dispersion

of scores around the mean render this a less powerful test
of the hypothesis. Therefor, the null hypothesis of no
unique contribution was not rejected in favor of the

alternative hypothesis.

III. The third hypothesis was concerned with the
relationship between anxiety and alexithymia. The STAI-Y
yields two measures of anxiety, trait and state, and, thus,
hypothesis three was tested in two parts; IIIa, relating to
trait anxiety and IIIb, relating to state anxiety. In
testable form, hypothesis IIIa reads:

IIIa Ha: Holding other variables constant, scores on
the SSPS measure of alexithymia will not
covary systematically with scores on the
STAI-Y measure of trait anxiety.

Partialling other variables, scores on the STAI-Y

trait anxiety measure made no unique contribution to
variation in SSPS scores, F(1, 139) = 0, p> 1.0. The
alexithymia by trait anxiety t-test also failed to discover
significant differences between groups, t = .360, df = 148,
p = .719. Hypothesis IIIa asserted the null hypothesis of

no unique contribution and was upheld.

In testable form, hypothesis IIIb reads:

IIIb Ha: Holding other variables constant, scores on
the SSPS measure of alexithymia will not
covary systematically with scores on the
STAI-Y measure of state anxiety.

Partialling other variables, scores on the STAI-Y

105
state anxiety measure did not make a unique contribution to

variation in SSPS scores, F(1,139) = .74, p> 1.0. The

alexithymia by state anxiety t-test found no significant
differences between groups, t = .131, df = 148, p = .896.
Hypothsis IIIb asserted the null hypothesis of no unique

contribution and was supported.

IV. The fourth hypothesis concerned itself with the
relationship between depression and alexithymia. In
testable form it reads:

IV Ha: Holding other variables constant, scores on
the SSPS measure of alexithymia will not
covary systematically with scores on the BDI
measure of depression.

Partialling other variables, scores on the BDI made a

unique contribution to variation in SSPS scores,

F(1, 139) = 11.67, p< .01. The alexithymia by BDI t-test
did not reveal significant differences between groups,

t = 1.098, df = 148, p = .274. Again, the bivariate test
is a less powerful test due to strains on the underlying

assumptions. Hypothesis IV asserted the null hypothesis

and was rejected.
V. Hypothesis V concerned the relationship between
alexithymia and the demographic variables of age, sex,

race, and education.

Hypothesis Va pertained to the realationship between

106
Va Ha: Mean scores on the SSPS measure of

alexithymia will not differ between subjects
falling in different age categories.
The analysis of variance found no significant
differences between age levels on the mean SSPS scores,

F(4, 145) = .658, p = .622. Hypothesis Va asserted the

null hypothesis of no differences and was supported.

Hypothesis Vb concerned the relationship between

alexithymia and sex. In testable form it reads:

Vb Ha: Mean scores on the SSPS measure of
alexithymia will not differ between males and
females.

A two-tailed t-test found no significant difference

between males and females on the mean SSPS scores, t =

1.54, df = 148, p = .125. Hypothesis Vb asserted the null

hypothesis of no differences and was upheld.

Hypothesis Vc concerned itself with the relationship
between race and alexithymia. In testable form it reads:
Vc Ha: Mean scores on the SSPS measure of
alexithymia will not differ between subjects
falling in different racial groups.
A two-tailed t-test found no significant
difference between whites and others on the mean SSPS
scores, t = 1.452, df = 148, p = .149. Hypothesis Vc

asserted the null hypothesis of no differneces and was

supported.

107
Hypothesis Vd pertained to the relationship between

level of education and alexithymia. In testable form it
reads:

Vd Ha: Mean scores on the SSPS measure of
alexithymia will not differ between subjects
differing in educational level.

An analysis of variance was conducted and discovered
significant differences between levels of eduation on the
mean SSPS scores, F(4, 145) = 4.619, p = .002. Hypothesis
Vd asserted the null hypothesis of no differences between
groups and was rejected.

Further analysis of the paired contrasts was conducted
in order to better understand the relationship between
alexithymia and education. Scheffe’s correction formula
was applied (Furguson, 1981). Table 4-18 presents the
pairwise differences by level of eduation. An examination
of Table 4-18 shows that only comparisons 1 8 3, 1 8 4, and
1 8 5 were found to be significant at the .05 level using
the Scheffe correction, F’=9.72. While these results would
appear to suggest that achieving an educational level at
the highschool level or above is related to lower levels of

alexithymia, the small number of subjects in level 1 (n=8)

calls this conclusion into question.

108
Table 4-18

Pairwise Contrasts for Education x SSPS

 

 

Educational Educational Level
Level 1 2 3 4 5
1 0.00
2 3.29 0.00
3 15.58* 4.19 0.00
4 10.69* 2.27 0.07 0.00
5 10.06* 1.67 0.68 0.19 0.00

 

Note: * = signficant at .05 level

All-Subsets Analysis

An all-subsets regression analysis was conducted
post-hoc on all of the independent variables. The
all-subsets analysis examines all possible combinations of
variables in order to find the model which produces the
overall best ‘fit’. The combinations that were examined
were from single variables to different combinations of all
six variables.

None of the resulting subsets produced a better R
squared value than the original model. The subset R
squared vaules ranged from .000133 to .29298. Therefor, no
unique solutions were found which provided better
predictive power than the original hierarchical model which

produced an R squared value of .309.

Stepwise Regression Analysis
A stepwise regression analysis was stringently

conducted in order to determine the single best predictor

109
in the model by setting the alpha level for entry or

removal of variables at .01. In order for any variabe to
appear in this model, it must contribute to the multiple R
at or beyond the .01 level of significance. This analysis
produced the following model:

SSPS = C + SMI-A
The value of R in this model was .477 and the R squared
value was .227. As can be seen, only the mental imagery
variable contributed significantly at the .01 level with
all of the other variables dropping out of the model. The
SMI-A accounted for 22.7% of the total variance in SSPS

scores 0

Summagy

A number of hypotheses were tested in order to
investigate the relationship between the dependent variable
of alexithymia and several independent variables. The
independent variables included mental imagery, affective
differentiation, anxiety, depression and the demographic
background variables of age, sex, race and educational
level. Hypotheses I through IV were tested by conducting
F-tests on the incremental unique contribution of
successive variables to the total linear regression model
for both hierarchical and stepwise analyses. Hypotheses I
through IV were also tested using the bivariate analyses
conducted with the Alexithymia variable. Hypotheses Va

through Vb were tested by conducting bivariate analyses on

110
the grouped demographic variables.

1. Hypothesis I predicted that a relationship would
exist between alexithymia and mental imagery. The null
hypothesis of no unique contribution of variation on the
SSPS scores was rejected beyond the .001 level, thereby
supporting the prediction.

2. Hypothesis II predicted that a relationship would
exist between alexithymia and affective differentiation.
The null hypothesis of no unique contribution of variation
on the SSPS scores was not rejected at the .05 level,
thereby refuting the prediction.

3a. Hypothesis IIIa predicted that a relationship
would not exist between alexithymia and trait anxiety. The
null hypothesis of no unique contribution of variation on
the SSPS scores was not rejected at the .05 level, thereby
supporting the prediction.

3b. Hypothesis IIIb predicted that a relationship
would not exist between alexithymia and state anxiety. The
null hypothesis of no unique contribution of variation on
the SSPS scores was not rejected at the .05 level, thereby
supporting the prediction.

4. Hypothesis IV predicted that a relationship would
not exist between alexithymia and depression. The null
hypothesis of no unique contribution of variation on the
SSPS scores was rejected at the .01 level. Therefor, the
prediction was not supported.

5a. Hypothesis Va predicted that a relationship would

111
not exist between alexithymia and age. The null hypothesis

of no differences between age categories on the SSPS scores
was not rejected at the .05 level, thereby supporting the
prediction.

5b. Hypothesis Vb predicted that a relationship would
not exist between alexithymia and sex. The null hypohtesis
of no differences between males and females on the SSPS
scores was not rejected at the .05 level, thereby upholding
the prediction.

5c. Hypothesis Vc predicted that a relationship would
not exist between alexithymia and race. The null
hypothesis of no differences between whites and others on
the SSPS scores was not rejected at the .05 level, thereby
supporting the prediction.

5d. Hypothesis Vd predicted that a relationship would
not exist between alextihymia and educational level. The
null hypothesis of no differences between educational
levels on the SSPS scores was rejected at the .002 level,
thereby failing to support the prediction.

A summary of the study together with a discussion of
the results, the limitations of the study, and the
implications for future research will be presented in

Chapter Five.

CHAPTER V

SUMMARY AND CONCLUSIONS

The major purpose of the study was to examine the
relationship between alexithymia and it’s primary
psychological concomitants, affective differentiation and
mental imagery. A secondary purpose was to examine the
relationship between alexithymia and manifestations of
anxiety and depression. In Chapter Five, an overall
summary of the study is provided followed by a discussion
of the findings. The limitations of the study are next
with a discussion of the implications for future research

closing the chapter.

Summapy
The primary purpose of the present study was to

investigate the relationship between the presence of
alexithymia and it’s two essential psychological
concomitants, poor affective differentiation and limited
mental imagery. A secondary purpose was to examine the
possible relationship between alexithymia and anxiety,
depression, and demographic background variables.

While it has long been recognized that psychosomatic
conditions present a significant drain on health care
services, the available clinical literature has suggested
that these conditions are difficult to diagnose and to

treat effectively. Early models of psychosomatic

112

113
functioning centered around the Freudian concept of

"conversion" which was based on an intrapsychic conflict
resolution model of psychic functioning. There flourished
a rich anecdotal clinical literature, based largely on
individual analytic case studies, to back up the conversion
and specificity hypotheses.

The work of Selye (1946), Cannon (1920), Alexander
(1943)and others, began to broaden the horizons of
psychosomatic thinking. A second large group of persons
with psychophysiological complaints were seen as suffering
from the effects of prolonged exposure to stress rather
than converting their psychological conflicts into physical
illnesses. This conceptual shift pointed physchosomatic
researchers in the direction of better understanding the
psychophysiological interfaces of the "mind/body" issue.

In the spirit of this latter perspective, Sifneos
(1967) began to describe a group of psychosomatic patients
with a clinical presentation that he labeled "alexithymic".
These patients typically demonstrated a marked inability to
describe their internal emotional experiences or to engage
in forms of mental imagery. Their thinking tended to be
very flat and detail oriented and, while they were
relatively well adjusted in their social roles, they tended
to demonstrate poor interpersonal relations.

Since the introduction of the concept of alexithymia,
there has risen a large number of competing theories

regarding the meaning and etiology of the alexithymic

114
condition. Nemiah, (1977) provided a beginning

organization of these models and basically divided them
into Psychological and Neuroanatomical theories. The
former tend to revolve around the development and
functioning of the mind or psyche while the latter tend to
revolve around the development and functioning of the brain
and nervous system proper.

Within the Psychological realm of theory, the most
detailed articulation of the alexithymic problem comes from
the object relations school of contemporary psychoanalysis.
Clinical writers from this perspective suggest that
alexithymia represents a fundamental developmental arrest
in the transition from early, undifferentiated affective
states of the infant to the more clearly pronounced and
differentiated affective experiences of the young todler.
This developmental arrest is seen as arising primarily
through a disturbance in the relationship with the primary
caretaker. This prevents the youngster from being able to
tolerate highly emotionally charged experiences without
being somatically overwhelmed. As a result, the child is
not able to "metabolize" the affective experience,
integrate the experience using newly forming language
skills, or develop internal mental "representations" or
images to comfort and soothe him or herself. This early
developmental difficulty is then viewed as laying the
foundation for a later tendency toward "somatization" as an

adult. When faced with a stressful situation or the

115
elicitation of a strong emotional response, the adult is

more likely to cope with the stimulus through bodily
changes than through language, mental imagery, and adaptive
behavior based on insight into internal psychological
states and needs.

The Neuroanatomical theories are most clearly
articulated in regard to a structural or functional
disturbance in the relations between the right and left
hemispheres of the brain. It has been proposed that
alexithymia is primarily a result of poor cross-hemispheric
connections that prevent the contents of the right
hemisphere (spacial relations, imagery productions, certain
affective experiences) to be translated into the language
of the left hemisphere for logical representation and
processing. Developmentally, there has been a suggestion
that alexithymia may represent an inherited genetic defect
in the relations between the hemispheres or a defect in the
structures that produce emotional reponses themselves.

While many competing theories abound, Lesser (1981)
and Nemiah (1977) point out the speculative nature of both
the Psychological and the Neuroanatomical perspectives and
call for an integrative approach. Lesser has gone on to
criticize the literature for a premature emphasis on the
etiology of a condition that has been inadequately
validated as a viable clinical entity. It was in the
spirit of this call for the empirical validation of the

alexithymia concept that the current study was conducted.

116
A total of 150 subjects, drawn from the Isla Vista

Medical Clinic and the Montecito Medical Offices,
volunteered for full participation in the study.

The Schalling-Sifneos Personality Scale was used as
the measure of alexithymia and was administered at the time
that subjects volunteered for the project. Adminstration
of the Beck Depression Inventory, the State-Trait Anxiety
Inventory-Form Y, the Differential Emotions Scale-IV, and
the Survey of Mental Imagery-Form A were scheduled and
conducted in the months of June and July, 1987. All 150
subjects completed all measures.

The study was of an observational, non-experimental
nature using a hierarchical multiple regression analysis to
test the first four research hypotheses. All independent
variables were ordered hierarchically prior to the
regression analysis.

The initial hierarchical linear model presented for
testing was: Alexithymia = Constant + Sex + Mental Imagery
+ Affective Differentiation + Anxiety + Depression. An
F-test was performed in order to test the prediction of
alexithymia by the overall model. F-tests were performed
on the incremental contribution of each variable to the
overall R squared in order to test the individual
hypotheses. An exploratory all-subsets regression analysis
was conducted post-hoc in order to refine the model. A
stepwise regression analysis was performed in order to

determine the best single predictor in the model

117
In order to test Hypotheses Va through Vd, bivariate

analyes were performed on the demographic variables of age,
sex, race, and education by treating them as grouping
variables and contrasting the mean scores on measured
variables by level of demographic information. A bivariate
analysis was also performed using a new variable,
Alexithymia (ALEX), which was created using the cutoff
score of 45 on the SSPS. The bivariate analyses using the
ALEX variable provide an additional test of the major
research hypotheses. The .05 level of significance was
used for all hypothesis tests. Bartlett's Chi-Square Test
for the homogeneity of variances was performed prior to all
bivariate analyses and Scheffe's post-hoc correction was
calculated where appropriate for pairwise contrasts.

Hypothesis I predicted that a relationship would exist
between alexithymia and mental imagery. The hypothesis was
supported beyond the .001 level of significance.

Hypothesis II predicted that a relationship would
exist between alexithymia and affective differentiation.
This hypothesis was not supported at the .05 level of
significance.

Hypothesis IIIa predicted that a relationship would
not exist between alexithymia and trait anxiety. The
hypothesis was supported at the .05 level of significance.

Hypothesis IIIb predicted that a relationship would
not exist between alexihtymia and state anxiety. The

hypothesis was supported at the .05 level of significance.

118
Hypothesis IV predicted that a relationship would not

exist between alexithymia and depression. This hypothesis
was not supported at the .05 level of significance.

Hypothesis Va predicted that a relationship would not
exist between age and alexithymia. This hypothesis was
supported at the .05 level of significance.

Hypothesis Vb predicted that a relationship would not
exist between sex and alexithymia. This hypothesis was
supported at the .05 level of significance.

Hypothesis Vc predicted that a relationship would not
exist between alexithymia and race. This prediction was
supported at the .05 level of significance.

Hypothesis Vd predicted that a relationship would not
exist between alexithymia and educational level. This
prediction was not supported at the .05 level of
significance.

The all—subsets analysis demonstrated that no
alternative model using the independent variables included
in the study provided a better correlational ‘fit’ than the
original hierarchical model.

A stepwise regression analysis conducted at the .01
level of significance demonstrated that the measure of
Mental Imagery, the SMI-A, was the single best predictor in

the model presented.

119
Discussion

A discussion of the conclusions based on the results
of the study will be presented below.
Hypothesis I

First, and foremost, the study provides empirical
support for the imagery component of the alexithymia
construct. It would appear that the alexithymic
individuals in this sample, indeed, demonstrated a limited
capacity for mental imagery compared to their
non-alexithymic counterparts. The degree to which the
SMI—A measures the "characteristic lack" of Marty and
M’Uzan (1963) or the absence of fantasy production
described by other workers in this field (Sifneos, 1967;
Stephanos, 1975; Krystal, 1979) is still an open question.
However, these results clearly demonstrate that the imaging
capacity of alexithymic respondents is less well developed
than non-alexithymics.

While the sample was too small to include the SMI-A
subscale scores in the initial analysis, some interesting
differences were observed between alexithymics and
non-alexithymics on various modality subscales. For purely
exploratory purposes, the modality subscale scores were
examined. Appendix I contains the means and standard
deviations on the modality subscales. Bartlett Chi-squares
were performed and where non-signficant at the .01 level,
t-tests were conducted on the group mean differences.

Considering the intercorrelation of the subscales along

120
with the inflation of Type I error using this method, some

interesting results were still obtained. The summary of
t-tests is presented in Appendix J. Significant
differences were found beyond the .001 level on the
following subscales; visual Vividness, auditory Vividness,
olfactory control, olfactory Vividness, gustatory
Vividness, tactile Vividness, somesthetic control,
somesthetic Vividness, and kinesthetic Vividness.

The heavy emphasis on the Vividness subscales suggests
that future research might pare the overall SMI-A to only
the Vividness subscales. The somesthetic, tactile and
kinesthetic differences are interesting to ponder given the
tendency toward somatization noted in alexithymic
populations (Sifneos, 1967, 1977; Stephanos, 1975; Krystal,
1979; Martin, 1984). An interesting followup might include
the administration of a physical symptom checklist and the
Hypocondriasis and Hysteria subscales of the MMPI in order
to examine the relationship between these imagery
modalities and tendencies toward somatization.

Hypothesis I;

The second hypothesis regarding the relationship
between alexithymia and affective differentiation was not
supported by the F-test performed on the incremental
contribution of the DES-IV to the variance in SSPS scores.
However, the t-test performed on the mean DES-IV scores
between alexithymics and non-alexithymics found significant

differences between the groups in the predicted direction.

121
As Chronbach points out (1972), results that do not

converge based on different methods of analysis point to
the possibility of method-specific differences rather than
data-specific differences. In this case, it was felt that
the violation of the assumptions underlying the t-test were
serious enough to weaken its statistical power. The most
prudent conclusion would be that, while the relationship
between alexithymia and affective differentiation (as
measured by the DES-IV) was not strong enough to reach
statistical significance, there does still appear to be a
weak and inverse relationship between the two.

There could be several reasons why the F-test was not
significant. First, as mentioned in Chapter One, it may be
that affective differentiation does pp; reflect an
underlying independent status from cognition and, by
extension, is not a significant concomitant of alexithymia.
This would suggest that the lack of affective
differentiation might possibly be overshadowed by, or
embedded in, the general impoverishment of the capacity to
access inner experience via fantasy and cognition. This
conclusion would support the cognition-arousal theory of
affect (Schacter 8 Singer, 1962). An alternative
explanation is that affective differentiation lg a viable
component of the alexithymia construct but that it was
poorly measured by the DES-IV. This seems like a more
plausible explanation given the lack of psychometric
evaluation of the instrument. It may simply be that the

122
original "state” aspect of the instrument is not

sufficiently transformed into a "trait" measure only by
changing the wording of the instructions. As a result, the
demand characteristics of the DES-IV may elicit more of the
experience of affective states rather that the capacity for
differentiated expression of those states. Further, the
bulk of the individual items on the DES-IV elicit responses
pertaining to negative affect which may, in fact, result in
the DES-IV measuring more of general factor of
psychological distress than a capacity for emotional
differentiation. The tendency for the DES-IV scores to
correlate highly with the other measures of emotional
distress supports this possibility. Rosenberg (1984) used
an adaptation of the Response To Situations Test (RTS) in a
psychoeducational study involving college students and
found significant results on the Differentiation of Affects
subscale. Perhaps this instrument would be an improvement
over the DES-IV. However, imbeddedness of affect in
imagery remains a problem with this instrument as well.
Similarly to the SMI-A, the sample was too small to
meaningfully include all of the DES-IV subscales in the
statistical analysis. Again, however, there were
interesting, if speculative, differences between
alexithymics and non-alexithymics. The DES-IV subscale
means and standard deviations are summarized in Appendix K.
Again, Bartlett Chi-squares were calculated and, when found

to be non-significant at the .01 level, t-tests were again

123
performed. In this case only two subscales demonstrated

signficant differences beyond the .001 level; interest and
surprise. Differences were found on two further subscales
but at nonsignificant levels: guilt, p = .051; and shame,
p = .093.

Interestingly, the non-alexithymic individuals showed
higher levels of guilt and shame along with higher levels
of interest and surprise which seems to suggest a generally
broader range of affective experience and expression for
non-alexithymic individuals. This would be consistent with
the theory. The significant differences on the interest
and surprise subscales point, speculatively, toward higher
levels openness and emotional flexibility, also consistent
with the theory (Sifneos, 1967; Krystal, 1979).

Hypothesis III

The overall results of the study support the findings
reported elsewhere (Blanchard, et. al., 1981; Taylor, et.
al., 1981; Martin, et. al., 1984) of no significant
relationship between alexithymia and State or Trait
anxiety. The fact that scores on the SSPS were correlated
the least with the state-anxiety measure of the STAI-Y of
any of the variables lends support to the trait conception
of alexithymia. It would suggest that alexithymia is
relatively impervious to state the fluctuations in anxiety.
As psychological defense mechanisms are erected in order to
fend off the experience of anxiety, one might expect

alexithymia to be high when subjective anxiety is low and

124
alexihtymia to be low when anxiety is high if alexithymia

were simply a different name for defense mechanisms such as
repression or denial. This was not found to be the case
which supports Nemiah's contention (1977) that alexithymia
represents a more profound and global condition. This
leaves open the question, however, of whether alexithymia
is primarily a psychological phenomenon or a
neuroanaotomical one, or both. The above result could
support both conclusions equally well. The addition of
physiological measures would be necessary to address this
differentiation.

Hypothesis 1!

Hypothesis IV concerned the relationship between
alexithymia and depression. Several previous studies have
found an insignificant relationship between depression and
scores on the SSPS (Blanchard, et. al., 1981; Taylor, et.
al., 1981; Martin, et. al., 1984). Recently, however,
Bagby, et. al. (1986) conducted a study involving college
students in an attempt to validate yet another alexithymia
measure. They found that the BDI correlated .596 with the
new instrument, the Toronto Alexithymia Scale. The results
of the present study were mixed as the BDI was found to add
significantly to the variation in SSPS scores on the
F-test, yet no significant differences were discovered
between alexithymics and non-alexithymics on the bivariate
analyses. To the extent that the significant results

reflect a method-dependent finding, they are inconclusive.

125
The bivariate analysis was a weaker secondary test of the

hypothesis. The prudent conclusion remains that, in the
present sample, depression (as measured by the BDI) showed
a moderate and positive relationship with alexithymia.

Again, several possibilities exist. Perhaps the
literature is insufficient to support a reliable hypothesis
of the independence of alexithymia and depression. The
results of the Bagby study (1986) showed a high correlation
between alexithymia and the BDI as well as a correlation of
.466 between alexithymia and the depression subscale of the
Basic Personality Inventroy (BPI). Both of these
correlations were significant beyond the .001 level. The
results of the present study would support the Bagby
findings.

Another possibility would be to suggest that the BDI
is a poor measure of depression. Although the BDI has been
used extensively in psychological research it has been
criticised for its sensitivity to the effects of social
desireability (Beck, in press).

Yet another possibility is that this particular sample
is anomalous in respect to the coexistence of depression
and alexithymia. Perhaps another sample would not
demonstrate this relationship. The mixed results of the
present study reflect the mixed results of the literature
on this point and further clarification is left to future

study.

126
Hypothesis y

Hypothesis V pertained to the relationship between
alexithymia and demographic background variables. The data
analysis supported hypotheses Va through Vc in the
prediction of no significant relationship between
alexithymia and age, sex, and race respectively. These
findings support the previous literature in this regard
(Lesser, 1981).

The results for the education variable showed a
significant difference between levels of education on the
SSPS scores. This overall finding refuted hypothesis Vd.
Upon closer pairwise examination it was discovered that the
only significant differences were between educational level
1 (less than high school) and levels 3 (partially completed
college), 4 (completed 4 year degree), and 5 (professional
or graduate school). It would be inviting to speculate
about the implied effects of education on alexithymia
except for the fact that level 1 contained only eight
subjects and there were not significant pairwise
differences between levels 2 8 3, 3 8 4, or 4 8 5. If the
implication of a positive relationship existed in fact, one
might expect a consistent difference between levels as
education increased or at least some other confirmatory
difference. This was not the case. The prudent conclusion
is that the significant finding for education by
alexithymia in this sample was an artifact of the sampling

procedure which produced too few subjects in the lower end

127
of the educational range to provide confidence in the

result.
All-Subsets Analysis

The results of the all-subsets analysis indicated that
no alternative ordering of the independent variables
included in the study provided a better correlational ‘fit’
than the original hierarchical model tested. There is, of
course, no implication regarding alternative models using
other variables that might have contributed in such a
manner to provide a better overall predictive model. In
relation to the present study, the conclusion is supported
that the hierarchical model ordered a prioi based on the
existing theory was the best possible ordering of those
variables studied.
Stepwise Analysis

The result of the stepwise analysis indicated that the
SMI-A was the single most powerful predictor in the model.
It contributed 22.7% of the total variation in the scores
on the SSPS.

Clearly, alexithymia is a multidimensional construct.
The capacity for mental imagery appears to account for one

meaningful dimension in its overall understanding.

Limitations g; the Study

 

A number of limiting factors were inherent in the
present study and will be summarized in the following

section.

128
Several limitations in respect to the sample

characteristics deserve attention. Of some concern was the
size of the overall sample. There were six independent
variables included in the initial hierarchical regression
analysis. Following Cohen 8 Cohen (1983), a power analysis
was conducted using a moderate effect size for the
population of .15 and a 99% confidence level. This yielded
a desired N of 159.37, very close to the sample size of the
study. One particular advantage of a larger sample would
have been the opportunity to include the subscales of the
DES-IV in the regression analysis. It is quite possible
that, while the correlation between the SSPS scores and the
DES-IV total scores proved to be insignificant, meaningful
subscale correlations with the SSPS might have emerged.
This would have provided useful information regarding the
affective patterning of the alexithymic individuals in the
sample. While it is always desireable to maximize the
power of our statistical tests by maximizing the sample
size, the reality facing most non-funded applied researcher
involves logistical and resource limitations directly
influencing sample size. Such was the case with the
present study.

Of greater concern, in this instance, was the biased
nature of the sample in question. The overwhelming
majority of subjects were white, female, well—educated,
young and, most likely, middle class. The physical

proximity of the participating medical clinics to the

129
University of California, Santa Barbara campus suggests

that the study sample was largely constituted of UCSB
students. The sample demographics support this conclusion.
The resulting restriction in the generalizability of the
conclusions offered by the study is considerable. However,
the existing literature on the measurement of alexithymia
is similarly biased and the current study is within the
demographic parameters established by earlier work in this
area. Still, a broader sampling would be desireable.

In addition to demographic bias, the sample may have
been biased by the self-selection effects of volunteering
for participation in the study. While volunteerism is
generally considered a threat to the external validity of a
study, in this case it may be even more critical by virtue
of the nature of the variables under study and the sample
selection procedure. Participants were asked to fill out
the SSPS at the time of volunteering for the project.

While the SSPS is not considered a particularly
psychologically threatening instrument, a number of the
items potentially suffer from the effects of social
desireability. Social desireability might have influenced
response patterns or the choice to participate in the study
at all. Unfortunately, these potential social
desireability effects are likely to influence racial groups
differentially in respect to the trust level necessary for
cooperating with a predominantly white, middle class

medical facility in the conduct of this type of research.

130
This may have further hampered the recruitment of non-white

participants.

Another consideration involves the effects of time.
While the time lapse between the administration of the SSPS
and the remainder of the instruments was relatively brief,
generally varying between three to ten days, it is possible
that maturation may have had a uneven effect on the more
state-oriented measures such as the STAI-Y State Anxiety
measure, the DES-IV and the BDI. Administration of all
measures at one time would perhaps increase the consistency
of response patterns across instruments and across
subjects.

Several considerations are in order regarding the
measures use in the study. All of the measures were of the
self-report variety which often suffer uduly from the
effects of social desireability. The researcher is at the
mercy of the triangulation of socio-cultural values,
consciously determined honesty and unconsciously determined
defensiveness on the part of the individual subjects.
Internal and external press from these three sources may
have unevenly affected the response patterns of the
participants. In addition, two of the instruments, the
DES-IV and the SMI-A are new and more psychometric research
is needed to establish the validity and reliability of
each. Further, the constructs of mental imagery and
affective differentiation are instrospective and unstable

by nature and, as such, difficult to reliably quantify

131
given a single administration. It may be that mental

imagery and affective differentiation, like blood pressure,
are sufficiently labile to require the averaging of several
observations over time in order to determine a relatively
stable range. If this were the case, intra-individual
variation might exceed inter-individual variation with a
single observation, providing very misleading results.

The SSPS, while being currently the most widely used
instrument in the assessment of alexithymia, has
accumulated limited normative data and is continually under
revision (Blanchard, 1981; Martin, 1984, 1986, 1987; Shipko
8 Noviello, 1984; Sifneos, 1986). Sifneos (1986) has
recently noted the tendency in respondents to avoid the
extremes of the response categories and has proposed a
yes-no format in place of the original four point likert
scale. Much work is needed in bolstering the validity and
reliability of this instrument.

Another limitation of the study was the categorical
coding of the demographic variables of age and education.
This prevented their inclusion in the regression analysis,
necessitating separate bivariate statistical treatment
which was a less powerful test of their respective
relationships with the dependent variable, alexithymia.

In addition to the threats to the external validity
posed by the sample characteristics, there exist threats to
the internal validity of the study in respect to both the

regression analyses and the bivariate analyses.

132
Several factors may have affected the size of the

correlations between the dependent variable and the
independent variables. The distributions of the variables
departed considerably from normal. When the skewness of
distributions takes the opposite direction, reduction in r
is markedly increased along with the heteroscdasticity of
the residuals. The reliability of the individual measures
will also affect the size of r and, as discussed above, the
reliability estimates of the instruments used in the study
are not high. A limitation related to reliability,
particularly under the condition of heteroscedastic
residuals, is one of restriction of range on any of the
variables due to the sampling procedure. A wider range of
scores as a result of a less restricted sample may have
provided a more reliable measure of the variables under
consideration and higher r's.

Finally, there were threats to the assumptions
underlying the confident interpretation of the bivariate
analyses. The lack of experimental controls such as
randomization and balancing allowed for the violation of
the assumptions of the normality of the distributions and
the homogeneity of variances across groups. Hays discusses
the importance of these violations in the fixed-effects
model of the analysis of variance and asserts that
meaningful results can still emerge if the sample is of
sufficient size and the results are presented

conservatively (Hays, 1981, pg.347.).

133
Implications for Future Research

The results of the present study show partial
validation of the alexithymic construct in a sample of
primarily college-age, white females presenting for medical
services.

There are several implications for continued study.

Of serious concern is the validity and reliability of the
measures used to tap the constructs of mental imagery and
affective differentiation. While the present study found
significant results with the SMI-A, this was not the case
with the DES-IV. The results should not be taken, directly
or indirectly, as a validation of the instruments used. A
larger sample would allow the examination of the
contribution of the subscales of the DES-IV to variation in
SSPS scores. While the present study found no significant
correlation between the SSPS and the DES-IV total score, it
is possible that individual or groups of subscale scores
might have correlated highly with the dependent measure,
providing useful insights into affective patterning and
alexithymia. However, the demand characteristics of the
DES-IV may focus more on emotional experience rather than
emotional expression and its use in future research on
alexithymia should be approached cautiously. Perhaps use
of the RTS Differentiation of Affect subscale (Rosenberg,
1984) would be an improvement over the DES-IV in the
ability to measure acutal expression of differentiated

affective experience.

134
The results of the study do suggest that the continued

use of the SMI is warranted in future research on
alexithymia. Perhaps paring the instrument to the
Vividness subscales would be prudent. Focusing
specifically on the kinesthetic, somesthetic and tactile
modality subscale scores together with the addition of a
symptom checklist might provide interesting insights into
the psychophysiological phenomenon of somatizastion.

A related measurement issue pertains to a single
observation vs multiple observations of the mental imagery
and affective differentiation constructs. As indicated in
the previous section, the constructs may be sufficiently
labile to obscure the person x Situation interaction.
Future researchers might consider taking several measures
over several day’s time in order to smooth out the state
(situation) fluctuations which is a within-subject factor.
This would allow for a clearer delineation of the trait
(person) component of mental imagery and affective
differentiation as stable capacities over time, or
between-subjects factor. It is this between-subjects trait
factor that should be most highly correlated with the
measurement of alexithymia.

A number of investigators have worked with the
physiological concomitants of alexithymia (Kaplan 8 Wogan,
1976/77; Martin, 1984, 1986; Hoppe, 1977, Nemiah, 1975,
1977). Future research might include physiological

measures so that the psychophysiological interplay can be

135
more clearly articulated. Martin (1986) has measured

frontal EMG, digital blood volume pulse amplitude and heart
rate. These measures could be built into future studies.
However, the problem of the effect of instrumentation is
increased whenever the complexity of the measuring devices
is increased. Physiological measurements obscure the
natural phenomena that they seek to isolate simply by
virtue of their artificiality and uneven impact across
individuals. Unlike eduational research, where it can be
expected that the results can be generalized to populations
accustomed to paper and pencil testing situations, it's
difficult to imagine a population that could be expected to
routinely encounter frontal EMG measurement.

Another group of implications cluster around sampling
procedures. Unless we plan to restrict our applications of
research findings to college counseling centers only,
future study must take place with a broader sample base.
The college sophomore analogue study has severe limitations
in its generalizability. The present study attempted to
broaden this sample base by recruiting subjects off campus.
It's aim was only minimally achieved. It is recommended
that future study be conducted with samples that more
nearly approximate the broader population to which we can
usefully generalize our results.

Summaxy
In summary, the results of the present study

contributed to the empirical understanding of the

136
alexithymic construct. Validation of the mental imagery

component was achieved. The findings of previous workers
in relation to the independence of alexithymia from anxiety
were supported. The results suggested that depression may
be related to alexithymia in the present sample. This
finding is contrary to several previous studies and
possible explanations were discussed. The relative
independence of alexithymia from the background variables
of age, sex, and race was also supported. It was found
that level of education may be related to alexithymia but
findings were weak. In this study, the validation of the
affective differentiation component of alexithymia was not
achieved. Possible explanations for this finding were

offered.

REFERENCES

Achterberg, J. 8 Lawlis, G.F. (1978). Imagexy 9; cancer:
A diagnostic tool for the process 2; disease.
Champaign, Ill: Institute for Personality and Ability
Testing.

8 Lawlis, G.F. (1984). Imagexy pg disease:
A diggnostic tool for behavioral medicine. Champaign,
Ill: Institute for Personality and Ability Testing.

 

. (1984). Imagery and Medicine:
Psychophysiological speculations. Journal 93
Mental Imagery, g, 1-14.

 

Ahsen, A. (1982). Imagery in perceptual learning and
clinical application. Journal pg Mental Imagery, 9,
157-186.

. (1984). ISM: The triple code model for imagery
and psychophysiology. Journal 9; Mental Imagery, g,
15-42.

 

Alexander, F. (1943). Fundamental concepts of psychosomatic
research: Psychogenesis, conversion, specificity.
Psychosomatic Medicine, 5, 205-210.

. (1950). Psychosomatic Medicine. New York:
Norton 8 Co.

 

American Psychological Association. (1982). Ethical
Principles ix the conduct 9; research with human
participants. Washington: APA.

American Psychiatric Association. (1980). Diagnostic and
statistical manual 1 III. Washington: APA.

 

Ammon, G. (1979). Psychosomatic illness as a result of a
a deficit in egostructure under consideration of the
genetic, dynamic, structural, and groupdynamic point
of view. Psychotherapy and Psychosomatics, 3;,
179-189.

Apfel, R.T. 8 Sifneos, P.E. (1979). Alexithymia: Concept
and measurement. Psychotherapy and Psychosomatics,
3;, 180-190.

B.M.D.P. (1983). Statistical Software Manual. Los Angeles:
UCLA Press.

Bagby, R.M., Taylor, G.J. 8 Ryan, D. (1986). Toronto
alexithymia scale: Relationship with personality and
psychopathology measures. Psychotherapy and
Psychosomatics, 45, 207-215.

137

138

Barker, B.M., Barker, H.R. 8 Wadsworth, A.P., Jr. (1977).
Factor analysis of the items of the State-Trait
Anxiety Inventory. Journal 9; Clinical Psychology,
33, 450-455.

Beck, A.T., Ward, C.H., Mendelson, M., Mock, J. 8
Erbaugh, J. (1961). An inventory for measuring
depression. Archives 9; General Psychiatry, 3,
561-571.

, Rush, A.J., Shaw, B.F. 8 Emery, G. (1979).
Cognitive therapy 92 depression. New York: Guilford
Press.

 

, Steer, R.A. 8 Garbin, M.G. (in press).
Psychometric properties of the Beck Depression
Inventory: 25 years of evaluation. Review 2; Clinical
Psychology.

 

Bell, I.R. 8 Schwartz, G.E. (1975). Voluntary control and
reactivity of human heart rate. Psychophysiology, 33,
339-348.

Blanchard, E.B., Arena, J.G. 8 Pallmeyer, T.P. (1981).
Psychomentric properties of a scale to measure
alexithymia. Psychotherapy and Psychosomatics, 35,
64-71.

Blumberg, S.H. 8 Izard, C.E. (1985). Affective and
cognitive characteristics of depression in 10- and 11-
year old children. Journal g; Personality and Social
Psychology, 33, 149-202.

. (in press). Discriminating
patterns of emotions in 10- and 11- year old
children's anxiety and distress. Journal 93
Personality and Social Psychology.

 

Buchanan, D.C., Waterhouse, G.J. 8 West, S.C. (1980). A
proposed neurophysiological basis of alexithymia.
Psychotherapy and Psychosomatics, 33, 248-255.

Burns, D. 8 Beck, A.T. (1978). Modification of mood
disorders. in Cognitive and behavioral therapy.
Foreyt, J. and Rathjic, D. (Eds.). New York:
MacMillan.

 

Cannon, W. (1920). Bodily changes 3p painy hunger, fear,
and rag . New York: D. Appleton and Co.

Cattell, R.B. 8 Scheier, I.H. (1963). Handbook for the
IPAT anxiety scale. (2nd Ed.) Champaign Ill.:
Institute for Personality and Ability Testing.

139

Cohen, J. 8 Cohen, P. (1983). Applied multiple regression/
correlation analysis for the behavioral sciences.
Hillsdale: Lawrence Erlbaum Associates.

Cohen, S. (1979). Updating the model for psychotherapy of
psychosomatic problems. Psychotherapy and
Psychosomatics, 33, 72-90.

Cronbach, L.J. (1972). The Dependability pg Behavioral
Measures. New York: John Wiley and Sons.

Darwin, C. (1904). The expression 2; emotion 3p man and
animals. London: Murray (original work published in
1872)

Davidson, R.J. (1984). Hemispheric asymmetry and emotion.
in Approaches pg Emotion. Scherer 8 Ekman (Eds.).
Hillsdale: Lawrence Erlbaum Associates.

Delay, J., Pichot, P., Lemperiere, T., 8 Mirouze, r.
(1963). Classification of depressive states: Agreement
between etiology and syptomology: 2. Results of Beck's
Questionnaire. Encephale, 33, 497-505.

Ekman, P. 8 Friesen, W. V. (1969). The repertoire of non-
verbal behavior: Categories, origins, usage and
coding. Semiotica, 3, 49-98.

 

. (1975). Unmasking the Face: 3 Guide pg
Recognizing Emotions from Facial Clues. Englewood
Cliffs: Prentice Hall.

. (1980). Facial signs of emotional experience.
Journal 93 Personality and Social Psychology, 33,
1125-1134.

 

. (1982). Emotion 3p the human face. New York:
Cambridge University Press.

(1983). Autonomic nervous system activity
distinguishes among emotions. Science, 221,
1208-1210.

 

Engel, G.L. 8 Schmale, A.H. (1967). Psychoanalytic theory
of somatic disorder: Conversion, specificity, and the
disease onset situation. Journal 2; the American
Psychoanalytic Association, 33, 344-365.

Fenichel, O. (1945). The psychoanalytic theory 23 neurosis.
New York: Norton and Co.

Ferguson, G.A. (1981). Statistical Analysis 3p Psychology
and Education. New York: McGraw-Hill Co.

 

140

Freeman, R.D. (1979). Developmental Neurobiology 22 Vision.
New York: Plenum Press.

Friedman, S. (1978). A psychophysiological model for the
chemotherapy of psychosomatic illness. Journal 93
Nervous and Mental Disease, 166, 110-116.

Freud, S. (1895). Collected papers. London: Hogarth Press.

Gaudry, E. 8 Poole, C. (1975). A further validation of the
state-strait distinction in anxiety research.
Australian Journal 93 Psychology, 3, 119-125.

, Spielberger, C.D. 8 Vagg, P. (1975).
Validation of the state-trait distinction in anxiety
research. Multivariate Behavioral Research, 33,
331-341.

 

Gottschalk, L.A. 8 Gleser, G. (1969). The measurement 2;
psychological states through the content analysis 93
verbal behavior. Los Angeles: University of California
Press.

 

Greenough, W.T. (1975). Experiential modification of the
developing brain. American Scientist, 33, 37-46.

, (1978). Development and memory: The synaptic
connection. in Brain and Learning. T. Teyler (Ed.).
Stamford: Greylock.

 

Harrell, E. Lambert, P. 8 Achterberg, J. (1981). The
effects of electrical stimulation of the hypothalamus
on macrophagic activity in the rat. Physiological
Psychology, 3, 193-196.

Hays, W.L. (1981). Statistics. New York: CBS College
Publishing.

Heiberg, A.N. 8 Heiberg, A. (1978). A possible genetic
contribution to the alexithymic trait. Psychotherapy
and Psychosomatics, 33, 205-210.

Holt, R.R. (1964). Imagery: The return of the ostracized.
American Psychologist, 33, 254-264.

Hoppe, K.D. (1977). Split brains and psychoanalysis.
Psychoanalytic Qparterly, 33, 220-244.

Hull, C.R. 8 Render, G.F. (1984). The effects of
relaxation, fantasy journeys and free imagery on
control and Vividness of imagery. Journal 93 Mental

Imagexy, 3, 67-78.

141

Ikemi, Y. (1965). Psychological desensitization in
allergic disorders. in J. Lassner (Ed.). Hypnosis and
Psychosomatic Medicine: Proceedings 93 333
International Congress for Hypnosis and Psychosomatic
Medicine. Berlin: Springer-Verlag.

Izard, C.E. (1971). The face 93 emotion. New York: B.
Appleton.

, (1972). Patterns of emotions: A new analysis of
anxiety and depression. New York: Academic Press.

 

 

, Daugherty, P.E., Bloxom, B.M. 8 Kotch, W.E.
(1974). The differential emotions scale: 3 method 93
measuring the subjective experience 93 discrete
emotions. Unpublished manuscript, University of
Deleware.

 

 

, (1977). Human emotions. New York: Plenum
Press.

 

, 8 Schwartz, G.M. (1986). Patterns of emotion.
in Depression in young people. Rutter, Izard and Read
(Eds. ). New York: Guilford Press.

 

James, W. (1950). The principles of psychology, Vol 3. New
York: Dover (original work published in 1890)

Kaplan, C.D. 8 Wogan, M. (1976/77). Management of pain
through cerebral activation: An experimental analogue
of alexithymia. Psychotherapy and Psychosomatics, 33,
144-153.

Kellner, R. (1975). Psychotherapy in psychosomatic
disorders: A survey of controlled studies. Archvies
23 General Psychiatry, 33, 1021-1028.

(1979). Psychotherapeutic strategies in the
treatment of psychophysiological disorders.
Psychotherapy and Psychosomatics, 33, 91-100.

 

Kendall, P.C., Finch, A.J., Jr., Auerbach, S.M., Hooke, J.
8 Mikulka, P. (1976). The State-Trait Anxiety
Inventory: A systematic evaluation. Journal 93
Consulting and Clinical Psychology, 33, 406-412.

King, D. L. (1973). An image theory of classical
conditioning. Psychological Reports, 33, 403-411.

. (1974). An image theory of instrumental
conditioning. Psychological Reports, 33, 1115-1122.

 

Kolers, P.A. (1983). Perception and representation. Annual
Review 93 Psychology, 33, 129-166.

142

Kotch, W.E., Gerbing, D.W. 8 Schwartz, L.E. (1982). The
construct validity of the Differential Emotions Scale
as adapted for children and adolescents. in Measuring
emotions 3p infants and children. Izard, C.E. (Ed.).
Cambridge, England: Cambridge University Press.

Krystal, H. (1962). The study of withdrawal from narcotics
as a state of stress. Psychiatric Quarterly

Supplement, 33, 53.

, 8 Raskin, H. (1970). Drug dependence.
Detroit: Wayne State University Press.

 

. (1971). Trauma: Consideration of severity and
chronicity. in Psychic traumatization. Krystal, H. and
Niederland, W. (Eds.). Boston: Little and Brown.

 

. (1975a). The genetic development of affects and
affect regression. Annual 93 Psychoanalysis, V. II.
New York: International Universities Press.

 

. (1975b). Affect tolerance. Annual 93
Psychoanalysis, V. III. New York: International
Universities Press.

 

. (1977). Aspects of affect theory. Bulletin 93
the Menninger Clinic, 33, 1-26.

 

. (1979). Alexithymia and psychotherapy.
American Journal 93 Psychotherapy, 33, 17-31.

 

LaBarre, W. (1947). The cultural basis of emotions and
gestures. Journal 9; Personality, 33, 49-68.

Lazarus, R.S. (1984). Thoughts on the relations between
emotion and cognition. in Approaches pg emotion.
Scherer 8 Ekman (Eds.). Hillsdale, N.J.: Lawrence
Erlbaum Assoc.

Leach, E. (1972). The influence of cultural context on
nonverbal communication in man. in R. Hinde (Ed.).
Nonverbal communication. Cambridge: Cambridge
University Press.

Leibovich, M.A. (1973). Psychogenic vomiting:
Psychotherapeutic considerations. Psychotherapy and
Psychosomatics, 33, 263-268.

Lesser, I.M. (1981). A review of the alexithymic concept.
Psychosomatic Medicine, 33, 531-543.

Leventhal, H. 8 Tomarken, A. (1986). Emotion: Today’s
problems. Annual Review 9; Psychology, 33, 565-610.

143

Levitan, H. (1976/77). Psychological factors in the
etiology of ulcerative colitis: Objectlessness and
rage. International Journal 3; Psychiatry 33
Medicine, 1, 221-228.

Locke, S.E. 8 Honig-Rohan, M. (1983). Mind and immunity:
Behavioral immunologyy gp annotated bibliography,
1976-1982. New York: Institute for the Advancement of
Health.

Lolas, F., Parra, G., Aronson, S. 8 Collins, C. (1980).
On the measurement of alexithymic behavior.
Psychotherapy and Psychosomatics, 33, 139-146.

Lorenz, K. (1965). Evolution and modification 33 behavior.
Chicago: University of Chicago Press.

Marks, D.F. (1973). Visual imagery differences in the
recall of pictures. British Journal 33 Psychology,
33, 17-24.

Martin, J.B., Pihl, R.o. 8 Dobkin, P. (1984). Schalling-
Sifneos personality scale: Findings and
recommendations. Psychotherapy and Psychosomatics,
33, 145-152.

Marty, P. 8 M’Uzan, M. (1963). La 'pensee operatiore'.
Review 9; French Psychoanalysis, (supplement), 1345.

, 8 Pihl, R.O. (1986). Influence of alexithymic
characteristics on physiological and subjective stress
responses in normal individuals. Psychotherapy and
Psychosomatics, 33, 66-77.

. (1987). Personal Communication.

Mead, M. (1975). Darwin and facial expression. Journal
3; Communication, 33, 209-213.

Mitscherlich, M. (1977). The significance of the
transitional object for psychosomatic thingking.
Psychotherapy and Psychosomatics, 33, 272-277.

Moran, P.W. 8 Lambert, M.J. (1983). A review of current
assessment tools for monitoring changes in depression.
in The Assessment 33 Psychotherapy Outcome. Lambert,
M.S., Christensen, E.R., 8 Dejulio, S.S. (Eds.). New
York: Wiley.

Musaph, H. (1974). The role of aggression in somatic
symptom formation. International Journal 3;
Psychiatry 33 Medicine, 3, 449-460.

144

Nelson, D. (1985). Impact 9; negative life events upon
psychological functioning: Comparison 33 cumulative,
event-specifigy and reporting-bias effects.
Unpublished doctoral dissertation, University of
Delaware.

 

Nemiah, J. (1974). Conversion: Fact or chimera?
International Journal 3; Psychiatry 33 Medicine, 3,
443-448.

. (1975). Denial revisited: Reflections on psycho-
somatic theory. Psychotherapy and Psychosomatics, 33,
140-147.

 

. (1977). Alexithymia: Theoretical considerations.
Psychotherapy and Psychosomatics, 33, 199-206.

 

Paivio, A. (1971). Imagery and Verbal Processes. Toronto:
Holt, Rinehart 8 Winston.

Plutchik, R. (1980). Emotion: A psychoevolutionary
synthesis. New York: Harper and Row.

 

. (1984). Emotions and imagery. Journal 33 Mental
Imagexy, 3, 105-112. ‘

Pribram, K.H. (1984). Emotion: A neurobehavioral analysis.
in Approaches pp Emotion. Scherer 8 Ekman (Eds.).
Hillsdale: Lawrence Erlbaum Associates.

Price, A.D. (1975). Heart rate variability and respiratory
concomitants of visual and nonvisual "imagery" and
cognitive style. Journal 33 Research 33 Personality,
3, 341-355.

Pylyshyn, Z.W. (1973). What the mind's eye tells the mind’s
brain: A critique of mental imagery. Psychological
Bulletin, 33, 1-24.

Richardson, A. (1983). Imagery: Definitions and types. in
A.A. Sheikh (Ed.). Imagery: Current Theory, Research
and Applications. New York: Wiley.

 

Rodenhauser, P., Khamis, H.J. 8 Faryna, A. (1986).
Alexithymia and handedness: A pilot study.
Psychotherapy and Psychosomatics, 33, 169-173.

Rogers, M.P., Dubey, D. 8 Reich, P. (1979). The influence
of the psyche and the brain on immunology and disease
susceptibility: A critical review. Psychosomatic
Medicine, 33, 147-165.

145

Rosenberg, J. (1984). Evaluation 33 g prggram 33 preventive
mental health: ”Building 3 competent self". Doctoral
Dissertation, Michigan State University.

 

 

Schacter, S. (1957). Pain, fear and anger in hypertensives
and normotensives. Psychosomatic Medicine, 33, 17-29.

. 8 Singer, J. (1962). Cognitive, social and
physiological determinants of emotional state.
Psychological Review, 33, 379-399.

 

Schmale, A.H. (1964). A genetic view of affects: With
special reference to the genesis of helplessness and
hopelessness. Psychoanalytic Study 33 the Child, 33,
287-310.

Schneider, J., Smith, C.W. 8 Whitcher, S. (1983). The
relationship of mental imagery to white blood cell
(neutrophil) function: Experimental studies of normal
subjects. Summarized in Achterberg 8 Lawlis, 1984.

Schneider, P. 8 wulhiemier, F. (1979). The psychotherapy of
the psychosomatic patient. Psychotherapy and
Psychosomatics, 33, 112-117.

Schnurr, R., Hoaken, P.C.S. 8 Jarrett, F.J. (1976).
Comparison of depression inventories in a clinical
population. Canadian Psychiatric Association Journal,
3;, 473-476.

Schreter, R. (1978/79). Treating the untreatables: Identi-
fying the somatizing borderline patient.
International Journal 33 PsychiatrY 12 Medicine, 3,
207-216.

Schur, M. (1955). Comments on the metapsychology of
somatization. Psychoanalytic Study 33 the Child, 33,
119-164.

 

Selye, H. (1946). General adaptation syndrome and diseases
of adaptation. Journal 33 Clinical Endocrinology, 3,
117-230.

Shands, H. (1977). Suitability for psychotherapy II.
Unsuitability and psychosomatic disease.
Psychotherapy and Psychosomatics, 33, 28-35.

Sheehan, P.W. (1967). A shortened form of Bett’s
Questionnaire upon Mental Imagery. Journal 33
Clinical Psychology, 33, 386-389.

146

, Ashton, R. 8 White, K. (1983). Assessment of
mental imagery. in Imagery: Current Theory, Research,
333 Application. Sheikh, A.A. (Ed.). New York:
Wiley.

Sheikh, A.A. 8 Jordan, C.S. (1983). Clinical uses of mental
imagery. in Imagery: Current Theory, Research, 333
Application. Sheikh, A.A. (Ed.). New York: Wiley.

 

. 8 Kunzendorf, R. (1984). Imagery, physiology
and psychosomatic illness. International Review 33

Imagexy, 3, 95-138.

Shipko, S. 8 Noviello, N. (1984). Psychometric properties
of self-report scales of alexithymia. Psychotherapy
and Psychosomatics, 33, 85-90.

 

Sifneos, P. (1967). Clinical observation on some patients
suffering from a variety of psychosomatic diseases.
Proceedings 33 the 7th European Conference pp
Psychosomatic Research. Basel: Karger Publishing.

. (1974). A reconsideration of psychodynamic
mechanisms in psychosomatic symptom formation in view
of recent clinical observations. Psychotherapy and
Psychosomatics, 33, 151.

 

. (1975). Problems of psychotherapy of patients
with alexithymic characterstics and physical disease.
Psychotherapy and Psychosomatics, 33, 68.

 

, Apfel-Savitz, R. 8 Frankel, F. (1977). The
phenomenon of "Alexithymia": Observations in neurotic
and psychosomatic patients. Psychotherapy and
Psychosomatics, 33, 47-57.

 

. (1983). Psychotherapies for psychosomatic and
alexithymic patients. Psychotherapy and
Psychosomatics, 33, 66-73.

 

. (1986). The Schalling-Sifneos Personality Scale
Revised. Psychotherapy and Psychosomatics, 33,
161-165.

 

Spector, N.H. 8 Korneva, E.A. (1979). Neurophysiology,
immunology and neuroimmunomodulation. in
Psychoneuroimmunology. R. Adler (Ed.). New York:
Academic Press.

Spielberger, C.D., Vagg, P.R., Barker, L.R., Donham, C.W. 8
Westberry, L.G. (1980). The factor structure of the
State-Trait Anxiety Inventory. in Stress and Anxiety
(vol 7), 1.6. Sarason 8 C.D. Spielberger (Eds.). New
York: Hemisphere/Wiley.

147

. (1983). Manual for the State-Trait Anxiety
Inventory. Palo Alto: Consulting Psychologists Press.

 

Stein, M., Schiavi, R. 8 Camerino, M. (1976). Influence of
brain and behavior on the immune system. Science,
191, 434-440.

Stephanos, S. (1975a). A concept of analytic treatment for
patients with psychosomatic disorders. Psychotherapy
and Psychosomatics, 33, 178-187.

 

. (1975b). The object relations of the
psychosomatic patient. British Journal 33 Medical
Psychology, 33, 257-266.

 

. (1976). Ambulatory analytical psychotherapy for
the treatment of psychosomatic patients: A report on
the method of "relaxation analytique". British
Journal 33 Medical Psychology, 33, 305-313.

 

. (1978). Pathological primary identifications and
their effects on the psychosomatic economy of the
individual. Psychotherapy and Psychosomatics, 33,
56-67.

 

. (1979). Libidinal cathexis and emotional growth
in the analytic treatment of psychosomatic patients.
Psychotherapy and Psychosomatics, 33, 101-111.

 

Suler, J.R. (1985). Imagery ability and the experience of
affect by free associative imagery. Journal 33 Mental
Imagegy, 3, 101-110.

Switras, J.E. (1978). An alternate-form instrument to
assess Vividness and controllability of mental imagery

in seven modalities. Perceptual and Motor Skills, 33,
379-384.

 

. (1979). Survey 33 Mental Imagery: Test Manual.
Manuscript available from the author.

Taylor, G., Doody, K. 8 Neuman, A. (1981). Alexithymic
characteristics in patients with inflammatory bowel
disease. Canadian Journal 33 Psychiatry, 33, 470-474.

Taylor, J.A. (1953). A personality scale of manifest
anxiety. Journal 33 Abnormal and Social Psychology,
33, 285-290.

Tomkins, S. (1962). Affect, imagery and consciousness, Vol.
I. New York: Springer Publishing.

. (1963). Affect, imagery and consciousness, Vol.
II. New York: Springer Publishing.

 

148

. (1975). The phantasy behind the face. Journal
33 Personality Assessment, 32, 551-562.

 

 

. (1978). Script theory: Differential
magnification of affects. in Nebraska Symposium 33
Motivation, 33. Howe 8 Dienstbier (Eds.). Lincoln:
University of Nebraska Press.

 

. (1982). Affectl imagery and consciousness, Vol.
III. New York: Springer Publishing.

 

. (1984). Affect theory. in Approaches 33 Emotion.
Scherer 8 Ekman (Eds.). Hillsdale: Lawrence Erlbaum
Associates.

 

 

Vagg, P.R., Spielberger, C.D. 8 O’Hearn, T.P. (1980). Is
the State-Trait Anxiety Inventory multidimensional?
Personality and Individual Differences, 3, 207-214.

Von Rad, M. (1979). Comments on theory and therapy of
psychosomatic patients with a follow-up study.
Psychotherapy and Psychosomatics, 33, 118-127.

, 8 Lolas, F. (1978). Empirical evidence of
alexithymia. Psychotherapy and Psychosomatics, 33,
91-102.

 

Weiner, F. (1977). The concept of alexithymia and the
future of psychosomatic medicine. Psychotherapy and
Psychosomatics, 33, 376-388.

Winnicott, D.W. (1953). Transitional objects and transi-
tional phenomena. International Journal 33
Psychoanalysis, 33.

Wittkower, E. (1974). Historical perspective of
contemporary psychosomatic medicine. International
Journal 33 Psychiatry 33 Medicine, 3, 309-319.

Wolff, H. (1968). The psychodynamic approach to
psychosomatic disorders: Contribution and limitations
of psychoanalysis. British Journal 33 Medical
Psychology, 33, 343-348.

Zajonc, R.B. (1984). On primacy of affect. in Approaches 33
Emotion. Scherer 8 Ekman (Eds.). Hillsdale: Lawrence
Erlbaum Associates.

Zuckerman, M. (1960). The development of an affect
adjective check list for the measurement of anxiety.
Journal 33 Consulting Psychology, 33, 457-462.

APPENDICES

APPENDIX A

SCHALLING-SIFNEOS PERSONALITY SCALE-REVISED

l 50
APPENDIX A.

1) Does not apply at all.
2) Does not apply very much.
3) Applies very much.
4) Applies completely.
Please indicgfie only ONE: 1.2.3 or 4
1. I find it hard to describe how I feel about people. . . .

2. Mhen something unpleasant happens. I feel as if I
COUld expl°de O o o o o o o o o o o o o o o o o o o o o o

3. I often remember the content of dreams very vividly. . .
4. Feelings are what make life worthwhile. . . . . . . . . .

5. I prefer movies with action rather than psychological
dramas. o o o o o o o o o o o o o o o o o o o o o o o o o

6. I do not tend to examine my own feelings. . . . . . . . .

7. I think it is not worthwhile discussing how one feels.
I prefer to act. . . . . . . . . . . . . . . . . . . . .

8. I find life pretty boring most of the time. . . . . . .
9. One may say that I lack imagination. . . . . . . . . . .
10. I hardly ever cry except when Ifm frustrated. . . . . . .
11. It is important to find out how one feels about people. .
12. I daydream rarely. . . . . . . . . . . . . . . . . . .
13. I find it hard to find the right words for my feelings. .

14. I spend much time daydreaming whenever I have nothing
else to do. 0 O O O O O O I O O O O I O O O I O O O I 0

15. I prefer taking action rather than thinking. . . . . . .
16. I don't seem to get really excited about anything. . . .
17. It is easy for me to describe events in detail. . . . . .

18. I like people better than things. . . . . . . . . . . . .

Code §

APPENDIX B

DIFFERENTIAL EMOTIONS SCALE-IV

1.51
APPENDIX B

DES-IV

Take a few moments and think over the past month or so, the events
in your life and your reactions to them. After you've done so,
mark your responses to the following items in terms of how often
you have felt the way indicated in the item.

Rarely Hardly Some- Very
or Never Ever times Often Often

1. Feel regret, sorry about

something you did............... 1 2 3 4 5
2. Feel sheepish, like you do

not want to be seen............. 1 2 3 4 5
3. Feel glad about something....... 1 2 3 4 5
4. Feel like something stinks.

puts a bad taste in your mouth.. 1 2 3 4 5
5. Feel you can't stand yourself... 1 2 3 4 5
6. Feel embarrassed when anybody

sees you make a mistake......... 1 2 3 4 5
7. Feel unhappy, blue. downhearted. 1 2 3 4 5

8. Feel surprised, like when some-
thing suddenly happens you had

no idea would happen............ 1 2 3 4 5
9. Feel like you are blushing...... 1 2 3 4 5
10. Feel like somebody is a low-

life, not worth the time of day. 1 2 3 4 5
11. Feel like you are not worth

anythingOOOOOIOOOOOOIOOOIOIOOOO. 1 2 3 a 5
12. Feel shy, like you want to‘hide. 1 2 3 5
13. Feel like what you're doing

or watching is interesting...... 1 2 3 4 5
14. Feel scared. uneasy, like

something might harm you........ 1 2 3 4 5
15. Feel mad at somebody............ 1 2 3 4 5
16. Feel mad at yourself............ 1 2 3 4 5

Code #

1_52

DES-IV, Page 2
Rarely Hardly Some- Very
or Never Ever times Often Often

17. Feel ashamed because you do
not know what to do............ 1 2 3 4 5

18. Feel happyCICOOOIOI... ....... O. 1 2 3 u 5

19. Feel like somebody is a
u gOOd-fOP-nOthing" o I 0 o o o o o o o o o o 1 2 3 4 5

20. Feel you are a
"good-for-nothing"............. 1 2 3 4 5

21. Feel like someone made you
look like a fool............... I 2 3 4 5

22. Feel so interested in what
you're doing that you're
caugrlt upinitOIOOlOOOOOOOOOOO 1 2 3 a 5

23. Feel amazed, like you can't A
believe what's happened, it

was so unusual................. 1 2 3 4 5
24. Feel fearful. like you're

in danger, very tense.......... 1 2 3 4 5
25. Feel like you are dumb......... 1 2 3 4 5

26. Feel like screaming at
somebody or banging on

somethmgOOOOIOOOOOOOO0.0.00...1 2 3 a 5
27. Feel sad and gloomy, almost

like crying.................... I 2 3 4 5
28. Feel like you did something

mongOOOOOIOOOOOOOOOOOOOOOIOOOO1 2 3 a 5
29. Feel bashful, embarrassed...... 1 2 3 4 5

30. Feel disgusted, like
something is sickening......... 1 2 3 4 5

31. Feel joyful. like everything
is going your way, everything
is rosyOO...OOOOIOOOOOOOOIOOOO. 1 2 3 a 5

32. Feel like people laugh at you.. 1 2 3 4 5

Code #

DES-IV, Page 3

Rarely Hardly Some- Very
or Never Ever times Often Often

 

33. Feel like things are so
rotten they could make you

SiCKeooooooo00090000000000.0000 1 2 3 u’ 5
34. Feel sick about yourself....... 1 2 3 4 5
35. Feel worried abou the way

you IOOKOCOOOOOOOOOOOOI.0...... 1 2 3 a 5
36. Feel like you are better

thm somebOdyOOOOO0.00.0.0.0... 1 2 3 L‘ S
37. Feel you are no good. a nobody. 1 2 3 4 5

38. Feel like you ought to be
blamed for something........... 1 2 3 4 5

39. Feel like whatever you do will
not be very good............... 1 2 3 4 5

40. Feel the way you do when
something unexpected happens... 1 2 3 4 5

41. Feel alert. curious. kind of
excited about something........ 1 2 3 4 5

42. Feel angry. irritated.
annoyed with somebody.... ..... . 1 2 3 4 5

43. Feel angry and annoyed
with yourself.................. 1 2 3 4 5

44. Feel like you cannot say what
you want to say as well as
otherSOeoooooooo00000090000000. 1 2 3 a 5

45. Feel discouraged. like you
can't make it. nothing's going

rigtOOOCOOOOIOOOOOOOIOOIICOOOC 1 2 3 a 5
46. Feel ashamed, like you want

to disappearooooooooooooooocoo. 1 2 3 I" 5
47. Feel afraid.................... 1 2 3 4 5

48. Feel like people always look
at you when anything goes
wrongOOOOOOOOOOOOOOOOOOII0.0... 1 2 3 a 5

49. Feel lonely.................... 1 2 3 4 5

Code #

1 5L}
APPENDIX C

SURVEY OF MENTAL IMAGERY: FORM A

Devised by Joseph E. Switras

Instructions to examinees:

The following is a questionaire designed to determine the
type of mental images that you are able to produce and
manipulate. But first. what are mental images? In the past
they have been called pictures in the mind. but actually images
can be tastes. sounds. feelings. sensations. as well as visual
scenes. They can also be combinations of sights. tastes.
feelings. etc. An image can be something that you see when
your eyes are closed; something that may look as if you can
just reach out and pick it up. but which is really not there.
An image can be the taste of an orange when you have not
actually eaten one. An image may be the smell of a flower when
you try to remember what one smells like. As a last example.
the picture in your mind of your home as you try to recall what
it looks like. is also an image. An image can be experienced
as a photograph. a movie or as if you are really there and it
is really happening. .

In responding to this questionnaire you will be asked to
imagine that a variety of things are actually happening. You
may be asked to close your eyes and try to see a flowerpot. one
with a large red flower growing out of it. You may be asked if
you can smell the flower and maybe even water it. All this
will occur in your thoughts only. but at the time may seem as
real as the chair in which you are seated. For most people
this is a new and exciting experience. one that proves quite
interesting.

With the actual image proposals. please choose the answer
that is closest to describing what it is that you are
experiencing. Answer every question. even if the answers do
not express precisely how you feel. Mark your answers on the
booklet as indicated on the next page.

 

C Copyright. Joseph E. Switras 1975

All rights reserved

 

Code #

155

The following sections will deal with actual images that
you will attempt to produce. These will involve seeing.
hearing. smelling. tasting. feeling and doing things that
occur in your mind and imagination only. Respond to each
test item in the following manner. First read the item. then
close your eyes and try to have the proposed experience.
Whatever the task indicated by the item (visualizing a
picture. tasting a fruit etc.). pay attention to two elements
of what is occurring: (1 How well you can control or manipplate
the image. and (2) How vivid or real is the scene. taste.
sensation. etc. By control is meant experiencing the scene as
close to the item instructions as possible. For example. being
able to form a mental picture of a squirrel eating an acorn.

Each item is followed by two opportunities to respond.
(1) Since each item is in the form of a question. it is possible
to respond with either a 'yes'. 'no'. or 'unsure'. On the answer
sheet black blacken over the number three (ypg) if ou produced
the proposed image: blacken over the number pp; (52 if you did
not produce the image. If you are really not sure if the image
was there. blacken over the number 312 (unsure).
(2) Next. five numbers follow preceded by the word 'vividness'.
Each number tells how vivid or real the image was as you experienced
it. On the answer booklet blacken in the number pp; (1) if there
is absolutel no ima e. and all that is happening is that you are
thinking of the scene. odor. sound. etc. Blacken over the number
two (2) if you are uncertain of the image. if the image is
Indistinct. vagpe. ambi ous. gig. has . doubtful. etc. Blacken
over the number three 3 if the image is limited or moderatel
clear. vivid. and perceptible. Blacken over the number four (i)
if the proposed experience is reasonably unobscure. vivid. and
clear. Finally. blacken over the number five (55 if the experience
(image) seems as if it is really happening. Here the image should
be distinct. 'photoggaphic'. and perfectly clear and vivid; exactly
the experience proposed.

 

Be sure that you attempt each item. and respond to both
questions that follow the item.

An example of how to respond:
Q. Can you visualize a book?

1. 1-no Z-unsure 4r;

2. Vividness 1 2 3

Responses one and two indicate that the image of a book did
occur. and that it seemed as if a real book was actually there.

Code y

1n56

Example two:

Q. Can you aste sour milk?
3. 1-no unsure 3-Kes
1+. Vividness 1" 3 5

Responses three and four indicate that this person could not be
sure if the image was there. However. some indistinct. vague.
trace of a taste occurred that seemed to resemble sour milk.

When ready. you pay begin. Be sure to respond to each item.

VIVIDNESS SCALE:

1) Absolutely no image.

2) Indistinct. vague. ambiguous. dim. hazy. doubtful.

3) Limited. moderately clear. vivid and perceptible.

u) Reasonably unobscure. vivid and clear.

5) Really happening: distinct. photographic. perfectly clear

_ -afld-VlVid; __________________________

I. ATTEMPT THESE ITEMS WITH YOUR EYES CLOSED. TRY TO GET A
MENTAL PICTURE (VISUALIZE) OF WHAT IS PROPOSED IN EACH ITEM.

Can you see the color red?
1. l-no 2-unsure 3-yes
2. Vividness 1 2, 3 a 5

Can you see a horse standing alone?
3. 1-no 2-unsure 3-yes
4. Vividness 1 2 3 4 5

Can you see the horse trot away?
5. 1-no 2-unsure 3-yes ,
6. Vividness 1 2 3 4 5

Can you see a bird sitting on a telephone wire?
7. 1-no Z—unsure 3-yes
8. Vividness 1 2 3 4 5

Can you see the bird jump from the wire and fly to the ground?
9. 1-no 2-unsure 3-yes
10. Vividness 1 2 3 u 5

Can you see the bird fly up to and land on the branch of a tree?
11. 1-no Z-unsure 3-yes
12. Vividness 1 2 3 4 5

Can you see a bottle on a picnic table?

13. l-no 2-unsure 3-yes
1a. Vividness 1 2 3 h 5

Code #

1n57

VIVIDNESS §Q§;§:
Absolutely no image.
2) Indistinct. vague. ambiguous. dim. hazy. doubtful.
3) Limited. moderately clear vivid and perceptible.
h) Reasonably unobscure vivid and clear.
5) Really happening: distinct. photographic. perfectly clear
_ -afld.Y1Ylda .........................
Can you see the same bottle on the picnic table. filled with
a colored liquid?
15. i-no 2-unsure 3-yes
16. Vividness 1 2 3 u 5

Can you now see the same bottle with a different colored liquid?
17. 1-no Z-unsure 3-yes
18. Vividness 1 2 3 4 5

Can you see a girl with red hair eating a green apple?
19. 1-no 2-unsure 3-yes
20. Vividness 1 2 3 h 5

Can you see a tobacco pipe?
21. 1-no 2-unsure 3-yes
22. Vividness 1 2 3 h 5

Can you visualize the number 123 written on a blackboard?
23. 1-no 2-unsure 3-yes
2a. Vividness 1 2 3 u 5

Can you visualize a circle with the letter '8' inside?
25. i-no Z-unsure 3-yes .
26. Vividness 1 2 3 b 5

Can you see a dog dancing?
2?. 1-no Z-unsure 3-yes
28. Vividness 1 2 3 h 5

Can you see a bird reading?
29. 1-no Z-unsure 3-yes
30. Vividness 1 2 3 h 5

Can you see a woman lifting an automobile over her head?
31. 1-no Z-unsure 3-yes
32. Vividness 1 2 3 h 5

II.ATTEIPT THE FOLLOWING ITEMS WITH youn axes CLOSED. TRY
TO app; THE souun paorosso IN EACH ITEM.

Can you hear the voice of a woman talking to someone?
33. 1-no Z-unsure 3-yes
34. Vividness 1 2 3 h 5

Can you hear a woman's voice in the distance yelling
something out loud? .

35. 1-no 2-unsure 3-yes

36. Vividness 1 2 3 h 5

b Code #

1n58

VIVIDNESS SCALE:
1} ASsqutely no image.
2) Indistinct. vague. ambiguous. dim. hazy. doubtful.
) Limited. moderately clear vivid and perceptible.
) Reasonably unobscure vivid and clear.
5) Reallyvthpening: distinct. photographic. perfectly clear
an .

 

Can you hear a masculine voice humming a tune?
37. 1-no 2-unsurs 3-yes
38. Vividness 1 2 3 4 5

Can you hear the sound of a train whistle?
BO. 1-no 2-unsure 3-yes
. Vividness 1 2 3 h 5

Can you hear the sound of a police siren?
#1. 1-no 2-unsure 3-yes
#2. Vividness 1 2 3 b 5

Can you hear the sound of a record being played loudly?
fig. i-no 2-unsure 3-yes
. Vividness 1 2 3 h 5

Can you hear someone lower the volume on the record player?
#5. 1-no Z-unsure 3-yes
#6. Vividness 1 2 3 b 5

Can you hear a trumpet being played?
#7. 1-no 2-unsure 3-yes
#8..Vividness 1 2 3 h 5

Can you hear a bathtub filling with water?
49. 1-no 2-unsure 3-yes
so. Vividness 1 2 3 it 5

,Can you hear a child crying?
51. i-no 2-unsure 3-yes
52. Vividness 1 2 3 u 5

Can you hear someone with heavy leather boots walking across
a wooden floor? '

53. 1-no 2-unsure 3-yes

5h. Vividness 1 2 3 h 5

Can you hear two people whistling while a third person sings?
55. 1-no 2-unsure 3-yes
56. Vividness 1 2 3 b 5

Can you hear water splashing?
57. 1-no Z-unsure 3-yes
58. Vividness 1 2 3 u 5

III..AGAIN WITH YOUR EYES CLOSED. ATTEMPT TO SMELL THE FOLLOWING
ODORS AND FRAGRANCES.

5 Code #

159

VIVIDNESS SCALE:
so uter no image.

2) Indistinct. vague. ambiguous. dim. hazy. doubtful.

) Limited. moderately clear vivid and perceptible.

) Reasonably unobscure vivid and clear.
5) Really happening: distinct. photographic. perfectly clear
- .FBQ_VEV1SE .........................
Can you smell the odor of a gasoline station?
59. 1-no 2-unsure 3-yes
6o. Vividness 1 2 3 b 5

Can you smell a raw onion?
61. 1-no 2-unsure 3-yes
62. Vividness 1 2 3 h 5

Can you smell a rose?
6 . i-no 2-unsure 3-yes
6 . Vividness 1 2 3 A 5

Can you smell an odor that you really like?
65. 1-no 2-unsure 3-yes
66. Vividness 1 2 3 u 5

Can you smell the odor of a freshly mown lawn?
67. 1-no 2-unsure 3-yes
68. Vividness 1 2 3 u 5

Can you smell a hambuger?
69. i-no 2-unsure 3-yes
70. Vividness 1 2 3 h 5

Can you smell the odor of a new pair of shoes?
71. 1-no 2-unsure 3-yes
72. Vividness 1 2 3 4 5

Can you smell the scent of a new bar of soap?
73. 1-no 2-unsure 3-yes
74. Vividness 1 2 3 u 5

Can you smell incense burning?
75. 1-no Z-unsure 3-yes
76. Vividness 1 2 3 4 5

Can you smell the odor of sausage frying?
77. i-no 2-unsurs 3-yes
78. Vividness 1 2 3 u 5

Can you smell the strong odor of amonia?

79. 1-no 2-unsure 3-yes

80. Vividness 1 2 3 h 5

IV. WITH EYES CLOSED. ATTEMPT TO EXPERIENCE THE PROPOSED TASTES.
Can you taste fresh raw lemon Juice?

81. 1-no 2-unsure 3-yes

82. Vividness 1 2 3 u 5

6 Code #

160

VIVIDNESS SCALE:

1) Absolutely no image.

2) Indistinct. vague. ambiguous. dim, hazy. doubtful.

3) Limited. moderately clear vivid and perceptible.

A) Reasonably unobscure vivid and clear.

5) Rezllyvhgppening: distinct. photographic. perfectly clear
an .

 

Can you taste salt?
33. i-no 2-unsure 3-yes
. Vividness 1 2 3 A 5

Can you taste something sweet?
85. i-no 2-unsure 3-yes
86. Vividness 1 2 3 A 5

Can you taste a chocolate bar?
87. 1-no 2-unsure 3-yes
88. Vividness 1 2 3 A 5

Can you taste jelly?
89. 1-no 2-unsure 3-yes
90. Vividness 1 2 3 A 5

Can you taste an apple?
91. l-no 2-unsurs 3-yes
92. Vividness 1 2 3 A 5

Can you taste soup?
9E. 1-no 2-unsure 3-yes
9 . Vividness 1 2 3 A 5

Can you taste fried chicken?
95. i-no 2-unsure 3-yes
96. Vividness 1 2 3 A 5

Can you taste salad dressing?
9?. 1-no 2-unsure 3-yes
98. Vividness 1 2 3 A 5

Can you taste a piece of pizza?
99. 1-no 2-unsure 3-yes
100. Vividness 1 2 3 A 5

Can you taste Coca-Cola?
101. 1-no 2-unsure 3-yes
102. Vividness 1 2 3 A 5

Can you taste a pear?
10 . 1-no 2-unsure 3-yes
1O . Vividness 1 2 3 A 5

Can you taste fried eggs?

105. 1-no 2-unsure 3-yes
106. Vividness 1 2 3 A 5

7 Code #

1161

VIVIDNESS SCALE:
I) ASsquter no image.
2) Indistinct. vague. ambiguous. dim. hazy. doubtful.
a) Limited. moderately clear vivid and perceptible.
) Reasonably unobscure vivid and clear.
5) Reallyvhappening: distinct. photographic. perfectly clear
an

_ _ ...d- _ .dJ. -------------------------

v. NEXT. SEE IF YOU CAN FEEL THE FOLLOWING PROPOSED EXPERIENCES.
ONCE AGAIN. ATTEMPT TO'RAVE THESE EXPERIENCES WITH YOUR EYES CLOSED.

Can you feel a toothbrush rubbing against your gums and teeth?
10?. 1-no 2-unsure 3-yes
108. Vividness 1 2 3 A 5

Can you feel long cool grass against the bottom of your bare feet?
109. 1-no 2-unsure 3-yes
110. Vividness 1 2 3 A 5

Can you now feel a rough scouring-pad rubbing over your fingertips?
111. 1-no 2-unsure 3-yes
112. Vividness 1 2 3 A 5

Can you feel a feather tickling your nose?
11 . 1-no 2-unsure 3-yes .
11 . Vividness 1 2 3 A 5

Can you feel a hand on your shoulder?
115. 1-no Z-unsure 3-yes
116. Vividness 1 2 3 A 5

Can you feel fingers scratching your scalp?
117. 1-no 2-unsure 3-yes
118. Vividness 1 2 3 A 5

Can you feel a warm cup pressed against your lips?
119. 1—no 2-unsure 3-yes
120. Vividness 1 2 3 A 5

Can you feel your hand on a doorknob?
121 . 1 -no 2-unsure 3-yes
122. Vividness 1 2 3 A 5

Can you feel fur-lined gloves on your hands?
122. i-no 2-unsure 3-yes
12 . Vividness 1 2 3 A 5

Can you feel warm soup in your mouth?
125. 1-no 2-unsure 3-yes
126. Vividness 1 2 3 A 5

VI. WITH EYES CLOSED. TRY TO IMAGE (EXPERIENCE) THE FOLLOWING
Eflz§ICAL §§N§ATIONS.

Can you imagine yourself being extremely hungry?

127. 1-no 2-unsure 3-yes
128. Vividness 1 2 3 A 5

8 Code #

162

VIVIDNESS SCALE:

1) Absolutely no image.

2) Indistinct. vague. ambiguous. dim. hazy. doubtful.

3) Limited. moderately clear vivid and perceptible.

A) Reasonably unobscure vivid and clear.

5) Really happening: distinct. photographic. perfectly clear

Can you imagine (feel yourself) becoming sick to your stomach?
129. l-no 2-unsure 3-yes
130. Vividness 1 2 3 A 5

Can you feel your mouth become dry?
131. 1-no 2-unsurs 3—yes
132. Vividness 1 2 3 A 5

Can you feel your mouth now become very moist?
13 . i—no 2-unsure 3-yes
13 . Vividness 1 2 3 A 5

Can you feel a headache?
135. 1-no 2-unsure 3-yes
136. Vividness 1 2 3 A 5

Can you now feel your body surge with energy?
137. i-no 2-unsurs 3-yes
138. Vividness 1 2 3 A 5

Can you feel a tickle in your arm?
1 9. 1-no 2-unsure 3-yes
1 . Vividness 1 2 3 A 5

Can you feel a numbness in your foot?
1A1. 1-no 2-unsure 3-yes
1A2. Vividness 1 2 3 A 5

Can you feel the numbness move up to your hand?
1A3. 1—no 2-unsure 3-yes
1AA. Vividness 1 2 3 A 5

Can you feel an itch on your left cheek?
1A5. 1-no 2-unsure 3-yes
1A6. Vividness 1 2 3 A 5

VII. WITH EYES CLOSED. TRY TO EXPERIENCE THE FOLLOWING
MOVEMENT§. AS IF YOU WERE ACTUKLLY DOING THEM.

Can you feel yourself running down some stairs?
1A7. 1-no 2-unsure 3-yes
1A8. Vividness 1 2 3 A 5

Can you feel yourself jumping up and down?

1A9. i-no 2-unsure 3-yes
150. Vividness 1 2 3 A 5

9 Code #

163

VIVIDNESS SCALE:

1) AEsolutely no image.

2) Indistinct. vague. ambiguous. dim. hazy. doubtful.

3) Limited. moderately clear vivid and perceptible.

A) Reasonably unobscure vivid and clear.

5) Really happening: distinct. photographic. perfectly clear
and vivid;

Can you feel yourself throwing a heavy rock?
151. 1-no 2-unsure 3-yes
152. Vividness 1 2 3 A 5

Can you feel yourself drawing a triangle?
15 . 1-no 2-unsure 3-yes
1 . Vividness 1 2 3 A 5

Can you feel yourself writing your name?
155. 1-no 2-unsure 3-yes
156. Vividness 1 2 3 A 5

Can you feel yourself kicking a football?
15?. 1-no 2-unsure 3—yes
158. Vividness 1 2 3 A 5

Can you feel yourself swinging a baseball bat?
159. i-no 2-unsure 3-yes
160. Vividness 1 2 3 A 5

Can you feel yourself tying a rope knot?
'161. 1-no 2-unsure 3-yes
162. Vividness 1 2 3 A 5

Can you feel yourself swinging on a park swing?
16 . 1-no 2-unsure 3-yes
16 . Vividness 1 2 3 A 5

Can you feel yourself shuffling a deck of playing cards?
165. 1-no 2-unsure 3-yes
166. Vividness 1 2 3 A 5

Can you feel yourself bending down to pick up a dime?
167. i-no 2-unsure 3-yes
168. Vividness 1 2 3 A 5

Can you feel yourself standing up from a seated position?
169. i-no 2-unsure 3-yes
170. Vividness 1 2 3 A 5

Can you feel yourself singing a song?

171. 1-no 2-unsurs 3-yes
172. Vividness 1 2 3 A 5

10 Code #

APPENDIX D

STATE-TRAIT ANXIETY INVENTORY-FORM Y

APPENDIX C

SURVEY OF MENTAL IMAGERY-FORM A

16A

APPENDIX D

SELF -EVALUATION QUESTIONNAIRE

Developed by Charles D. Spielberger

in collaboration II”)

R. L. Gorsuch. R. Lushene. P. R. Vagg, and G. A. Jacobs

STAI Form Y-l

 

Name Date
Age Sex: M F. _.

DIRECTIONS: A number of statements which peOple have used to
describe themselves are given below. Read each Statement and then
blacken in the appropriate circle to the right of the statement to indi- ,
cate how you feel right now . that is. a! this moment. There are no right ’/

or wrong answers. Do not spend too much time on any one statement
but give the answer which seems to describe your present feelings best.

-)

)9.

‘2”.

I am tt-nst' ....................................

_

.—

—

—

.—

~

~—

_

—

_

lt‘l'l t.tlm ..................................................

)(‘t‘l \l'klll't' ...,. ........................

lt-t'l stittilit-tl ..................................
lt't'l .tl l‘.l\&' .................................

lt't'l ll})\t'l ...................................

um pt‘t-st-tttlt iturrung mcr ptNSlhlt‘ mislmtutit-s

lk‘t‘l sailisht-tl ..................................
let-l lrighlcnrtl ...............................
it'L'l ttilllitirlublt' ..............................
(eel st'll-t'tinfidem .............................
let'l Ilcrsuus ..................................
.llll llllcl‘" ....................................
ltt'l Illtlt't islu' ................................

.nn relaxed .................................

Ht)IHlll&'|ll .. . .....

.llll mnrlul ..............................
lecl Lunfuscd ............ . ..................
l't't'l SM'J()\ ............................

leel pleasant .............................

..............

..............

I,” I,
’1
/ 4‘
i,’ 4.! l
’l/ / I//
I
I, / f
/ J /
I / ’

I I I t, s,
l 1 4
d
t I d
1 3 d
3 e
1 J a
7 2 l 4
I 3 4
I l J 4
) 2 3 e
l 2 ) A
I 2 i e
l 2 3 e
l J a
I 3 e
4
J A
i 1 j e
i 2 J I
1 ’5 e

Consulting Psychologists Press Code #
577 College Avenue. Palo Alto. California 94306

Name

165

SELF-EVALUATION QUESTIONNAIRE
STAI Form Y-2

Date

 

 

DIRECTIONS: A number of statements which people have used to

describe themselves are given below. Read each statement and then

blacken in the appropriate circle to the right of the statement to in- 4;,

dicate how you generally feel. There are no right or wrong answers. Do

not spend too much time on any one statement but give the answer
which seems to describe how you generally feel.

2U.
30.
3 l.

32.

39.

40.

"I.

I
I
I

_—_———-—_—_

~

feel pleasant ................................................
feel neryous and restless .....................................
feel satisfied with myself .....................................
\ylsh I could be as happy as others seem to be .................
feel like a failure ...........................................
feel rested .................................................
.Illl "taint. tool. and collected" ...............................
feel that difficulties are piling up so that I cannot overcome them

worry too much over Something that really doesn't matter ......
am happy ..................................................
haye disturbing thoughts ....................................
lack self-confidence .........................................
feel secure .................................................
make decisions easily ........................................
feel inadequate .............................................

am content .................................................

Some unimportant thought runs through my mind and bothers me

I

take disappointments so keenly that I can't put them out of my

mind ........................................................

I am a steady person .........................................

I get in a state oftension or turmoil as I think over my recent concerns

and interests .................................................

( “ping/II I‘NM‘ I‘ll-7 ht ( hurl: \ I) Spa/h: rc: r R:pr~:lmlmn ”I flu: l: \I :H‘ rm: poi/Ion Ih: rm]
In um ,pnur u gill/In“! u'rllh'u I» rmlulun n] (In Pub/Ml"! n Pro/uhlhrf

’/
' ’/
_ r
1'J
I
I ’1
x ’/ x
’ ' "I
I l
4.
i 7
(D 3
CD (1‘
I: 3
T I
A f“
g, 2
T 3
"r i
1‘ x?
CD 3
T 3
A [--
xi! 3
CU 3
C" 2
3 3
3 i
I 3
3 3
1‘ 3
Code #

Hi

1:.)

.u)

1 all

‘9‘:

in)

tail

I“)

u‘l

IUl

0)

.9:

I.)

D)

5)

bl

1.)

,a:

w)

‘ 9)

5)

bl

It;

,5)

APPENDIX E

BECK DEPRESSION INVENTORY

166
APPENDIX E

The Beck Depression Inventory

Please read each item carefully on this multiple-choice questionnaire and circle

the number next to the answer that best reflects how you have been feeling during

the past week. If you cannot decide between two answers. circle the higher

nmmnr.
1. do not feel sad.

fsallnd.

am sad all the time and I can't snap out of it.

am so sad or unhappy that I can't stand it.

am not particularly discouraged about the future.

feel discouraged about the future.

feel I have nothing to look forward to.

feel that the future is hopeless and that things cannot improve.

do not feel like a failure.

feel I have failed more than the average person.
I look back on my life. all I can see is a lot of failures.
feel I as a complete failure as a person.

get as much satisfaction out of things as I used to.

don't enjoy things the way I used to.

don't get real satisfaction out of anything anymore.

am.dissatisfied or bored with everything.

don't feel particularly guilty.

feel guilty a good part of the time.

feel quite guilty most of the time.

feel guilty all of the time.

don't feel I am being punished.

feel I may be punished.

expect to be punished.

feel I am being punished.

don't feel disappointed in myself.

am.disappointed in myself.

am disgusted with myself.

hate myself.

don’t feel I as: any worse than anybody else.

am critical of myself for my weaknesses or mistakes.

blame myself all the time for my faults.

blame myself for everything had that happens.

don't have any thoughts of killing myself.

have thoughts of killing myself. but I would not carry them out.

would like to kill myself.

would kill myself if I had the chance.

don't cry any more than usual.

cry more now than I used to.

cry all the time now.

used to be able to cry. but now I can't cry even though I want to.

:HHHHHHHHHH

10.

0‘
O
UNPOUNHOUNHOUNPOUNHOUNPOUNPOUNHOUNHOUNPO

HHHHHHHHHHHHHHHHHHHHHHHHHHHHH

Code #

167

The Beck Depression Inventory page 2

11. am no more irritated now than I ever an.

get annoyed or irritated more easily than I used to.

feel irritated all the time now.

don't get irritated at all by the things that used to irritate me.
have not lost interest in other people.

am less interested in other people than I used to be.

have lost most of my interest in other people.

have lost all of my interest in other people.

make decisions about as well as I ever could.

put off making decisions more than I used to.

have greater difficulty in making decisions than before.

can't make decisions at all anymore.

don't feel I look any worse than I used to.

am worried that I am looking old or unattractive.

feel that there are permanent changes in my appearance that make
me look unattractive.

I believe that I look ugly.

I can work about as well as before.

It takes an extra effort to get started at doing something.

I have to push myself very hard to do anything.

I can't do any work at all.

I can sleep as well as usual.

I don't sleep as well as I used to.

I wake up l-2 hours earlier than usual and find it hard to get back
to sleep.

I wake up several hours earlier than I used to and cannot get back
to sleep. - .

I don‘t get more tired than usual.

I get tired more easily than I used to.

I get tired from doing almost anything.

I am too tired to do anything.

My appetite is no worse than usual.

My appetite is not as good as it used to be.

My appetite is much worse now.

have no appetite at all anymore.

haven't lost much weight. if any. lately.

have lost more than 5 pounds.

have lost more than 10 pounds.

have lost more than 15 pounds.

am purposely trying to lose weight by eating less. Yes;_;__ he
an no more worried about my health than usual.

mm worried about physical problems such as aches and pains: or upset
stomach; or constipation.

I am very worried about physical problems and it's hard to think of
much else.

I am so worried about my physical problems. that I cannot think about
anything else.

I have not noticed any recent change in my interest in sex.

I am less interested in sex than I used to be.

I am much less interested in sex now.

I have lost inteieet in sex coupletely.

12.

l3.

l4.

NHOWNHOUNHOUNHO
HHHHHHHHHHHHHHH

15.

16.

NPOUNHOU

U

17.

18.

19.

UNHOUNHOUNHO

 

20.

N HO
HHHHHHHH

U

21.

HMO-‘0

Code #

APPENDIX F

STUDY INTRODUCTION

168

APPENDIX F

STUDY INTRODUCTION

Dear clinic user,

You are being invited to spend a few minutes while you wait for your
appointment filling out the attached questionnaire as part of a
research project being conducted in cooperation with the Isla Vista
Medical Clinic and Michigan State University. We are looking into

the role of imagination in health care management. By voluntarily
filling out the questionnaire and providing your name and phone number
you would be giving permission to be contacted in the next few days
for a more complete description of the project and an invitation to
participate further. Filling out this questionnaire gggg pg: obligate
you in any way to participate beyond the follow up phone call. Your
name and phone number will be held in strictest confidence whether

or not you decide to participate fully in the project.

Please feel free to direct any questions you might have to Mr. Bradley
at the phone number given below.

Thank you very much for your time and consideration.

Please return the completed questionnaire to the receptionist.

Terence L. Bradley. M.A.
Project Director
968-h455

 

name’fplease prinff

 

d
phone number Co e #

 

day a time I can most
conveniently be reached

APPENDIX G

CONSENT FORM

169
APPENDIX G

Consent Form

v

This is to acknowledge that l have been fully informed about the
research project being conducted by Terence L. Bradley and that

I understand the proposal and my participation to.my satisfaction.

Nith the understanding and assurance that my name and/or identifiable
test responses will be held in strictest confidence, I freely agree
to participate in the study as outlined to me. However, I also

understand that I may withdraw from the project at any time without

penalty.
I further understand that there are no guarantees of benefits

offered as a result of my participation and that, within the
restrictions of confidentiality, the general results of the study
and my questionnaire results will be made avialable to me upon

request.

 

 

witness Participant

 

 

Date Date

Project Director:

Terence L. Bradley

22 w. Micheltorena

Santa Barbara, CA 93101 Code #
965-1915

APPENDIX H

DEMOGRAPHIC SHEET

170
APPENDIX H

DEMOGRAPHI C SHEET

 

 

 

 

AGE: 15-19 SEX: Male

20-2h Female
_. 25-29

30-3“ RACE: Black
35-39 Hispanic
uo-uu White
b5-q9 Asian
50-5“ Native American
55-59 Other
60 +

EDUCATION: Less than high school
._____High school graduate
Partially completed college
Completed four year college degree
Graduate or professional school

If you would like to receive a brief written summary of the results
of the study. please provide your mailing address below.

 

street and*#

 

city. state. zip

Thank you for your time and cooperation.

Code i

APPENDIX I

MEANS & STANDARD DEVIATIONS ON SMI-A
SUBSCALES x ALEXITHYMIA

171
APPENDIX I

Means & Standard Deviations on SMI-A
Subscales x Alexithymia

 

 

SMI-A Alexithymic Non-Alexithymic
Subscales Mean S.D. Mean S.D.
Visual Cont. 40.067 8.642 45.042 5.159
Visual Viv. 50.600 17.093 62.458 12.942
Auditory Cont. 30.233 7.290 36.092 3.282
Auditory Viv. 35.900 11.751 49.375 10.050
Olfactory Cont. 23.500 7.551 28.100 5.425
Olfactory Viv. 26.700 9.487 35.850 11.219
Gustatory Cont. 31.233 8.232 35.367 5.299
Gustatory Viv. 37.133 13.117 46.683 12.673
Tactile Cont. 23.400 6.112 28.158 3.095
Tactile Viv. 27.067 9.336 38.858 8.866
Somesthetic Cont. 22.000 5.146 26.308 3.759
Somesthetic Viv. 26.433 8.858 34.875 8.091
Kinesthetic Cont. 31.000 8.238 37.167 3.158
Kinesthetic Viv. 37.133 13.006 52.117 9.574

 

APPENDIX J

T-TESTS FOR SMI-A SUBSCALES x ALEXITHYMIA

T-tests for

172
APPENDIX J

SMI-A Subscales x Alexithymia

 

 

SMI-A Subscale Chi-Square t p
Visual Viv. .046 4.193 <.000
Auditory Viv. .273 6.344 <.000
Olfactory Cont. .017 3.818 <.000
Olfactory Viv. .272 4.112 <.000
Gustatory Viv. .814 3.666 <.000
Tactile Viv. .723 6.447 <.000
Somesthetic Cont. .024 5.188 <.000
Somesthetic Viv. .532 5.014 <.000
Kinesthetic Viv. .028 7.101 <.000

 

APPENDIX K

MEANS & STANDARD DEVIATIONS ON DES-IV
SUBSCALES x ALEXITHYMIA

173
APPENDIX K

Means & Standard Deviations on DES-IV

Subscales x Alexithymia

 

 

DES-IV Alexithymic Non-Alexithymic
Subscales Mean S.D. Mean S.D.
Guilt 5.633 1.752 6.525 2.315
Shame 20.433 6.129 22.717 6.724
Inner-Hostil. 12.033 4.476 12.500 4.476
Anger 7.700 2.409 8.075 2.450
Fear 5.833 2.679 5.775 2.842
Sad.-Dist. 9.400 3.558 9.800 3.592
Interest 9.367 2.710 11.867 2.070
Enjoyment 9.267 3.619 12.008 2.213
Surprise 6.367 2.785 8.275 2.394
Disgust 4.900 2.468 5.000 2.257
Contempt 5.667 2.040 6.358 2.289
Shyness 5.233 2.096 5.867 2.219

 

APPENDIX L

T—TESTS FOR DES-IV SUBSCALES x ALEXITHYMIA

T-tests for DES-IV Subscales x Alexithymia

174

APPENDIX L

 

 

DES-IV Subscale Chi-Square t p
Guilt .074 1.971 .051
Shame .537 1.692 .093
Interest .054 5.542 <.000
Surprise .290 3.777 <.000