'—

  

‘Wl

1)

 

W

H

 

(I

l

   

THESts

”33m l\\\L\\\\\\1\\l\\;l3llMllulllglglglll

Michigan State
Unlversity

 

 

 

 

This is to certify that the
thesis entitled
At CS6" Mafia cat/«i HOQQHY 1 v‘
jammed“ A11“ ”—5

presented by

"DQUt‘A (0‘76

 

has been accepted towards fulfillment
of the requirements for

AAL—degree in MW
W

Major professor

Date 4121:}?

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

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1“ WW“

 

ALEXITHYMIA AND HOSTILITY IN JAMAICAN ADULTS
By

David Rose

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

1998

ABSTRACT
ALEXITHYMIA AND HOSTILITY IN JAMAICAN ADULTS
By

David Rose

Alexithymia is the inability to identify or describe one’s feelings and is associated
with a wide variety of psychopathology and behavioral problems. However, while the
theoeretical literature argues that high alexithymia levels should predict high levels of
hostility, little empirical research currently exists in this area. This study used the
Toronto Alexithymia Scale-20, a self-report measure of alexithymia, and the Hostility
Dimension of the BSI to survey 352 Jamaican adults. Simultaneous regressions revealed
that high alexithymia levels predicted high levels of hostility. In addition, alexithymia
was higher among lower SES Jamaicans as opposed to higher SES Jamaicans, and
alexithymia was higher among Jamaican males than among Jamaican females. Further,
only males with lower SES had significantly higher levels of alexithymia. Results
suggest that teaching adult Jamaicans with alexithymia how to correctly identify and
label their hostile feelings may help them to gain control over and reduce their frequency

of hostile and violent behaviors.

Dedicated to the memory of my grandmother, Mary Olive Rose, for always believing in

me and for the delicious chocolate chip cookies.

ACKNOWLEDGMENTS

I’d like to thank the following individuals for their assistance with this project:
Michael C. Lambert for his support and for pushing me to reach my full potential;
Gershen Kaufman for always being there when I needed encouragement; Bertram P.
Karon for his wisdom and sense of humor", Alytia Levendosky for helping me to gain
confidence and believe in myself; Bob Caldwell for his warm empathy and guidance;
Richard P. Deshon for the clarity and conciseness of his advice regarding statistical
methodology; and Mikhail Lyubansky and Cheryl-Lynn Podolski for cheerfully helping
me with my data analysis. In addition, I’d like to thank J eff Hanneman, Tom Araya, and

Kerry King for providing a never dying source of motivation for my work.

iv

TABLE OF CONTENTS

CHAPTER 1

ALEXITHYMIA”
What IS Alexithymia? Definition and Descnptlon
Why 15 Alexithymia Important? Comorbidity and Consequences
What Factors Lead to Alexithymia? The Etiology of Alexithymia"

CHAPTERZ
HOSTILITY..

What IS Hostility? Definition and Descnptlon
CHAPTER3
WHYMIGHT ALEXITHYMIA BE ARISK FACTOR FOR HOSTILITY?
CHAPTER4
ALEXITHYMIAANDHOSTILITYINJAMAICAN CULTURE
CHAPTERS
PREVIOUS STUDIES THAT HAVE EXAMINED THE RELATIONSHIP
BETWEENALEXITHYMIAANDHOSTILITY
CHAPTER6
METHOD”

Research Participants

Measures”

Toronto Alex1thym1a Scale (TAS-ZO)

Brief Symptom Inventory (BSI) Host111ty D1mens1on . ....

Method of Data Analysis.

CHAPTER 7
RESULTS.

Confirmatory Factor Analys1s
Exploratory Factor Analysis“

Data Analyses”

.. vii

\OQMU‘

Primary Alexithymia: Neuropsychological Factors .. 10
Secondary Alexithymia: Proactive Trauma
Secondary Alexithymia: Retroactive Trauma

Secondary Alexithymia: Social and Family Factors

12
12

14
..14

17

22

27
29

30
.....30
....30

..30

31

‘5‘!

.. DJ

35
35
....35

.. 35

First Wave ofRegressions
Second Wave ofRegressions

CHAPTER 8

AppendixAzTablel
AppendiszTableZ
AppendiszTable3 .
AppendixDzTable4

.. 38
..43
.46
... 47
.48

.49

Table

LIST OF TABLES

Item Loadings of the TAS-ZO onto the Confusion About Feelings Factor

Item Loadings of the Hostility Dimension of the BSI from Principle
Components Analysis

Summary of First Wave of Simultaneous Regression Analyses
(standardized beta weight) and Interactions for Independent Variables
Predicting scores of the Confusion About Feelings (CAP) Factor

Summary of Second Wave of Simultaneous Regression Analyses

(standardized beta weight) and Interactions for Independent Variables
Predicting scores of the Hostility Dimension of the BSI

vii

46

47

48

INTRODUCTION

Externalizing and antisocial behaviors such as oppositional defiant disorder
(ODD), conduct disorder (CD), antisocial personality disorder (APD), wife battering,
child abuse, and violent criminal behavior are alarmingly frequent and thus are a major
societal problem. One violent crime occurs every 17 seconds in the United States
(Tsytsarev & Grodnitzky, 1995) and more than 2 million people are the victim of
physical attack (Kassinove & Sukhodolsky, 1995). A wife is battered every 15 seconds
(Geffner & Rosenbaum, 1990) and 30% of couples will experience at least one episode
of marital aggression during their lifetime. The United States has a higher homicide rate
than any other industrialized, Western nation (Tsytsarev & Grodnitzky, 1995). The
prevalence rates of conduct disorder are estimated to be 6-16% for boys and 2-9% for
girls, while prevalence rates of antisocial personality disorder are estimated to be 3% for
males and 1% for females in community samples (American Psychiatric Association,
1994)

Hostility, anger, and aggression are major components of all categories of
externalizing behavior disorders (Feindler, 1995). A number of psychiatric disorders
such as oppositional defiant disorder (ODD), conduct disorder, and antisocial personality
disorder are associated with hostility and anger. The major characteristics of ODD are: a
pattern of hostile and defiant behavior, frequent outbursts of temper, anger and
resentment, and a tendency to argue with adults. In conduct disorder, the violation of the
rights of others accompanies the hostile and angry behaviors of ODD. Both ODD and

CD apply only to children who are age 17 or younger. When the symptoms of CD

continue to occur past the age of 17, a diagnosis of antisocial personality disorder is
given (American Psychiatric Association, 1994). Empirical research has demonstrated
that psychopaths, individuals with antisocial traits, have elevated levels of hostility
(Blackburn & Lee-Evan, 1985). The following excerpt of an individual with antisocial
personality disorder highlights the accompanying high levels of hostility:

I had so much anger. . .I was just like a bomb. . . it’s just ticking, and it’s just a

matter of seconds before this bomb blows up, and you’re trying to get out of

there... and the way I’m going, that bomb was going to blow up in me. I wouldn’t
be able to get away from it... going to be a lot of people hurt... I’m not going out

without taking somebody with me. .. (Barlow & Durand, 1995, p. 533).

In addition, extemalyzing behaviors such as spouse battering and child abuse,
which are not considered to be psychiatric disorders, are also associated with high levels
of hostility and anger. A consistent finding in the research literature is that male batterers
have higher levels of hostility (Else, Wonderlich, Beatty, Christie, & Staton, 1993;
Gavazzi, Julian, & McHenry, 1996; Tolman & Bennett, 1990) and general trait anger
(Deffenbacher, Oelting, Lynch, & Morris, 1996) than non-batterers. Leonard and
Senchak (1996) found that both husband and wife hostility predict marital aggression one
year later. Men who commit acts of domestic violence have also been found to have
anti-social personality traits (Else et al., 1993). Anger control problems have been found
in perpetrators of physical child abuse (Barnett, Miller-Penn, & Perrin, 1997). Widom
(1978) found that female criminal offenders also had elevated levels of hostility. In
summary, individuals with high levels of hostility are at an increased risk for a number of

negative outcomes including psychopathology, violent physical altercations, frequent

verbal arguments, poor communication abilities, strained interpersonal relationships,

destruction of property and belongings, occupational maladjustment (Defi‘enbacher et al.,
1996) and low self-esteem (Tsytarev & Grodnitzky, 1995).

Despite overwhelming evidence which highlights the problematic nature and
clinical relevance of high levels of hostility and anger, relatively little research has been
conducted specifically on hostility and anger compared with to other emotional states.
Kassinove and Sukhodolsky (1995), in a search of the PSYC INFO data base, discovered
that the negative feelings of depression and anxiety have been the focus of much more
research than either hostility or anger. The authors speculate that this is because hostility
and anger are more diffith to behaviorally operationalize and measure than depression
and anxiety. Further, they pointed out that this is an excellent example of how research
in the laboratory falls short of providing useful information for clinicians who are
working in the field.

Because hostility and anger create severe personal, family, and vocational
distress, they often become the primary focus of clinical intervention. “In fact, disruptive
behavior disorders represent the largest referrals for psychotherapy intervention
(F eindler, 1995, p. 233). Unfortunately, treatment outcomes for hostility and anger
related behavior problems are generally much poorer than for intemalyzing disorders
(Robins, 1979). Patterson (1996) reports that the effectiveness of treatment interventions
for conduct disorder rapidly decline after the age of 8. A review of the literature on
conduct disorder concludes that clinical interventions are generally of limited
effectiveness (Offord & Bennett, 1994). In regards to treatment of antisocial personality
disorder (APD), Barlow and Durand (1995) state that because treatment outcomes are so

poor, most of these individuals are incarcerated to prevent them from banning others. In

a 10-year survey of treated and untreated male batterers, Dutton (1995) found that the
treated men had reduced subsequent arrests for assault by a mere 3%, subsequently the
effectiveness of abuse counseling remains debatable (Barnett et al., 1997).

Clearly, more research needs to be conducted on hostility and anger because of
their clinical importance and because of the currently poor treatment effectiveness of
behaviors associated with hostility and anger. This study looks at the relationship
between alexithymia (the inability to identify and describe one’s feelings) and hostility.
If alexithymia is found to predict hostility levels, then treatments designed to help hostile,
angry individuals to correctly label and accurately communicate their feelings may help
to reduce levels of hostility and anger and thereby improve treatment outcomes.

First, I will define and describe alexithymia, discuss why alexithymia is important
to understand, and present some of the literature regarding the etiology of alexithymia.
Second, I will define and describe hostility and discuss the rationale for why alexithymia
may be an important factor in the etiology of hostility. Third, I will discuss the literature
that has directly looked at the relationship between alexithymia and hostility. Fourth, I
will discuss why it is important to study alexithymia and hostility in adult Jamaicans.
Fifih, I will discuss why this study of alexithymia and hostility is important in terms of
contributing to the knowledge base in the literature and specific hypotheses for this study
will be given. Finally, I outline the methods and analyses that I will perform in order to

discover the relationship between alexithymia and hostility.

Chapter 1
ALEXITHYMIA
M15 Alexithymia? Definition and Description

Alexithymia is a widespread, but little known condition which affects the way
pe0ple experience and express emotion. More specifically, alexithymia, which literally
means “no words for feelings” (Sifneos, 1972), is the inability to identify, describe, and
distinguish between subjective feelings. The prevalence of alexithymia in psychiatric
outpatients is estimated to be 39.8% (Taylor, Parker, Bagby, & Acklin, 1992).
Individuals with alexithymia have difficulty in the three following areas: 1) they have
difficulty identifying feelings, 2) they have difficulty describing and expressing feelings,
and 3) they have an externally oriented cognitive style which reflects the inability to
think about or reflect upon their own internal affective experiences (King, Mallinckrodt,
& Coble, 1996). As a result, individuals with alexithymia often can’t tell whether they
feel sad, angry, or anxious, despite being aware of sensations of negative arousal
(Krystal, 1988).

The externally oriented thinking associated with alexithymia is believed to stem
from a cognitive condition known as ‘pensee operatoire’ {thought operating} (Montreuil,
J ouvent, Carton, Bungener, & Widlocher, 1991). Because the alexithymic can’t
understand their inner envirronment, their perception of the external environment
changes and consequently they tend. to focus on concrete objects and the surface
dimensions of objects. As a result these individuals usually present as dull, mundane,

unimaginative, utilitarian and their verbal descriptions resemble a sequential recitation of

concrete facts. Not surprisingly, individuals with high levels of alexithymia demonstrate
impaired creativity and fantasy (Clerici, Albonnetti, Pupa, Penata, & Invenizzi, 1992).

Alexithymics also suffer from aprosody, or lack of affect in speech, and a deficit
in the generation of spontaneous facial expressions. Alexithymics can accurately identify
various emotional stimuli, but these same emotional stimuli do not produce subjective
feelings (Berenbaum & Prince, 1994). According to Buck (1988), emotion has the
following three components: 1) subjective feelings, e.g., “I feel sad”, 2) physiological
arousal, e.g., increases in heart rate or skin conductance, and 3) facial and bodily
elements of expression, e. g, frown, smile, crying, laughter, and voice tone. Thus,
alexithymia can be defined as a deficit of the ability to identify and communicate the first
component of emotion, subjective feelings.

Alexithymia is a rather recently defined construct, and there is still much debate
as to what alexithymia actually measures. On the one hand, some theorists argue that the
key characteristic of alexithymia is that there is a cognitive deficit in the symbolic
representation of emotion (Clerici et al., 1992). This implies that subjective feelings are
“normal” in terms of intensity and degree of differentiation and the problem is primarily
a linguistic deficit in which subjective feelings are not semantically and verbally
encoded.

Conversely, theorists such as Krystal (1988) believe that alexithymia is the result
of a self-protective, homeostatic mechanism against excessive psychological distress and
pain According to this theory, if the level of psychological pain that one consciously
experiences is too great they will die, as in cases of psychogenic death. Therefore, as the

level of psychological pain reaches a certain threshold, a mechanism is triggered that

causes a freezing of behavior and darnpens the amount of pain that is experienced.
However, this mechanism tends to stay in place because the individual is motivated not
to experience these painful feelings. As a result, the individual feels less feelings.
According to this theory then, alexithymia is not caused by a deficit in the ability to
liguistically encode feelings, but rather is caused by feelings that are weak in intensity or
absent altogether and therefore are difficult to differentiate or even detect. The small
amount of empirical evidence that is available supports the hypothesis that alexithymia
results from the diminished intensity of feelings. Quinland (1993) found that individuals
with alexithymia did not suffer from difficulties in linking non-verbal images, such as
emotions, from sensory modalities with their verbal linguistic code. This indicates
alexithymia is not related to deficit in the ability to link words with feelings. Therefore,
it is most likely that alexithymia is related to attenuated affective intensity.
Why is Alexithymia Immrtant? Comorbiditv and Comuences

Further support for Krystal’s hypothesis (1988) that alexithymia develops as a
form of protection against negative affect comes from the research literature which
examines the comorbid psychopathology which is associated with alexithymia.
Alexithymia tends to be found in populations that are exposed to painful and distressing
circumstances. Alexithymia has been found in high levels in brutally traumatized
populations with post-traumatic stress disorder (PTSD) such as concentration camp
survivors (Krystal, 1988), military combat veterans (Hyer, Woods, Summers,
Boudewyns, & Hanison, 1990), and survivors of repeated sexual abuse (Zeitlin,

McNally, & Cassiday, 1993).

Hyer et a1. (1990) studied the relationship between PTSD and alexithymia in a
sample of Vietnam Veterans (N = 227). They found 85% of the PTSD patients to be
alexithymic. In addition, these researchers observed clinically that PTSD patients and
alexithymics share several similar characteristics. PTSD patients cannot read their own
emotions and they tend to have many physical complaints, much like patients with
alexithymia. Other research has shown that alexithymia and PTSD patients have similar
patterns of neurotransmitters (Henry, Haviland, & Cummings, 1992). Both groups were
found to have abnormally high norepinephrine levels, suggesting greater activation and
effort of the left hemisphere compared to normals.

Alexithymia has also been found in high levels in populations that have been
exposed to less acute traumatic situations, which are nonetheless psychologically painful
such as in depression and anxiety disorders (Taylor et al., 1992), dysfunctional families
and abusive relationships (King et al., 1996), and in stages of racial identity believed to
involve poorer adjustment and greater distress (Davis-Dinsmore & Mallinckrodt, 1996).

Because individuals who have high alexithymia levels have difficulty identifying
emotion, it is difficult for them to use feelings as cues for self-regulation and the
maintenance of homeostasis. Accordingly, alexithymics have difficulty regulating
glucose levels, and high levels of alexithymia positively correspond with high rates of
diabetes, bulimia nervosa, and anorexia nervosa (Abramsorr, McClelland, Brown, &
Kellner, 1991; Taylor et al., 1992). The difficulty that alexithymics have in identifying
their feelings also leads to poor functioning in interpersonal relations, especially those
that are intimate (Swiller, 1988). Additionally, alexithymia predicts poor treatment

outcomes, especially when psychotherapeutic interventions emphasizing the insight to

At;

and expression of emotions are used According to Krystal ( 1982), alexithymia may be
the single most important impediment in treatments involving psychoanalysis and
psychodynamic psychotherapy.

As previously mentioned, though alexithymics can not identify and describe their
feelings, they do experience undifferentiated and unpleasant sensations of negative
arousal. It is suggested that individuals with alexithymia may attempt to modulate these
negative sensations through food, the use of psychoactive substances, and compulsive
behaviors. This may help to explain the high comorbidity of alexithymia with eating
disorders, substance abuse, psychopathic deviance, impulsive and acting out behaviors,
and pathological gambling (King et al., 1996). Further, Greenberg, Rice, and Elliott
(1993) argue that psychopathology originates, in part, because of the inability to identify
and express emotion. Thus, alexithymia can be seen as detrimental to the ability to
process information regarding one’s immediate environment.

Rates of psychosomatic disorders are also high among patients with alexithymia
(Krystal, 1988). Headaches, numbness, constipation, pain in joints, dry mouth, and
heartburn all correlate positively with alexithymia. Psychosomatic disorders are
approximately twice as common among alexithymics as compared to the general
population (Kauhanen, Kaplan, Julkunen, Wilson, & Salonen, 1993; Taylor et al., 1992).
Additionally, alexithyrrria correlates positively with impairment of the immune system
(Todarello et al., 1994).

What Factors Lead to Alexithymia? The Etiology of Alexithymia
Alexithymia has two primary sources. Primary alexithymia has a biological basis

which involves disconnection of either or both: 1) interhemispheric areas of the

 

neocortex via the corpus callosum, and 2) cortico-limbic pathways. Secondary
alexithymia is related to psychosocial factors, which are believed to result in a
psychogenic “functional disconnection” of these neuroanatomical pathways (TenHouten,
Hoppe, Bogen, & Walter, 1985). There are three possible sources of secondary
alexithymia noted in the literature. First, as previously mentioned, exposure to traumatic
situations in the past, or proactive trauma, is related to alexithymia. Second, fear of
psychological distress in the future, or retroactive trauma, can give rise to alexithymia.
Third, social factors and family environments impact the incidence of alexithymia.
Primary Alexithvmfleuropsvchological Factors. Neuropsychological research
suggests that alexithymia is related to a deficit in communication between the two
cereme hemispheres. Historically, alexithymia was noted to have occurred in the “split-
brain” patients of the 1960’s. These patients had their corpus callosums severed in an
attempt to lessen the severity of intense epileptic seizures. Since the corpus callosum
connects the right and left hemispheres of the brain at the level of the cortex, it was
hypothesized that alexithyrrria was caused by a lack of communication of information
between the two hemispheres (TenHouten et al., 1985). Buck (1994) states that humans
lack connections between the two hemispheres at the subcortical level and thus, the
corpus callosum of the cortex is the primary route for interhemispheric communication.
Zeitlin, Lane, and O’Leary ( 1989) used the tactile finger location task to determine the
level of callosal transfer in non-neurologically impaired individuals. They found a
positive correlation between high alexithymia scores and deficits in interhemispheric

transference. Dewaraja and Sasaki (1990) found that alexithymics suffer from a deficit

10

in transferring nonlinguistic information from the right to the left hemisphere, but not
from a deficit in the callosal transfer of linguistic information.

In the paralimbic cortex of the right hemisphere is a structure known as the
insula. Ross, Homan, and Buck (1994) propose that the insula may be a node for
emotional comprehension, much like Wemicke’s area is a node for semantic
comprehension in the left hemisphere. Of all the cortical areas known, the insula has the
most intense reciprocal connection with the amygdala a structure associated with the
experience and expression of emotion. Lesions to the insula lead to deficits in the ability
to emotionally respond to painful or threatening stimuli (Ross et al., 1994). It is possible
that in alexithymia the emotional information processes by the insula does not reach the
lefi hemisphere and therefore cannot be communicated This also helps to explain why
alexithymics can identify emotional stimuli (because there is no interhemispheric
language deficit), but do not have subjective feelings in response to these stimuli
(because there is a deficit in the transference of nonlinguistic information from the right
to the left hemisphere).

Secondng Alexithymi_a: Proactive Tm As previously mentioned, there is
much evidence to suggest that previously occurring psychological distress and trauma, or
proactive trauma, is a factor in the etiology of alexithymia The intensity of an emotion
that can be tolerated by anyone at anytime is finite. People who have experienced
emotional distress beyond a certain threshold often experience automatization,
robotization, freezing and momentary paralysis, or even psychogenic death in extreme
cases (Krystal, 1988). This going beyond the threshold is trauma, and it is associated

with the degree to which one feels completely helpless in a certain situation. Therefore,

11

as one approaches this threshold, one is motivated to initiate activity to block or evade
conscious recognition of their emotions.

The blocking of intense affect that is associated with alexithymia may actually
serve an adaptive function in some situations. In environments that are emotionally
painful or where emotional expression is potentially harmful, alexithymia may act as a
buffer from the environment and may allow the individual to cope with an environment
wliich the individual is unable to change (Davis-Dinsmore & Mallinckrodt, 1996).
However, if the blocking of painful affect causes permanent and long-lasting change in
one’s own frmctioning, then a psychopathological state will ensue (Krystal, 1988). Once
the affect barrier is in place, it is extremely difficult to remove because the individual is
highlir motivated to avoid re-experiencing the intense and negative affect that was
associated with the traumatic event.

Secondagrv Alexithymia: Retroagive Trauma. Retroactive trauma is the fear of
overwhelming psychological distress in the future. The individual believes they are
completely helpless to avoid the anticipated negative affect. Alexithymia is found in
patients who are diagnosed with terminal illnesses and medical conditions, such as
cancer. Fukinishi, Saito, and Ozaki (1992) found that alexithymia increased in patients
after the initiation of hemodialysis therapy. The researchers speculate that the increase in
alexithymia may be due to the activation of denial-defense mechanisms against
psychological distress engendered by the hemodialysis treatment.

Secondary Alexithymia; Socia_l_ and Family Factors. Research has also shown that
alexithymia is related to social factors. In a massive study of middle-aged men in

Finland (N = 2,682), alexithymia correlated positively with lower levels of education,

12

less annual income, poor social support, and lower occupational status (Kauhanen et al.,
1993). A possible explanation for these results is that all of these factors are associated
with greater levels of psychological distress and that alexithymia develops as a self-
protective coping mechanism. Thus, alexithymia may develop as a way to fend off the
chronic negative affects associated with lack of status, lack of power, and social isolation
in one’s social environment.

Berenbaum and James (1994) found that certain childhood family environments
correlate highly with alexithymia. The best predictor of alexithymia was having grown
up in a home where little positive and supportive communication existed Alexithymia
rates were found to be high in people that grew up in family environments where farme
members were not permitted or felt unsafe expressing their feelings. Thus, this is another
example where alexithymia is believed to result fiom exposure to a psychologically
distressing environment. These findings can also be attributed to a lack of parental role
models who expressed emotion and to a lack of practice identifying and describing one’s

feelings as a child.

13

Chapter 2
HOSTILITY
m Hostility? Definition gm Description

In the 1950’s, Silvan Tomkins conducted a series of studies which examined
emotion across different cultures. On the basis of this research, Tomkins identified nine
primary affects, which he argues are innate in all human beings. Tomkins further argues
that these affects are the primary motivators of human behavior and that they have
survival value for both the individual and the community. Each affect has a nonlinguistic
message attached to it, instructing the organism to behave very quickly in a rather
specific fashion. According to Tomkins (1963 ), anger is triggered when the individual
perceives that something bad is occurring. Anger is the affect which says, “Do
Something”, as a response to the aversive stimulus. For example, when an individual
experiences hurt feelings, especially when those feelings are attributed to the actions of
another, anger is likely to follow in response. Consequently, anger occurs primarily in an
interpersonal context (Tsytsarev & Grodnitzky, 1995).

Anger can be conceived of as either state anger or trait anger. State anger is a
state or condition where an individual experiences physiological arousal, stereotyped
body movements and facial expressions, and subjective feelings that vary in intensity
from mild irritation or annoyance to intense rage and fury (Speilberger, Reheiser, &
Sydeman, 1995). Trait anger refers to the frequency and intensity with which anger is
experienced over time.

Hostility refers to trait anger and thus is considered to be a personality trait

evidenced by cross-situational patterns of anger, possibly in combination with criminal

l4

aggression (Tsytsarev & Grodnitzky, 1995). Individuals who score high on measures of
hostility and trait anger differ from low hostility individuals in three important ways.
First, persons high in hostility are more likely to experience more frequent and intense
anger in situations perceived as frustrating or annoying. Second, high hostility
individuals perceive a greater number of situations, especially those of an interpersonal
nature, to be anger provoking (Spielberger et al., 1995). Third, individuals high in
hostility are more likely to respond in an uncontrolled“ manner. Spielberger et al. (1993)
state that, “Although hostility usually involves angry feelings, this concept has the
connotation of a complex set of attitudes that motivate aggressive behaviors directed
toward destroying objects or injuring other people.” Therefore, the distinction between
anger and hostility is important and also is supported by the literature.

Closely related to anger and hostility is the concept of aggression. “While anger
and hostility refer to feelings and attitudes, the concept of aggression generally implies
distinctive or punitive m directed towards other persons or objects” (Spielberger,
Jacobs; Russell, & Crane, 1983, p. 16). Aggression is generally divided into two major
types. Hostile aggression refers to aggressive behavior which is motivated by the urge to
attack when one is feeling angry whereas instrumental aggression refers to goal directed
behavior which 'is canied out for an extrinsic purpose when the behavior is not motivated
by hostility (Spielberger et al., 1995). Dodge (1991') also makes a similar distinction
between reactive and proactive aggression. Reactive aggression is characterized by high
levels of hostility, rage, a hot temper, is triggered in response to perceived threat, and is
' similar to hostile aggression. Proactive aggression, like instrumental aggression, differs

from reactive aggression in that it is associated with low arousal, a lack of hostility, and

15

object acquisition. Moyer (1976) uses the analogy of the behavior of a cat to
differentiate reactive from proactive aggression. In reactive aggression, the cat arches its
back, hisses and growls, its hair stands on end, and the cat may attack by scratching and
biting. In proactive aggression the cat makes little emotional display as it slinks close to

the floor while stalking a mouse.

16

Chapter 3
WHY MIGHT ALEXITHYMIA BE A RISK FACTOR FOR HOSTILITY?

Some of the research suggests that genetic factors account for a small percentage
of the variance associated with the development of hostility (Miller, Smith, Turner,
Guijarro, & Hallet, 1996), however the bulk of the research has identified that a history
of physical abuse and exposure to family violence as a child is a major risk factor for the
development of high levels of hostility (Hoglund & Nicholas, 1995; Leonard & Senchalk,
1996; Tolman & Bennett, 1995). Research studies indicate that a history of physical
abuse as a child is a risk factor for marital aggression among male batterers (Else et al.,
1993; Tolman & Bennett, 1990). Leonard and Senchak (1996) found that hostility is
mediating factor in the relationship between history of family violence and marital
aggression.

What is the process by which exposure to physical abuse or family violence as a
child leads to high levels of hostility? Much of literature has focused on cognitive
processes which are altered in hostile and reactive aggressive individuals in such a
manner that hostile intentions are attributed to others in ambiguous social situations
(Blackburn & Lee-Evans, 1985; Dill, Anderson, Anderson, & Deuser, 1997; Dodge,
1991; Feindler, 1995). Dodge has undertaken a program of research which shows that
reactive aggression is associated with a history of trauma, abuse, and deprivation (Dodge,
1991), and that 9M reactive aggressive children show a hostile attributional bias (Crick
& Dodge, 1996). Reactive aggressive boys displayed a bias toward attributing hostile
intentions to peers. They also demonstrated deficits in accurately interpreting others’

intentions, a deficit in linking interpretations to behavior responses, and hypervigilance

17

toward hostile cues. The hostile attributions are exacerbated under conditions of threat in
reactive aggressive individuals, but not in controls (Dodge & Somberg, 1987).
According to Weiss & Dodge (1992), harsh discipline (beyond spanking or slapping)
leads to this maladaptive social information processing style, which in turn leads to
hostile aggression. Indeed, a hostile attributional bias predicts levels of hostile
aggression (Crick & Dodge, 1994). Additionally, Dodge and Somberg’s (1987)
explanation for hostile attributional bias in aggressive boys is that they engage in
preemptive processing, where a specific cue leads to abandoning formal logic and
adopting a less sophisticated, more impulsive response pattern In a relaxed condition,
the hostile aggressive child uses logic to infer intent from facial expressions. Conversely,
under threat, the aggressive child invokes a preemptive rule. This rule says that all
provocations are to be met with a defensive and retaliatory response.

From this, it follows that cognitive interventions aimed at invalidating hostile
beliefs should be effective in reducing levels of aggressive and externalizing behaviors.
However, the link between change of social cognitions and changes in aggressive
behaviors has not been demonstrated (Ofi‘ord & Bennet, 1994). “To date, efforts to
produce such effects have not been successful, or those that seem successful could not be A
replicated (Patterson, 1996, p. 86).” Patterson (1996) suggests that the reason for the
lack of treatment effectiveness of aggression levels with cognitive approaches may be
because the variance accounted for in the hostile attributional bias model is typically less
than 10%. Therefore, it is important to identify other factors which may be involved in

the development of hostility. One factor very well may be alexithymia.

18

High levels of alexithymia have been found in emotionally dysfirnctional families
(Lumley, Mader, Gramzow, & Papineau, 1996). Further, Berenbaum (1996) found that
child abuse is related to alexithymia. An effect size of (d = .58) was found for the
prediction of the inability to identify feelings from a history of child abuse. Berenbaum
(1996) interprets his findings as indicating that alexithymia is associated with the degree
to which individuals felt unsafe physically and emotionally during childhood. These
studies establish that a history of child abuse predicts alexithymia

In addition, much literature indicates that alexithymia is related to hostile
behaviors. Clinical reports indicate high levels of alexithymia are typically associated
with psychopathy and sociopathy (Sifireos, Apfel-Savitz, & Frankel, 1977; Sifi1eos 1991,
1994, & 1996). Sifneos et a1. (1977) state that this may be related to an increased
tendency for alexithymics to act impulsively and to be action-oriented. Krystal (1982)
noted that alexithymics have violent bursts of emotional behavior, but are unaware of the
feelings expressed during such outbursts. Hofeller (1982) states that male batterers tend
to lack verbal communication skills. Yelsma (1996) found that both male and female
perpetrators of marital violence were significantly more alexithymic than nonabusive
individuals.

A number of authors have suggested that developing a greater and more accurate
understanding of one’s feelings is important in the treatment of hostile behaviors. In a
reView of the literature on male batterers, Tolman and Bennett (1990) conclude that
developing a greater awareness of feelings may be important for the treatment of these
individuals. Feindler (1995) notes that aggressive children and adolescents show an

excess of nonverbal, direct action-oriented solutions in response to social problems

19

which is due to the inability to accurately identify and label one’s feelings. “Aggressive
children have a distorted affect-labeling process, and are inclined to label any situation
that elicits physiological arousal as an anger situation, rather than consider it as a
sadness, fear, guilt, or anxiety situation” (Feindler, 1995, p. 185). Consequently,
treatment of hostile aggressive children should consist of helping the child to label
arousal properly. Similarly, Marohn (1992) identifies alexithymia as a factor
contributing to adolescent violence. Accordingly, treatment should include increasing
the understanding of affects so one can understand oneself better and communicate with
others better. Marohn (1992) argues that properly labeling and communicating the
feeling of anger helps to convert violent motor behavior to verbal behavior.
Berkowitz (F inman & Berkowitz, 1989) has deveIOped a theory on the cognitive
processes associated with hostility.
Employing depth of processing notions, Berkowitz suggests that the rudimentary
feelings initially activated by an unpleasant event can be affected or even
transformed by additional cognitive processing involving appraisals, attributions,
or other interpretations. This higher order, controlled processing does not always
operate but has to be activated by an incentive to think more extensively and
“deeply” about the available information (Harkness, DeBono, & Borgida, 1985).
Once activated, however, this processing can differentiate the initial experience,
intensify or suppress the feelings that come into focal awareness, determine what
ideas come to mind, and regulate the overt displays of feelings and behavior.
This controlled processing thus influences the extent to which the negative affect
generated anger, hostility, and instigation to aggression are revealed openly
(F inman & Berkowitz, 1989, p.72).
Therefore, in order to activate these cognitive regulatory mechanisms, one must
pay attention to one’s feelings in order to increase the awareness of these feelings.

Research by Finman and Berkowitz (1989) supports this argument. The greatest display

of overt hostility occurred when participants were distracted so as to minimize their

20

awareness of their feelings. Further research indicates that awareness of hostility and
aggression reduces hostile feelings, whereas not being aware of expressed hostility
maintains hostile feelings (Caprara, Passerini, Pastorelli, Renzi, & Zelli, 1986).
Therefore, it seems likely that alexithymia would be a risk factor for the development of
hostility and the expression of hostility related behaviors and that research looking at the

relationship between these two constructs may very useful and instructive.

21

Chapter 4
ALEXITHYMIA AND HOSTILITY IN JAMAICAN CULTURE

Hostile and aggressive behaviors are currently widespread and constitute severe
social problems in Jamaica. The Jamaica Constabulary Force (1998), the police force in
Jamaica, reports that in January of 1998 alone there were 74 homicides in Jamaica. At
this rate, Jamaica will have nearly 900 homicides in 1998, which is an extremely high
rate, considering the population of Jamaica is only 2.5 million people (Barrow, 1996).
Perhaps even more distressing, the number of homicides associated with domestic
violence have skyrocketed 58% from January 1997 to January 1998 (The Jamaica
Constabulary Force, 1998). However, the high levels of violence and aggression found in
Jamaica have not yet been the focus of empirical research. Clearly, it is important to
discover the etiological factors involved in these behaviors, so that they can be
effectively prevented and treated Two factors that may play major roles in the
development of violence in Jamaica are alexithymia and hostility.

Social and family factors may play a major role in the development of
alexithymia among Jamaican adults. As children, Jamaicans are taught to be quiet and
stay out of the way of adults (Barrow, 1996). Thus, starting at an early age there is a lack
of verbal communication in the typical Jamaican home. This also includes the
expression and sharing of feelings with other family members. In addition, Jamaican
youth are reinforced for passive and submissive behaviors (Lambert, Weisz, & Knight,
1989). Therefore, Jamaicans may learn to inhibit expression of feelings as children and

may continue to suppress their feelings as adults.

22

Also, child discipline is very harsh and severe in Jamaica. Children in Jamaica
are commonly beaten and flogged for such things as crying too much and interrupting or
talking back their parents (Barrow, 1996). What typically happens in Jamaican families
is the mother, who is generally the primary caregiver, loses her temper and hits the child
with a stick, strap, or a belt. The father, who tends to act in a more peripheral role in the
Jamaican family, generally uses severe threats to discipline children. However, when
Jamaican fathers do physically punish their children, the beatings tend to be very severe
(Barrow, 1996). Also, Jamaican teachers tend to use severe corporal punishment within
the classroom (M. C. Lambert, personal communication, March 27, 1998). Jamaican
teachers commonly use straps and belts to discipline students and bnrising often occurs.
In one case, a Jamaican child was beaten with a belt so severely by a teacher that he lost
an eye.

Therefore, the disciplining practices of the Jamaican family and Jamaican schools
may give rise to both alexithymia and hostility. Alexithymia may develop in Jamaican
children because they are discouraged from and feel threatened by expressing their
feelings in their family environments. In addition, a history of severe child disciplining
practices and the potential for child abuse predicts alexithymia (Berenbaum, 1996). A
history of physical abuse and exposure to family violence as a child is also a strong
predictor of hostility. In fact, history of family violence predicts marital aggression and
this relationship is mediated by hostility. This provides an explanation for a potential
link between Jamaican child disciplining practices and the high levels of domestic
violence found in Jamaica. Also, because hostile attribitions increase under threatening

conditions, the fear of harsh physical punishment experienced by many Jamaican

23

children may result in increased levels of hostility (Dodge & Somberg, 1987). Barrow
(1996) points out that many Jamaican boys are abandoned by their families because they
are uncontrollable. Perhaps this is an illustration of children who become more hostile
because they feel threatened living with their families. Therefore, it is very reasonable to
speculate that conditions exist within Jamaican society which are risk factors for the
development of pathological levels of alexithymia and hostility.

There are three main hypotheses as to why child beating is so common in Jamaica
(Barrow, 1996). First, since most Jamaicans were African slaves taken by the British,
harsh child treatment may originate in the harsh treatment Afiican-Jamaicans were
exposed to as slaves. Second, mothers and fathers may displace their frustration and
anger towards their children. Low SES families tend to experience greater levels of
stress and thus may be expected to exhibit more severe child discipline practices. For
example, the greater of an the economic burden the children are on the family, the more
severe their discipline tends to be (Barrow, 1996). Therefore, higher rates of both
alexithymia and hostility may be associated with low SES in J amaica.

Religious background and practices may also contribute to physically harsh
methods of child discipline, resulting in alexithymia. Jamaicans are highly religious
(Barrow, 1996) and most identify themselves as Christians who follow the teachings of
the Bible. Moreover, the Bible warns followers if they spare the rod and then they will
spoil the child. Thus, pithy folk “wisdom” may unintentionally be responsible for
increased severity of child discipline practices and subsequent increases in violence.

In addition to differences in alexithymia and hostility by SES, there also may be

gender differences associated with these variables in Jamaica Research on J arnaican

24

children indicates that boys have greater numbers of externalizing disorders, while girls
have greater numbers of internalizing disorders (Lambert, Knight, Taylor, & Achenbach,
1996; Lambert, Knight, Taylor, & Newell, 1993). This seems to be a consistent finding
across cultures (Lambert et al., 1996). Because externalizing behaviors are harshly
rebuked in Jamaica, it is likely that boys are physically punished more severely than girls.
Also, while Jamaican girls are encouraged to express their feelings, Jamaican boys are
discouraged fiom demonstrating and expressing their feelings (Lambert et al., 1997,
July). Because of these childhood experiences, it is likely that Jamaican males have
higher alexithymia and hostility levels than females.

Currently, most studies on Jamaican psychopathology focus on comparing child
behavior problems in J arnaica with child behavior problems in other countries. Lambert
and his colleagues have undertaken a systematic program of research which indicates that
that while children in Jamaica and the United States have about the same total problem
levels, adolescents (ages 12 - 18) and clinic referred children in Jamaica tend to have
greater numbers of intemalyzing versus extemalyzing behavior disorders as compared
with their counterparts in the United States (Lambert, Lyubansky, & Achenbach, in press;
Lambert et al., 1989). Lambert et a1. (1989) suggest that this pattern occurs because the
British and Afiican traditions which currently influence Jamaican culture both are more
intolerant of extemalyzing and undercontrolled behaviors, whereas in the United States, a
certain degree of nonconfonnity and brashness is expected.

The previously discussed distinction between reactive/hostile aggression and
proactive/instrumental aggression can help us to understand the paradox of high rates of

violent crime in Jamaica, despite a culture that is very disapproving of undercontrolled

25

and extemalyzing behaviors. As previously mentioned, proactive and instrumental
aggression arise from operant conditioning. Proactive and instrumental aggression
develop when an individual has a history of being rewarded for the use of aggressive
behaviors to achieve goals and obtain objects (Dutton, 1995). Therefore, proactive and
instrumental aggression should be culturally sensitive; levels of proactive and
instrumental aggression should vary to the degree that they are suppressed or facilitated
by the culture.

Conversely, reactive and hostile aggression does not develop from operant
conditioning, but rather is influenced more by rage associated with physical abuse,
shame, and abandonment (Dutton, 1995 ), and is therefore less influenced by the cultural
reinforcement of desired behaviors. Instead, hostile aggression is related to the frequency
and severity of child abuse in a culture. It is likely that hostile/reactive aggression is high
in Jamaica, because of the harsh and severe child discipline practices which occur there.
This is consistent with statistics from Jamaica which suggest that aggressive behavior
associated with violent crime is much more of the hostile and reactive type as opposed to
the instrumental and proactive type (The Jamaica Constabulary Force, 1998). Thus, the
attempt by Jamaican parents to reduce extemalyzing behaviors may unintentionally have
the opposite effect and increase extemalyzing behaviors. The methods used to discipline
Jamaican children are so severe that they may inadvertently increase hostile aggression
by way of increasing alexithymia. Because there is good reason to suspect that
alexithymia and hostility play a role in the etiology of the high rates of violent crime and
homicide that currently exist in Jamaica, it is important review the literature to see what

studies have been done in these areas.

26

Chapter 5
PREVIOUS STUDIES THAT HAVE EXAMINED THE RELATIONSHIP BETWEEN
ALEXITHYMIA AND HOSTILITY

Currently, very few studies exist which examine Jamaican adult psychopathology
(Lambert et al., 1997, July). The studies on adult Jamaican psychopathology which do
exist usually look at the experiences of Jamaican immigrants in other countries (Hohn,
1996). No studies currently exist in the literature which examine alexithymia in adult
Jamaicans. A recent study found that confusion about one’s feelings predicted elevated
levels of psychopathology in a sample of adult Jamaicans (Lambert et al., 1997, July).
Much more research needs to be done in all areas of Jamaican adult psychopathology,
and it seems especially important to focus on alexithymia in Jamaica because it’s
development may be facilitated by the culture.

Relatively few studies have examined the relationship specifically between
alexithymia and hostility. Several studies conducted in the 19805 yielded conflicting
results, probably because the methods used to assess hostility and alexithymia in these
studies have poor psychometric properties. For example, Keltikangas-Jarvinen (1982)
used the TAT and the Rorschach to assess both hostility and alexithymia She concluded
that hostility was lower in alexithymics versus controls. Conversely, a Finnish study
found alexithymia and hostility were postively correlated (J ulkunen, Hurri, & Kankainen,
1988). These researchers used the Rorschach to assess both alexithymia and hostility. A
member of these earlier studies applied the Gotschalk-Glaser method content analysis to
particpants responses to measure both alexithymia and hostility. Because the Gottschalk-

Glaser method is not psychometrically sound (von Rad, Drucke, Knauss, & Lolas, 1979),

27

it should not be surprising all three studies using this method came to different
conclusions. One study found a positive correlation between alexithymia and hostility
(Ten Houten et al., 1985), another found a negative correlation (von Rad et al., 1979),
while a third found no differences in hostility between high and low alexithymia groups
(Taylor & Doody, 1995).

A more recent study by F ukunishi (1994) is the only study in the literature which
uses relatively sound psychometrically measures of both alexithymia and hostility. In
this study of Japanese college students (15 = 215), the Toronto Alexithymia Scale (TAS)
was used to assess alexithymia and the MMPI Hostility scale was used to assess hostility.
The results of this study indicate a significant and positive relationship between
alexithymia and hostility (r =.33, p < .01). The author concludes that alexithymia may be
an important risk factor for the development of hostility. A major limitation of this study
was that the TAS-26 was used as opposed to the newer version, the TAS-20 (Bagby,
Parker, & Taylor, 1994). The TAS-26 consisted of several items that have little apparent
relationship to emotional introspection, for example, “daydreaming is a waste of time,”
“I like to let people know where I stand on this,” and “Its not enough for me that
something gets the job done: I need to know why and how it works.” A newer 20 item
version of the TAS has been constructed because of the poor psychometric properties of
the TAS-26. The TAS-20 is currently the most psychometrically sound measure of
alexithymia in existence. Therefore, I propose a study examining the relationship
between alexithymia and hostility using the newer and updated version of the Toronto

Alexithymia Scale, the TAS-20, should be conducted to add to the knowledge base.

28

Hypotheses

This study will have four key purposes. Since no study currently exists where the
TAS-20 is used to measure alexithymia in relation to hostility, this study aims to use the
TAS-20 to study alexithymia and hostility. Second, the study will be conducted with a
J arnaican population because of the alarmingly high frequency of violence that currently
exists in the country and the current dearth of research regarding psychopathology among
adult Jamaicans. I predict that a positive correlation between alexithymia and hostility
will be found Third, because low SES children may be exposed to harsher disciplinary
practices than higher SES children, and because severe corporal punishment is related to
alexithymia, I predict a negative correlation between SES and alexithymia Fourth, I
predict that men are more likely to have high alexithymia scores than women because
Jamaican boys may be treated more harshly that girls and because Jamaican boys are

dissuaded fiom expressing their feelings more than girls.

29

Chapter 6

METHOD
Research Particimnts

We sampled 223 Jamaican men ages 15 to 64 (M = 29.4, S_D = 8.3) and 129

Jamaican women ages 17 to 67 (M = 30.5, Si), = 8.7). A random sample of participants
was recruited from persons visiting a large hospital in Kingston and fiom other
government agencies in Jamaica. Measures of SES such as income which are commonly
used in North America do not work in Jamaica because Jamaicans have several sources
of non-monetary income that they do not report (Lambert et al., 1997, July).
Accordingly, socioeconomic status (SES) was determined by looking at what appliances
and material goods one has in one’s home (e. g, quality of toilet facilities, where water is
obtained, quality of indoor lighting, type of stove, etc.). Because SES has been found to
be negatively correlated with psychopathology (Dohrenwend, B. P. et al., 1992), we
regressed our total SES (M = 16.2, SQ = 3.2) score and several other potential indices of
SES onto the Global Severity Index of the Brief Symptom Inventory (BSI; Derogatis,
1993 ). Results indicated that the formula for calculating SES in our Jamaican sample
was predictive of the global severity of psychopathology as other measures of SES have
done in other populations; higher SES scores predicted lower global psychopathology
scores.
Measures

Toronto Alexithunia Scale (TAS-20L The TAS-20 is a 20-item self-report

 

questionnaire designed to assess alexithymia (Bagby, Parker, et al., 1994). The 20 items

are rated on a 5-point scale, where 1 = strongly disagree and 5 = strongly agree. The

30

TAS-20 consists of the following three subscales: 1) difficulty identifying feelings, i.e., “I
am often confused about what emotion I am feeling”, 2) difficulty describing feelings,
i.e., “It is difficult for me to reveal my innermost feelings, even to close friends”, and 3)
externally oriented thinking, i.e., “I prefer talking to people about their daily activities
rather than their feelings”. The internal consistency reliability for the three factors were
found to be .78, .75, and .66 respectively in a large sample (11 = 965) of Canadian
undergraduates. The test-retest reliability is .77 for the total scale score with a 3-week
interval in between testing was found using a separate and smaller sample (l_\l_ = 72) of
Canadian college undergraduates. The TAS-20 also demonstrates good convergent
validity. In a sample of (N = 85) Canadian undergraduates, the TAS-20, as predicted,
was negatively correlated with measures of psychological mindedness (r = -.68, p < .01),
need for cognition (_r_ = -.55, p < .01), openness for experience (1; = -.49, p < .01), and
positive emotions (r = -.36, p < .01) and positively correlated with a measure of
neuroticism (r = .27, p < .01) (Bagby, Taylor, & Parker, 1994). Using this same sample,
the TAS-20 was found to have adequate discriminant validity as it was not significantly
related to either measures of agreeableness (_r_ = .09) or conscientiousness (r = -.21)
(Bagby, Taylor, et al., 1994).

Brief Symptom Inventory (BSI) Hostility Dimension The BSI is a self-report

 

symptom inventory with 53 items designed to assess psychopathology. Each item is rated
on a five point Likert scale from 0 to 4, where 0 indicates no distress and 4 suggests
extreme distress. The BSI is currently widely used for both research and clinical
assessment with such diverse topics as cancer patients, psychoneuroimmunology, pain

assessment and management, therapeutic intervention, HIV research, hypertension, and

31

student mental health (Derogatis, 1993). The BSI has a total of nine symptom
dimensions of which one is Hostility. The other eight dimensions include: Somatization,
Obsessive-Compulsive, Interpersonal Sensitivity, Depression, Anxiety, Phobic Anxiety,
Paranoid Ideation, and Psychoticism. The internal consistency reliability for the nine
factors were found to be very good, ranging from .71 on the Psychoticism dimension to
.85 on the Depression dimension in a sample (N = 719) of psychiatric outpatients
(Derogatis, 1993). The internal reliability for the Hostility dimension was .78. The BSI
also demonstrates excellent test-retest reliability. The stability coefficient for the Global
Severity Index was found to be .90, and coefficients for the nine BSI dimensions ranged
from .69 for the Somatization dimension to .91 for the Phobia dimension, in a sample of
nonpatients (N = 60) tested across a two-week interval. The Hostility dimension of the
BSI had a test-retest coefficient of .81. The dimensions of the BSI also show impressive
convergent validity with major measures of psychopathology, including the MMPI and
the SCL-90-R Derogatis, 1993). In a sample of symptomatic volunteers (N = 209), the
Hostility dimension of the BSI was positively correlated (r = .42) with the Depression
scale of the MMPI.

The BSI Hostility dimension has five items which include: “Feeling easily
annoyed or irritated”, “Temper outbursts that you could not control”, “Having urges to
beat, injure, or have someone”, “Having urges to break or smash things”, and “Getting
into frequent arguments”. The BSI Hostility subscale was chosen as the measure of
hostility in this project for several reasons. First, it is very quick and easy to administer,

which was necessary for data collection. Second, the BSI has very solid psychometric

32

properties. Finally, the BSI is easily understood by J amiacans, and therefore it could be
administered without revision.
Method of Data Analysis

First, because both the TAS-20 and the BSI were normed using North American
samples, I will first conduct Confirmatory Factor Analyses (CF A) on both measures to
determine of their factor structure is the same in Jamaica as in North America The
measures must meet all of the following goodness of fit criteria in order for the original
factors to be retained First, the Goodness of Fit Index (GFI) must exceed .90. Second
the chi-square must be non-significant, (p > .05). Third, the Root Mean Square Error of
Approximation (RMSEA) Estimate, which is similar to the mean square error in a
regression, must be significant, (p < .05). If the structure of the measures is the same for
the Jamaican sample, then these factors will be used in further analyses. If the North
American factors are not replicated in the Jamaican sample, then Exploratory Factor
Analyses (EF A) will be performed to determine item loadings on the Jamaican factors.
Principle components analyses would be performed on both the TAS—20 and the BSI.
The criteria for retention of factors will be an eigenvalue > or = 1.0. These factors will
then be rotated to a varimax solution.

Second, I will test for main effects of age, gender, and SES by conducting a wave
of simultaneous regressions that controls for the effects of each variable, using each
factor of the TAS-20 as a criterion variable to be considered separately. This will allow
me to test if, as predicted, alexithymia is negatively correlated with SES and if males
score higher on alexithymia than females. Then I will test if there are interactions

between the predictor variables of age, SES, and gender and the alexithymia factors,

33

which are the criterion variables. This will allow me to test the effects of the predictor
variables as well as control them.

Third, a second wave of regressions will be conducted where age, SES, gender,
and the alexithymia factors will be the predictor variables which will be regressed onto
the criterion variable of hostility. This will test our main hypothesis to see if alexithymia
and hostility are positively correlated and it will also allow for detection of any direct
relationships between SES, gender, and age. The predictor variables will be tested for
interactions, and predictor variables which do interact with alexithymia will be entered

into a stepwise regression to control for them.

34

Chapter 7
RESULTS

Data Analyses

Confirmatorv Factor Analysis. Confirmatory Factor Analyses were performed on
both the TAS-20 and the BSI to determine if the factor structure of these instruments is
the same in Jamaica as in North America. In order for the North American factors to be
replicated in the Jamaican sample, all three of the following goodness of fit criteria must
be met: the GFI must exceed .90, the chi-square must be non-significant (p > .05), and the
RMSEA Estimate must be significant (p < .05) (Loehlin, 1992; McDonald, 1985).

The results indicate that neither the factor structure of the TAS-20 nor the B81
was replicated in the J arnaican sample. For the TAS-20, only one of the three criteria for
goodness of fit was met, as the GFI was .91, whereas the chi-square was significant
(p < .0001 ), and the RMSEA was not significant (p_ < .06). None of the criteria for
goodness of fit was met by the BSI: the GFI was .77, the chi-square was significant
(p < .0001), and the RMSEA was not significant (p < .07).

Exploratory Factor Analysis, Because the North American factors/dimensions of
both the TAS-20 and the BSI were not replicated, exploratory factor analyses were
performed to determine item loadings on the Jamaican factors. Principal components
analyses were performed on both the TAS-20 and the BSI to identify their factor
structures for J arnaicans. An eigenvalue > or = 1.0 was used as criteria to determine how
many factors to retain. The factors were then rotated to the varimax orthogonal criterion.

For the TAS-20, only one factor with an eigenvalue of 1.0 or greater emerged.

Twelve items with factor loadings of .30 or above were identified (see Table 1). Most of

35

these items are related to a confusion or lack of understanding of one’s own feelings or
bodily sensations. Therefore, this factor was labeled Confusion About Feelings.

Six factors with an eigenvalue of 1.0 or greater emerged fiom the BSI. An
orthogonal varimax rotation was performed on the 6 factors and the factors were
comprised of items that had the greater loading on each factor. One of these 6
dimensions was a new hostility factor, which included the original 5 hostility items and
three additional items (see Table 2). The three new items on the hostility dimension are:
“feeling others are to blame for most of your troubles”, “having to check and double
check what you do”, and “feeling so restless you couldn’t sit still”.

First Wave of Remssions

Gender, age, and SES were used as predictor variables and Confuslion About
PM was criterion variable in the first wave of simultaneous regressions (see Table
3). As predicted, Jamaican males experienced significantly higher Confusion About
1mg; than Jamaican females (:9 < .02). Also as hypothesized, SES negatively
predicted Confusion Abou_t Feelirgs (p < .0003). Thus, persons of low SES had high
ratings on Confusion About Feelings. However, a significant three-way interaction
between gender, SES, and age was found (9 < .004). The three-way interaction was.
analyzed by breaking it down into three separate two-way interactions. All three of these
interactions were significant, including gender X age (p < .04), SES X age (p < .02), and
SES X gender (9 < .0001). The one-way components of SES and age were not examined
because both are continuous variables. However, the one-way components of both SES
X gender, and age X gender were analyzed No significant effects were found in the age

X gender interaction. In the SES X gender interaction, no significant effects were found

36

for females (9 < .18), however males with lower SES had significantly higher levels of
Confion About Feelings (p < .0009).
Second Wave of Regressions

A second wave of regressions was conducted where gender, SES, age, and
Confrysion Abofleelings were the predictor variables and the new Hostility dimension
was the criterion variable (see Table 4). As predicted, Confusion About Feelings
significantly positively predicted Hostility (p < .0001). Neither gender, SES, nor age
were significant predictors of hostility. However, a significant three-way interaction
between these variables and hostility was found (2 < .03). The three-way interaction was
then broken down into three separate two-way interactions. There was a significant
interaction between gender and age (p < .02) and between SES and age (p < .03 ), while
no interaction was found between gender and SES. The two-way interaction between age
and SES could not be examined because both are continuous variables. Further analyses
of the two-way interaction between gender and age indicated that younger females
experienced higher levels of hostility than older females (1; < .02), whereas age did not

have a significant impact upon hostility in males.

37

Chapter 8
DISCUSSION

The main hypothesis, which argued that high alexithymia levels predict high
levels of hostility in Jamaican adults was supported, as Confusion About Feelings
significantly predicted @3dein (_r; = .30). This finding supports the notion that a lack of
awareness of one’s feelings leads to an increase in hostility and is consistent with
research that indicates overt hostility increases as one pays less attention to their feelings
(F inman & Berkowitz, 1989). Other research indicates that awareness of expressed
hostility reduces hostile feelings, but that lack of awareness of one’s own expressed
hostility maintains hostile feelings (Capara et al., 1986). Since alexithymia is the
inability to identify or describe one’s feelings, these individuals lack the ability to
introspect their hostile feelings. As a result, they are more likely to act out the hostile
feelings than employ cognitive mechanisms to help self-regulate hostile behaviors. The
greater the awareness of one’s feelings, especially anger and hostility in this case, the
more they have the ability to modulate and control them (Marohn, 1992 ).

A second hypothesis predicted that Jamaican males would have higher
alexithymia levels than Jamaican females. This hypothesis was supported, as Jamaican
males reported experiencing significantly higher Confusion About Feelings than
Jamaican females. There are two potential explanations for this finding, both based on
differences in child rearing practices of boys and girls in Jamaica First, Jamaican boys
are discouraged from expressing and demonstrating their feelings while Jamaican girls
are encouraged to express their feelings (Lambert et al., 1997, July). Second, it is likely

Jamaican boys are punished more severely than Jamaican girls because Jamaican boys

38

exhibit greater numbers of extemalyzing behaviors than Jamaican girls (Lambert, Knight,
Taylor, & Achenbach, 1996; Lambert, Knight, Taylor, & Newell, 1993), and
extemalyzing behaviors tend to be punished more severely and harshly than intemalyzing
behaviors. Being the recipient of harsh physical punishment as a child is both a risk
factor for the development of alexithymia (Berenbaum, 1996), and hostility (Hoglund &
Nicholas, 1995; Leoneard & Senchalk, 1996; Tolman & Bennett, 1995). Thus, Jamaican
boys are exposed to more risk factors for the development of alexithymia and hostility
than are Jamaican girls.

The third hypothesis argued that low SES scores should predict high levels of
alexithymia. This hypothesis was also supported, as individuals with lower SES scores
reported higher Confusion About Feelings than individuals of high SES. Although this
study does not test for SES, stress, and punishment effects, it might support the notion
that lower SES Jamaican families employ more physically harsh methods of child
disciplining than high SES J arnaican families because low SES parents experience
greater levels of stress (Barrow, 1996). This stress is then displaced by the low SES
parent onto their children in the form of severe corporal punishment, which in turn leads
to high alexithymia scores.

One should interpret this finding with caution, as the SES effect was moderated
by an SES X gender interaction. This finding indicates that SES predicted high
alexithymia levels only in males, and not in females. As previously noted, Jamaican boys
are much more likely to be the recipients of severe corporal punishment as compared to
Jamaican girls. The low SES Jamaican parent is more likely to experience greater levels

of stress. By virtue of a culture that endorses corporal punishment, especially in boys,

39

lower SES parents are more likely to use physically harsh disciplinary practices than high
SES Jamaican parents when they discipline their sons. The interaction of these two
factors helps to explain why only boys are at an increased risk for alexithymia in low SES
J arnaican families.

While the results from this study are intriguing, they must be interpreted with
caution because of the limitations of this study. First, both of the measures used in the
study relied on the self-report of participants. Research has shown that self-report
measures fail to distinguish between the psychologically healthy and defensive deniers
(Shedler, Mayman, & Manis, 1993 ). This is an especially salient issue because
individuals who experience high in alexithymia are likely to have difficulty introspecting
and self-reporting accurately. Second, while the TAS-20 and the BSI were tested for
equivalent factor structures in the Jamaican sample and were subjected to EFA’s when
equivalence was refuted, the original items were generated to reflect the concerns of
North Americans. The measures were not originally designed for Jamaicans, and thus
they may lack the breadth of emotions, experiences, and problems experienced by
Jamaicans. Third, the sampling was limited to Kingston and the surrounding areas.
Thus, these results may not represent the entire country of Jamaica Fourth, because the
study is a cross-sectional design, age effects cannot be solely attributed to developmental
processes, as cohort effects cannot be ruled out as a confounding variable.

Further research is recommended to overcome the shortcomings of this study. In
order to get a better understanding of the influence of age on alexithymia across time,
future studies should employ a longitudinal design. Sampling in future studies should be

conducted more widely throughout Jamaica, so as to ensure a more representative sample

40

of all of Jamaica It would be helpful to utilize measures of alexithymia and hostility that
do not rely on self-report. Perhaps alexithymia could be measured by using emotion
provoking pictures or slides and asking research participants to describe their feelings in
response to these stimuli. Hostility could be measured by depicting ambiguous social
situations on videotape and asking for participants’ interpretations of the motives of the
actors in a manner similar to Dodge’s studies (Dodge & Somberg, 1987). Also, the items
for these measures should be specifically generated fi'om the Jamaican population in an
exploratory manner and then exploratory factor analyses should be conducted to identify
the underlying Jamaican factor structures of these new Jamaican alexithymia and
hostility measures.

Despite these limitations, this study makes several valuable contributions to our
understanding of the relationship between alexithymia and hostility. It suggests that
addressing alexithymia in Jamaicans may be an important step in reducing their hostility
levels. For example, teaching adult J arnaicans with alexithymia how to correctly identify
and label their hostile feelings may help them to gain control over and reduce their
frequency of hostile and violent behaviors. This step may be especially important in
Jamaica where the rates of hostile and violent behavior are extremely high (J arnaica
Constabulary Force, 1998).

Additional findings fi'om this study indicate that lower SES Jamaicans and
especially lower SES Jamaican males are at greater risk for developing high levels of
alexithymia Further, results indicate that low SES predicted high alexithymia levels
only in Jamaican males. Since other research (Barrow, 1996) suggests that lower SES

parents utilize harsh disciplinary practices and these practices in turn result in higher

41

alexithymia scores, interventions aimed at improving parenting skills may be appropriate.
Also important is stress management intervention for lower SES parents who are prone to
use severe punishment, especially when they discipline boys.

'In summary, this study provides important contributions to our knowledge of
alexithymia, and hostility and violence in Jamaica This was the first study conducted
using the TAS-20 to study the relationship between alexithymia and hostility. It was
found that alexithymia has a moderately strong effect on hostility in Jamaican adults.
Additionally, the study suggests that low SES, Jamaican males are at greatest risk for
developing pathologically high levels of alexithymia Despite its limitations, this study
serves as an important foundation for future research on alexithymia and hostility in

Jamaica.

42

APPENDICES

43

APPENDIX A
Table 1

Item Loading of the TAS-20 onto the Confusion About Feelings Factor

 

 

TAS-20 Item Item Loading
1. I am often confused about what emotions I am feeling. .62“
2. It is difficult for me to say how I feel. .60*
3. I have physical feelings that even doctors don’t understand .5 2*
4. I am able to talk about my feelings easily. .23

5. I prefer to think through problems rather than just talk about them. .02

6. When I am upset, I don’t know if I’m sad, frightened, or angry. .4l"‘
7. I am often confused about sensations in my body. .67*

8. I prefer to just let things happen rather than to understand why they turned

out that way. .32“
9. I have feelings that I don’t understand. .62*
10. Being in touch with your feelings is a must. .12
11. I find it hard to say howl feel about people. .46*
12. People tell me to say more about my feelings. 43*
13. I don’t know what’s going on inside me. .46*
14. I often don’t know why I am angry. .49*

15. I prefer talking to people about what they are doing than about how they feel. .22

Table 1 (cont’d)

16. I prefer to watch “light” entertainment shows rather than serious shows.

17. It is difficult for me to tell my innermost feelings, even to close friends.

18. I can feel close to someone, even when there is no talking.
19. I find looking into my feelings helps me solve my personal problems.

20. Looking for hidden meanings in movies or plays takes away the fun.

.13

.41*

.10

.09

.26

 

Values greater that 0.3 have been flagged by an ‘*’.

45

APPENDIX B
Table 2

Item Loadings of the Hostility Dimension of the BSI from Pfincisle Components

 

 

m

Hostility Dimension Item Item Loading
4. Feeling other people are to blame for most of your troubles. .42

6. Feeling easily annoyed or irritated .56
13. Temper outbursts that you could not control. .52
26. Having to check and double-check what you do. .51
40. Having urges to beat, injure, or harm someone. .58
41. Having urges to break or smash things. .58
46. Getting into frequent arguments. .52
49. Feeling so restless you couldn’t sit still. .49

 

APPENDIX C
Table 3
Summary of First Wave of Simultaneous Regression Analyses (standardized beta_weigl;11)

and Interactions for Independent Vaflbles Predicting scores of the Confssion About

 

 

 

Feel'mgsmAF) Factor
Predictors B E B 12
Gender -2.251 .995 -.117 024*
Age .007 .056 .007 .899
SES -.542 .150 -. 186 0003*"
Interactions
Gender X Age X SES -.004 .973 -. 149 .004“
Gender X Age -.O49 .023 -. 107 0001“"
Gender X SES -.192 .047 -.200 0001*"
Age X SES -.006 .003 -. 120 02*

 

*p<.05 **p<-01 ***p<.001

47

APPENDIX D
Table 4
Summary of Second Wave of SimultagousRegession Analyses (standardized beta

weight) and Interaflons for Independent Vg'ables Predicting scores of the Hostility

 

 

 

Dimension of the BSI
Predictors _B_ _S_E_-B B p
CAF .172 .030 .297 0001*"
Gender .079 .573 .007 .891
Age -.084 .032 -.132 .009"
SES .125 .087 .074 .151
Interactions
Gender X Age X SES -.002 .001 -.110 .034“
Gender X Age -.031 .014 -.118 .023
Gender X SES -.021 .028 -.O37 .460
Age X SES -.003 .002 -.113 030*

 

*p<.05 **p<.01 ***p<.001

48

REFERENCES

49

REFERENCES

Abramson, L., McClelland, D. C., Brown, D., & Kelner, S. J. (1991). Alexithymic
characteristics and metabolic control in diabetic and healthy adults. J ousnal of Nervorss
and Mental Dissa_ses, 179, 490-494.

American Psychiatric Association (1994). Diagpostic and statistical manual of
mental disorders (4tltl ed). Washington, DC: Author.

Bagby, R M., Parker, J. D. A., & Taylor, G. J. (1994). The twenty-item Toronto
Alexithymia Scale-I. Item selection and cross-validation of the factor structure. Journal
of Psychosomatic Research, 38, 23-32.

Bagby, R. M., Taylor, G. J ., & Parker, J. D. A. (1994). The twenty-item Toronto
Alexithymia Scale-II. Convergent, discriminant, and concurrent validity. Journal of
Psychosomtic Resejarch, 38, 33-40.

Barlow, , D. H & Durand, V. M. (1995). Abnormal mchology: An integtive
approach. Pacific Grove, CA: Brooks/Cole.

Barnett, O. W., Miller-Perm, C. L., & Penin, R D. (1997). Family violence across
the lifespan An introduction. Thousand Oaks, CA: Sage Publications.

Barrow, C. (1996). Family in the Caribbean: Themes and parsm'ves. Kingston,
Jamaica: Ian Randle Publishers & Oxford: James Currey Publishers.

Berenbaum, H. (1996). Child abuse, alexithymia and personality disorder. Journal
of Psychosomatic Research. 41. 585-595.

Berenbaum, H, & James, T. (1994). Correlates and retrospectively reported
antecedents of alexithymia. Psychosomatic Medicine, 56. 353-359.

Berenbaum, H., & Prince, J. P. (1994). Alexithymia and the interpretation of
emotion-relevant information. granition a_r_1_d Emotion, 8. 231-244.

Blackburn, R. & Lee-Evan, J. M. (1985). Reactions of primary and secondary

psychopaths to anger-evoking situations. British Journal of ClinicAPsvchology, 24, 93-
100. '

Buck, R. (1988). Human Motivation and Emotion. New York: John Wiley &
Sons

50

Buck, R (1994). The neuropsychology of communication: Spontaneous and
symbolic aspects. Joursal of Pragmatics, 22, 265-278.

Caprara, G. V., Passerini, S., Pastorelli, C., Renzi, P. & Zelli, A. (1986).
Insti gating and measuring interpersonal aggression and hostility: A methodological
contribution. Aggressive Behavior. 12, 237-247.

Clerici, M., Albonetti, S., Pupa, R, Penata, G., & Invenizzi, G. (1992).
Alexithymia and obesity. Psychotherapy and Psychosomatics, 57, 88-93.

Crick, N. R, & Dodge, K. A. (1994). A review and reformulation of social
inforrnation-processing mechanisms in children’s social adjustment. Psychological
Bulletin, 115, 74-101.

Crick, N. R, & Dodge, K A (1996). Social infonnation-processing mechanisms
in reactive and proactive aggression. Child Develtipmenfi67, 993-1002.

Davis-Dinsmore, B., & Mallinckrodt, B. (1996). Emotional self-awareness
(alexithymia) and racial identity attitudes in Afiican-American women. (Unpublished
manuscript).

Deffenbacher, J. L., Oelting, E. R, Lynch, R. S., & Morris, C. D. (1996). The
expression of anger and its consequences. Behaviour Research and Therapy, 34, 575-590.

Derogatis, L. R (1993). BSI: Brief symptom inven_torv, administration scoring
and mocedures manual. Baltimore, MD: Clinical Psychometric Research.

Dewaraja, R., & Sasaki, Y. ( 1990). A left to right hemisphere callosal transfer
deficit of nonlinguistic information in alexithymia. Psychotherapy and Psychosomatics,
21, 201-207.

Dill, K B, Anderson, C. A, Anderson, K. B., & Deuser, W. E. (1997). Effects of
aggressive personality on social expectations and social perceptions. Journal of Research
on Personality. 31, 272-292.

Dodge, K. A. (1991 ). The structure and function of reactive and proactive
aggression. In D. J. Pepler & K. H. Rubin (Eds), The developmentand treatment of
childhood agggsion (pp. 201-218). Hillsdale, NJ: Erlbaum.

Dodge, K. A., & Somberg, D. R. (1987). Hostile attributional biases among
aggressive boys are exacerbated under conditions of threat to the self. Child
Development. 58, 213-224.

51

Dohrenwend, B. P., Levav, I., Shrout, P. E., Schwartz, S., Naveh, G., Link, B. G.,
Skodol, A. E., & Stueve, A. (1992). Socioeconomic status and psychiatric disorders: The
causation-selection issue. Scienca, 255, 946-952.

Dutton, D. G. (1995). The batterer: A fipsychological profile. New York: Basic
Books.

Else, L., Wonderlich, S. A, Beatty, W. W., Christie, D. W. & Staton, R D.
(1993). Personality characteristics of men who physically abuse women. Hospital and
CommunitLPsychiatry, 44, 54—5 8.

Feindler, E. L. (1995). Ideal treatment package for children and adolescents with
anger disorders. In H. Kassinove (Ed), Anger disorders: Definition, d1'_agnosis.,a_n<_l
treatment. Washington, DC: Taylor & Francis, pp. 173-196.

F inman, R., & Berkowitz, L. (1989). Some factors influencing the effect of
depressed mood on anger and overt hostility toward another. Journal of Research in
Persopality. 2; 70-84.

Fukunishi, I. (1994). Social desirability and alexithymia Psychological Repprts,
15, 835-83 8.

Fukunishi, I., Saito, S., & Ozaki, S. (1992). The influence of defense mechanisms
on secondary alexithymia in hemodialysis patients. Psychotheram and Pachosomatics,
5_7_, 50-56.

Gavazzi, S. M., Julian, T. W., & McHenry, P. C. (1996). Utilization of the brief
symptom inventory to discriminate between violent and nonviolent male relationship
partners. Psychological Reports. 79, 1047-1056.

Geffner, R. & Rosenbaum, A. (1990). Characteristics and treatment of batterers.
Behavioral Sciencesand the Lava 8. 131-140.

Greenberg, L. 8., Rice, L. N., & Elliott, R. (1993). Facilitating Emotional Change:
The Moment-Bv-Moment Process. New York: Guilford

Harkrress, A., DeBono, K., & Borgida, E. (1985). Personal involvement and
strategies for making contingency judgments: A stake in the dating game makes a
difference. Journal of Personalig and Social Psychology, 49, 22-32.

Henry, J. P., Haviland, M G., & Cummings, M. A. (1992). Shared

neuroendocrine patterns of posttraumatic stress disorder and alexithymia. Psychosomatic
Medicine. 54. 407-415.

52

Hofeller, K. H. (1982). Social, psychological, and situational factors in wife
abuse. Palo Alto, CA: R & E Research Associates.

Hoglund, C. L. & Nicholas, K. B. (1995). Shame, guilt, and anger in college
students exposed to abusive family environments. J ournsa; of Family Violence, 10. 141-
157.

Hohn, G. E. (1996). The effects of family functioning on the psychological and
social adjustment of Jamaican immi gmt children. Unpublished doctoral dissertation,
Columbia University.

Hyer, L., Woods, M. G., Summers, M. N., Boudewyns, P., & Harrision, W. R.
(1990). Alexithymia among Vietnam veterans with posttraumatic stress disorder.
Juournal of Clinical Psychiat_ry, 51, 243-247.

The J arnaica Constabulary Force. The Jamaica Constabulary Force Home Page-
Crime Review. [Online] Available http://www.infochancom/jcf/crimerev.htrn, February
2 l , 1998.

Julkunen, J ., Hurri, H., & Kankainen, J. (1988). Psychological factors in the
treatment of chronic low back pain: Follow-up study of a back school intervention.
chhotherapy and Psychosomatics. 50, 173-181.

Kassinove, H. & Sukhodolsky, D. G. (1995). Anger disorders: Basic science and
practical issues. In H. Kassinove (Ed), Anger disorders: Definition, diagposis, and
treatment. Washington, DC: Taylor & Francis, pp. 1-26.

Kauhanen, J ., Kaplan, G. A., Julkunen, J ., Wilson, T. W., & Salonen, J. T. (1993).
Social factors in alexithymia. Comprehensive Psychiasy. 34, 330-335.

Keltikangas—Jarvinen, L. (1982). Alexithymia in violent offenders. Journal of
Persopalsity Assessment, 46, 462-467.

King, J. L. Mallinckrodt, B., & Coble, H. M. (1996). Family dysfimction,
alexithymia, (emotional self-awareness), and clients’attachment to therapist.
(unpublished manuscript).

Krystal, H. (1982). Alexithymia and the effectiveness of psychoanalytic
treatment. International Journal of Psychoafllytic Psychotherapy, 9, 353-37 8.

Krystal, H. (1988). Integration and Self-Healing: Affect, Trauma. Alexithm
Ann Arbor: The Analytic Press.

 

53

Lambert, M C., Knight, F. H, Taylor, R, & Achenbach, T. M. (1996).
Comparisons of behavioral and emotional problems among children of Jamaica and the
United States. Journjal of Cross-Cultrrrathsychology, 27, 82-97.

Lambert, M. 0, Knight, F. H., Taylor, R, & Newell, A. L. (1993). Further
comparisons of teacher and parent ratings of behavioral and emotional problems in
Jamaican children. Interaational J ouraal of Intercultural Relations 17 1-18.

 

Lambert, M C., Lyubansky, M, & Achenbach, T. M. (in press). Behavioral and
emotional problems among adolescents of J arnaica and the United States: Parent,
teacher, and self reports for ages 12-18. Journal of Emotional and Behavioral Disorders.

Lambert, M. C., Sarnms-Vaughan, M. E., Hannah, S. D., Lyubansky, M., Rowan,
G. T., Grandison, T., & Holness, A. (1997, July). Identification of emotionsalg
emotional confusion in Jamaican men versus Jamaican women: Do they paedict severigg
gin/Des of psychopathology? Poster session presented at the )O(VI Interamerican
Congress of Psychology, Sao Paulo, Brazil.

 

Lambert, M C., Weisz, J. R, & Knight, F. (1989). Over- and undercontrolled
clinic referral problems of Jamaican and American children and adolescents: The culture
general and the culture specific. Journal of Coasultingand Clinical Psychology. 57. 467-
472.

Leonard K E., & Senchak, M. (1996). Prospective prediction of husband marital
aggression within newlywed couples. Journal of Abnormal Psychology, 10; 369-3 80.

Loehlin, J. (1992). Latent variable models: An introduction to factor, % and
structural analysis. Hillsdale, NJ: Erlbaum.

Lumley, M. A, Mader, C., Gramzow, J. & Papineau, K. (1996). Family factors
related to alexithymia characteristics. Psychosomatic Medicine. 58, 211-216.

Marohn, R C. (1992). Management of the assaultive adolescent Hospital and
Community Psychiam, Q 622-624.

McDonald, R. P. (1985). Factor analysis and related methods. Hillsdale, NJ:
Erlbaum.

Miller, T. Q., Smith, T. W., Turner, C. W., Guijarro, M. L., & Hallet, A J. (1996).
A meta-analytic review of research on hostility and physical health. Psychological
Bulletip, 119, 322-348.

Montreuil, M., Jouvent, R, Carton, S., Bungener, C., & Widlocher, D. (1991).

Parallel information processing test: An experimental assessment of alexithymia.
Psychotherapy and Psychosomap'cs. 56, 212-219.

54

Moyer, K. E. (1976). me psychobiology of gggession. New York: Harper &
Row.

Offord, D. R. & Bennett, K. J. (1994). Conduct disorder. Long-terrn outcomes and
intervention effectiveness. Journal of the American Academy of Child and Adolescent

chhiag, 33, 1069-1078.

Patterson, G. R (1996). Some characteristics of a developmental theory for early-
onset delinquency. In M. F. Lenzenweger & J. J. Haugaard (Eds), Frontiers of
developmental mychopghology (pp. 81-124). Oxford: Oxford University Press.

Quinland, M. A. (1993). Is “Alexithymia” but a dimension of the concept known
as referential activity: A study to define the psychological phenomenon known as
alexithymia. Dissension Abstracts International, 53(9—B ), 4966.

Robins, L. N. (1979). Follow-up studies. In H C. Quay & J. S. Werry (Eds),
chhomthological disorders of childhood (3“1 ed). New York: Wiley.

Ross, E. D., Homan, R. W., & Buck, R. (1994). Differential hemispheric
lateralization of primary and social emotions: Implications for developing a
comprehensive neurology for emotions, repression, and the subconscious.

Neuropgchiapy, Neuropsychology, and Beflviogral Neurology, 7, 1-19.

Shedler, J ., Mayman, M., & Manis, M. (1993). The illusion of mental health.
American Pachologig, 48, 1117-1131.

 

Sifneos, P. E. (1972). Short-Term Psychotherapy and Emotional Crisis.
Cambridge: Harvard Press University.

Sifneos, P. E. (1991). Affect, emotional conflict, and deficit: An overview. .
Psychotherapy and Psychosomatics, 56, 116-122.

Sifneos, P. E. (1994). Affect deficit and alexithymia. New Trends in
Eypgrimental and Clinical Psychiapy, 10, 193-195.

Sifneos, P. E. (1996). Alexithymia: Past and present. American Journal of
Psychartry. 153, 137-142.

Sifneos, P. E., Apfel-Savitz, R. & Frankel, F. H. (1977). The phenomenon of
‘alexithymia’: Observations in neurotic and psychosomatic patients. Psychotherapy and

Psychosomatics, 28, 47-57.

Spielberger, C. D., Jacobs, G., Russell, 8., & Crane, R S. (1983). Assessment of
anger: The state-trait anger scale. In J. N. Butcher & C. D. Spielberger (Eds), Advances
in persoflitv assessment (Vol. 2, pp. 161-190). Hillsdale, NJ: Erlbaum.

55

Speilberger, C. D., Reheiser, E. C., & Sydeman, S. J. (1995). Measuring the
experience, expression, and control of anger. In H. Kassinove (Ed), Anger disorders:
Defimfiomdiagosis, and treatment. Washington, DC: Taylor & Francis, pp. 49-68.

Swiller, H. I. (1988). Alexithymia: Treatment utilizing combined individual and
group psychotherapy. International J ouraal of Group Psychotherapy, 47-61.

Taylor G. J ., & Doody, K. (1995). Verbal measures of alexithymia: What do they
measure. Psychotherapy and Psychosomatics. 43, 32-37.

Taylor, G. J ., Parker, J. D. A, Bagby, R. M., & Acklin, M. W. (1992).
Alexithymia and somatic complaints in psychiatric out-patients. Journal of
Psychosomatic Research, 36. 417-424.

TenHouten, W. D., Hoppe, K. D., Bogen, J. E., & Walter, D. O. (1985).
Alexithymia and the split brain, I. Pachotherapy and Psychosomatics. 43, 202-208.

Todarello, O., Casamassima, A, Marinaccio, M., La Pesa, M W., Caradonna, L.,
Valentino, L., & Marinaccio, L. (1994). Alexithymia, immunity, and cervical
intraepithelial neoplasia: A pilot study. Psychotherapy and Psychosomatics 61 199-204.

 

Tolman, R. M., & Bennett, L. W. (1990). A review of quantitative research on
men who batter. Journal of Interpersonal Violence. 5. 87-118.

Tomkins, S. S. (1963). Affect, imageg, and consciousness: The negative affects
(Vol. 2). New York: Springer.

Tsytsarev, S. V., & Grodnitzky, G. R (1995). Anger and criminality. In H
Kassinove (Ed), Anger disorders: Definition, diagosis, and treatment. Washington, DC:
Taylor & Francis, pp. 91-108.

von Rad, M, Drucke, M., Knauss, W., & Lolas, F. (1979). Alexithymia: Anxiety
and hostility in psychosomatic and psychoneurotic patients. Pachotherapy and
Psychosomatics, 31, 223-234.

Weiss, 8., Dodge, K A., Bates, J. E., & Pettit, G. S. (1992). Some consequences
of early harsh discipline: Child aggression and a maladaptive social information
processing style. Child Development. 63. 1321-1335.

Werner, B. K, Love, S. L., & Shafii, B. (1998). Comparison of seasonal
development of tuber specificity gravity among seven potato clones. American Journal of
Potato Research, 75, 121-128.

Widom, C. S. (1978). An empirical classification of female offenders. Criminal
Justice and Behavior, 5. 35-52.

56

Yelsma, P. (1996). Affective orientations of perpetrators, victims, and functional
spouses. J ouAal of htememonal Violence. 11. 141-161.

Zeitlin, S. 3., Lane, R D., & O’Leary, D. S. (1989). Interhemispheric transfer
deficit and alexithymia. American Journal of Psychgtrv. 14d 1434-1439.

Zeitlin, S. B., McNally, R J ., & Cassiday, K L. (1993). Alexithymia in victims of

sexual assault: An effect of repeated traurnatization? American Journal of PsychiaEL,
_15_O, 661-663.

57

"‘illllllillrlill’3