University of Southern California Dissertations and Theses

The link between the expression of negative emotion and future well-being: the role of disclosure modality in an East Asian American population

(USC Thesis Other)

The link between the expression of negative emotion and future well-being: the role of disclosure modality in an East Asian American population

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

THE LINK BETWEEN THE EXPRESSION OF
NEGATIVE EMOTION AND FUTURE WELL-BEING:
THE ROLE OF DISCLOSURE MODALITY IN AN
EAST ASIAN AMERICAN POPULATION

by

Lauren Christina Ng

A Thesis Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF ARTS
(PSYCHOLOGY)

December 2008

Copyright 2008 Lauren Christina Ng
ii

Table of Contents

List of Tables iii

List of Figures iv

Abstract v

Preface: Specific Aims vi

Chapter 1: Introduction 1
Experimental Disclosure 3
Cultural Differences in Emotional Expression 8
Implications of Cultural Differences on Experimental Disclosure
Narrative 12

Chapter 2: Method 14
Participants 14
Design 15
Outcome Measures 17

Chapter 3: Results 21
Analysis Plan 24
Testing the T3 Mediation Model 27
Supplementary Analyses: Testing the T2 Mediation Model 30
Supplementary Analyses: Does Cultural Orientation Moderate the
Relationships in the Model? 39

Discussion 41

References 47

Appendix: Changes in Outcome Measures Over Time 56

iii

List of Tables

Table 1: Demographic Variables of Study Population (N = 42) 21

Table 2: Intercorrelations Between Constructs 33

Table 3: Negative emotion acting as a suppressor of the relationship between
disclosure modality and T2 well-being 35

Table 4: Disclosure modality acting as a suppressor of the relationship
between use of negative emotion and well-being 39

Table A1: Pairwise comparisons of the change in well-being over time 56
iv

List of Figures

Figure 1: Distribution of T1 Distress 22

Figure 2: Distribution of T1 Self-Reported Health Symptoms 23

Figure 3. Model to be Tested 25

Figure 4. Relationships among constructs in the model at T3 28

Figure 5. Relationships among variables at T2 after controlling for T1
well-being 31

Figure 6. Relationships between variables at T2 after including suppressor
variables in the model 38

Figure A1. Change in distress over time by disclosure modality 57

Figure A2: Change in self-reported health symptoms over time by disclosure
modality 58
v

Abstract
One explanation for the effectiveness of experimental disclosure is
Exposure Theory (Frattaroli, 2006) which hypothesizes that future well-being is the
result of the expression of negative emotion. The goal of this study was to assess
whether writing or talking facilitated emotional expression and to determine
whether increased emotional expression then predicted improved well-being.
Given cultural differences between European Americans and East Asian Americans
regarding the utility of, and comfort with, verbalizing negative emotional
experiences (Kim, 2002), we hypothesized that writing would evoke more negative
emotional expression and would therefore lead to greater improvements in well-
being. The results indicated that writing did elicit more negative emotional
expression than talking. However, use of negative emotion was not related to
improvements in well-being.
vi

Preface: Specific Aims
Aim 1: To determine whether disclosure modality impacts emotional expression
leading to differences in future well-being.
Hypotheses:
a. East Asian American participants would be more likely to express negative
emotion in a written rather than a verbal format;
b. Writing would result in improved well-being six-months later compared to
talking;
c. Expression of negative emotion would be associated with improved well-
being six-months later; and
d. Expression of negative emotion would mediate the relationship between
disclosure modality and future well-being.
Aim 2. To investigate whether Asian or European American orientation impacts the
strength of these relationships.
Question of interest:
a. Does acculturation moderate the relationships between disclosure modality
predicting well-being, disclosure modality predicting emotional expression,
or emotional expression predicting well-being?
1

Chapter 1: Introduction
Research shows that disclosing traumatic events can improve a participant’s
future well-being as measured by behavioral, psychological, social, and immune
functioning indicators (Pennebaker & Chung, 2007). Since 1986, James
Pennebaker and colleagues have investigated the effects of writing about traumatic
experiences in informal, self-selected, unstructured formats, through a simple
experimental disclosure paradigm (Pennebaker & Beall, 1986). While results have
been mixed, generally researchers find that participants who disclose traumatic and
stressful emotional experiences, by either writing or talking, exhibit long-term
improvements in mood and physical health, compared to individuals who disclose
neutral events (Francis & Pennebaker, 1996; Frattaroli, 2006; Pennebaker, et al.,
1988; Pennebaker, et al., 1990; Pennebaker & Chung, 2007; Richards, et al., 2000).
Although the overall measurable, therapeutic effect of disclosure in experimental
settings is small to moderate (r=.075, d.=.15; Frattaroli, 2006), utilizing or
encouraging the use of various disclosure techniques would be reasonable given
that this procedure is easily implemented (individuals can perform it without
assistance), fast (the time is about an hour total), and generally well received by
participants (Frattaroli, 2006).
While the therapeutic benefits of experimental disclosure have been
extensively researched, the mechanisms through which disclosure leads to
therapeutic benefits are not understood well. Additionally, research on cultural
2

differences in behavior and communication suggest that culture and ethnicity may
play an important role in determining participants’ degree of comfort with, and the
overall effectiveness of, experimental disclosure. Although Exposure Theory is
expected to generalize across groups, cultural differences in the appropriateness
and utility of emotional expression may impact its effectiveness. Given cultural
and ethnic variation in the value and role of verbal communication and emotional
expression (Chen & Davenport, 2005; Hsu, 1985; Kim, 2002), it is possible that
individuals who are more or less oriented to European American or East Asian
culture will respond differentially to the experimental disclosure paradigm, and
these disparities may provide insight into the mechanisms behind the efficacy of
the intervention.
Research suggests that compared to European Americans, East Asian
Americans appear to be less likely to disclose negative emotional material, and that
when disclosed, the chosen mechanism does not tend to be verbal (Kim, 2002).
While research does not indicate significant differences in the efficacy of
experimental disclosure modality (writing versus talking) among European
American samples, there may be differences for East Asian Americans. By
analyzing the content of experimental disclosure narratives and comparing them to
self-reported measures of well-being, the study sought to: clarify the proposed link
between expression of negative emotion and improved well-being, determine
whether disclosure modality impacts emotional expression to predict future well-
3

being, and investigate whether Asian or European American orientation effects the
strength of this relationship.
We hypothesized that 1) East Asian American participants would be more
likely to express negative emotion in an written rather than a verbal format; 2)
Writing would result in improved well-being at six-months compared to talking; 3)
Expression of negative emotion would be associated with well-being, and 4)
Expression of negative emotion would mediate the relationship between disclosure
modality and future well-being. Additionally, as a question of interest, we explored
the potential impact of Asian and European American orientation on the
relationships in the proposed model.
Experimental Disclosure
The experimental disclosure paradigm involves participants (usually healthy
college students) being randomly assigned to write about their most stressful and
traumatic experiences (experimental condition) or neutral events (control
condition). While verbal disclosure has been investigated in a handful of studies
(Esterling, et al., 1994; Murray & Segal, 1994), almost all research in this area has
focused on written disclosure (Frattaroli, 2006). Participants write (usually in the
lab) for three to five sessions of 15-20 minutes each. Investigators measure
participant well-being after the initial sessions and then a few weeks to several
months later (Frattaroli, 2006).

4

Mechanisms and Proposed Theories
Several theories explaining the effectiveness of experimental disclosure
have been proposed including Inhibition Theory, A-to-D (Analog-to-Digital)
Emotion Theory (i.e., Cognitive-Processing Theory), Social Integration Theory,
and Exposure Theory (Frattaroli, 2006; Pennebaker & Chung, 2007). Inhibition
Theory suggests that the cathartic release of thoughts and feelings related to an
upsetting experience leads to health benefits. A-to-D Emotion Theory argues that
the act of organizing, gaining insight into and making sense of a traumatic
experience, and ultimately integrating the experience into one’s self-schema are the
mechanisms through which expressive writing is beneficial. Self-Regulation
Theory posits that experimental disclosure gives individuals the opportunity to
create mastery over an experience by observing themselves controlling and
expressing their emotions. Individuals feel that they have more power and control
over traumatic and stressful events, and this increase in self-efficacy leads to
positive health changes. Social-Integration Theory argues that experimental
disclosure impacts the way participants interact with their social world, and these
social interactions improve their well-being and health. Finally, Exposure Theory
suggests that when an individual repeatedly confronts and relives the feelings and
thoughts associated with a negative experience, the exposure eventually
extinguishes the negative thoughts and feelings. Exposure Theory is the only
theory that assumes a linear relationship between the expression of negative
5

emotion and improved future well-being: the more one is exposed to negative
information, the more one habituates, the greater the improvement in well-being.
While support for all of these theories has been mixed (Frattaroli, 2006), this study
will focus on the validity of Exposure Theory by examining whether increased
expression of negative emotion leads to reduced psychological and physical
symptoms, and whether cultural differences moderate this relationship.
Exposure Theory
Exposure Theory proposes that as participants write or talk about adverse
emotional events they habituate to the negative emotions, and repeated exposure to
negative emotion extinguishes the link between traumatic events and the
participants’ reactions to them (Frattaroli, 2006). Exposure Theory is worth
investigating because it parallels empirically supported exposure interventions used
to treat PTSD symptoms (Bootzin, 1997). Exposure theory is also supported by
research that finds that experimental disclosure reduces avoidant and intrusive
thoughts about an event (Klein & Boals, 2001). However, other studies have failed
to find such benefits (de Moor, et al., 2002; Frattaroli, 2006; Lepore, 1997).
The hypothesis that emotional labeling and expression are the keys to
expressive disclosure outcomes is supported by a study that compared the outcomes
of disclosing traumatic events by hand versus by typing. This study assigned
college students to either type or write by hand about stressful events or neutral
events. Investigators found that writing by hand led to increased negative affect
6

and self-rated openness, compared to typing. Crucially, participants who wrote by
hand had better health outcomes than people who typed (Brewin & Lennard, 1999).
Brewin and Lennard suggest that the improved health outcomes were attributable to
the increased emotional expression in the writing by hand condition (Brewin &
Lennard, 1999). They posit that the increased cognitive strain imposed by typing
reduces individuals’ ability for self-focus and emotional involvement, leading to
lower levels of emotional expression. Moreover, the authors suggest that the
increased emotional expression from writing by hand could also be due to the fact
that individuals are simply more experienced at writing about emotional material
by hand than by typing (Brewin & Lennard, 1999). Given that Exposure Theory
hinges on exposure to, or disclosure of, negative emotions, disclosure modalities
that facilitate emotional expression (in this case writing in long hand versus typing)
would be expected to lead to greater health benefits.
Writing Versus Talking
Only a handful of studies have directly compared the therapeutic efficacy of
experimental disclosure through writing or talking, but those that have investigated
differences in disclosure method have generally found that writing and talking (into
a tape recorder, without feedback) are equally effective at reducing mental and
physical health symptoms (Murray and Segal, 1994; Pennebaker, Kiecolt-Glaser, &
Glaser, 1988), and improving life satisfaction and social functioning (Lyubomirsky,
Sousa, & Dickerhoof, 2006) at follow-up. Additionally, studies that have
7

compared expressive writing to speaking with a psychotherapist have found
comparable benefits (Donnelly & Murray, 1991; Murray, Lamnin, & Carver,
1989). Moreover, equivalent efficacy was found when participants were asked to
write alone or talk to a listener about their life goals. However, talking was found to
be a more difficult task than writing, although talking also resulted in higher
positive mood and lower negative mood than writing (Harrist, et al., 2007). While
writing and talking effects are comparable across most studies, one study did report
more overt emotional expression from talking (Murray & Segal, 1994).
When differences in the efficacy of writing and talking have been found,
vocal expression seems to be more successful at reducing physical health
symptoms than written expression (Esterling, et al., 1994). One study compared
the impact of talking and writing about a stressful event on Epstein-Barr Virus
antibody titers (markers of immune system control and response), and found
talking resulted in significantly better cellular immune control than writing.
Moreover, participants who talked were rated as: having achieved greater cognitive
change regarding their stressful experiences, and showing more improved self-
esteem and more adaptive coping strategies, than participants who wrote about
stressful experiences (Esterling, et al., 1994). However, increased use of negative
emotion did not appear to be correlated with these benefits from talking. In fact,
writing about stressful events resulted in significantly more expression of negative
emotion than talking about stressful events (Esterling, et al., 1994). While results
8

are mixed, generally writing and talking have been equally effective at reducing
symptoms and improving well-being at follow-up.
Cultural Differences in Emotional Expression
According to Exposure Theory, the repeated expression of negative
emotional material ultimately leads to a reduction in negative thoughts and
feelings. While this model is expected to generalize across ethnic groups, cultural
differences in the appropriateness and utility of emotional expression may affect
certain parameters in the model. Research indicates that compared to European
Americans, East Asian Americans are more likely to value emotional moderation,
consider emotional disclosure inappropriate, and hide negative or “disruptive”
thoughts and feelings (Chen & Davenport, 2005; Hsu, 1985; Kim, 2002). One
theory suggests that in East Asian cultures, expressing negative emotions increases
social discord, whereas self-restraint helps maintain social harmony (Kao, et al.,
1997; Kim, 2001). A second theory is that in collectivistic cultures, emotional
expression does not carry as much social significance as it does in individualistic
cultures; hence, emotional displays are not necessarily discouraged or suppressed,
but rather they are ignored (Potter, 1988).
Cultural differences in the value of emotional expression coincide with
disparities in the actual amount and type of emotion expressed. Studies comparing
communication patterns of East Asian Americans and European Americans have
found that East Asian Americans are less verbally and emotionally expressive than
9

European Americans (Kao, et al., 1997), and believe that it is more appropriate to
express negative emotions in an indirect or understated manner (Toyokawa &
Nedate, 1996). Furthermore, East Asian Americans are more likely to somaticize
their emotions than European Americans (Kleinman & Sung, 1979; Marsella, et al.,
1973; Soto, et al., 2005; Sue & Sue, 1974). Research suggests that individuals of
Chinese ancestry often use physical complaints to mask emotional pain, which is
considered shameful (Sue & Sue, 1999). Furthermore, comparative studies have
found that among Chinese Americans, expressing emotional stress and
psychological pain through physical complaints is both common and culturally
accepted (Sue, 1997; Sue & Sue, 1999). In situations where emotional expression
is encouraged, such as in therapy, East Asian Americans may be more reluctant to
express strong emotions (Leong &Lau, 2001).
Importantly, despite cultural differences in emotional expression, there
appear to be no cultural differences in the physiological arousal that often
accompanies the experience of emotion (Soto, et al., 2005). In one study
examining startle response in Chinese American and Mexican American
participants (selected because Chinese culture has been found to value emotional
control and moderation and Mexican culture has been found to value free and open
emotional expression), researchers found that across startles, Chinese American
participants reported experiencing significantly less positive and negative affect
than Mexican American participants. However, the two groups showed no
10

differences in physiological measures of experienced emotion (Soto, et al., 2005).
Thus, while physiological emotional response was uncorrelated with cultural
background, there appeared to be a relationship between culture, self-reported
emotion, and the outward expression of emotion. Moreover, for negative emotion,
the more closely oriented individuals were to their culture of origin, the more likely
they were to have emotional behavior consistent with the literature (more emotion
for Mexican Americans and less emotion for Chinese Americans) (Soto, et al.,
2005).
Other studies have demonstrated that although East Asian Americans and
European Americans show no differences in autonomic arousal in response to
stimuli, large differences are still found in self-reported emotions, with East Asian
Americans reportedly experiencing less emotion in response to the same stimuli
(Tsai & Levensen, 1997; Tsai, Levensen & Cartensen, 2000). Thus, culture seems
to exert a powerful influence, even in response to a simple startle stimulus;
however, this difference does not extend to physiological response. Compared
European Americans, East Asian Americans appear to be less likely to disclose
emotional material, but experience the same physiological responses. Given the
reduced likelihood that East Asian Americans will disclose emotional material, it is
important to find a disclosure modality that may facilitate emotional expression for
this population.
11

Western and East Asian cultures have different social norms and
expectations about the ideal amount and quality of verbal expression (Soto, et al.,
2005; Kim, 2002). In Western cultures, verbal expression is considered a positive
act that expresses individuality and “…can create, change and signify thinking…”
(Kim, 2002, p. 828). In contrast, East Asian cultures emphasize silence and
introspection as the path to higher levels of thought. Furthermore, individuals from
East Asian backgrounds tend to emphasize and utilize indirect and nonverbal
communication more than individuals from Western cultures (Kim, 2002).
Researchers have found differences in communication patterns in infants as
young as seven months, with Chinese infants vocalizing less in response to
laboratory stimuli than their European American counterparts (Kim, 2002).
Research has demonstrated that Japanese children make fewer utterances (Minami,
1994) and use verbal language to communicate emotion less frequently (Caudill &
Schooler, 1973) than American children. Importantly, East Asian participants who
were born and raised in the United States have demonstrated similar differences in
beliefs about the importance and utility of talking (Kim, 2002), although cultural
differences in attitudes towards, and presentation of, verbal and nonverbal
expression may be related to an individual’s acculturation status.
One example of the differences in the role and utility of verbal expression
across cultures can be seen in a study by Kim (2002) in which East Asian
Americans and European Americans were asked to think aloud while solving
12

reasoning problems. Compared to problem-solving in silence, European
Americans’ performance was enhanced when they were asked to problem-solve
aloud, whereas East Asian Americans’ performance was impaired. The
investigator posited that European Americans were using internal speech when
solving the problems in silence, and therefore articulating problem-solving steps
further enhanced their skills. In contrast, East Asian Americans were not utilizing
internal speech to solve the problems to the same degree, and therefore verbal
expression interfered with task performance.
In a subsequent experiment, Kim (2002) had participants work silently or
engage in an articulatory-suppression task (saying the alphabet out loud) designed
to impede internal speech by distracting verbal problem-solving. As expected,
European Americans’ performance was significantly impaired by the articulatory-
suppression task, whereas East Asian Americans’ performance was unchanged.
The finding that use of internal speech differs across cultures was somewhat
elucidated by the finding that European Americans reported 1) more positive
beliefs about verbal expression, 2) relying more on language in thinking, and 3)
engaging in more verbal interactions than East Asian American participants (Kim,
2002).
Implications of Cultural Differences on Experimental Disclosure Narrative
Given cultural differences between European Americans and East Asian
Americans concerning use of and comfort with verbal expression, they may be
13

expected to react differently to experimental disclosure. In particular, while
disclosure modality has not been differentially effective in European American
samples, it may be more relevant in an Asian American sample. Specifically, given
that verbal expression tends to interfere with problem solving for East Asian
American students, verbal disclosure of emotional material (already less likely in
this population) may hinder emotional expression, thereby diluting, or perhaps
eliminating, the effective components of experimental disclosure according to
Exposure Theory. For East Asian Americans, writing about traumatic experiences,
a task that does not require verbal expression, may be a more efficacious way to
elicit emotional expression, thereby leading to greater reductions in mental and
physical health symptoms.
This study sought to explore the effectiveness of experimental disclosure in
East Asian Americans. The study is particularly concerned with the differential use
of negative emotion by disclosure modality in this population, and how that
impacts the reduction of physical and psychological symptoms at follow-up. By
analyzing the content of experimental disclosure narratives and comparing them to
self-reported measures of well being, the study seeks to (1) provide more
information about the assumption that exposure to negative emotion leads to
increased future well-being, (2) determine whether this assumption generalizes to a
new participant population, (3) and analyze whether disclosure modality impacts
this assumption.
14

Chapter 2: Method
Participants
Participants were 42 undergraduate students recruited as part of a larger
study examining the overall effects of experimental disclosure for East Asian
Americans. Fifty percent of the participants were in the writing condition and 50%
were in the talking condition. Multivariate analysis of variance (MANOVA) tests
for pre-treatment demographic variables of age, gender, and East Asian ethnicity
were run to ensure successful randomization. No significant group differences
between the writing and talking conditions were found.
Participants were recruited through the undergraduate psychology subject
pool, announcements made during undergraduate classes, e-mail announcements,
flyers, and word of mouth. Eligible subjects from the undergraduate subject pool
were emailed information about the study and recruited for participation. Potential
recruits who learned about the study in other ways were required to contact study
administrators who explained the study in greater detail and conducted a phone or
email screen to determine eligibility.
Students were screened using a one-page demographics questionnaire that
asks about age, gender, ethnicity, country of birth, previous mental health
counseling, use of psychotropic medication, and previous traumatic life
experiences. Students who were age 18 or older and of East Asian background
were scheduled for an initial Time 1 (T1) assessment. Immediately prior to
15

conducting the initial assessment, informed consent was obtained. Graduate
student experimenters, project coordinators, and undergraduate researchers
administered the assessments and oversaw the experimental disclosure sessions.
Participants recruited through the undergraduate subject pool received four
Experimetrix extra credit points for the T1 assessment, disclosure sessions, and the
one-week follow-up (T2) assessment, and those recruited through other means
received $40-$50 for completion of the same assessments and disclosure sessions.
All participants received $10-$20 for completion of the six-month follow-up (T3)
assessment.
Design
Interested students were scheduled for an initial session and given informed
consent. Researchers explained privacy and confidentiality procedures in detail to
maximize participant comfort. As part of a larger study (n=84), participants were
randomly assigned to one of four conditions: (1) Written disclosure of traumatic
events, (2) Verbal disclosure of traumatic events, (3) Written disclosure of neutral
events, and (4) Verbal disclosure of neutral events. However, this study focused
only on those 42 participants assigned to written (n=21) and verbal (n=21)
disclosure of traumatic events. Randomization was denoted by sealed envelopes
after the T1 assessment and prior to the first disclosure session. The envelope was
opened in the presence of participants immediately following the T1 assessment.

16

After completing the informed consent, participants completed the T1
questionnaire, placed it in an envelope, sealed the envelope, and signaled the
experimenter that they were finished with the questionnaires. Participants then
began the experimental disclosure sessions. Participants wrote or talked four times
over the course of four to fourteen days, for 20 minutes each session. Writing or
talking took place in private rooms. Participants were told that they could be seen
but not heard by the experimenter, so that the experimenter could monitor
participants’ experiences and reactions during the sessions, and intervene, if
necessary, in cases of serious distress. Writing samples were collected from
participants in sealed envelopes. Talking was recorded, and audiotapes were
placed in sealed envelopes in front of participants.
Participants received the following instructions, which varied slightly based
on experimental condition:
(Session 1)
What I would like to have you write [talk] about for the next four days is
the most traumatic, upsetting experience of your entire life. When you
write [talk], I want you to really let go and explore your very deepest
emotions and thoughts. You can write [talk] about the same experience on
all four days or about different experiences each day. In addition to a
traumatic experience, you can also write [talk] about major conflicts or
problems that you have experienced or are experiencing now. Whatever
you choose to write [talk about], however, it is critical that you really delve
into your deepest emotions and thoughts. Ideally, we would also like you to
write [talk] about significant experiences or conflicts that you have not
discussed in great detail with others. Remember that you have four days to
write [talk]. You might tie your personal experiences to other parts of your
life. How is it related to your childhood, your parents, people you love,
who you are, or who you want to be. Again, when you write [talk],
examine your deepest emotions and thoughts.
17

(Session 2)
How did yesterday’s writing [talking task] go? Today, I want you to
continue writing [talking] about the most traumatic experience of your life.
It could be the same topic that you wrote [talked] about yesterday or it
could be something different. But today, I really want you to explore your
very deepest emotions and thoughts…

(Session 3)
You have written [talked] about important issues now for two days. You
only have today and tomorrow to finish. As with the first two days, I want
you to really explore your deepest thoughts and feelings.

(Session 4)
Today is the last day of writing [talking]. When you write [talk] today, I
again want you to explore your deepest thoughts and feelings about the
most traumatic experience of your life. Remember that this is the last day
and so you might want to wrap everything up. For example, how is this
experience related to your current life and your future? But feel free to go
in any direction you feel most comfortable with and delve into your deepest
emotions and thoughts. . .

When the 20 minutes were complete, administrators knocked on the door,
and told participants that they could stop writing or talking. Administrators
collected the sealed writing samples, or saved the talking samples and placed the
tapes in sealed envelopes. Writing sheets and tapes were coded so that no personal
information appeared on them.
Outcome Measures
The measures below were completed during T1 only.
Demographics Questionnaire. Information about age, race/ethnicity, place
of birth, age of immigration (if applicable), languages spoken, and parents’ place of
birth and occupation were obtained through a self-report questionnaire developed
for this study.
18

Asian American Multidimensional Acculturation Scale. Participants
completed the Asian American Multidimensional Acculturation Scale (AAMAS;
Chung, et al., 2004) to assess acculturation level. The AAMAS uses an orthogonal
model to measure acculturation on multiple dimensions. The scale consists of 31
items and uses a 6-point Likert type scale ranging from not very much to very much
to measure three cultural dimensions: culture of origin (CO), Asian American (AA)
and European American (EA) (e.g. “How well do speak the language of: a. your own
Asian ethnic group?, b. other Asian groups?, c. other non-Asian minority groups?, d.
English?). Individual items were grouped by associated subscale and then
averaged by participant, so that each participant had an average AAMAS composite
score for the Asian-American Subscale (AAMAS-AA), Country of Origin Subscale
(AAMAS-CO) and the European American Subscale (AAMAS-EA). All three of the
composite scores had high internal consistency: AAMAS-CO (α = .93), AAMAS-
AA (α = .90), AAMAS-EA (α = .89).
Asian orientation was operationalized as the composite of the standardized
Asian American (AAMAS-AA) and Country of Origin (AAMAS-CO) subscales of
the AAMAS, because these two scales were designed to tap into various aspects of
Asian orientation and were significantly correlated (r = .63, p<.001). Additionally,
the orthogonal construct of European American orientation was operationalized as
the AAMAS European American subscale.

19

The measures below were completed at T1, T2, and T3.
Brief Symptom Inventory. Participants completed the Brief Symptom Index
(BSI; Boulet & Boss, 1991) to measure psychiatric distress. The BSI is a 53-item
list describing a variety of complaints and problems (e.g., feeling lonely, trouble
remembering things). It was designed as a shortened version of the Revised
Symptom Checklist-90 (SCL-90-R). Items are rated on a 5-point Likert scale,
measuring degrees of distress from not at all to extremely. For scoring, the General
Severity Index (GSI) of the BSI was used, which is the mean of all completed items
on the BSI at each assessment time point. The internal consistency of the BSI GSI
composite score was .97 at T1, .97 at T2, and .95 at T3.
The following measure was completed at T1 and T3 only.
SMU Health Questionnaire. Information about current and past health
status was obtained using the SMU Health Questionnaire (SMU-HQ; Watson &
Pennebaker, 1989). The SMU-HQ is a symptom and illness questionnaire
measuring whether individuals have been diagnosed or treated for a variety of
health problems. For this study, the criteria for listing health information were
altered from the “previous year” to the “previous six-months”. Composite scores
were created by adding the total number of endorsed symptoms. The coefficient
alpha for the SMU-HQ at T1 was .75, and at T3 was .79.
20

Use of Negative Emotion
In order to measure participants’ use of negative emotion words during
disclosure sessions, transcribed talking sessions and typed writing sessions were
run through the Linguistic Inquiry and Word Count (LIWC) program, which
computed the percentage of words in each disclosure session that were negative
emotion words (Pennebaker, et al., 2001). A composite score of negative emotion
use was calculated by collapsing the percent negative emotion score across the four
disclosure sessions, resulting in a mean percent of negative emotion for each of the
forty-two participants
21

Chapter 3: Results
Characteristics of Participants
Participants averaged 21 years of age and included students in each college
class, as well as graduate students. The majority of participants identified
themselves as Chinese or Chinese American (71.4%), and most of the other
participants identified themselves as Korean, Japanese, or Vietnamese. Sixty-two
percent of participants were foreign-born. Most indicated that they were extremely
comfortable speaking (69%) and writing (60%) in English (Table 1).
Table 1: Demographic Variables of Study Population (N = 42)
%* Mean (SD) n
Age 20.88 (3.25) 42

Gender
+

Female 66.7% 28
Male 28.6% 12

Year in College
Freshman 21.4% 9
Sophomore 28.6% 12
Junior 16.7% 7
Senior 7.1% 3
Senior + 4.8% 2
Graduate Student 21.4% 9

Asian Ethnicity
Chinese/Chinese American 71.4% 30
Korean/Korean American 11.9% 5
Japanese/Japanese American 4.8% 2
Vietnamese/Vietnamese American 2.4% 1
Other 9.5% 4

Born in the United States
Yes 38.1% 16
No 61.9% 26

For those not born in the USA:
Age of immigration 15.85 (7.46) 26

22

Table 1, Continued
Comfort Writing in English 4.43 (.77) 42
1 Not at all comfortable 0% 0
2 0% 0
3 Somewhat comfortable 16.7% 7
4 23.8% 10
5 Extremely comfortable 59.5% 25

Comfort Speaking in English 4.55 (.74) 42
1 Not at all comfortable 0% 0
2 0% 0
3 Somewhat comfortable 14.3% 6
4 16.7% 7
5 Extremely comfortable 69.0% 29
Note. Due to rounding, percentages may not equal 100%
+
Item data missing for two participants

Investigation of the distribution of scores on the T1 well-being measures (Figures 1
and 2) revealed that one participant reported significantly more health symptoms
upon entering the study, and was an outlier.
Figure 1. Distribution of T1 Distress

T1 Distress
3.0
2.5
2.0
1.5
1.0
0.5
0.0

23

Figure 2. Distribution of T1 Self-Reported Health Symptoms

T1 Self-Reported Health Symptoms
14
12
10
8
6
4
2
0

Comparability of Groups at Pre-treatment
Participants were randomly assigned to either write or talk about traumatic
events. Analysis of variance (ANOVA) tests for pre-treatment variables showed no
significant group differences for any demographic, acculturation, or psychosocial
variables.
Missing Data
All participants completed the T1 questionnaire and the first disclosure
session. However, two participants declined to be audiotaped and so disclosure
content information was missing for them. In addition, two participants failed to
complete one measure each during the T1 assessment. Two participants dropped
out of the study after the first disclosure session. One participant failed to complete
24

the T2 questionnaire, but completed T3. Five participants were either unable to be
located or declined to complete the T3 follow-up. Finally 3 participants failed to
complete some of the 6-month follow-up measures. For all data, missing
composite scores were replaced using regression imputation by condition.
Analysis Plan
Figure 3 outlines the mediation model to be tested, in which the relationship
between disclosure modality and well-being is mediated by use of negative
emotion, after controlling for T1 well-being.

25
Figure 3. Model to be Tested

Disclosure
Modality

Well-Being
(distress,
health
symptoms)

Expression
of Negative
Emotion

Figure 3. Mediation model to be tested in which well-being at T3 and T2 is mediated by disclosure
modality. It was predicted that writing would lead to increases in well-being at T3 compared to talking, and
that the expression of negative emotion would mediate this relationship.

T1 Well-Being
(distress,
health
symptoms)

26
Additionally, we hypothesize that writing will lead to more expression of
negative emotion, and will therefore be more effective at reducing physical and
psychological symptoms at follow-up. While this study tested outcomes at one
week and six-months following disclosure, the primary outcomes of interest are
those at six-months, as this has been the time frame when beneficial effects have
been found in the past. Supplementary analyses examining the relationships in the
model at one-week follow-up were conducted as well. However we anticipate that
one-week will not be enough time to detect positive health benefits, and in fact,
expect to see an initial increase in psychological symptoms, rather than a decrease,
given the proximity to traumatic disclosure. Accordingly, results will be reported
first for well-being at six-months, and then will be followed by supplementary one-
week results.
To test the proposed mediation model, a series of residualized regressions
were run with each independent outcome (psychiatric distress and self-reported
health status) being individually predicted by disclosure modality and use of
negative emotion, after controlling for T1 well-being. Additionally, regressions
were run with disclosure modality as the predictor and use of negative emotion as
the outcome, controlling for T1 well-being as well, in order to ensure that
differences in the use of negative emotion was related to disclosure modality rather
than differences in T1 well-being. In cases where all three relationships were

27
significant, Sobel mediation tests were run to determine if any of the models
indicated that mediation was occurring through use of negative emotion.
Finally, the literature suggests that an area of interest to explore is whether
or not orientation to Asian or European American culture moderates any of the
relationships in the mediation models and if so, how. To explore this question,
moderation analyses were included as well. In order to investigate whether Asian
or European American orientation moderates these relationships, a series of
regressions were run in which the corresponding centered T1 score, centered key
predictor and the centered constructs of Asian orientation, European American
orientation, and being foreign-born were included as covariates and predictors in
the model, along with their interaction terms. Moderation was indicated if the
interaction terms were significant.
Testing the T3 mediation model
To test the T3 mediation model, each path was first tested independently,
and then if all three paths were significant, the significance of the overall mediation
model was assessed (Figure 4).

28
Figure 4. Relationships among constructs in the model at T3

Distress: r = .01
Health symptoms = -.04
Disclosure
Modality
Well-Being
(distress,
health
symptoms)

Expression
of Negative
Emotion

Distress: β =.03
Health symptoms: β = .21

β = -.52**

Distress: β = 05
Health symptoms: β = .17
Figure 4. At T3, after controlling for T1 well-being, the only significant relationship in the mediation model is the
relationship between disclosure modality and use of negative emotion, such that writing elicits more negative emotion
than talking. Disclosure modality is dummy-coded with Writing = 0 and Talking = 1.
** β is significant at the <.001 level

T1 Well-Being
(distress,
health
symptoms)
Distress: β = .62**
Health symptoms: β = .73**
Distress: β = .16
Health symptoms: β = .18

29

Path 1: Relationship between disclosure modality and well-being at T3
In order to address whether disclosure modality predicted health outcomes
at T3, a series of regressions were run. Whether participants wrote or talked about
their traumatic experiences did not lead to differences in self-reported well-being
six-months later.
Path 2: Relationship between disclosure modality and use of negative emotion
words
A regression was run to determine if disclosure modality predicted use of
negative emotion words. Disclosure modality significantly predicted use of
negative emotion words. Participants who wrote about traumatic events used a
greater percentage of negative emotion words than those who talked (β=-.52, p <
0.001).
Path 3: Relationship between use of negative emotion words and health at T3
Regressions were run to predict T3 well-being from use of negative
emotion. After controlling for T1 pre-treatment scores, use of negative emotion
words did not predict well-being at T3 (p’s >.05). Thus, across treatment
conditions, participant use of negative emotion words was not related to well-being
six-months following disclosure.
Summary of Mediation at T3
The adjusted βs of two of the mediation paths failed to reach significance,
and the model was therefore ineligible to be tested for mediation (Figure 4).

30

Summary of T3 Results
The mediation model to be tested, in which greater use of negative emotion
words mediated the effect of disclosure modality on outcomes, was not supported.
For T3 outcomes, only the relationship between disclosure modality and use of
negative emotion was significant, such that writing about traumatic events resulted
in significantly greater use of negative emotion words (as a percent of all words
used during experimental disclosure) compared to talking about traumatic events.
No other significant relationships existed in the model.
Supplementary Analyses: Testing the T2 Mediation Model
Path 1: Relationship between disclosure modality and well-being at T2
Following the same analysis strategy utilized for T3 outcomes, disclosure
modality’s prediction of T2 well-being was assessed. Disclosure modality was not
a significant predictor of T2 outcomes (Figure 5).

31
Figure 5. Relationships among variables at T2 after controlling for T1 well-being

Figure 5. At T2, after only controlling for T1 well-being, the only significant relationship in the mediation model is the relationship
between disclosure modality and use of negative emotion, such that writing elicits more negative emotion than talking. Disclosure
modality is dummy-coded with Writing = 0 and Talking = 1.
** β is significant at the <.001 level

Disclosure
Modality

Well-Being
(distress)

Expression
of Negative
Emotion

Distress: β =.16

Distress: β = .07

β = -.52**

T1 Well-Being
(distress)
Distress: r = .01

Distress: β = .81**

Distress: β = .19

32

However, given that the direction of the direct relationship between disclosure
modality and T2 outcomes was positive, and that the direction of the indirect
relationship was negative (Figure 5, Table 2), the possibility of a suppression effect
was indicated.

33
Table 2. Intercorrelations Between Constructs

Use of Negative
Emotion
Asian
Orientation
European American
Orientation
T1
Distress
T2
Distress
T3
Distress
T1 Health
Symptoms
T3 Health
Symptoms

Disclosure
Modality

-.57** .02 -.24 .01 .17 -.05 -.04 -.08

Use of Negative
Emotion

-- -.01 .08 .19 .22 .165 .23 .33*

Asian Orientation

-- -- -.14 -.07 -.15 .05 -.17 -.38*

European American
Orientation

-- -- -- -.29 -.40** -.36* -.06 .32*

T1 Distress

-- -- -- -- .81** .62** .19 .32*

T2 Distress

-- -- -- -- -- .70** .39* .40**

T3 Distress

-- -- -- -- -- -- .21 .08

T1 Health
Symptoms

-- -- -- -- -- -- -- .73**
** Correlation is significant at the 0.01 level (2-tailed)
* Correlation is significant at the 0.05 level (2-tailed)

34

To test for suppression, use of negative emotion was included as a covariate in the
model. After controlling for use of negative emotion and T1 scores, disclosure
modality significantly predicted distress (β = .30, p = 0.007) and depressive affect
(β = .28, p = .02), with talking leading to greater symptoms than writing. By
adding an indirect effect into the regression, the direct effect was strengthened,
indicating that use of negative emotion may be acting as a suppressor.
According to Tzelgov and Henik (1991), a suppressor exists if β
multivariate
>
r
criterion,predictor
. The inclusion of the suppressor variable reduces the dependent-
irrelevant part of the predictor’s variance, leading to an increase in the beta weight
of the prediction of the dependent variable. Essentially the suppressor acts as an
“irrelevant-variance ‘cleaner’” (Tzelgov and Henik, 1991). Identifying the
potential suppression effects is necessary, because partialing-out the irrelevant
variance through multiple regression makes the true effect of disclosure modality
clearer (Tzelgov and Henik, 1991). If negative emotion acts as a suppressor
variable, then negative emotion will clear out the variance associated with negative
emotional expression from that measuring the impact of disclosure modality.
In order to test whether, and for which regressions, negative emotion acts as
a suppressor, the test of suppression as outlined in Tzelgov and Henik (1991) was
used, in which suppression is indicated by β
predictor
> r
cp
when all variables have
been standardized and disclosure modality is scored such that r
criterion ,disclosure modality

is positive. All T2 variables were standardized and disclosure modality was

35

dummy coded to ensure that the correlation between disclosure modality and
outcome would be positive. Finally, beta weights from the residualized regressions
above -- in which disclosure modality was the predictor, T2 well-being measures
were the dependent variables, use of negative emotion was the potential
suppression variable, and T1 scores were covariates -- were compared to the
corresponding bivariate correlation coefficients (Table 3).
Table 3. Negative emotion acting as a suppressor of the relationship between
disclosure modality and T2 well-being

r
cp

Writing = 0
Talking = 1

β
††
of
disclosure
modality

Difference between
β and r
cp

Suppression
Indicated
(Y/N)
Distress
†

.17

.30** .14 Y
†
Scores have been standardized into Z-scores for comparison
††
After controlling for T1 well-being and use of negative emotion (suppressor)
** Significant at the 0.01 level
* Significant at the 0.05 level
Results indicated that all three relationships between disclosure modality
and T2 distress were being suppressed by use of negative emotion. Disclosure
modality was a significant predictor of distress at T2 after removing the criterion-
irrelevant variance attributable to use of negative emotion (the suppressor). It
should be noted that use of negative emotion also became significant in this
equation, as simple reflection of the variables of disclosure modality and use of
negative emotion in the residualized regression “converts the predictor to a
suppressor and the suppressor to a predictor” (Conger & Jackson, 1972). As the
regression coefficients of use of negative emotion and disclosure modality both

36

increase in multivariate regression compared to their respective validities, a pattern
of reciprocal suppression emerged (Tzelgov and Henik, 1991). Thus, disclosure
modality was a significant predictor of distress one week after disclosure, such that
when use of negative emotion is held constant, talking leads to more distress than
writing.
Path 2: Relationship between disclosure modality and use of negative emotion
words
As noted earlier, disclosure modality significantly predicted use of negative
emotion words (Figure 5). Participants who wrote about traumatic events used a
greater percentage of negative emotion words than those who talked (β = -.52, p <
0.001).
Path 3: Relationship between use of negative emotion words and health at T2
Using the same process described for T3 outcomes, regressions were run to
predict T2 well-being from use of negative emotion. Use of negative emotion
words did not predict T2 outcomes (Figure 3; p’s >.05). To test for reciprocal
suppression, bivariate correlation coefficients between negative emotion and well-
being were compared to standardized beta weights of negative emotion predicting
well-being after including disclosure modality in the multivariate regressions
(Table 4).

37

Table 4. Disclosure modality acting as a suppressor of the relationship between use
of negative emotion and well-being

r
cp
Use of Negative
Emotion

β
††
of
Negative
Emotion

Difference
between
β and r
cp

Suppression
Indicated
(Y/N)

Distress
†

.22 .25* .03 Y
†
Scores have been standardized into Z-scores for comparison
††
After controlling for T1 well-being and use of negative emotion (suppressor)
** Significant at the 0.01 level
* Significant at the 0.05 level
Using standardized variables, disclosure modality was added to Block 2 of the
above regression equations, and negative emotion (scored so that r
criterion,negative
emotion
is positive) was the key predictor in Block 3.
A suppression effect was found, with disclosure modality suppressing the
relationship between use of negative emotion predicting T2 distress. While this
relationships was previously not significant, when disclosure modality was
included in the regressions, this relationship became significant, with greater use of
negative emotion predicting greater distress (β=0.25, p=0.03). Therefore, it appears
that disclosure modality and use of negative emotion have a cooperative
suppression relationship in the prediction of distress.
Summary of Mediation at T2
After accounting for the criterion-irrelevant invariance by including the
suppressor variables in the model, the adjusted βs of the relationships between the
constructs in the T2 mediation model all reached significance (Figure 6) and were
therefore eligible to be tested for mediation.

38
Figure 6. Relationships between variables at T2 after including suppressor variables in the model

Disclosure
Modality

Well-Being
(distress)

Expression
of Negative
Emotion

Distress: β = .30**

Distress: β = .25*

β = -.52**

Figure 6. After accounting for suppression effects, all relationships in the mediation model are significant at T2. Disclosure modality is
dummy-coded with Writing = 0 and Talking = 1. Writing elicits more negative emotion than talking; use of negative emotion results in
increased distress at T2; and talking leads to more distress than writing at T2.
** β is significant at the <.001 level

T1 Well-Being
(distress)
Distress: r = .01
Distress: β = .19

Distress: β = .76**

39

In order to test for mediation, the Sobel test was run using unstandardized
coefficients of the regressions between disclosure modality, use of negative
emotion, and distress, after controlling for T1 scores. The mediation model was
not supported.
Summary of T2 Results
One week after experimental disclosure sessions concluded, greater use of
negative emotion words significantly predicted increased psychological distress.
This path might suggest that one week after disclosure, writing about traumatic
events would also lead to greater distress than talking about traumatic events.
However, the opposite effect was found. While both disclosure modality and use
of negative emotion words were significant predictors of psychological distress at
T2, after accounting for reciprocal suppression, use of negative emotion appears to
be primarily a suppressor rather than a predictor. The direct effect of talking
leading to greater distress than writing overrode the indirect effect of: (1) writing
leading to more negative emotion which (2) leads to more distress. Additionally,
although requirements for testing mediation were met for T2 distress, the Sobel
mediation test failed to reach statistical significance.
Supplementary Analyses: Does Cultural Orientation Moderate the Relationships in
the Model?
To assess whether Asian or European American orientation moderates the
relationship between disclosure modality and T3 well-being outcomes, regressions

40

were run using each of the centered T3 and T2 health measures as outcomes and
the orientation interaction terms as key predictors. There were no moderation
effects for any T3 or T2 health measures. The procedure was repeated when
exploring whether orientation moderated the relationship between disclosure
modality and use of negative emotion words and the relationship between use of
negative emotion words and T3 well-being. There was no moderation effect of
Asian or European American orientation on (1) the relationship between disclosure
modality and use of negative emotion words or (2) the relationship between
negative emotion and health.

41

Discussion
This study explored the relationship between disclosure modality and future
well-being and whether use of negative emotion mediated this relationship in an
East Asian American population. Additionally, the study investigates how
disclosure modality interacts with Asian and European American orientation to
impact efficacy. According to Exposure Theory, the expression of negative
emotion is critical to understanding how experimental disclosure improves well-
being (Frattaroli, 2006). While experimental disclosure studies have found writing
and talking to be equally effective disclosure modalities (Murray and Segal, 1994;
Pennebaker, Kiecolt-Glaser, & Glaser, 1988), European Americans and East Asian
Americans differ in their comfort with verbal and emotional expression (Chen &
Davenport, 2005; Kim, 2002). This difference could alter the efficacious
component and moderate the effectiveness of experimental disclosure. Given these
cultural dynamics, one goal of the study was to determine whether disclosure
modalities differ in their effects on emotion expression and well-being for East
Asian Americans.
Given cultural norms and values that de-emphasize verbal expression (Kim,
2002), it was hypothesized that East Asian Americans would be more likely to
express negative emotions when asked to write about stressful events versus talking
about these events. The hypothesis was supported, as the results indicated that
writing elicited more negative emotional expression than talking did. However, the

42

second hypothesis, that writing, as the better elicitor of negative emotion, should
therefore lead to greater improvements in future well-being compared to talking,
was not supported. Despite writing leading to more negative emotional expression,
writing was not more effective than talking at improving well-being at the six-
month follow-up.
Although the lack of difference in the efficacy of writing versus talking was
not hypothesized, it is in line with previous research that found no difference in the
efficacy of writing or talking during experimental disclosure (Lyubomirsky, Sousa,
& Dickerhoof, 2006; Murray and Segal, 1994; Pennebaker, Kiecolt-Glaser, &
Glaser, 1988). It is also important to note that even though the two interventions
were not differentially effective at improving well-being, well-being did change
over time (see appendix) it is possible that both conditions led to improvements in
well-being, but as they were only compared to each other and not to a neutral
condition, we cannot discern the ultimate impact of these interventions.
The third hypothesis predicted that the increased well-being following
writing would be a result of the increased use of negative emotion during writing.
In other words, it was hypothesized that use of negative emotion would mediate the
relationship between disclosure modality and future well-being. This hypothesis
was not supported as writing and talking were not differentially effective at
predicting well-being six-months later. One conclusion would be that the
mediation model as hypothesized by Exposure Theory is not supported; however,

43

this conclusion is premature as this study had two significant limitations in addition
to the one mentioned above about the lack of a control condition. First, the study
relied exclusively on self-report outcomes. Most studies using experimental
disclosure have found significant six-month effects using more objective health and
well-being data such as participant visits to a health center, grades, immune
functioning as measured by blood samples, or future employment (Frattaroli,
2006). As this study relied exclusively on self-report measures, significant effects
may not be as detectable. Second, given that the average effect size (based on
objective measures) is small to moderate (Frattaroli, 2006), a total sample size of
42 may not be sufficient to detect differences in self-reported well-being between
two intervention conditions.
Although the one-week analyses were supplementary, a difference in well-
being by disclosure modality was found at T2. While both disclosure modality and
use of negative emotion words were significant predictors of psychological distress
one week following experimental disclosure, use of negative emotion was primarily
a suppressor rather than a predictor (Tzelgov and Henik, 1991), and the direct
effect of talking leading to greater distress overrode the indirect effect of writing on
negative emotion. Therefore, overall, talking about traumatic events resulted in
increased distress one week after disclosure compared to writing about traumatic
events.

44

While it was anticipated that writing and talking about traumatic events
would lead to an initial increase in symptoms at T2, the pattern of change and the
differences by disclosure modality are noteworthy. Not surprisingly, one week
after traumatic disclosure, greater use of negative emotion words significantly
predicted increased psychological distress. However, although writing elicited
significantly more negative emotion than talking, talking resulted in more T2
distress than writing. This counterintuitive result parallels that found by Esterling
and colleagues (1994), in which increased use of negative emotion did not appear
to be correlated with well-being, and that although writing resulted in significantly
more expression of negative emotion than talking, talking was the more effective
intervention. Writing and talking appear to have intrinsic qualities that differentiate
the degree of distress that they cause, such that talking about traumatic events is
significantly more upsetting than writing, even though writing elicits more negative
emotion.
While increased distress is certainly not a positive result, it was not
unexpected, as participants were asked to disclose personal and highly upsetting
events only a few days before the T2 assessment, and one week may not be a long
enough time after disclosure of traumatic events for a reduction in symptoms to
occur. However, it is notable that there were differences by disclosure modality in
the degree of reported distress. As a mode of disclosure, talking about traumatic
events was more upsetting. Although there were no differences in the ability of

45

these two interventions to improve well-being six-months following disclosure, if
the interventions prove to be more effective at improving well-being using
objective outcome measures, writing will be the more useful intervention, as it not
only elicited the most negative emotion, but was also the least upsetting of the two
disclosure modalities.
While some of the hypotheses originally proposed in this study were not
supported, the goal of finding a disclosure modality that would facilitate emotional
expression in an East Asian American sample was successful. Writing elicited
significantly more negative emotional expression than talking; moreover, writing
was less upsetting to participants than talking. The finding that writing may be a
more beneficial and acceptable mode of disclosure of traumatic events for East
Asian Americans indicates that culture and ethnicity have a place in determining
how to best implement an intervention. While more work needs to be done to
clarify these complex findings, the study demonstrated that the generalizability of
interventions across ethnic and cultural groups should not be taken for granted, as
certain aspects may be moderated by group differences.
Future Work
The limitations outlined above are being addressed in the broader study, of
which this study was a part. The larger study includes two control conditions
which will allow us to determine if the two intervention conditions discussed here
were effective at improving future well-being, but were simply not more effective

46

than each other. Additionally, the larger study includes objective indicators of
health, as measured by visits to the campus health center, and an objective measure
of academic functioning as measured by changes in G.P.A. across semesters. The
larger study also includes twice the number of participants, which may improve
power.
The broader study will allow us to address some of the significant
limitations in this study design, however, there are two concerns that will need to
be attended to in future work. One is that writing and talking may not be different
enough disclosure modalities to result in differential effects. One modality that is
currently being used to expose participants to negative emotional information
regarding a trauma is virtual reality (Parsons & Rizzo, 2008). Use of an exposure
modality, such as virtual reality, which is very different than writing or talking,
may lead to greater differentiation of well-being benefits. Additionally, European
Americans should be included in future work to better assess cultural effects.

47

References
Applebaum, M.I, & McCall, R.B. (1983). Design and analysis in developmental
psychology. In P.H. Mussen (Series Ed.) & W. Kessen (Ed.), Handbook of
Child Psychology: History, Theory, and Methods (Vol. 1, pp. 415-476).
New York: John Wiley.

Atkinson, D.R. & Gim, R.H. (1989). Asian-American cultural identity and
attitudes toward mental health services. Journal of Counseling Psychology,
36(2): 209-212.

Barrett, L.F., Gross, J., Christensen, T.C., & Benvenuto, M. (2001). Knowing what
you’re feeling and knowing what to do about it: Mapping the relation
between emotion differentiation and emotion regulation. Cognition and
Emotion, 15, 713-724.

Barry, D.T. (2003). Cultural and demographic correlates of self-reported
guardedness among East Asian immigrants in the US. International
Journal of Psychology, 38, 150-159.

Barry, D.T. & Mizrahi, T.C. (2005). Guarded self-disclosure predicts
psychological distress and willingness to use psychological services among
East Asian immigrants in the United States. The Journal of Nervous and
Mental Disease, 193, 535-539.

Beck, A.T., Ward, C.H., Mendelson, M., Mock, J., & Erbaugh, J. (1961). An
Inventory for Measuring Depression. Archives of General Psychiatry, 4,
561-571.

Bootzin, R.R. (1997). Examining the theory and clinical utility of writing about
emotional experiences. Psychological Science, 8, 167-169.

Bond, M.H., (1993). Emotions and their expressions in Chinese culture. Journal
of Nonverbal Behavior, 17, 245-262.

Boulet, J. & Boss, M.W. (1991). Reliability and Validity of the Brief Symptom
Inventory. Journal of Consulting and Clinical Psychology, 3, 433-437.

Brewin, C.R. & Lennard, H. (1999). Effects of mode of writing on emotional
narratives. Journal of Traumatic Stress, 12, 355-361.

48

Brown, E. J. & Heimberg, R. G. (2001). Effects of writing about rape: Evaluating
Pennebaker’s paradigm with a severe trauma. Journal of Traumatic Stress,
14, 781-790.

Cameron, L.D. & Nicholls, G. (1998). Expression of stressful experiences through
writing: Effects of a self-regulation manipulation for pessimists and
optimists. Health Psychology, 17, 84-92.

Campbell, R.S. & Pennebaker, J.W. (2003). The secret life of pronouns: Flexibility
in writing style and physical health. Psychological Science, 14, 60-65.

Caudill, W. & Schooler, C. (1973). Child behavior and child rearing in Japan and
the United States: An interim report. Journal of Nervous and Mental
Disease, 157, 323–338.

Chen, S.W. & Davenport, D.S. (2005). Cognitive-behavioral therapy with Chinese
American clients: Cautions and modifications. Psychotherapy Theory:
Research, Practice, Training, 42, 101-110.

Chung, R.H.G., Kim, B.S, & Abreu, J.M. (2004). Asian American
Multidimensional Acculturation Scale: Development, factor analysis,
reliability, and validity. Cultural Diversity and Ethnic Minority
Psychology, 10, 66-80.

Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of perceived
stress. Journal of Health and Social Behavior, 24, 385-396.

Conger, A.J. & Jackson, D.N. (1972). Suppressor variables, predictions, and the
interpretation of psychological relationships. Educational and
Psychological Measurement, 32, 579-599.

de Moor, C., Sterner, J., Hall, M., Warneke, C., Gilani, Z., Amato, R., et al. (2002).
A pilot study of the effects of expressive writing on psychological health
and behavioral adjustment in patients enrolled in a Phase II trial of vaccine
therapy for metastatic renal cell carcinoma. Health Psychology, 21, 615-
619.

Donnelly, D.A. & Murray, E.J. (1991). Cognitive and emotional changes in written
essays and therapy interviews. Journal of Social and Clinical Psychology,
10, 334-350.

49

Esterling, B.A., Antoni, M.H., Fletcher, M.A., Margulies, S., & Schneiderman, N.
(1994). Emotional disclosure through writing or speaking modulates latent
Epstein-Barr virus antibody titers. Journal of Consulting and Clinical
Psychology, 62, 130-140.

Falsetti, S.A., Resnick, H.S., Resick, P.A., & Kilpatrick, D.G. (1993). The
Modified PTSD Symptom Scale: A brief self-report measure of
posttraumatic stress disorder. Behavior Therapist, 16, 161-162.

Faul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (in press). G*Power 3: A
flexible statistical power analysis program for the social, behavioral, and
biomedical sciences. Behavior Research Methods.

Foa, E.B., Riggs, D.S., Dancu, C.V., & Rothbaum, B.O. (1993). Reliability and
validity of a brief instrument for assessing post-traumatic stress disorder.
Journal of Traumatic Stress, 6, 459-473.

Frattaroli, J. (2006). Experimental disclosure and its moderators: A meta-analysis.
Psychological Bulletin, 132, 823-865.

Friedman, L. & Wall, M. (2005). Graphical views of suppression and
multicollinearity in multiple linear regression. The American Statistician,
59, 127-136.

Frisina, P.G., Borod, J.C., & Lepore, S.J. (2004). A meta-analysis of the effects of
written emotional disclosure on the health outcomes of clinical populations.
The Journal of Nervous and Mental Disease, 192, 629-634.

Gim, R.H., Atkinson, D.R., & Whiteley, S. (1990). Asian-American acculturation,
severity of concerns, and willingness to see a counselor. Journal of
Counseling Psychology, 37, 281-285.

Graybeal, A., Sexton, J.D., & Pennebaker, J.W. (2002). The role of story-making
in disclosure writing: The psychometrics of narrative. Psychology and
Health, 17, 571-581.

Greenberg, M.A. & Stone, A.A. (1992). Emotional disclosure about traumas and
its relation to health: Effects of previous disclosure and trauma severity.
Journal of Personality and Social Psychology, 63, 75-84.

50

Greenberg, M.A., Wortman, C.B., & Stone, A.A. (1996). Emotional expression
and physical health: Revising traumatic memories or fostering self-
regulation. Journal of Personality and Social Psychology, 71, 588-602.

Harrist, S., Carlozzi, B.L., McGovern, A.R., & Harrist, A.W. (2007). Benefits of
expressive writing and expressive talking about life goals. Journal of
Research in Personality, 41, 923-930.

Ho, M.K. (1984). Social group work with East Asian Americans, Social Casework,
57, 195-201.

Hsu, F.L.K. (1985). The self in cross-cultural perspective. In A. Marsella, G.
DeVos, & F. Hsu (Eds.), Culture and Self (pp. 24-55). New York:
Tavistock.

Kao, E.M., Nagata, D.K. & Peterson, C. (1997). Explanatory style, family
expressiveness, and self-esteem among Asian American and European
American College Students. The Journal of Social Psychology, 137, 435-
444.

Keltner, D., Locke, K.D., & Audrian, P.C. (1993). The influence of attributions on
the relevance of negative feelings to personal satisfaction. Personality and
Social Psychology Bulletin, 19, 21-29.

Kim, H.S. (2002). We talk, therefore we think? A cultural analysis of the effect of
talking on thinking. Journal of Personality and Social Psychology, 83, 828-
842.

Kim, B.S.K., Atkinson, D., & Yang, P.H. (1999). The Asian Values Scale:
Development, factor analysis, validation, and reliability. Journal of
Counseling Psychology, 46, 342-352.

Klein, K. & Boals, A. (2001). Expressive writing can increase working memory
capacity. Journal of Experimental Psychology: General, 130, 520-533.

Kleinman, A. & Lin, T. (1981). Normal and abnormal behavior in Chinese
culture. Boston: Reidel.

Kleinman, A. & Sung, L.H. (1979). Why do indigenous practitioners successfully
heal? Social Science & Medicine, 13B, 7-26.

51

Krantz, A. & Pennebaker, J.W. (1995). Bodily versus written expression of
traumatic experience. Unpublished manuscript.

Leong, F.T. (1986). Counseling and psychotherapy with Asian-Americans: Review
of the literature. Journal of Counseling Psychology, 33, 196-206.

Leong, F.T.L. & Lau, A.S.L. (2001). Barriers to providing effective mental health
services to East Asian Americans. Mental Health Services Research, 3,
201-214.

Lepore, S.J. (1997). Expressive writing moderates the relation between intrusive
thoughts and depressive symptoms. Journal of Personality and Social
Psychology, 73, 1030-1037.

Liem, R., Lim, B., & Liem, J. (2000). Acculturation and emotion among East
Asian Americans. Cultural Diversity and Ethnic Minority Psychology, 6,
13-31.

Lin, Y.N. (2002). The application of cognitive-behavioral therapy to counseling
Chinese. American Journal of Psychotherapy, 56, 46-58.

Littrell, J. (1998). Is the reexperiencing of painful emotion therapeutic? Clinical
Psychology Review, 18, 71-102.

Lum, R.G. (1982). Mental health attitudes and opinions of Chinese. In E. Jones &
S. Korchin (Eds.), Minority mental health (pp. 165-189). New York:
Praeger.

Lyubomirsky, S., Sousa, L., & Dickerhoof, R. (2006). The costs and benefits of
writing, talking, and thinking about life’s triumphs and defeats. Journal of
Personality and Social Psychology, 90, 692-708.

Marsella, A.J., Kinzie, D., & Gordon, P. (1973). Ethnic variations in the
expression of depression. Journal of Cross-Cultural Psychology, 4, 435-
458.

MacKinnon, D.P., Krull, J.L., & Lockwood, C.M. (2000). Equivalence of the
mediation, confounding and suppression effect. Prevention Science, 1, 173-
181.

52

Meads, C. & Nouwen, A. (2005). Does emotional disclosure have any effects? A
systematic review of the literature with meta-analyses. International
Journal of Technology Assessment in Health Care, 21, 153-164.

Minami, M. (1994). English and Japanese: A cross-cultural comparison of parental
styles of narrative elicitation. Issues in Applied Linguistics, 5, 383–407.

Murray, E.J. & Segal, D.L. (1994). Emotional processing in vocal and written
expression of feelings about traumatic experiences. Journal of Traumatic
Stress, 7, 391-405.

Parsons, T.D. & Rizzo, A.A. (2008). Affective outcomes of virtual reality
exposure therapy for anxiety and specific phobias: A meta-analysis.
Journal of Behavior Therapy and Experimental Psychiatry, 39, 250-261.

Pennebaker, J.W. (1997). Writing about emotional experiences as a therapeutic
process. Psychological Science, 8, 162-166.

Pennebaker, J.W. & Beall, S.K. (1986). Confronting a traumatic event: Toward an
understanding of inhibition and disease. Journal of Abnormal Psychology,
95, 274-281.

Pennebaker, J.W., & Chung, C.K. (2007). Expressive writing, emotional
upheavals, and health. In H.S. Friedman and R.C. Silver (Eds.)
Foundations of Health Psychology (pp. 263-284). Oxford University Press:
New York.

Pennebaker, J.W. & Francis, M.E. (1996). Cognitive, emotional, and language
processes in disclosure. Cognition and Emotion, 10, 601-626.

Pennebaker, J.W., Francis, M.E., & Booth, R.J. (2001). Linguistic Inquiry and
Word Count. LIWC2001 Manual. Erlbaum Publishers, Mahwah, NJ.

Pennebaker, J.W., Kiecolt-Glaser, J., & Glaser, R. (1988). Disclosure of traumas
and immune function: Health implications for psychotherapy. Journal of
Consulting and Clinical Psychology, 56, 239-245.

Pennebaker, J.W., Mehl, M.R. & Niederhoffer, K.G. (2003). Psychological aspects
of natural language use: Our words, our selves. Annual Review of
Psychology, 54, 547-577.

53

Pennebaker, J.W. & Susman, J.R. (1988). Disclosure of traumas and
psychosomatic processes. Social Science and Medicine, 26, 327-332.

Plake, B.S., Impara, J.C., & Murphy, L.L. (Eds.), (2001). The Fourteenth Mental
Measurements Yearbook, The Buros Institute of Mental Measurements.
The University of Nebraska Press, Lincoln.

Potter, S.H. (1988). The cultural construction of emotion in rural Chinese social
life. Ethos, 16, 181-208.

Redeker, G. (1984). On difference between spoken and written language. Discourse
Processes, 7, 43-55.

Root, M. (1985). Guidelines for facilitating therapy with Asian American clients.
Psychotherapy, 22, 349-356.

Shapiro S.S., & Wilks, M.B. (1965). An analysis of variance test for normality
(complete samples). Biometrika, 52, 591-611.

Sloan, D.M. & Marx, B.P. (2004). A closer examination of the structured written
disclosure procedure. Journal of Consulting and Clinical Psychology, 72,
165-175.

Sloan, D.M. & Marx, B.P. (2004). Taking pen to hand: Evaluating theories
underlying the written disclosure paradigm. Clinical Psychology: Science
and Practice, 11, 121-137.

Sloan, D.M., Marx, B.P., and Epstein, E.M. (2005). Further examination of the
exposure model underlying the efficacy of written emotional disclosure.
Journal of Consulting and Clinical Psychology, 73, 549-554.

Smyth, J.M. (1998). Written emotional expression: Effect sizes, outcome types,
and moderating variables. Journal of Consulting and Clinical Psychology,
66, 174-184.

Smyth, J., True, N., & Souto, J. (2001). Effects of writing about traumatic
experiences: The necessity for narrative structuring. Journal of Social and
Clinical Psychology, 20, 161-172.

Soto, J.A., Levenson, R.W. & Ebling, R. (2005). Cultures of moderation and
expression: Emotional experience, behavior, and physiology in Chinese
Americans and Mexican Americans. Emotion, 5, 154-165.

54

Sue (1997). Counseling strategies for Chinese Americans. In C.C. Lee (Ed.)
Multicultural issues in counseling: New approaches to diversity (2
nd
ed., pp.
173-187). Alexandria, VA: American Counseling Association.

Sue, S. & Kitano, H.H. (1973). Stereotypes as a measure of success. Journal of
Social Issues, 29, 83-98.

Sue, S., & Sue, D.W. (1974). MMPI comparisons between Asian-American and
non-Asian students utilizing a student health psychiatric clinic. Journal of
Counseling Psychology, 21, 423-427.

Sue & Sue (1977). Barriers to effective cross-cultural counseling. Journal of
Counseling Psychology, 24, 420-429.

Sue, D.W. & Sue, D. (1999). Counseling East Asian Americans. In D.W. Sue &
D. Sue, Counseling the culturally different: Theory and practice (pp. 255-
271). New York: Wiley.

Sue, S., Wagner, N., Davis, J.A., Margullis, C., & Lew, L. (1976). Conceptions of
mental illness among Asian and Caucasian-American students.
Psychological Reports, 38, 703-708.

Tabachnick, B.G. & Fidell, L.S. (2001). Using Multivariate Statistics (4
th
ed.).
Needham Heights, MA: Allyn and Bacon.

Toyokawa, T., & Nedate, K. (1996). Application of cognitive behavior therapy to
interpersonal problems: A case study of a Japanese female client. Cognitive
and Behavioral Practice, 3, 289-302.

Tsai, J.L. & Levenson, R.W. (1997). Cultural influences on emotional responding:
Chinese American and European American dating couples during
interpersonal conflict. Journal of Cross-Cultural Psychology, 28, 600-625.

Tsai, J.L., Levenson, R.W., & Carstensen, L.L. (2000). Autonomic, expressive, and
subjective responses to emotional films in older and younger Chinese
American and European American adults. Psychology and Aging, 15, 684-
693.

Tzelgov, J. & Henik, A. (1991). Suppression situations in psychological research:
Definitions, implications and applications. Psychological Bulletin, 109,
524-536.

55

Tzelgov, J. & Stern, I. (1978). Relationships between variables in three variable
linear regression and the concept of suppressor. Educational and
Psychological Measurement, 38, 325-335.

Ullrich, P.M. & Lutgendorf, S.K. (2002). Journaling about stressful events:
Effects of cognitive processing and emotional expression. Annals of
Behavioral Medicine, 24, 244-250.

van Minnen, A., Wood, I., Dijkstra, T., & Roelofs, K. (2002). Changes in PTSD
patients’ narratives during prolonged exposure therapy: A replication and
extension. Journal of Traumatic Stress, 15, 255-258.

Walker, D.A. (2003). Suppressor variable(s) importance within a regression
model: An example of salary compression in career services. Research in
Brief, 44, 127-133.

Watson, D., Clark, L.A., & Tellegan, A. (1988). Development and validation of
the brief measures of positive and negative affect: The PANAS scales.
Journal of Personality and Social Psychology, 54, 1063-1070.

Watson, D., & Pennebaker, J.W. (1989). Health complaints, stress, and
distress: Exploring the central role of Negative Affectivity. Psychological
Review, 96, 234-254.

Wu, D.Y.H. (1982). Psychotherapy and emotion in traditional Chinese medicine.
In A.J. Marsella & G.M. White (Eds.), Cultural Conceptions of Mental
Health and Therapy, (pp. 285-301). Dordrecht, Netherlands: D. Reidel.

Yan, S. (1996). Traditional Chinese emotional therapy and its application.
Psychological Science China, 19, 265-268.

56

Appendix
Changes in outcome measures over time
In order to examine changes in outcome measures over the three
assessments (pre-treatment, one-week following disclosure, and six-months
following disclosure), repeated measures ANOVAs were run on distress and self-
reported health status. Results indicated that mean distress levels were
significantly lower at each assessment period (Table A1, Figure A1).
Table A1. Pairwise comparisons of the change in well-being over time
Mean Difference
Distress
Self-Reported
Health

T1 to T2

.28* ---

T2 to T3

.25* ---

T1 to T3

.53* -.013
Note. Based on estimated marginal means
* The mean difference is significant at the .05 level.

57

Figure A1. Change in distress over time by disclosure modality
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
T1 T2 T3
BSI Scores
Writing
Talking

There was no difference between T1 and T3 measures of self-reported health status
(Table A1, Figure A2).

58

Figure A2: Change in self-reported health symptoms over time by disclosure
modality
3.5
3.6
3.7
3.8
3.9
4.0
4.1
4.2
4.3
4.4
4.5
T1 T3
SMU-HQ Scores
Writing
Talking

Abstract (if available)

Abstract One explanation for the effectiveness of experimental disclosure is Exposure Theory (Frattaroli, 2006) which hypothesizes that future well-being is the result of the expression of negative emotion. The goal of this study was to assess whether writing or talking facilitated emotional expression and to determine whether increased emotional expression then predicted improved well-being. Given cultural differences between European Americans and East Asian Americans regarding the utility of, and comfort with, verbalizing negative emotional experiences (Kim, 2002), we hypothesized that writing would evoke more negative emotional expression and would therefore lead to greater improvements in well-being. The results indicated that writing did elicit more negative emotional expression than talking. However, use of negative emotion was not related to improvements in well-being.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Direct and indirect predictors of traumatic stress and distress in orphaned survivors of the 1994 Rwandan Tutsi genocide

PDF

Family environment as a moderator of the association between theory of mind and social functioning in people with schizophrenia

PDF

Genetic manipulation of receptor interacting protein (RIP140) uncovers its critical role in the regulation of metabolism, gene expression and insulin signaling in skeletal muscle cells

Asset Metadata

Creator Ng, Lauren Christina (author)

Core Title The link between the expression of negative emotion and future well-being: the role of disclosure modality in an East Asian American population

School College of Letters, Arts and Sciences

Degree Master of Arts

Degree Program Psychology

Publication Date 12/03/2008

Defense Date 10/21/2008

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Asian,culture,experimental disclosure,narrative,negative emotion,OAI-PMH Harvest,talking,trauma,well-being,Writing

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Huey, Stanley J., Jr. (committee chair), Margolin, Gayla (committee member), McArdle, John J. (committee member)

Creator Email lauren.c.ng@gmail.com,laurenng@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m1860

Unique identifier UC1279753

Identifier etd-Ng-2525 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-128346 (legacy record id),usctheses-m1860 (legacy record id)

Legacy Identifier etd-Ng-2525.pdf

Dmrecord 128346

Document Type Thesis

Rights Ng, Lauren Christina

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu