EMOTIONAL EXPRESSION OF SPANISH-ENGLISH BILINGUALS IN TRAUMATIC

Document Sample
EMOTIONAL EXPRESSION OF SPANISH-ENGLISH BILINGUALS IN TRAUMATIC Powered By Docstoc
					  EMOTIONAL EXPRESSION OF
SPANISH-ENGLISH BILINGUALS IN
   TRAUMATIC NARRATIVES

       Paula Waisman, Ph.D.
The Hispanic Population in the
        United States

 This group makes up approximately
 13% of the population and is comprised
 of 35.3 million people.
 These figures increased by 57.9% from
 1990 to the year 2000.

 (U.S. Bureau of the Census, 2000, 2005).
      Access to Mental Health
            Resources
Hispanics have been consistently identified as
experiencing limited access to proper
assessment and treatment.

 This problem is confounded by variables that
include low SES and acculturation issues.

(Malagady & Zayas, 2001, Prieto, McNeill, Walls & Gomez, 2001).
 Definitions of Bilingualism

Compound bilinguals : those who have
learned two languages in the same context
(e.g.,   English   and    Spanish     learned
simultaneously at home in childhood).

Corresponding words in the two languages
are associated with the same concept.
  Definitions of
  Bilingualism
Coordinate bilinguals : those who have
learned language in different settings (e.g.,
Spanish at home and English at school).

Corresponding words are associated with
different concepts.
(Hakuta, 1992).
FRAME SWITCHING:

  A Cultural Concept
         Frame Switching

Language, like customs or contexts are part
of the cultural cues that are internalized by a
bilingual individual.




               (Hong, Morris, Chiu, & Benet-Martinez, 2000)
        Frame Switching

Frame Switching may occur as a response to
cues such as context (home or school) and
symbols (language) that are psychologically
associated with one culture or another.



              (Hong, Morris, Chiu, & Benet-Martinez, 2000)
  Trauma exposure in the
    Hispanic population
A greater prevalence of PTSD is noted
in the Hispanic population. This could
be explained by:
  Immigration-specific experiences
  Low SES
  Racism and discrimination
  Culture or race-specific vulnerability factors
  (such as somatization and dissociation)
CREATION OF MEMORY:

PTSD is often referenced as a disorder
of memory. How Does Trauma impact
         memory formation?
  Brewin’s Dual Representation
             Model


VAM: Verbally Accessible Memory

SAM: Situationally Accessible Memory

                            ( Brewin, 2003)
          Creation of Memory
                  EPISODIC MEMORY
          (autobiographical with a sense of time and self)


      HIPPOCAMPUS                        AMYGDALA
          (memory)                          (emotion)

Long term storage(association cortex/ frontal lobe)

              Semantic memory (factual memory)
   Brewin’s Dual Representation
              Model
                         THOUGHTS
           VAM          RUMINATION

                     PRIMARY/SECONDARY
                         EMOTIONS



TRAUMA                                 CONTENTS
          MEANING
STIMULI                                   OF
          ANALYSIS                   CONCIOUSNESS


                         FLASHBACKS

                          RELIVING
          SAM              PRIMARY
                          EMOTIONS
                                           (Brewin, 2003)
CREATION OF NARRATIVE:


 How does traumatic memory impact
 narrative formation?
        Trauma Narratives
Narratives are characterized by being:

 Disjointed
 Disfluent
 Incomplete sentences
 Repetition
 Higher use of sensory references
Statement of the Problem:
Use of a second language also might
produce    more      disfluency, more
fragmentation, and less use of complex
sensory description.

 How does use of language interact
with narrative valence (positive versus
traumatic narrative)?
                Design
N = 64
Repeated measures
2 (Language: English/Spanish) x 2 (Valence:
Positive/Trauma)
Variable Counts and word count corrected
analyses
            Procedure
Participants were first screened for
language competency.

Participants were then asked to tell a
personal story about both a traumatic
event and a non-traumatic event, each
in English and Spanish.
      Dependent Variables
Pennebaker & Francis’ Linguistic Inquiry and
Word Count (LIWC)

LIWC is a computerized word count approach
that searches its dictionary of over 2300
words and compares them to those in the
text sample.

The LIWC program has an equivalent Spanish
language dictionary.
        Dependent Variables
The linguistic dimensions included:
  The psychological processes

 Affective or Emotional Processes
 Cognitive processes
 Sensory and Perceptual Processes.
  Dependent Variables:
Measurement of Disfluency
Nonfluency (LIWC): a measure of the
amount of non-words such as “hmm, umm,
ahh”
Disfluency (Subjective rating): a
measure of frequency of self-
interruption with content deviant from
main theme
Dependent Variables: Sensory
          Coding
Coding System for Sensory Words

Includes a focus on five sensory
experiences related to the physical
sensations of: Sound, Sight, Smell,
Touch, and Taste.
     Potential Covariates
The Acculturation Rating Scale for
Mexican Americans-II (ARMSA-II).
             Cuellar, Harris & Jasso (1980)


Demographic Questionnaire items.
DEMOGRAPHICS
      Birthplace
14%                US

                   MEXICO

                   PERU
            38%



48%
        Gender
                  men
          29.7%
                  women



70.3%
                Education
                1.6%   3.1%
        10.9%
                                        Elementary School
                              17.2%
                                        Some High School
15.6%
                                        High School

                                        Technical Degree

                                        Jr college

                                        BA
                                15.6%
                                        MA or higher
        35.9%
RESULTS
Results : Narrative Content
Most Common Topics of Positive Narratives
included:
Work                14.1%
Having Children     21.9%
Vacation            7.8%
Marriage            6.3%
School              23.4%
       Results: Narrative Content
Most Common Topics of Trauma Narratives

 Personal Illness             7.8%
 Accident                     12.5%
 Reproductive Trauma          3.1%
 Death of Family Member       20.3%
 Illness of a Family Member   7.8%
 Family Problems              18.8%
 Domestic Violence            10.9%
            Word Count
Total number of words per narrative

All analyses conducted based on
  word count in original narrative and
  word count in category divided by total
  word count (word count corrected)
      Mean word counts
450
400
350
300
250                       Spanish
200                       English
150
100
 50
 0
      Positive   Trauma
               Results
Hypothesis 1: Predicted that bilinguals will
express greater emotion in their trauma
narratives as measured by the Affective and
Emotional Processes dimension of the LIWC
when evaluated in their first language
(Spanish) versus acquired language (English).


Not supported as a main effect
                 Results
 Main effect for Valence: positive narratives
 showed more affect than trauma narratives
(F (1,63) =13.11, p<. 001) partial eta²=.17

When corrected for word count, both effects
remained significant.

Interaction also emerged
(F(1,63) = 5.60, p < .05) partial eta²=.08
  Interaction of Valence of Narrative
    and Language on Total Affect
 0.05
0.045
 0.04
0.035
 0.03
                                  Spanish
0.025
                                  English
 0.02
0.015
 0.01
0.005
    0
         Positive     Trauma
   Follow-Up to Emotional
   Expression Hypotheses
Significantly more positive affect was
seen in Spanish than in English in the
trauma narratives (F)
Significantly more negative affect was
seen in Spanish than in English in the
positive narratives (F)
  Thus, Spanish narratives were more
  emotionally complex
                Results
Hypothesis 2: Predicted that bilinguals will
express greater cognitive organization in their
trauma narratives as measured by the
Cognitive Processes dimension of the LIWC
when evaluated in their acquired language
(English) versus their first language
(Spanish).

Not supported
                  Results
The results of Hypothesis 2 indicate that bilingual
individuals expressed greater organization of
traumatic narratives in the Spanish language rather
than in the English language as predicted.
(F (1, 63) = 53.14, p < .001) partial eta²=.46

More cognitive process words were also noted in the
trauma versus the positive narrative (F (1, 63) =
44.76, p < .001 ) partial eta²=.42

An interaction emerged in the word count corrected
analyses(F (1, 63) = 13.88, p < .001)
  Interaction of Valence of Narrative and
     Language on Cognitive Processes
0.06

0.05

0.04
                                      Spanish
0.03
                                      English
0.02

0.01

  0
         Positive        Trauma
                  Results
Hypothesis 2b: Predicted that there will be greater
organization or less speech nonfluencies in all
(Spanish and English) positive narratives versus
trauma narratives as measured by the nonfluencies
dictionary of the LIWC.

Not supported, however, an interaction appeared
(F(1,63) =13.02, p<.001) partial eta²=.17
                Results
English trauma narratives contained greater
nonfluencies than Spanish trauma narratives
(Simple effect for Language within Trauma on
word count corrected narratives: F (1, 63) =
12.91, p < .001)

Spanish positive narratives contained
marginally greater nonfluencies than English
positive narratives
  Interaction of Valence of Narrative and
         Language on Nonfluency
0.016
0.014
0.012
 0.01
                                      Spanish
0.008
                                      English
0.006
0.004
0.002
   0
          Positive       Trauma
              Results
The overall main effect did appear in
the subjective disfluency ratings

(F (1,55) = 21.15 p<. 001)
partial eta²=.28
                Results
H3: Predicted that bilinguals will use a higher
frequency of sensory words in their traumatic
narratives versus positive narratives, as
measured by a greater total score of sensory
images as determined by the Coding System
for Sensory Words.


Supported: (F (1, 63) 27.06, p<. 001)
 partial eta²=.30
               Results
There was a higher use of Sensory/Perceptual
words in the Spanish trauma narratives as
compared to the English trauma narratives
(F (1,63) = 60.8, p<. 001)



 Remained in the word count corrected
analysis.
     Mean use of sensory words
16
14
12
10
                            Trauma
8
                            Positive
6
4
2
0
              Sensory
EXPLORATORY FINDINGS
Change in emotional tone from Spanish to
       English trauma narratives
 For those who expressed fear in
 Spanish, 61.8% expressed it in English
 For those who expressed anger in
 Spanish, 60% expressed it in English
 For those who expressed sadness in
 Spanish, 83.8% expressed it in English
     Complexity in trauma narratives: High
            negativity + optimism
25

20

15
                                       Spanish
                                       English
10

5

0
                  Complex
     Complexity in Positive Narratives: High
     Positivity and Expression of one or more
                 Negative Emotions
35
30
25
20                                         Spanish
15                                         English

10
5
0
                Percent Complex
         Trauma Severity
Participants were divided into two groups:
(1) negative experiences, (2) traumatic
events as defined by the traumatic literature
( such as events that included a degree of
fear, helplessness or horror. )

The latter variable was rated by two raters on
a 1-5 scale with a reliability of .73.
       Trauma Severity
Traumatized group (n= 30)

Hypotheses were re-examined within
the traumatized sample.

Greater variance was accounted for in
Cognitive Mechanisms and Negative
Emotions (for the Trauma main effect).
            Conclusions
Spanish narratives were more emotionally
complex than were English narratives
Trauma narratives were more disfluent than
positive narratives
Trauma narratives included more use of
sensory words than did positive narratives,
supporting Brewin’s Dual Representation
Theory
         Clinical Implications
The option to conduct therapy with bilingual individuals in their
first language, especially when dealing with traumatic events,
appears to be a salient choice.

Other options to consider is the role of more non-verbal
treatments for trauma, such as EMDR, Somatic therapies,
Biofeedback, and Movement therapies.

Practitioners should consider the role of diagnostic
overshadowing when working with a traumatized bilingual
population—evaluating the possibility that trauma related
disfluency is not missed or misinterpreted as overall language
difficulty.
   Limitations of the Study
Issues of accessibility to the population

Possible volunteer biases

Fewer than half of the participants
(n=30) were rated to have been
affected by severe trauma
  Limitations of the Study
The research also should be replicated
in other bilingual groups

 This includes other Hispanic groups,
due to possible regional effects on
language use (accent, slang words).
         Future Research
Physiological component of traumatic
narratives

Disfluency

Nonverbal analysis of emotional expression
(e.g.,Facial Action Coding System, Ekman,
Friesen, & Hager, 2002)
       Future Research
More research that focuses on both
cultural elements such as language, and
traumatic phenomena particular to this
population is needed.