Sing-Kiat Ting 1
Effect of language and culture on bilingual Chinese verbal expression of depression1
Rachel Sing-Kiat Ting, PhD & Alvin Dueck, PhD
Fuller Graduate School of Psychology, Pasadena, CA, U.S.A.
Mailing Address: 4343 Ocean View Blvd Apt. #177, Montrose, CA 91020, USA
1. Brief Overview of Literatures
Language can be a powerful means of activating associated cultural constructs, judgments,
self-assessments, and memories (Bond, 1983; Trafimow et al., 1997). Recent research has shown
that bilinguals’ use of language may influence cognitive styles: when speaking a language
associated with a more individualistic (collective) culture, bilinguals produce more
individualistic (collective) narratives (Marian & Kaushanskaya, 2004; Ross, Xun, & Wilson,
2002). This language priming effect has been advocated as a promising strategy in cultural
psychology (Oyserman, Coon & Kemmelmeier, 2002). It is reasonable, therefore, to assume that
switching between spoken languages will cause a parallel change in the cultural framework used
by a bilingual person. In other words, the use of the Chinese language may increase the
accessibility of traditional Chinese cultural knowledge and beliefs when this language is spoken.
Conversely, the use of English may increase the accessibility to Western cultural constructs.
Do bilingual Chinese express depressive emotion differently when using different languages?
It has been argued that Westerners conceive of depression as an intra-psychic, existential
experience, whereas in Chinese and many other non-Western societies, it is most frequently
experienced somatically and in terms of interpersonal dysfunction (Kleinman, 1985; Ying et al.,
2000). Despite a long history and interest in clinically depressed Chinese, there is a scarcity of
research that delves into verbal expression of depressive emotion in the general Chinese
population. With regard to emotional experience, some empirical findings suggest that Chinese
use more somatic expressions and social words than Americans (Tsai et al., 2004). However, past
studies are limited in terms of their generalizability due to disparities in measurement tools (BDI,
CES-D, Hamilton etc.), geography of the samples (U.S.A., China, Hong Kong, and Taiwan),
response types (interview versus self-report measurement), language conditions (Chinese versus
English), language proficiency (ability in speaking versus writing), and pool of participants
(clinical versus non-clinical).
To build upon previous studies on bilingualism and depression among the Chinese, this
study focused on Chinese who can speak, read, and write in both Chinese and in English, and
examines how they respond in English or Chinese to depressive images and stories. A pool of
Caucasian Americans was recruited to form the comparison group. Culture and language were
controlled for in this study by means of several preconditions distinctive to this study: (a)
randomly assigning bilingual participants into different language conditions to control for
acculturation, citizenship, country of birth, and other cultural variables; (b) requiring bilingual
Chinese to have adequate language proficiency in both languages, by adopting a narrower
Presented at the First World Congress of Cross-cultural Psychiatry, Beijing, China, Sept.
Sing-Kiat Ting 2
definition of bilinguality; (c) controlling the degree of acculturation of bilingual Chinese with an
acculturation scale; (d) utilizing both non-verbal (images) and verbal (stories) stimuli to evoke
participants’ emotional responses; (e) integrating both quantitative and qualitative approaches in
gathering Chinese viewpoints on depression; and (f) employing computer software (LSA) to
analyze the language of written responses.
The present study was the first attempt to examine depressive feelings using a language
priming effect. The aim was to deepen our understanding of the language of depression through a
combination of qualitative and quantitative approaches.
2. Research Hypothesis
This study hypothesized that the use of the Chinese language may increase the accessibility
of traditional Chinese cultural knowledge and beliefs, including expressions of negative emotion.
Existing literature shows that the Chinese do not view depression the way Westerners do. Instead,
they associate depression more with interpersonal issues, and bodily symptoms. Therefore, when
asked to describe the depressive pictures, bilingual Chinese who responded in Chinese were
expected to refer more to interpersonal difficulties, somatization and to express less depression
language than bilingual Chinese who responded in English and the Caucasian American group.
Two hypotheses were generated for two different studies reported in this paper. The former
is a study on general depressive emotion expression; whereas the latter is a study on the
conceptualization of clinical depression. First, when asked to describe a set of depressive pictures,
bicultural Chinese who responded in Chinese were expected to be more likely to project
interpersonal difficulties and somatization, and less likely to project intrapsychic clinical
depression than the bicultural Chinese who responded in English and the Caucasian Americans
who served as a control group. Second, when asked to conceptualize cases of depression given 2
short vignettes, a similar pattern of responses was expected.
3. Instruments and Measures
This study utilized both projective (images) and written (vignettes) forms of stimuli to
access depressive languages from participants, while comparing the effects of language and
culture. The administration protocol consisted of seven parts (which were prepared in both
Chinese and English versions), beginning first with demographic information, open-ended
self-description, Pre-Post bi-directional Likert Mood Scale, and four emotional Chinese images.
Participants were asked to write a story on each of the four pictures, using their imagination to
focus on the feelings and thoughts of the main character. In the next section, there were two
vignettes portraying Chinese individuals exhibiting depressive symptoms. Participants were
asked to answer four accompanying questions that attempted to solicit their thoughts and feelings
regarding the scenario and the characters. In the final section of the questionnaire, participants
were given the Beck Depression Inventory-II and the Suinn-Lew Self-Identity Scale (only for
Chinese participants). On average, it required approximately 45-60 minutes to complete the
4. Subject Recruitment and Random Assignment
Participants with the following characteristics were recruited through the snowballing
technique. The participants consisted of two samples: first, bilingual Chinese: Chinese-born
participants who displayed equal proficiency in speaking, reading and writing Chinese and
English and who currently resided in the United States; and second, a comparison group of
Caucasian Americans with English as their primary language and currently residing in the United
Sing-Kiat Ting 3
States. There were 74 self-reported bilingual Chinese who participated in this study and who
were verified as bilinguals from the Language Proficiency Scale. Among these participants, 36
of them were randomly assigned to answer the questionnaire in Chinese translation, whereas 38
of them were randomly assigned to answer the questionnaire in English translation. There were
no statistically significant differences in language proficiency between these two groups. Both
groups were highly proficient in reading, writing and speaking Chinese and English. There were
37 Caucasian American who were only proficient in English recruited to answer the
questionnaire in English.
In terms of gender, there were 32 males and 41 females in the bilingual Chinese group, and
20 males and 17 females in the Caucasian group. With respect to demographic indicators, the two
bilingual Chinese groups did not differ significantly from each other. The mean age for the two
Chinese groups was approximately 32 and 31 years respectively. The mean age for the Caucasian
participants was approximately 28 years. Age did not differ significantly across the three
conditions. With the exception of one participant who was born in the U.S., all Chinese
participants were born in Asia—Taiwan (33), Hong Kong (6), Mainland China (28), Malaysia (4),
and Singapore (2). Their first language was either Mandarin Chinese or a local Chinese dialect
and their average length of stay in the U.S. was approximately 7 years (range: 1-28 years).
As anticipated, participants across all three groups experienced a change in mood after
viewing the depressive pictures, and were affected significantly in the negative direction (p
< .001). However, to make sure that extent of mood change was not different among these three
groups, a one-way ANOVA was conducted. ANOVA results revealed no significant differences
across groups, in terms of their mood change. This suggests that all participants were
emotionally involved with the depressive pictures to a similar degree. These results also
demonstrate the construct validity of the depressive pictures selected. The total BDI score did not
differ across groups, implying that the variance of emotional projection was not confounded by
individual symptoms of depression. Furthermore, there were no significant differences between
Group 1 (Bilingual Chinese in Chinese condition) and Group 2 (Bilingual Chinese in English
condition) in terms of their degree of acculturation (SL-ASIA). The average degree of
acculturation for both groups was below 3 on a 5-point Likert scale (3 is equal to a bicultural
identity, the lower the score, the less acculturated to American culture an individual would be).
These results support the randomness of assignment—bilingual Chinese were evenly distributed
into two different language conditions so that the degree of acculturation and other cultural
variables were controlled for.
5. Linguistic Analysis
Two trained bilingual Chinese individuals, who were proficient in both Chinese and English
writing, and familiar with American culture, assisted in translating the Chinese responses.
Together, they arrived at a consensus for the translation glossary. Given the sensitivity towards
emotional language, the aesthetic-poetic translation was employed to preserve the sentimental
content of the original language (Chinese). Chinese writings are renowned for their rich use of
idioms and traditional sayings in expressing thoughts and feelings. To prevent the loss of such
depth and intensity, the translators paid particular attention to the context of the sentences, and
translated the sayings into meaningful English phrases as opposed to their literal meaning.
After translating all the Chinese written responses into English, the transcripts were
analyzed quantitatively with a computational language program, Latent Semantic Analysis (LSA).
LSA was first applied to cognitive psychology by computer programmers, linguists and
psychologists at the University of Colorado. It has been proven to be a useful analytical tool that
Sing-Kiat Ting 4
simulates human cognitive processing of vocabulary, word sorting, category judgments,
estimation of passage coherence, and even the quality of knowledge in an essay (Landauer et al.,
1998; Kintsch & Bowles, 2002). It is a powerful tool to analyze qualitative data with rigorous
quantitative analysis. The present study is the first attempt of utilizing LSA in cross-cultural
research to compare subjective textual responses using objective semantic analysis.
LSA was used to analyze similarities between verbal depressive expressions and three
different dimensions of clinical syndromes from the Symptom Checklist-90-R (Derogatis, 2000).
This study employed the items from three factors of the SCL-90-R: Somatization (SOM),
Interpersonal Sensitivity (I-S), and Depression (DEP). The Somatization dimension reflects
distress arising from perceptions of bodily dysfunction. It is comprised of complaints of
discomfort in the autonomic system and the gross musculature and additional somatic
equivalents of anxiety. The Interpersonal Sensitivity dimension focuses on feelings of inadequacy
and inferiority in the context of relationships. Marked discomfort during social interactions, such
as self-doubt and self-deprecation, is highlighted in this scale. Lastly, symptoms of the
Depression dimension reflect the scope of clinical depression, which includes dysphoric mood,
lack of interest or motivation, feelings of hopelessness, suicidal thoughts, and so on. These three
dimensions served as “proxies” for the construct of depression in both cultures.
To test the hypotheses, correlational analyses were employed because it was assumed that
there was a rank order of differences across cultural conditions. Instead of an analysis of mean
difference, researchers were more interested in examining the “direction” of differences across
conditions. If language is a powerful mean to prime cultural values, then the responses from
bilingual Chinese assigned to the English condition would fall between the continuum of Chinese
culture and American culture. To test this hypothesis, the group was coded and ranked in the
order that reflected their direction on the continuum. Hence the group variable was expected to
correlate positively with the variables Somatization and Interpersonal, but correlate negatively
6.1 Findings from Study 1
In this part of the study, participants were asked to write stories in response to 4 depressive
pictures. There was a significant negative correlation (r = -.246, p < .01) between Group ranking
and Interpersonal language on the SCL-90-R. This result supported the hypothesis that bilingual
Chinese who responded in Chinese (Group 1) would use interpersonal language (e.g., “Feeling
critical of others,” “Feeling that people are unfriendly”) more often than bilingual Chinese who
responded in English (Group 2), and Caucasian Americans (Group 3), in projecting their
depressive feelings. However, the assumption that Chinese responding in Chinese (Group 1)
would use more language of somatization and less language of intrapsychic depression was not
6.2 Findings on Study 2
In this part of study, participants were asked to respond to the case vignettes of depression.
The group was coded and ranked in the order that reflected their direction on the continuum of
each dependent variable. There was a significant positive correlation (r = .208, p < .05) between
Group ranking and Depression language in SCL-90-R. These results supported the assumption
that the Caucasian American group would use more intrapsychic depressive language (e.g.,
“Feeling no interest in things,” “Feeling hopeless about the future”) than bilingual Chinese who
respond in English and Chinese. However, the correlations between group and interpersonal
difficulties and somatization were not significant. In another word, Bilingual Chinese responding
Sing-Kiat Ting 5
in Chinese are less likely to view depressive vignettes from a intrapsychic perspective than
In summary, the major findings were that: 1) bilingual-Chinese who responded in Chinese
used more interpersonally sensitive language in describing depressive pictures than did their
counterparts; 2) bilingual-Chinese who responded in English and the Caucasian Americans when
describing the distressed cases, used terminology more typical of Western views of depression
than did Bilingual Chinese who responded in Chinese.
7. Implications for cultural psychology and multicultural therapy
7.1. Chinese depression
The results indicate that the choice of language has a significant priming effect on how
depressive emotion in bilingual Chinese individuals is expressed. As expected, Caucasian
Americans responded linguistically in ways similar to the Bicultural Chinese who responded in
English rather than in Chinese. In Chinese culture, where the individual’s identity is fostered
through interdependence with their community, it is not surprising that emotions are
fundamentally influenced by social norms and relationships with other individuals in the
community, and to a large extent can only be described by factors external to the individual. In
contrast, in Western culture where the independent self is valued, emotion often becomes a
personal, cognitive matter and is defined primarily by the individual’s interpretation.
In the past, researchers have tended to generalize their findings from studies of depressed
Chinese patients to Chinese populations generally, assuming that on the whole, individuals in
Chinese culture “somatize” emotional expression. The results of this study do not support such
an assumption. Instead the results suggest that emotion words used by Chinese are embedded in
intersubjective and relational semantic networks which are different from those of individualistic
Caucasian American semantic networks.
In addition, there is heterogeneity in the definition of somatization by researchers, and
sometimes the definition is overly inclusive, ranging from neurasthenic somatic complaints (e.g.,
Kleinman, 1985) to sensory and perceptual processes (Tsai et al., 2004). For instance, this study
adopts a working definition of somatization as the one measured by the SCL-90-R. This raises
the issue whether the term “somatization” has scientific value if researchers cannot come to a
consensus about its meaning. In the future, it will be crucial to clarify the definition of
“somatization” in order to reduce the confusion surrounding this term, since it varies from
discipline to discipline when used by psychiatrists, anthropologists, cultural psychologists, or
7.2. Therapy with Chinese populations
Given the fact of globalization and immigration, it is obvious that there will be an increased
number of bilingual and bicultural citizens around the world. On the other hand, it is worthwhile
to investigate how the Chinese community in America may perceive and respond to negative life
events differently from the dominant American society considering the many traumatic
experienced by Chinese in last century. This study used a sample of bilingual Chinese in America
to examine the power of language in emotional expression. Our understanding of the nature of
depression can be heavily influenced by the culture from which we view it. The implications are
wide-ranging. Clinically, we advocate an approach to therapy that is culturally and linguistically
sensitive to the semantic and emotional frameworks of what is understood to be depression. A
culturally competent therapist may not need to speak the same language as the client, but will
need to be able to identify the cultural and linguistic variations of bicultural clients. Health
Sing-Kiat Ting 6
practitioners should be aware of major linguistic and cultural contrasts in the course of their
interactions with Chinese clients. Also, how we develop practices and research methodologies in
light of this fact will determine the effectiveness of multicultural clinical practices in the future.
The present study revealed that one of the “markers” for depression diagnosis is interpersonal
depressive language, such as “feeling that people are unfriendly.” Therefore it is critical for
clinicians to listen for emotions embedded in relationships when treating Chinese clients. This
also suggests that therapists should be more in tune with Chinese clients’ “interpersonal
language” rather than focusing solely on their internal “emotional language” or “symptom
Bond, M. H. (1983). How language variation affects inter-cultural differentiation of values by
Hong Kong bilinguals. Journal of Language and Social Psychology, 2, 57-66.
Derogatis, L. R. (2000). SCL-90-R. In A. E. Kazdin (Ed.), Encyclopedia of psychology (Vol. 7,
pp. 192-193). Washington, DC: American Psychological Association.
Kintsch, W., & Bowles, A. R. (2002). Metaphor comprehension: What makes a metaphor
difficult to understand? Metaphor & Symbol, 17, 249-263.
Kleinman, A., & Kleinman, J. (1985). Somatization: The interconnections in Chinese society
among culture, depressive experiences and the meaning of pain. In A. Kleinman & B.
Good (Eds.), Culture and depression: Studies in the anthropology and cross-cultural
psychiatry of affect and disorder (pp. 429-490). Berkeley, CA: University of California
Landauer, T. K., Foltz, P. W., & Laham, D. (1998). Introduction to Latent Semantic Analysis.
Discourse Processes, 25, 259-284.
Marian, V., & Kaushanskaya, M. (2004). Self-construal and emotion in bicultural bilinguals.
Journal of Memory & Language, 51, 190-201.
Oyserman, D., Kemmelmeier, M., & Coon, H. M. (2002b). Cultural psychology, a new look:
Reply to Bond (2002), Fiske (2002), Kitayama (2002), and Miller (2002). Psychological
Bulletin, 128, 110-117.
Ross, M., Xun, W. Q. E., & Wilson, A. E. (2002). Language and the bicultural self. Personality
and Social Psychology Bulletin, 28, 1040-1050.
Trafimow, D., Silverman, E. S., Fan, R. M., & Law, J. S. F. (1997). The effects of language and
priming on the relative accessibility of the private self and the collective self. Journal of
Cross-Cultural Psychology, 28, 107-123.
Tsai, J. L., Simeonova, D. I., & Watanabe, J. T. (2004). Somatic and social: Chinese Americans
talk about emotion. Personality and Social Psychology Bulletin, 30, 1226-1238.
Ying, Y.-W., Lee, P. A., Tsai, J. L., Yeh, Y.-Y., & Huang, J. S. (2000). The conception of
depression in Chinese American college students. Cultural Diversity and Ethnic Minority
Psychology, 6, 183-195.