Diversity in bilingual child language acquisition research: A commentary on

Abstract

Kidd and Garcia (2022) report that language acquisition studies are skewed towards monolingual and English-speaking populations. This commentary considers Kidd and Garcia’s arguments in light of our research on mother-preschooler discourse and non-verbal communication in Thai monolingual and Thai-English bilingual children. We discuss lessons learned from testing linguistically diverse children and underscore the importance of research on non-WEIRD populations. We advocate for the inclusion of children who speak understudied languages and those who speak multiple languages in developmental science.

Over 7,000 spoken and signed languages are used in the world today, each unique in its own way. Kidd and Garcia’s finding that only 1.5% of the world’s languages have been examined in the past 45 years of research published in the main child language acquisition journals is a wake-up call to all in the field of language development. Not only does it mean that we have limited knowledge about the acquisition of the other 98.5% of the languages, but that we most likely also have an incomplete and possibly inaccurate understanding of language acquisition in general. For if we only know how 1.5% of something works, do we really know how it works?

Kidd and Garcia (2022) present a compelling case for the need to diversify developmental science in order to build generalizable theories of language acquisition. In this commentary, we build on Kidd and Garcia’s points and put forward two main arguments. The first is to advocate for improving representation of non-WEIRD (Western, Industrialized, Educated, Rich, and Democratic) languages in child language acquisition research. The second is to advocate for improving representation of language acquisition of children who grow up with more than one language.

First, it is important that researchers study a variety of languages instead of focusing on a select few. As bilingualism researchers who study linguistically and culturally diverse populations, we strongly support the expansion of languages covered. Kidd and Garcia (2022, p. 20) identify the Thai language as one that has millions of speakers and the focus on which provides real opportunities for researchers interested in language development. By collecting data from understudied languages such as Thai, a tonal language of the Kra-Dai family with a distinct orthography, it is possible to test whether existing theories primarily based on English and Indo-European languages generalize to other languages. As Kidd and Garcia have posited, research topics have implications for the generalizability of the conclusions. Particularly, aspects of language that are informed by culture, including narrative discourse and non-verbal communication, lend themselves to theory testing and allow for researchers to examine whether specific linguistic phenomena are universal. By studying language acquisition across languages and cultures, we find that narrative patterns of English-speaking children are not always generalizable to Thai-speaking children.

As a case in point, we discuss findings from an ongoing large-scale project that examines dyadic interactions in mothers and their 4-year-olds in Thailand and the United States. Our research shows that the narrative patterns of Thai and English monolingual mothers and their preschoolers differ as a function of cultural background (Rochanavibhata & Marian, 2020, 2021, 2022). Specifically, dyads from the two cultures differ in the elaborateness of their conversations during naturalistic tasks including autobiographical reminiscing (Rochanavibhata & Marian, 2020), book sharing (Rochanavibhata & Marian, 2021), and toy play (Rochanavibhata & Marian, 2022). American English monolingual dyads are more likely to adopt the high-elaborative and story co-constructor styles, characterized by longer narratives and maternal elicitation strategies that encourage child participation. On the other hand, Thai monolingual dyads are more likely to exhibit the low-elaborative and story-audience styles, characterized by concise narratives and maternal scaffolding strategies that establish the mothers as authority figures who model appropriate behaviors for their children. These communicative patterns persist despite controlling for socioeconomic status and reflect culture-specific socialization goals. In individualistic cultures such as the American culture, children are typically treated by adults as equals in order to foster autonomy, individuality, and self-expression. In collectivist cultures such as the Thai culture, children are taught to respect and listen to adults, as well as to conform to social norms. These results demonstrate that patterns of narrative development are not homogenous across cultures and underscore the importance of studying language acquisition in societies where languages other than English are used.

Second, in addition to studying underrepresented languages, it is equally important to expand the scope of research to include children who are learning more than one language, considering that multilingualism is the norm rather than the exception in the world. However, as reported by Kidd and Garcia, existing cross-cultural research has primarily focused on monolinguals. Research on narrative discourse in bilingual mothers and children expands on the currently skewed knowledge of language acquisition by demonstrating that bilingual mothers and their preschoolers communicate differently in their two languages (Rochanavibhata, 2022). When the interactions of Thai-English bilingual mothers and children in Thailand are examined, bilingual mothers use more scaffolding strategies to elicit narrative contributions from their children when speaking in English and use more corrections and commands when speaking in Thai. These differences across languages align with the previously mentioned American and Thai norms associated with English and Thai respectively. To our knowledge, this study is the first to show cross-linguistic differences in communicative styles within the same bilingual speakers, particularly among children as young as preschool-age. These novel findings further strengthen the argument for a more diverse representation of world languages and populations in research.

Kidd and Garcia also find that non-verbal communication is the research topic with the fewest number of articles in the language acquisition literature. Recognizing that the communicative system of a spoken language is comprised of both verbal and non-verbal cues, Rochanavibhata et al. (2022) examine cross-linguistic differences in the gestural patterns of Thai-English bilingual mothers and children. Findings from the study reveal that bilinguals’ gesture use differs depending on the language and the communicative task. Gesture use is influenced by the speaker’s proficiency, language-specific features, and context demands. For example, bilingual mothers and children use more conventional gestures (e.g., nodding to indicate “yes”) in their second and less-fluent language, English, potentially to compensate for their lower proficiency. Bilingual dyads also adapt their gesture use to the activity in which they engage. For instance, mothers and children use more showing gestures during toy play (e.g., holding up a toy in front of their interlocutor) than during reminiscing and book sharing, but use more pointing gestures during book sharing (e.g., extending their index finger towards a picture of a frog) than during reminiscing and toy play. These results suggest that researchers should take a multifaceted approach to studying communication, including how meaning is conveyed non-verbally by individuals from linguistically diverse groups, in order to gain a more complete understanding of language acquisition.

So how do we move the field forward? One way to increase language coverage and representation is incentivizing collaborative research with colleagues in countries where understudied languages are used, as well as incentivizing recruitment, acceptance, and support of students from diverse backgrounds. In turn, these colleagues and students can bring unique perspectives and knowledge about, as well as access to, speakers and signers of different linguistic profiles. Recruitment of underrepresented individuals should be a priority at institutions as part of their diversity, equity, and inclusion mission.

In sum, considering the skewness of existing child language data, challenging the status quo is long overdue. It is critical that researchers correct the misguided “norms” built around exclusion of marginalized populations. We must move away from the common practice of narrowly studying WEIRD groups and move towards inclusivity and diversity that reflect the human condition.