Developing and Using Big Data Archives to Quantify Disfluency and Stuttering in Bilingual Children

Worldwide, bilingualism is the rule rather than the exception, and yet we have surprisingly little research data on the fluency development of bilingual children, and even less information on their potential risk for stuttering. Many variables influence a bilingual child's language, speech, and fluency development (e.g., amount of exposure to each language); controlling these variables in research studies necessitates large numbers of bilingual participants. The frequency and types of typical disfluencies in the speech of young children are also varied. In addition, stuttering is also variable in its presentation, and when we assess bilingual children for the presence of stuttering we are adding yet another layer of complexity. This article reviews research on typical disfluencies in monolingual and bilingual speakers, and how this information might be useful clinically. We provide examples from our laboratory to illustrate how computerized language analysis (CLAN) can be used over time to track the behaviors of research participants. We also present data on the identification of stuttering in bilingual children. We discuss challenges to studying bilingual speakers and how big data initiatives such as TalkBank address these challenges to increase our understanding of bilingual fluency development.