Exploring Multilingual Minds: A Study of Laboratory-induced Language Code-switching in Four Languages

Background and Objectives: Bilinguals deal with more than one language at a given time and depending on their linguistic environment, they constantly must decide which language they need to use in each conversation and situation. This requires them to switch between languages frequently, sometimes in a single conversation. This phenomenon is called code-switching which is a continuous stream of words in a different language in a given conversation. This study aims to examine bilingual code-switching across four languages in an experimental setting. Additionally, this study focuses on determining if the nature of a language or the linguistic background of the speakers influence laboratory-induced code-switching.

Methods: Bilingual speakers of English and either Spanish, French, German or Arabic participated in an extensive interview in their heritage language and completed an online survey containing 87 questions about their linguistic background. At a predetermined time during the interview, a monolingual English speaker interrupted the interview and remained in the room for the rest of the interview.

This study was done at the Department of Psychology, Queens University of Charlotte, January 2015 to March 2018.

Each language group included 20 subjects for a total of 80 subjects (39 men, 41 women, age range 18-77 years). The subjects included both college students and members of the community. The interviewer only spoke in one of the heritage languages and instructed the participants to only speak in that heritage language. The interview was videotaped and any occurrence of code-switchers after the interruption was recorded.

Results: This study confirmed that code-switching can be induced under laboratory conditions for more than one heritage language. Consistent with the initial prediction, several of the French, Spanish and Arabic subjects code-switched after the interruption. Those subjects who had learned their heritage language earlier in life or were more fluent in their heritage language were found to be less likely to code-switch in the presence of the interrupter (P = .02; P = .03). Subjects with a strong family heritage linguistic background were also less likely to code-switch (P = .006).

Conclusion: Overall, laboratory induced code-switching was found in three out of four languages. Furthermore, across the four heritage languages, the linguistic background and the level of fluency of bilinguals were predictive of code-switching behaviours. Based on the current findings, it is reasonable to assume that more fluent bilinguals learn their heritage language early from their parents and they are less likely to revert to code-switching in the presence of monolingual speakers.