Poster E59, Saturday, August 18, 3:00 – 4:45 pm, Room 2000AB

Using the EAR to Track Bilingual Language Use

Alessandra Macbeth1, Michelle Bruni1, Emily N. Mech1, Justin T. Sarkis1, Alexander Karan1, Megan L. Robbins1, Christine Chiarello1;1University of California, Riverside

There is great diversity in the daily language experiences of bilinguals. Investigations of the cognitive and neural correlates of bilingualism rely on self-report language history questionnaires (LHQs) to estimate the extent to which each language is used in various situations, and with various conversational partners, the frequency of language switching, etc. Because it is difficult to assess the validity of such questionnaire responses, unwanted variance may be introduced in our investigations of the cognitive and neural bases of diverse bilingual experiences. We will report on real world language use in a linguistically diverse sample of university students by employing an experience sampling method using the EAR (electronically activated recorder [1]). This initial investigation had two objectives: (1) to document the variety and frequency of multilingual language use in a linguistically diverse sample; and (2) to assess the correspondence between self-reported and actual language use. The findings will provide the foundation for a broader research program to explore the relationship between daily bilingual language experiences and their cognitive and neural correlates. UCR (mainly English dominant) students wore the EAR (cell phone with the EAR app) for 4 consecutive days as they went about their daily lives, and completed the LHQ [2] upon returning to the lab. The app makes 40-second auditory recordings, every 12 minutes during waking hours. Participants pause the device if they wish to have a private conversation, but are unaware of when the device is recording. Each participant’s speech was transcribed and coded to quantify the frequency of use of various languages, the settings in which the speech occurred, the frequency of language switching, etc. We also coded the language used by the participant’s conversational partner(s). Participant speech recordings were acquired at home (50%), in a public place (28%), in transit (19.2%), or in class (2.8%). Initial data (N = 15) indicates that our bilingual participants, and their conversational partners, used a language other than English (Spanish, Burmese, German, Korean, Mandarin, Vietnamese) approximately 10% of the time, although this differed for self-reported English dominant vs non-English dominant participants (4% vs 32%). Self-rated proficiency was somewhat greater for English than for another language (6.95 vs 5.86; 7-point scale). There was no relationship between self-rated proficiency and the amount of time the bilinguals actually spoke either of their languages (r’s < .16). However, self-report of the amount of non-English speech did predict the extent of participant (r = .55) and partner (r = .62) actual use of the language. Our bilingual participants only engaged in language switching 3% of the time, and this was only weakly related to their self-report of language switching frequency (r = .26). Our preliminary findings suggest that the validity of self-report bilingual questionnaires is variable. More in-depth analyses of the transcript data should reveal additional nuances of bilingual language experiences. The EAR method appears to be a promising tool for investigating bilingual language use “in the wild.”

Topic Area: Multilingualism