Kiran, S. 1 , Pena, E. 2 , Bedore, L. 2 & Sheng, L. 2
1 Boston University Speech Language and Hearing Sciences
2 University of Texas at Austin Communication Sciences and Disorders
Introduction: This project examines the validity of a language use questionnaire to be used across language combinations and across clinical populations (bilingual SLI, bilingual aphasia). The main goal of the project is to identify specific predictors of language proficiency from the analysis of the relationship between language use history and objective performance based measures. Specifically, the relationship between specific aspects of the language use questionnaire (reading, spoken production) is measured with comparable lexical access measures.
Methods: 137 participants across seven language combinations (Russian-English, Spanish-English, Turkish-English, Kannada-English, Hindi-English, Mandarin-English) completed a category generation task using three categories (FOOD, ANIMALS, CLOTHING) in their two languages. The order of presentation of language and category was counterbalanced across participants to control for order effects. The number of responses produced within two minutes was tabulated and we measured the total number of different words (NDW) produced, the number of cognates, the number of translation equivalents for each category across the two language. Each participant also completed a multi-dimensional language use questionnaire that includes questions focusing on the manner and time of acquisition for both languages (exposure, education, parental language background) as well as use patterns over time with an emphasis on use and proficiency. Further, participants completed an estimation of time (calculated in percentage) spent in each language during a typical weekday and weekend.
Results: Across language combinations, participants tended to produce different numbers of items (NDW) based on the language use and proficiency (Wilks lambda=.41305, F(10, 242)=13.454, p=0.00000). For instance, Hindi-English and Spanish English bilinguals provided more NDW in English than in their other language, whereas Russian-English bilinguals and Mandarin English bilinguals on average produced more NDW in their native language than in English. These results highlight the importance of understanding patterns of language acquisition and use in interpreting bilingual lexical access.