Poster #112 - An Investigation of the Bilingual Advantage in a Novel Population of Refugee Children in Canada.

• In Event: 1-005 - Poster Session 01
  In Poster Session: PS 01 Section - Language, Communication

Integrative Statement

Previous research has shown that bilinguals have stronger executive functions (EFs) than monolinguals, an effect called the “bilingual advantage” (e.g., Bialystok, 1999). However, new research challenges the existence of the bilingual advantage, with some researchers suggesting demonstrated EF differences are due to voluntary circumstances leading an individual to become bilingual, such as taking language classes or choosing to immigrate (e.g., Morton & Harper, 2007).

We examine the presence of the bilingual advantage in a novel population of 7- to 12-year-old, primarily refugee children (n = 43) who are Arabic-dominant bilinguals, enrolled in an English as a Second Language program (see Table 1.). Participants were recruited and tested in local schools. This population has not previously been studied and is of particular interest because they are becoming bilingual due to circumstances beyond their control rather than by choice. We compare the EF performance of these children, as measured by the flanker task and dimensional card change sort task (DCCS) from the NIH Toolbox: Cognitive Battery (Bauer & Zelazo, 2014), to an age- and gender-matched convenience sample, tested in the lab and from the same geographic area (n = 43). Participants chose whether to hear the flanker and DCCS task instructions in English or Arabic, and the tasks themselves were non-verbal. We also collected demographic data from participants’ parents, including parental education and family income which was used to create a socioeconomic status (SES) composite score.

The NIH Toolbox produces normed and standardized age-corrected scores for flanker and DCCS performance that account for accuracy and reaction time (RT; NIH Toolbox, 2012). A planned comparison of age-corrected flanker and DCCS scores between monolingual and bilingual participants showed that, on both measures, monolinguals significantly outperformed bilinguals (ps < .001).

One-way ANOVAs were used to compare the accuracy and RT scores between groups for each task. For both tasks, there was no difference in accuracy (ps < .223) but a significant difference in RT on executive trials existed (ps < .001), with bilinguals responding more slowly than monolinguals. These results suggest the observed differences in age-corrected EF scores are due to RT differences.

To determine whether performance differences could be accounted for by SES, we examined SES and Language Group as predictors of flanker and DCCS age-corrected scores using a hierarchical regression analysis with bootstrapping. SES was a significant predictor, explaining 23.3%, p < .001, of variance in the flanker age-corrected score and 28.3%, p < .001, of the variance for the DCCS age-corrected score (see Table 2.). But interestingly, Language Group explained a further 13.9% and 13.8% of unique variance on each task respectively.

These results suggest that the bilingual advantage may not exist in language groups of dominant bilinguals who do not choose to become bilingual, such as refugee children, and that the development of executive functions of refugee children may be vulnerable to delay. Future directions for research include longitudinal studies exploring the direction of effects between second language learning and EF development. Implications for teachers of English as a Second Language will be discussed.

Authors

• Serena McDiarmid, University of Waterloo
  Presenting Author

• Heather Henderson, Department of Psychology, University of Waterloo
  Non-Presenting Author