You are viewing the SNL 2018 Archive Website. For the latest information, see the Current Website.

Poster C58, Friday, August 17, 10:30 am – 12:15 pm, Room 2000AB

Bilingualism modulates L1 word processing in the developing brain

Olga Kepinska1, Myriam Oliver1, Zhichao Xia1,2, Rebecca Marks3, Leo Zekelman1, Roeland Hancock1,4, Stephanie Haft1, Priscilla Duong1,5, Yuuko Uchikoshi Tonkovich6, Ioulia Kovelman3, Fumiko Hoeft1,4,7;1University of California, San Francisco, 2Beijing Normal University, 3University of Michigan, 4University of Connecticut, 5Palo Alto University, 6University of California, Davis, 7Haskins Laboratories

While there is increasing consensus that the core neural phenotype of language is more universal than once thought (Rueckl et al., 2015), it is still considered gold-standard that language development requires additional neural resources to be recruited. In bilinguals, evidence suggests that second language (L2) is processed using the first language’s (L1) network (Tan et al., 2003). Recent findings also show an interaction between the L1 and L2, suggesting that L2 acquisition also influences L1 processing (Costa & Sebastián-Gallés, 2014). With the majority of studies on bilingualism and L2 acquisition concentrating on L2 representations and processing in the brain, here, we investigated whether experience with additional language(s) in childhood exerts influence on the processing of the L1. Typically developing kindergartners were recruited from public schools at the San Francisco Unified School District and participated in the study. During acquisition of functional MRI (fMRI) using a gradient-echo echo-planar pulse sequence, children performed an auditory L1 (English) word match task blocked alternating with periods of rest. In the task upon hearing two words, they indicated with a button press whether the words were the same or different. Based on task accuracy (>60%) and motion (<30% of time-points removed due to movement), fMRI data from 57 children (23 males, 5 left-handed, 1 ambidextrous) were subjected to further analyses. These kindergarteners were 5-6 years old (M=5.90, SD=0.36), and had varying levels of L2 experience, operationalized as the numbers of years of exposure to any language but L1, which ranged between 0 to 6.77 years (M=2.67, SD=2.45). We performed two whole-brain fMRI analyses of the data acquired during the L1 auditory task: (1) with years of exposure to L2 as a covariate of interest, and controlling for children’s age, gender and handedness; and (2) with additional covariates representing children’s socioeconomic status (SES), task performance and receptive English language skills, in order to test if the L2 exposure modulates L1 brain activity patterns above and beyond these variables of no-interest. The levels of brain activity during the task in L1 were modulated by L2 exposure: a cluster localized in the superior parietal lobule/precuneous showed higher levels of activation in children with longer L2 exposure, both when SES, task accuracy and L1 language skills were accounted for, and when they were not (Zmax=3.48, p=0.004, and Zmax=3.35, p=0.001, respectively, corrected for multiple comparisons using Gaussian Random Field theory, Z>2.0, cluster-wise p<0.005). Superior parietal lobule has been previously shown to be involved in (among others) reasoning, working memory, and attention processes (Wang et al., 2015). Our results suggest L2 exposure may impact L1 auditory word-related brain activation in a very young sample, and are in line with previous evidence showing that bilinguals recruit higher cognitive control regions when performing a language task (Bialystok, 2001). Our findings demonstrate the brain’s extraordinary ability to change in response to environmental experiences and point to the conclusion that the length of language exposure is a crucial factor that can affect the brain representation of even their native language.

Topic Area: Multilingualism