Poster E61, Saturday, August 18, 3:00 – 4:45 pm, Room 2000AB

Neurobiological signatures of L2 proficiency

Henry Brice1, W. Einar Mencl2, Stephen J. Frost2, Atira S. Bick1,4, Jay G. Rueckl2,3, Kenneth R. Pugh2,3, Ram Frost1,2;1The Hebrew University of Jerusalem, 2Haskins Laboratories, 3University of Connecticut, 4Hadassah Hebrew University Medical Center

This study presents data from the first epoch of a large-scale longitudinal neuroimaging investigation of second language (L2) acquisition. The study tracks parallel cohorts of English L1 speakers immigrating to Israel learning Hebrew as an L2, and Hebrew L1 speakers immigrating to the United States and learning English as an L2. The parallel structure of this design allowed us to pull apart language effects (English vs. Hebrew) and proficiency effects (L1 vs. L2). We examined the processing of words in both L1 and L2, in both auditory and visual modalities. Subjects performed animacy judgements on printed and spoken words. Following Perfetti and colleagues’ (2007) theoretical framework, we considered the neural changes associated with acquisition of L2 reading proficiency, as a process of accommodation, in which L2 circuits diverge from reading circuitry in L1, or assimilation, a process of convergence of L1 and L2 processing. Indeed, evidence from Chinese and English bilinguals has shown that greater L2 reading proficiency is associated with assimilation-greater convergence in the processing of L1 and L2 stimuli (Cao, Tao, Liu, Perfetti, & Booth, 2013). Similarly, Preston et al. (2015) have shown that a convergence of print and speech processing in L1 predicts reading ability in early readers. The measure of print and speech convergence across a network of reading areas in the brain has also been shown to be relatively invariant to cross linguistic differences (Rueckl et al., 2015), but has not been looked at in L2. In our study, the pattern of print/speech convergence across the parallel cohorts of L2 learners shows differences across the different regions of the reading network, with greater convergence for L2 in the left inferior frontal gyrus, a region tied to effortful language processing, but greater convergence for L1 in left-hemisphere posterior regions of the inferior parietal cortex and the fusiform gyrus, both tied to skilled reading processes. This is the first time that print/speech convergence has been examined in L2. Our findings suggest then that there is a shift in the weight of processing between L1 and L2. Whereas for L1 print/speech convergence is greater in posterior regions of the brain and is driven primarily by automatic skilled reading, for L2 print/speech convergence is greater in frontal regions of the brain, due to effortful processing in L2.

Topic Area: Multilingualism