Poster B52, Thursday, August 16, 3:05 – 4:50 pm, Room 2000AB
Network selectively responding to bilingual sentence comprehension vary as a function of the L2 age of acquisition
Sandra Gisbert-Muñoz1, Ileana Quiñones1, Manuel Carreiras1,2;1Basque Center on Cognition, Brain, and Language, 20009 Donostia, Spain, 2Ikerbasque, Basque Foundation for Science, 48013 Bilbao, Spain
Bilingual individuals prove that the human brain is capable of acquiring more than one language. However, how different languages are represented in the bilingual brain remains an open question. Previous studies on language processing in bilingual aphasics and neurotypical bilingual individuals have suggested that there is a single brain system underlying language, but that it can modulated by both age of second language acquisition and language proficiency. However, there are also studies that propose differential and functionally independent neuroanatomical substrates for the first and second language. To study if the neural substrate for language is shared in the bilingual brain, three groups of highly proficient Spanish-Basque bilinguals were recruited for our study. Basque and Spanish are both highly transparent languages that share a writing system and most phonemes and, while the sentence canonical order is different (SOV for Basque and SVO for Spanish) both languages accept both orders as correct. The three groups differed only on age of acquisition of Basque: in Group 1 (N=17) they were simultaneous bilinguals; in Group 2 (N=19), early bilinguals that acquired Basque at 3 years of age; and in Group 3 (N=13), late bilinguals with ages of acquisition for Basque ranging from 5 to 26. The three groups performed a grammaticality judgement task in Spanish and Basque in which they were asked to press a button with their right or left hand to indicate if the sentence was correct or if it contained an agreement violation. There were 180 sentences in Basque and 150 in Spanish, with sentence length ranging from 5 to 10 words. For each language, sentences were presented in three separate runs in a word by word paradigm. As predicted, a common network emerged for both Spanish and Basque sentence processing. The conjunction analysis for all correct sentences in both languages showed a bilaterally distributed network, involving the inferior frontal gyrus (triangularis and opercularis), insula, precentral and postcentral gyri, supplementary motor area, middle cingulum, middle and superior temporal gyri and inferior occipital lobe. When comparing Basque correct sentences to Spanish correct sentences, differences arouse between the three groups, with Group 1 showing greater activation than Group 2 and 3 in the calcarine sulcus and superior parietal lobe for Basque sentences and in the left fusiform and the left parahippocampus for Spanish sentences. Activations in the bilateral caudate, cerebellum, thalamus and left putamen were also modulated by Basque age of acquisition, showing a similar pattern in Groups 1 and 2 when compared to Group 3 in Basque sentences; and in the right angular gyrus and the middle and superior temporal gyri for Spanish sentences. These results support the existence of a common neural substrate for language that is modulated by age of acquisition. The differences between the two languages include not only the IFG described in previous literature or regions involved in L1 processing, but also other areas that seem more related to semantic than syntactic processing.
Topic Area: Multilingualism