Poster F62, Tuesday, March 28, 8:00 – 10:00 am, Pacific Concourse
Comprehension of code-mixed sentences in bilingual elders: An event-related potentials (ERP) study
Chia-Hsuan Liao1,2, Shiao-Hui Chan2; 1University of Maryland, 2National Taiwan Normal University
Code-mixing, switching languages back and forth in a sentence, is common in bilingual societies. Existing studies show that processing code-mixed sentences induces processing costs, which can be modulated by factors such as switching direction and language use (Moreno et al., 2008). However, previous studies mainly focused on performance in younger adults, leaving the performance in bilingual elders relatively untouched. To fill in the gap, the current ERP study recruited 24 healthy native Mandarin-Taiwanese bilinguals (mean age: 67), proficient in both languages but more dominant in Mandarin as assessed by pre-tests. Four sentential conditions were manipulated: non-switched Mandarin (MM), non-switched Taiwanese (TT), switched from Mandarin to Taiwanese (MT) and switched from Taiwanese to Mandarin (TM). The subjects had to listen to the stimuli and perform a word recognition task in 20% of the trials. The results revealed that, switching into a less dominant language (MT vs. MM) elicited N200, N400, and a sustained negativity, while switching into a more dominant one (TM vs. TT) did not. This finding is comparable to what we found in younger adults (Liao & Chan, 2016), indicating that language use might affect sentence processing in both younger and older adults, with switching in the dominant-to-nondominant direction taxing more cognitive resources than the other direction. However, while the LPC was present in younger adults regardless of switching directions, it was absent in the elderly, suggesting that the elderly might not predict an upcoming sentence structure as much as younger adults do and thus reanalysis is less prominent.
Topic Area: LANGUAGE: Development & aging