Effects of second language proficiency on neural organization for syntactic processing indexed by ERPs

Pakulak, E. & Neville, H.

University of Oregon

Several ERP and neuroimaging studies have reported that neural organization for syntactic processing is altered by delays in age of acquisition (AOA) as short as 4-6 years (e.g.,Wartenburger et al., 2003; Weber-Fox & Neville, 1996). However, as such delays in acquisition are typically associated with lower language proficiency (e.g., Johnson & Newport, 1989), it is difficult to assess whether differences in AOA or proficiency lead to these effects. In a previous ERP study of syntactic processing using auditorily presented phrase structure violations, we examined late learners of English matched for proficiency with a group of monolingual native speakers. Findings suggested that late learners recruited different neural mechanisms to achieve a level of proficiency similar to native speakers: violations elicited both an early anterior negativity and a posterior positivity (P600) in the monolingual group, but only a P600 in the bilingual group (Pakulak & Neville, in press). These results were supported by findings from a fMRI study using the same paradigm (Pakulak & Neville, 2009). Here we continue this investigation by examining two groups of late learners who vary in second language proficiency. Phrase structure violations elicited a significant P600 effect in both groups, suggesting that the recruitment of more controlled processes is less sensitive to proficiency differences. In the higher proficiency group violations also elicited a significant bilateral negativity over anterior sites which was maximal over right hemisphere medial sites. Consistent with some previous evidence (e.g., Weber-Fox & Neville, 1996), these findings suggest that some neural resources indexed by anterior negativity effects may be recruited by late learners with increases in proficiency. However, distributional differences suggest that these resources may differ from those recruited by native speakers. Future studies examining late learner groups with wider ranges of second language proficiency will shed further light on this issue.