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

Poster E60, Saturday, August 18, 3:00 – 4:45 pm, Room 2000AB

Electrophysiological correlates of first and second language: A within-subjects view of semantics and grammar

Sarah Grey1;1Fordham University

This study addressed two main gaps in electrophysiological research on second language (L2). First, most ERP research on L2 sentence processing has either examined L2 compared to 'L1 literature' (e.g., Batterink & Neville, 2013) or compared the L2 group to a separate L1 group (e.g., Bowden et al., 2013). Using these designs, deviations in L2 from L1 patterns have led to claims about qualitative and quantitative differences in L2 (Morgan-Short, 2014). However, even L1 processing deviates from 'the norm' (e.g., Grey et al., 2017; Pakulak & Neville, 2009) which complicates interpreting between-subjects L1/L2 differences as being due to a 'deviant' L2. This issue can be addressed using within-subjects designs, though little L2 research has done this. A second gap is that most L2 ERP sentence processing research has focused on grammar. There are few reports of L2 sentence-level semantic processing and conclusions from these studies are vague (e.g., Bowden et al., 2013; Ojima et al., 2005). The present study examined participants' L1 and L2 sentence processing, within-subjects, and using EEG/ERP. To add to the small amount of L2 ERP research that has probed sentence-level semantics, the study tested both semantics and grammar. Twenty English L1 speakers (M age = 19.4) who learned Spanish as a L2 were tested. EEG was recorded while participants read correct sentences or sentences with grammar or semantic errors. Participants completed the task in L1 and L2 (different sentences across L1/L2; lists and L1/L2 order counter-balanced across subjects). For grammar, results showed P600s for both L1 and L2. For semantics, results showed no ERP effects in either L1 or L2. Closer inspection of individual-level ERP patterns using a Response Dominance Index (RDI; Tanner & Van Hell, 2014) revealed more nuanced information. RDIs capture individual ERP patterns that may not be reflected in group patterns. RDIs for L1 semantics revealed that about half of participants showed an N400, but the other half showed a P600 in response to semantic errors. This pattern for semantics held true also for L2 - many participants showed N400s, but about half showed P600s. RDIs for L1 grammar showed highly uniform responses - most participants were P600-dominant, and this held true also for L2 grammar. The findings from this within-subjects study indicate that participants employed similar mechanisms for their L1 and L2 grammar (P600s) and that individuals were quite uniform - as shown by the RDI analysis - in employing P600-related mechanisms for both L1 and L2. For L1 and L2 semantics, RDI analysis revealed that some individuals showed the classic response to semantic processing (N400), but about half showed P600s, indicating a qualitatively different processing approach within these individuals. Overall, the findings have theoretical implications regarding how L2 and L1 compare across different domains, particularly regarding the processing routes that individuals take during sentence comprehension.

Topic Area: Multilingualism