L2 processing of filled gaps
Non-native brain activity not modulated by proficiency and working memory
This paper investigates how late L2 learners resolve filler-gap dependencies (FGD) in real-time and how proficiency and working memory (WM) modulate their brain responses in an event-related potential (ERP) experiment. A group of intermediate to highly proficient Mandarin Chinese learners of English listened to sentences such as “The zebra that the hippo kissed *the camel on the nose ran far away,” in which the extra noun phrase “the camel” created a ‘filled-gap’ effect. The results show that although L2 behavioral responses are comparable to native speakers and are positively correlated with proficiency and WM span, the brain responses to the filled gap are qualitatively different. Importantly, L2 processing patterns did not become more nativelike with higher proficiency levels or greater WM capacity. Specifically, while the native speakers exhibited a P600 typically observed for syntactic violations and repair, the L2 group produced a prefrontal-central positivity. Similar ERPs have previously been reported to reflect domain-general attentional and non-structural-based processes, suggesting that the L2 group has a reduced sensitivity to structural requirements for gap positing in the online resolution of FGDs. Our findings are discussed in light of various proposals accounting for L1-L2 processing differences, including the Shallow Structure Hypothesis.
- 1.1L2 processing of FGD: Is structural information underused in gap positing?
- 1.2ERP studies on the filled-gap effect and the present study
- 2.2Paper-and-pencil grammaticality judgment test
- 2.3ERP experimental materials
- 2.4.1EEG acquisition
- 2.4.2PCA-constrained derivation of time windows and electrode regions
- 3.1Paper-and-pencil grammaticality judgment test results
- 3.2Comprehension questions results
- 3.3ERP results
- 3.3.1Comparison of L1 and L2 speakers’ brain responses
- 3.3.2L1 participants’ brain responses to filled gaps: P600
- 3.3.3L2 participants’ brain responses to filled gaps: Prefrontal-central positivity