Changes in ERP signatures after 9 weeks of intensive second language learning are modulated by individual success

White, E. 1, 2 , Genesee, F. 1, 2 & Steinhauer, K. 2, 3

1 Department of Psychology, McGill University
2 Centre for research on language, mind and brain
3 School of Communication Sciences and Disorders, McGill University

This presentation will describe the results of a longitudinal ERP (event-related potential) study designed to address the following questions: What are the neuro-cognitive changes that accompany second language (L2) grammar learning in adults? Are L2 processing patterns related to an individual learner’s level of success or determined by his/her first language (L1) background? With these questions in mind, we examined the ERPs elicited in late L2 learners both before and after an intensive English-as-a-second-language course. Chinese and Korean learners of English read sentences that were grammatically correct or that contained a violation of regular past tense; a grammatical rule that operates differently in Korean and English and does not exist at all in Chinese. After 9 weeks of intensive English instruction, both groups of learners displayed a significant P600 component, an effect that was not present at the first testing session. Grouping participants according to their behavioural performance at each testing session showed that different neuro-cognitive processes were used depending on the learners’ knowledge of English past tense specifically. The L2 learners with poor behavioural performance at both testing sessions did not show a P600 at either session. Conversely, those that improved behaviourally showed a change in ERP effects that depended on their level of performance. Those with poor performance at session 1 and high performance at session 2 showed an N400 at session 1 and a P600 at session 2. Those with high performance at both sessions showed a P600 that increased in amplitude and duration by session 2. Together these findings suggest that the neuro-cognitive processes underlying the P600 (e.g. “grammaticalization”) are not restricted to grammatical structures that can be transferred from one’s L1; moreover they are modulated by individual levels of behavioural performance. Finally, ERP differences at low levels of proficiency (N400) may predict later success.