Poster C24, Friday, August 17, 10:30 am – 12:15 pm, Room 2000AB

Korean English L2ers’ sensitivity to information structure: An ERP study

Wonil Chung1, Myung-Kwan Park1;1Dongguk University

Cowles, Kluender, Kutas, and Polinsky (2007) found two types of ERP response in answers to wh-questions. One is that all words in focused position showed a large positivity understood as characteristic of sentence-final elements, and in fact the sentence-final words of the sentences containing them did. They suggested that focused elements might trigger integration effects, like those in sentence-final position. The other is that inappropriately-focused referents showed a right negativity. They suggested that this N400-like effect was elicited by comprehending structurally-encoded focus cues and discourse-level restrictions. Along the line of Cowles et al (2007), we are also to investigate ERP responses to violations of information structure (IS) in answers to wh-questions in Korean English speakers, where focus structure is incorrectly aligned in ‘it’-clefts. To this aim the experimental materials for our ERP study consisted of 60 sets with two types of stimuli (congruent and incongruent), adopted from Cowles, et al. (2007). Each trial contains a set-up context with the introduction of three discourse participants, and then a wh-question consisting of one participant as an agent and two participants as an undergoer of an event, and a target sentence that was constructed as an it-cleft, with its pivot marked for focus with a congruent or incongruent participant, schematically represented below. (1) set-up: Who did the queen silence with a word, the banker or the advisor? (2) congruent target: It / was/ the banker/ that/ the queen/ silenced/. (3) incongruent target: It /was/ the queen/ that/ silenced/ the banker/. Twenty Korean English L2ers with a high level of English proficiency participated in this experiment. ERPs were measured at the critical phrase (a cleft pivot: ‘the banker’ or ‘the queen’) and all the following expressions (i.e. words/phrases) in a sentence. We found that, first, all the expressions in cleft-pivot focus position registered a large positivity. Likewise, the final expressions in the congruent condition recorded a positivity, but those in the incongruent condition didn’t. Second, the expressions in cleft-pivot focus position in the incongruent relative to the congruent condition elicited N400 at right anterior regions and widespread P600, namely, a bi-phasic RAN-P600. The word immediately after the pivot (e.g., ‘that’) in the incongruent relative to the congruent condition elicited an ELAN, and the sentence-final expressions in the incongruent relative to the congruent condition evoked a sustained negativity. We take the results in this experiment to indicate that the N400 evoked at the cleft pivot in the incongruent condition reflects a violation of IS called for by the congruence between the preceding wh-question and its answer in a given context, and the P600 at the same position is a signature of syntactic integration difficulty due to the misfit of a non-focused constituent in a syntactic position reserved for focused expressions. At the same time, we suggest that the sustained negativity at the sentence-final elements in the incongruent condition is a neural correlate of increased syntactic complexity owing to the IS-wise mis-alignment of syntactic constituents.

Topic Area: Grammar: Syntax