통사-의미 역학의 신경언어학적 연구

Punctive verbs, such as jump, describe bounded events which occur within short periods of time (e.g. At 3 o’clock the rabbit jumped). Nonetheless, when placed within durative contexts, language comprehenders readily understand such verbs to mean that multiple events have occurred (e.g. For 10 minutes the rabbit jumped). This process is termed aspectual coercion. In the current experiment we used event related potentials (ERPs) to examine the online processing of aspectual coercion by Korean L2 learners of English. ERPs were measured as participants read sentences, presented word-by-word (400ms, 300ms ISI), that contain one of three contexts: 1) punctive (non-coerced, single iteration); 2) durative (coerced, multiple iterations); animacy-violating (inanimate subject + animate subject requiring verb, animacy violated). The result of the experiment is that when punctive verbs appeared in durative contexts, they evoked both a reduced negativity (N400) between 350-400msec and a positivity (P600) between 650-700msec than when they appeared in punctive contexts. Meanwhile, the animacy-violating condition produced both a largely-distributed negativity (N400) between 300-350msec and a largely-distributed positivity (P600) between 500-600msec, in comparison to the control condition. We suggest that among the representational hypotheses currently in the literature on aspectual coercion, our result is most consistent with an account where aspectual mismatch is repaired and resolved later in the syntactic component by Korean L2 learners of English rather than early in the semantic component found for L1 English speakers.