Poster D26, Friday, August 17, 4:45 – 6:30 pm, Room 2000AB

A neurophysiological investigation of translation and morphological priming in biscriptal bilinguals

Myung-Kwan Park1, Wonil Chung1, Say Young Kim2;1Dongguk University, 2Hanyang University

The Revised Hierarchical Model (Kroll & Stewart, 1994) assumes asymmetric lexical links between first language (L1) and second language (L2) in unbalanced bilinguals, supported by stronger priming effects from L1 primes to L2 targets than from L2 primes to L1 targets (e.g., Jiang & Forster, 2001). Recent ERP studies also supported these asymmetric links in N250 and N400 components (e.g., Schoonbaert et al., 2011). However, other studies provided mixed results regarding the N400 effect for translation priming (e.g., Christoffels et al., 2013; Midgley et al., 2009). In order to answer the question of how two languages in bilingual readers are connected to each other, the present study examines if the pattern of cross-language translation priming is consistent with the asymmetric links between L1 and L2, and if it occurs via morphological decomposition, using ERPs and a masked priming lexical decision paradigm with unbalanced Korean-English bilinguals. We hypothesize that the N400 effect will be significant or larger when the participants are presented with Korean L1 primes and English L2 targets. In addition, this cross-language activation will be maintained by morphological priming when the targets are compound words and the primes are their constituents. In Experiment 1, Korean-English late bilinguals performed a masked priming lexical decision task where targets were Korean (L1) compound word (e.g., 목선, ²mok-seon,² neckline), and primes were one of three English (L2) words; 1) translated whole word (neckline), 2) translated morphemic constituent (line), or 3) an unrelated word (work). Experiment 2 was the same as Experiment 1, except that the targets were in English (L2) and the primes were in Korean (L1). As for the behavioral results, both the translation priming effect and the morphological priming effect were significant only for Korean L1 primes and English L2 targets, but not for English L2 primes and Korean L1 target. In ERP results, the translation priming effect was found only for Korean L1 primes and English L2 targets on the reduced N150, P250, and N400. The morphological priming effect was found both for Korean L1 primes and English L2 targets, and for English L2 primes and Korean L1 target on the N400. Taken together, the results suggest that both cross-language translation priming and morphological priming occurs even between different scripts (between noncognate words), and the effects are stronger when L1 primed L2 as compared to when L2 primed L1. In addition, different time-course between translation priming and morphological priming suggests that cross-language morphological decomposition occurs after translation in bilingual readers.

Topic Area: Grammar: Morphology