You are viewing the SNL 2017 Archive Website. For the latest information, see the Current Website.

Poster A61, Wednesday, November 8, 10:30 – 11:45 am, Harborview and Loch Raven Ballrooms

Auditory and visual word processing in child and adult second language learners: Electrophysiological and behavioral evidence of cross-language interaction

Katharine Donnelly Adams1, Fatemeh Abdollahi1, Ping Li1, Janet G. van Hell1;1The Pennsylvania State University

When cognates (words that share semantics, phonology, and orthography across languages) are presented to bilinguals, these words are processed faster than noncognates (for a review, see Van Hell & Tanner, 2012). This cognate facilitation effect occurs even when bilinguals read words in only one language. Studies reporting cognate facilitation effects have typically presented cognates and noncognates visually (but see Lagrou et al., 2011). Visual presentation of words activates orthographic codes more strongly than phonological codes and, likewise, auditory presentation activates phonological codes more strongly. Cognates, even orthographically identical cognates, nearly always have (slightly) different phonological forms across languages. These language-specific auditory cues may constrain lexical access to only one language. Auditory presentation of cognates and noncognates, therefore, may reduce or even eliminate a cognate facilitation effect that is present in visually presented words. To test whether cross-language lexical activation is different with auditory versus visual presentation, we presented cognate and noncognates in auditory and visual go-no go tasks to adult second language (L2) Spanish learners while measuring ERPs. Cognates and noncognates were presented in L2 Spanish and in L1 English to examine whether cross-language activation depends on relative proficiency in L1 and L2. In a second experiment, we tested child L2 Spanish learners aged 7-9 years (both adults and children were native English speakers and classroom learners of L2 Spanish). Research in our lab has shown that cognate facilitation effects pattern differently in children and adults (Brenders et al., 2011; Poarch & Van Hell, 2012). For example, testing child L2 learners, Brenders et al. (2011) found that the cognate facilitation effect turned into a cognate inhibition effect when interlingual homographs were added to the list of cognates and noncognates, whereas this list manipulation did not affect the cognate facilitation effect in adult proficient bilinguals. In the visual word recognition task in L2 Spanish, adult L2 learners demonstrated an N400 effect for cognate status (increased negativity for noncognates relative to cognates with English), and a smaller, nonsignificant N400 effect for cognate status in the visual word recognition task in L1 English. In contrast, no significant cognate effects were observed in the auditory tasks, and this was found in both the L2 and L1 tasks. Experiment 2 showed that performance of the child L2 learners patterned with that of the adults. Visual presentation of L2 Spanish cognates and noncognates showed a (delayed) N400 effect, and visual presentation of L1 English words showed a small and negligible N400 effect. In contract, auditory presentation yielded basically overlapping waveforms for cognates and noncognates, both for Spanish and English words. These findings indicate that mode of presentation (visual or auditory) modulates the co-activation of languages, and that both child and adult L2 learners employ phonological cues to constrain access to one language. Furthermore, for visual word recognition in L2, parallel activation of the nontarget L1 is stronger than vice versa, which reflects the difference in L1 and L2 proficiency in L2 learners.

Topic Area: Multilingualism

Back to Poster Schedule