期刊:Linguistic Approaches to Bilingualism [John Benjamins Publishing Company] 日期:2024-05-13
标识
DOI:10.1075/lab.21016.lij
摘要
Abstract In the present study we aimed to obtain with trilinguals the cognate inhibitory effects reported earlier for bilingual speakers who performed L2 lexical decision tasks (LDTs) with non-identical cognates and controls. To that end, Polish–English–Dutch trilinguals performed two LDTs in their L3. In Experiment 1, the stimuli included two types of double non-identical cognates Polish–Dutch (e.g., SMAK–SMAAK ), and English–Dutch (e.g., BUTTER–BOTER ) as well as matched non-cognate Dutch controls (e.g., JASJE ). In Experiment 2, we tested triple non-identical cognates shared across Polish, English and Dutch (e.g., GRUPA–GROUP–GROEP ) and Dutch controls (e.g., BROEK ). We failed to find the bilingual inhibitory effects. In contrast, significant facilitation for English–Dutch and for Polish–English–Dutch cognates was found, even though no identical cognates were used. However, Polish–Dutch cognates yielded null results. The current findings emphasize the influence of the levels of form similarity on the observed cognate effects and point to the important role of stimulus list composition in cognate processing.