Lexical Representation and Processing in Cross-Script Urdu-English Bilinguals: The Case of Frequency-Balanced and Frequency-Unbalanced Cognates and Noncognates

摘要

The overall goal of this study was to examine the nature of lexical access and representation of frequency-balanced and frequency-unbalanced cognate and noncognate words in a previously unexamined cross-script language pair. More specifically, Experiment 1 was designed to determine if the cognate advantage obtained for same-script languages in the simple lexical decision task will also be obtained for the Urdu-English language pair. Both facilitation and inhibition effects were obtained for cognate words when participants were tested in English. This indicated nonselective lexical access and interconnectivity of the bilingual mental lexicon. However, when participants were tested in Urdu, a statistically significant cognate effect was not obtained. Experiment 2 was designed to examine whether the discrepancy in findings across cross-script studies in terms of the magnitude of the cognate and noncognate priming effect in a masked priming task can be attributed to frequency differences in the word stimuli as proposed previously. No significant priming effect was obtained for cognates or noncognates in any of the frequency-balanced conditions unlike the results from previous studies. However, a significant cognate and noncognate priming effect was found for some of the frequency-unbalanced conditions and again both facilitation and inhibition effects were suggested. The current version of the BIA+ model does not incorporate lateral inhibition effects at the phonological level for cross-script cognates. The findings from this study are explained within the BIA+ framework by allowing for lateral inhibition at the phonological level. In addition, the role of individual differences in language proficiency and processing strategy is also considered.