Cross-lingual phonological effects in different-script bilingual visual-word recognition

To isolate cross-lingual phonological effects during visual-word recognition, Arabic–Hebrew bilinguals who are native speakers of Spoken Arabic (SA) and proficient readers of both Literary Arabic (LA) and Hebrew, were asked to perform a visual lexical-decision task (LDT) in either LA (Experiment 1) or Hebrew (Experiments 2 and 3). The critical stimuli were non-words in the target language that either sounded like real words in the non-target language (pseudo-homophones) or did not sound like real words. In Experiment 1, phonological effects were obtained from SA to LA (two forms of the same language), but not from Hebrew to LA (two different languages that do not share the same script). However, cross-lingual phonological effects were obtained when participants performed the LDT in their second language, Hebrew (Experiments 2 and 3). Interestingly, while the within-language effect (from SA to LA) was inhibitory, the between-language effect (from SA to Hebrew) was facilitatory. These findings are explained within the Bilingual Interactive Activation plus (BIA+) model which postulates a fully interconnected identification system that provides output to a task/decision system.