Does working memory modulate cross-linguistic structural priming?

Abstract

Many studies argued shared syntactic representations between languages. However, less is known whether cross-linguistic syntactic processing rely on cognitive resources (e.g., working memory). We therefore conducted two structural priming experiments with a working memory task (i.e., a number series recall task between prime and target) for Mandarin-Cantonese-English multilinguals. When prime and target involved different actions (Experiment 1), cross-language priming was similar not only between different languages (Cantonese-to-Mandarin vs. English-to-Mandarin), but also between different working memory load conditions (high vs. low). When prime and target shared the same action (Experiment 2), cross-language priming was stronger in Cantonese-to-Mandarin condition and in low load condition. Importantly, such a translation-equivalent boost effect occurred in low load condition but not in high load condition. These findings suggest an automatic processing of abstract cross-linguistic structural priming and an important role of working memory on translation-equivalent boost effect, supporting learning-based models of shared syntax among multilinguals’ languages.