Options
Bilingual processing of formulaic sequences
Author
Tay, Linda Poh Ling
Supervisor
Green, Clarence
Abstract
Formulaic sequences (FSs), or multiword units, have been identified as a ubiquitous linguistic phenomenon that is critical in successful language acquisition; how FSs are processed and produced has therefore been the focus of studies that seek to understand the differences between native and non-native language use. One potential difference lies in how frequency and mutual information (MI) turned out as better predictors of non-native and native speaker processing of academic FSs respectively (Ellis, Simpson-Vlach, & Maynard, 2008). This study attempts to extend the investigation to spoken FSs and early bilinguals for whom English and Chinese were both acquired in childhood. In a phrasal-decision reaction- time experiment, participants judged the likelihood of visually presented FSs being found in English. The results show that differences were found not only in how academic and spoken FSs are processed, but also in how the three speaker groups process them. Mirroring the results in Ellis et al. (2008), a clear frequency effect for second-language speakers in both types of FSs was obtained, but for native speakers MI appeared as a predictor only for spoken FSs. While both frequency and MI did not appear to predict early bilingual reading of both types of FSs, the findings suggest that language-related factors such as language dominance and proficiency in English may play a role in how they, as well as second-language speakers, process FSs. In line with usage-based accounts of language acquisition, it is proposed that language experience – current and cumulative – accounts for these differences.
Date Issued
2019
Call Number
P118 Tay