Bilingualism, cognitive capacity and age: A computer-based study in L3 processing.

摘要

This doctoral research investigates from a psycholinguistic perspective third language development by bilingual Latino adults. It is one of a series of studies associated with The Latin Project, designed by Professor Cristina Sanz and graduate students Harriet Bowden and the researcher to investigate internal and external learner differences and their influence on non-primary language learning in adulthood. The present study examines prior language experience, working memory capacity and age and how these factors interact during processing of new linguistic input. Participants were native speakers of Spanish and users of English as a second language (L2). They varied in their degree of L2 proficiency and the age at which they began learning English and were classified into four experimental groups: Simultaneous Bilinguals, who began learning both Spanish and English from birth and maintained highly proficient use of both languages; Pre-adolescent Sequential Bilinguals, who began learning L2 English between ages 4 and 12 and maintained highly proficient use of both languages; Post-adolescent Sequential Bilinguals, who began learning L2 English after the age of 16 and maintained highly proficient use of both languages; Developing Sequential Bilinguals, whose L2 English learning began after the age of 16 and had not yet reached high proficiency at the time of the study. Participants were introduced to a third language (L3), Latin, by means of The Latin Project's interactive, computer-based lesson, practice and testing package. Results generally indicated more learning and maintenance of that learning by Developing Sequential Bilinguals and Simultaneous Bilinguals than by the other groups. In addition, cognitive capacity as measured by verbal working memory capacity was revealed as a significant predictor of successful L3 development. Prior language experience and age variables as operationalized in the present study were not significant predictors of successful L3 development. This may be due to the highly explicit nature of the treatment to which all participants were assigned. The findings of this study provide evidence that verbal working memory capacity is language-dependent and that it is implicated in non-primary language development in adulthood. They also suggest that explicit learning conditions may facilitate non-primary language learning in adulthood.