Doctoral Dissertation Research: Cross-linguistic activation of referential bias in second-language learners of English

双语是世界上许多地方的常态。因此，了解双语者是如何在语境中获取意义的，不仅是一个有趣的研究问题，即人类大脑如何在日常生活中适应两种或两种以上语言的常规使用，而且对应用语言专业人员，如口译员、教师和与双语者打交道的语言病理学家有直接的影响。先前的研究表明，双语者对一种语言的了解会影响他们用另一种语言表达和解释单词和句子的方式。这被称为跨语言激活。本论文研究的目的是帮助我们理解跨语言激活如何影响双语者在语境中理解语言的方式，即超越对单个单词和句子的理解。本研究将通过调查与韩语谓语相关的指称偏差强度如何影响韩语-英语双语者的英语指称加工，从而深入了解词水平上的跨语言激活对双语者话语水平加工的影响。基于动词的内隐因果关系（IC）会对下一个要提到的人产生偏见；例如，“约翰威胁比尔”暗示主语（“约翰”）是“威胁”事件的原因，使其最有可能成为语篇中的下一个指称物。除了IC动词外，韩语还有一类（主语偏向）句法使役结构，最好翻译为“使X成为Y”。一项初步研究表明，韩语使用者在使用SC结构时比使用常规IC谓词时具有更强的指称偏误。本研究旨在探讨韩语中与SC和IC谓词相关的偏差强度差异对韩语-英语双语者离线和在线英语参考加工的影响程度。该项目将使用书面句子补全（实验1）和视觉世界眼动追踪（实验2）两种范式，从说韩语的英语学习者和母语为英语的人身上收集数据。在这两个实验中，我们将评估英语语言能力和学习经验对参照加工中跨语言激活的潜在调节作用。

Bilingualism is the norm in many parts of the world. Understanding how bilingual speakers derive meaning in context is thus not only an interesting research question about how the human mind accommodates the regular use of two or more languages in daily life, but one that has direct implications for applied language professionals, such as interpreters, teachers, and speech-language pathologists working with bilinguals. Previous research has shown that bilinguals' knowledge of one language can affect how they say and interpret words and sentences in the other. This is called cross-linguistic activation. The goal of this dissertation research is to contribute to our understanding of how cross-linguistic activation affects how bilinguals make sense of language in context, that is, beyond understanding individual words and sentences. The proposed research will provide insight into the effect of cross-linguistic activation at a word level on bilinguals' discourse level processing by investigating how the strength of referential biases associated with Korean predicates affects Korean-English bilinguals' referential processing in English. Verb-based implicit causality (IC) creates biases for who will be mentioned next; e.g., 'John threatened Bill' implies the subject ('John') was the cause of the 'threaten' event, making it the most likely next referent in the discourse. Korean, in addition to IC verbs, has a class of (subject-biased) syntactic causative (SC) constructions, best translated as 'causing X to be Y'. A preliminary study has shown that Korean speakers have stronger referential biases with SC constructions than with regular IC predicates in Korean. This project investigates the extent to which differences in bias strength associated with SC and IC predicates in Korean influence Korean-English bilinguals' offline and online reference processing in English. The project will collect data from Korean-speaking learners of English and native English speakers, using both written sentence completion (Experiment 1) and visual-world eye-tracking (Experiment 2) paradigms. In both experiments, potentially modulating effects of English language proficiency and learning experience on cross-linguistic activation in referential processing will be assessed.