In this dissertation, I investigated how cognitive resources as well as formal, and informal language experience impact language learning in two studies. In the first study (Chapter 2), I examined the learning of Spanish grammatical gender by Chinese international students who were studying abroad in the US. The goal of that study was to uncover whether being immersed in their non-dominant language (English) to varying degrees and having ambient exposure to other languages in their environment would impact new language learning and modulate the relationship between cognitive control resources and learning. For the learning of grammatical gender, individual differences in the international students’ degree of exposure to English predicted higher accuracy in learning. Moreover, international students’ active use of an additional Chinese language modulated the relationship between cognitive control and reaction times for learning and generalization of grammatical gender in Spanish. These results suggest that both the learners’ current immersion context, their past language knowledge, and their cognitive resources come to shape new language learning in a third language.In the second study (Chapter 3), I investigated whether prior knowledge of a tone language as an L1, individual differences in cognitive control and musical experience predicted the ability to perceive tones in a new language. Speakers of tonal native languages (Vietnamese or Bantu languages) and speakers of non-tonal native languages (English or Dutch) were asked to identify Mandarin tones. Critically, Vietnamese has a tone system that is similar to Mandarin because it uses the same two tonal cues as Mandarin (pitch height and pitch direction). On the other hand, Bantu languages are more dissimilar because they only use one of the tonal cues from Mandarin (pitch height). All participants had no prior knowledge of Mandarin, but all were L2 speakers of English, also a non-tonal language. L1 speakers of a tone system that was similar to that of Mandarin (Vietnamese) were more accurate in identifying the Mandarin tones relative to speakers of nontonal languages (Spanish or Dutch). However, L1 speakers of Bantu languages, which have a more dissimilar tone system to that of Mandarin, were less accurate in identifying Mandarin tones compared to speakers of nontonal languages. Higher levels of musical experience also predicted higher Mandarin tone identification, but only for Spanish and Vietnamese groups. Cognitive control abilities, as indexed by performance on the Simon task, did not predict tone identification. These findings are consistent with the claim that speech L2 perception is largely influenced by native language tuning to L1 phonetic cues and with previous research showing that experience tracking pitch in music confers benefits to tracking pitch in a linguistic context.

An aim of research on bilingualism is to understand how the brain adapts to the use of more than one language. Although several important discoveries and insights about the consequences of bilingualism have been generated over the last several decades, concerns about replicability have narrowed the scope of inquiry and discussion to the application of prescriptions about sample size and method. In this dissertation, I critique this approach and reformulate its value by placing it in the broader context of science as a discovery process, in which incremental understanding, methodological and analytical diversification, the framing of our questions, and even underpowered studies, are essential to advance. I propose that this necessitates research practices and tools that: 1) focus on examining variation in the relation between language processing and cognitive functioning; 2) allow us to identify meaningful interactions rather than main effects only; and 3) provide a rich characterization of the participant sample to identify bilingual phenotypes: the adaptive variety induced by the interplay between biology and culture. In seeking to apply this framework, I then present two empirical studies investigating the cognitive mechanisms that enable fluent language use, and examine the hypothesis that proficient bilingualism is characterized by the active engagement of a regulatory mechanism that adapts to the demands of the language environment. To test this, I examine individual differences in language production and cognitive control across four groups of proficient bilingual phenotypes who differ with respect to the interactional contexts of language use. In Study 1, the results suggest that lexical access is mediated by a systematic interaction between regulation of the dominant language and cognitive control, but that the manifestation of this interaction depends on whether bilinguals are immersed in a second-language environment, a context in which active regulation is required. In Study 2, the results further confirm that dominant-language regulation is not analogous to proficiency per se; rather, it may more adequately reflect the coordination of language-related and domain-general resources that interactively contribute to the accessibility of information in the language network. Together, these studies suggest that language regulation is a fundamental feature of bilingual brains that engages a combination of language-related and domain-general cognitive resources to enable proficient language use.