Bilingual Language Control

University Of California, San Diego, La Jolla CA

Investigators

Linked publications & trials

Abstract

DESCRIPTION (provided by applicant): When speaking in just one language, bilinguals do not seem different from monolinguals. At other times bilinguals seem to be language acrobats, switching back and forth between languages, without apparent effort. Only on rare occasions do bilinguals produce a word in an unintended language - or cross-language intrusion errors (<1% Gollan et al., 2011; Poulisse, 1997), and perhaps for this reason intrusions have remained largely unexplored even though they hold potential to reveal the cognitive mechanisms fundamental to bilingual language control. The rarity of intrusion errors challenges an assumption that has become ubiquitous in bilingual research -that bilinguals face constant competition due to dual-language activation. We hypothesize that bilingual language control, including both the acrobatics associated with intentional language mixing, and the rarity of intrusion errors, reflects the joint operation of multiple control mechanisms including both language-specific and domain-general control mechanisms. The proposed studies will elicit intended and unintended language switches to investigate language-specific control mechanisms (Aim 1) including grammatical constraints, lexical accessibility, inhibitory links between translation equivalents, and error monitoring processes that rely on the language comprehension system. We hypothesize that bilinguals also rely on domain-general control mechanisms to achieve language selection, and investigate the possible roles of inhibition and attention for controlling single-language and mixed-language speech, the processing loci at which bilinguals use inhibition to maintain language selection, and contextual cues and strategies that improve language control by reducing switch costs and intrusion errors (Aim 2). Converging evidence for the classification of control mechanisms as language-specific versus domain-general will come from investigation of aging effects, and correlations with individual differences in measures of executive control (Aim 3). As we investigate factors that lead to more versus less effective ability to control language selection we develop new methods for studying speech errors using reading aloud with eye-tracking. The proposed work will establish test conditions that maximize bilingual proficiency, and will provide preliminary norms for young and aging bilinguals on a number of oft-used standardized tests. Broader questions addressed include the cognitive mechanisms underlying switching, deficits in cognitive aging, and how domain-general mechanisms function in concert with the language system to allow all speakers to resolve competition and produce error free speech.