We propose a series of studies that apply a cognitive neuropsychological approach to investigate the behavioral presentation of Alzheimer’s disease (AD) in Spanish-English bilinguals. We use models of bilingual language processing and cognitive decline in AD to motivate experimental manipulations that will reveal the mechanism/s underlying cognitive deficits in bilinguals with AD. We also address critical practical questions, aiming to determine if bilinguals should be allowed to use either language during cognitive assessment, if testing in one language hinders subsequent performance in the other, and more generally how to maximize test performance and test sensitivity to AD in bilinguals. Theoretical considerations begin with evidence that although bilinguals do not seem different from monolinguals when they speak in just one language, both languages always remain active (they cannot just “shut one language off”). Bilinguals also switch languages both across contexts (e.g., at home vs. at work), and within sentences (when conversing with other bilinguals). By virtue of using each language only some of the time, bilinguals also use each language less frequently than monolinguals who only use the one language they know. Thus, bilinguals face unique control requirements in their everyday language use: choosing which language to speak, switching languages, retrieving less frequently used linguistic representations, and managing competition between languages. Together, the proposed studies will examine how these differences affect some of the most commonly used cognitive tests (e.g., picture naming, list memory, verbal fluency). These studies will also examine experimental tasks designed to elicit more naturalistic connected speech, including a referential communication task and a read- aloud task that recently demonstrated sensitivity to AD in bilinguals. Each proposed study focuses on key factors that we hypothesize can affect test performance – especially in bilinguals with AD, including: (a) switching languages across testing blocks (order effects), (b) switching within a testing block (the “either language” option), and (c) switching with varying degrees of support from semantic and syntactic context. We hypothesize that relative to healthy controls, bilinguals with AD will exhibit larger testing order effects, and reduced benefit from the option to “use either language” (which invites voluntary language switching). Bilinguals with AD may also exhibit more prominent switching deficits and reduced ability to exert control over the dominant language when language switches are not supported by syntactic context. The proposed studies will develop unique tools for diagnosis of AD in bilinguals and will reveal which cognitive mechanisms are critical for managing activation, representation, and use of two languages in a single cognitive system. In turn, the proposed studies will constrain psycholinguistic models of bilingual language processing and wi...