DESCRIPTION (provided by applicant): Why do bilinguals selectively lose access to a language after neurological insult? Research in bilingual aphasia and more recent research using neuroimaging tools has found that both age of second language acquisition (AoA) and proficiency play a role in the neural representation of two languages. The current proposal is designed to extend this seminal work in two important directions. First, the work will look at early Spanish-English bilinguals, a group of participants who have been vastly understudied in both the behavioral and neuroimaging literature. Participants will be matched on English proficiency but will be placed in two groups which have either low or high proficiency in Spanish. They will be asked to make grammaticality judgments for two-word phrases or short sentences which have subject-verb agreement, number or grammatical gender violations. These represent a test of the neural correlates associated with morphological (the plural marking 's') violations in each language. Second, the proposed project will look at the influence of cross-language overlap (i.e. functions which transfer from one language to the other easily) on the differences across groups, a factor which has not been systematically manipulated in published fMRI research to date. Our hypothesis is that language proficiency and overlap will modulate neural activity. Cross-language differences in neural activity should be larger for low proficiency subjects than for high proficiency subjects. Increased activity will be observed for functions which do not overlap. This increased activity will be larger for low proficiency subjects. Finally, a group of monolingual Spanish speakers will be tested to insure that any differences between grammatical functions are not due to a difference in difficulty. The difference in the pattern of activity across groups and languages will provide insight into the cognitive mechanisms that are necessary to process morphological information in a less proficient language.