GOALS OF THE GROUP

We investigate how bilinguals acquire, perceive and process languages, and how grammars are represented in their minds. Our research focuses on the study of Basque/Spanish bilinguals differing in ages of acquisition and proficiency; given the significant typological differences between Spanish and Basque, we can also explore the impact of language distance on bilingualism, an issue that has been less widely studied to date. We combine knowledge from theoretical linguistics and methods from experimental psycholinguistics to study the main components of grammar.

.

SUBPROJECTS OF THE GROUP  

Subproject 1: Agreement processing in bilinguals.  This research inquires into the nature of grammatical processing in bilinguals, by studying agreement errors in Basque-Spanish and Spanish-Basque bilinguals. For all bilinguals, errors are induced by a constituent intervening at some point in the syntactic derivation, but not necessarily at surface position, and subject/object agreement are equally affected by intervention effects. However, Spanish-Basque bilinguals produce more agreement errors than Basque-Spanish bilinguals, even at highest proficiency. Non-native bilinguals produce more errors than native bilinguals in both subject-verb (an existing agreement relation in Spanish) and object-verb agreement (an agreement relation lacking in Spanish) (Figure 1). This suggests that, although non-native bilinguals’ agreement encoding is more prone to attraction errors, the presence or absence of a particular agreement relation in their native language is of no consequence to their linguistic performance. We also explore the processing of other types of agreement dependencies such as Negative Polarity Items that must meet both syntactic and semantic requirements to determine whether these dependencies behave in the same way as verb agreement and only display proficiency effects in bilingual processing.

Figure 1. Agreement error effects in Basque-Spanish vs. Spanish-Basque proficient bilinguals.

Subproject 2: Syntactic structure in bilinguals. Human language is built by means of constituents or phrases, where phrases are typically headed by a functional element. Languages can linearize phrases placing their heads first in the temporal signal (head first languages) or last in the temporal signal (head final languages). Spanish (head initial) and Basque (head final) represent clear examples of the two types of phrasal pattern found in language. Results from ERP and behavioral experiments undertaken by our group show that proficient bilinguals process these patterns equivalently regardless of which choice their native language makes. We explore these findings in more depth by studying the segmentation strategies and sentence-processing strategies/asymmetries in bilingual speakers of grammars that yield opposite values in these language variation domains. We employ artificial language paradigms to explore basic segmentation preferences, and fMRI methods to further investigate sentence processing in bilinguals.