Doctoral thesis

The main aim of this PhD dissertation is to contribute to the study of the acquisition of Spanish as a Foreign Language, which is key for the development of bilingual societies. For that purpose, this study will explore how task influences the linguistic production of Spanish bilinguals whose native language is English through the analysis of anaphora resolution (AR). The acquisition of AR is particularly challenging for bilinguals with that linguistic profile, since it is a key phenomenon at the syntax-discourse interface. Additionally, task effects on AR in Spanish bilinguals have not been previously studied. In order to fill the gap in the literature, the data analysis will be based on a learner corpus, which is a large digital database that collects the contextualized linguistic production of learners of a second language. Specifically, this study will employ the CEDEL2 corpus (http://cedel2.learnercorpora.com/), developed at the BilinguaLab laboratory in the University of Granada. CEDEL2 is the largest corpus of Spanish as a Foreign Language to date and includes numerous variables, such as the task used to elicit the linguistic production and the modality of the texts (spoken vs. written). Learner corpora are, therefore, powerful tools in the field of digital humanities.

This research aims to provide valuable methodological insights into how task type influences linguistic production, allowing researchers to improve future learner corpus designs and enabling professionals of Spanish as a Foreign Language to develop teaching materials based on the findings of this study. Furthermore, the CEDEL2 corpus will be expanded by collecting data from learners of Spanish and native English speakers in oral and written narrative tasks, contributing to the new version of CEDEL2 (v. 3). A new task will also be designed to explore how task complexity affects anaphora resolution in learners and natives. Statistical analyses will be conducted in R, employing simple and mixed linear regression models. The software UAM Corpus Tool will also be used for descriptive and inferential analysis, as well as for tagging data according to the syntactic-discursive factors that constrain anaphora resolution in each task. Ultimately, the aims of this thesis will be achieved through the use of digital tools and by contributing to the methodological development of learner corpora, with the project aiming to significantly contribute to digital humanities in bilingualism research.