Investigating Syntactic Representations in English. The Effect of Language Exposure and Experience on Syntactic Representations in English Late Bilinguals and Multilinguals

Abstract

This dissertation focuses on how L2 learners (Italian speakers of English and as well as other L1 speakers) process and produce complex syntactic structures, namely passives and double object dative sentences. The studies in this dissertation aimed at investigating four factors: (i) how L2 speakers and bilinguals process passive and double object dative constructions, (ii) how animacy interacts with the production of these constructions, (iii) the role of the lexical overlap in their production and finally (iv) how proficiency modulates these variables. The studies used a within-language syntactic priming paradigm to answer these questions whereby participants process a prime sentence (auditorily) while viewing an image and subsequently describe a different, unrelated image (written, by typing). An example of the stimuli is shown in (1a) and (1b) for the transitives and (2a) and (2b) for the datives.

(1a) The woman is pulled by the boat (1b) The boat is pulling the woman

(2a) The girl is giving flowers to the teacher (2b) The girl is giving the teacher flowers

Proficiency was measured using both objective and subjective measures of proficiency: the Michigan Test of English Language Proficiency (MTELP) and a self-rating of the four main linguistic abilities. These two measures were also combined to create a measure that took into consideration both the objective- and subjectiveness of the two tasks as proposed by Marian et al. (2007). Previous structural priming studies with late bilinguals suggest that the trajectory of second language (L2) syntax goes from item- and language-specific to shared abstract representations (Hartsuiker et al., 2004; Bernolet et al., 2013; Hartsuiker & Bernolet, 2017). The studies in this dissertation will discuss whether this holds across different populations of bilinguals and throughout L2 development, as a function of location testing context (English-immersed, English-non-immersed). It will also discuss if and how proficiency modulates priming effects.