UC San Diego

In three sets of experiments, this dissertation investigates the mental representation of syntactic structure. Chapter 1 aims to shed light on an ongoing debate between two models that aim to account for the regular production of an ungrammatical structure, resumption. Results support a production account rather than an audience-design account. Chapter 2 uses an artificial language learning paradigm to determine whether various long-distance dependencies are represented independently or as the same structure. Results indicate that the representation of these structures is unitary in some sense. Chapter 3 looks at individual differences in the production and comprehension of various structures to determine whether syntax is purely abstract, as generally assumed, or if it exists in a gradient representation space. Results reveal individual differences, suggesting that syntax does in fact have a gradient component. Together, these studies contribute to a growing body of work indicating that syntactic representations are complex and multifaceted, and require a more nuanced model than is often assumed.