The processing of pronouns and the production of referring expressions in L2 English

Psycholinguistic research investigating production and comprehension of referring expressions has shown that successful reference implies choosing/interpreting a form among a set of referring expressions (e.g., Arnold 2010 for a review). Native speakers are usually very fast and efficient in making these decisions in order to keep up with the flow of normal conversation, requiring them to comprehend and produce referring expressions at a fast pace. Research on second language (L2) acquisition has shown that, differently from native speakers, L2 learners may show residual indeterminacy in the L2 referential choice (e.g., Sorace & Filiaci, 2006), even at the highest levels of proficiency. Previous research on L2 comprehension and production of referring expressions has mainly focused on learners who acquire null subject languages (e.g., Spanish, Italian), and has primarily used offline experimental designs (see Wilson 2009, for a study on L2 German using eyetracking during listening, and Roberts et al., 2008, for a study on L2 Dutch using eye-tracking during reading). The present study contributes to filling this gap by examining the process of comprehension and production in L2 speakers of a non null-subject language (English) whose first language is nullsubject (Spanish), using both off-line and on-line methods. The paper is organized as follows. We will first present evidence from cross-linguistic research on anaphora resolution and referential choice showing the main differences between English and Spanish. Secondly, we will review evidence from the literature on L2 acquisition illustrating that L2 learners show a different pattern of anaphora resolution and referential choice than native speakers (L1). Then, we will present two main experimental studies that adopt off-line and online methodologies to address our research questions. We use eye-tracking measures during listening to examine the time course of pronoun resolution during language processing (Experiment 1) and behavioral measures of accuracy in a story telling task to test L2's production abilities (Experiment 2). To anticipate our findings, in Experiment 1 L2 speakers performed similarly to English native speakers on the eye-tracking task. In Experiment 2, we observed a difference in the type of referring expressions used by the two groups. When the native speakers prefer to use a full NP, L2 learners tend to frequently choose a pronoun, even in contexts where twoentities share the same biological gender, i.e., the use of a pronoun leads to ambiguity.