Proficiency and Animacy Effects on L2 Gender Agreement Processes During Comprehension

Abstract

This study examines whether adult second language (L2) learners of an ungendered first language (L1) are sensitive to gender congruency (grammatical feature absent in the L1) and noun animacy (semantic feature present in the L1) when processing L2 gender concord and whether L2 proficiency level determines such sensitivity. To address these questions, 63 Spanish monolinguals and 69 beginning and 64 intermediate Anglophone late learners of L2 Spanish completed a moving-window and a grammaticality judgment task with sentences with gender concord and discord with animate and inanimate nouns. The moving-window data reveal longer reading times in sentences with gender discord than concord and in those with animate than inanimate nouns in intermediates and Spanish monolinguals but not in beginners. Similarly, grammaticality judgments show that intermediates are more accurate in sentences with inanimate than animate nouns and are better than beginners in sentences with gender agreement violations. These results suggest that intermediate learners display targetlike patterns that are more qualitatively similar to those of natives than beginners, both in terms of semantic and grammatical features. In addition, these findings indicate that agreement with animate nouns is cognitively more demanding than with inanimate nouns both for intermediates and Spanish monolinguals, in line with lexical and syntactic accounts of gender.We would like to express our gratitude to Joyce Bruhn de Garavito, Pedro Guijarro-Fuentes, Jacqueline Toribio, Ludovica Serratrice, Bonnie Schwartz, the anonymous reviewers, and the audience of the 7th International Symposium on Bilingualism for their insightful comments and suggestions. Any shortcomings of the study are our sole responsibility.

Proficiency and sentence constraint effects on second language word learning

“…Based on previous studies about bilingualism, lower L2 proficiency results in weaker semantic processing (Ojima, Nakata, & Kakigi, 2005;Sagarra & Herschensohn, 2011) and syntactic processing (Ojima et al, 2005;Osterhout et al, 2008;Rossi, Gugler, Friederici, & Hahne, 2006;Sagarra & Herschensohn, 2010Steinhauer, White, & Drury, 2009). The efficiency of working memory is partially dependent on proficiency (van den Noort, Bosch, & Hugdahl, 2006).…”

The role of proficiency and working memory in gender and number agreement processing in L1 and L2 Spanish

“…While we do not want to exaggerate the native-like processing of our intermediate learners, their incipient sensitivity to discord points to development of target-like computation. 3 Indeed, they are more accurate and have faster RTs with inanimate (no semantic-pragmatic cue) than animate nouns (Sagarra and Herschensohn, 2011), an indication that they are developing the grammatical features [ugender] and [unumber]. This development favors accessibility over deficit approaches to representation (cf.…”

Gender agreement in L3 English by Basque/Spanish bilinguals

“…In both categories, heads are associated with features that can be interpretable or uninterpretable. Interpretable features convey semantic information and they can be related syntactically to grammatical uninterpretable features, such as gender agreement (Sagarra & Herschensohn, 2011). In minimalist terms, uninterpretable features have to be eliminated by means of the syntactic operation Agree through the corresponding interpretable feature (Chomsky, 1995).…”