Morphological variability in long-distance subject-verb agreement: A study of native and nonnative processing

Abstract

This study examined morphosyntactic variability in Spanish-speaking learners of English in order to determine the effects of two linguistic factors on the establishment of subject-verb agreement: structural distance and plural markedness. The study also investigated the effects of task demands on processing of agreement morphology. Participants completed an online reading task using a moving window self-paced reading paradigm. In order to explore whether increases in structural distance between agreeing elements leads to decreases in sensitivity to agreement violations (Clahsen & Felser, 2006; Gabriele et al., in press), agreement between subject and verb was established across a prepositional phrase (such as in warm southern Mexico) or a structurally more complex relative clause (who hunted in Mexico). Subject number (singular or plural) was also manipulated in the task in order to examine whether the marked plural feature facilitated agreement establishment even as distance increased (Wagers et al., 2009). In order to determine whether task effects cause native speakers to show learner-like patterns of agreement variability (MacDonald, 2006; Hopp, 2010; Lopez Prego, 2012), a group of native English speakers was placed under a concurrent digit load as they completed the self-paced reading task. The results revealed that L2 learner agreement was affected by structural distance, as learners became less sensitive to violations in the relative clause intervener condition. Weak effects of plural markedness emerged in the learner results as well, indicated by greater sensitivity to errors in pairwise comparisons to the plural subject-relative clause intervener condition over the singular subject counterpart. Finally, weak similarities in variability between the L2 learner and the native speaker group under a concurrent processing load tentatively suggest that learner variability may be caused by general processing limitations, not deficits in L2 grammatical knowledge.