The locus of processing for object relative clauses and the impact of methodology

Abstract

ABSTRACTTheories of sentence processing can make similar predictions about what sentences are difficult but distinct predictions about where the sentence is difficult. For instance, both surprisal-only [Levy, R. (2008). Expectation-based syntactic comprehension. Cognition, 106(3), 1126–1177. doi:10.1016/j.cognition.2007.05.006] and locality theories [Gibson, E. (1998). Linguistic complexity: Locality of syntactic dependencies. Cognition, 68(1), 1–76. doi:10.1016/S0010-0277(98)00034-1] predict object relative clauses to be more difficult than subject relative clauses, but each makes different predictions about the locus of processing difficulty. Most empirical research to date has supported the predictions of locality theories. In the current study, these sentences are studied using self-guided reading, a task in which a participant underlines sentences on a tablet device with a finger. Both reading times and acceleration adjustments initiate at the point predicted by surprisal theories, not that predicted by locality.