Do readers anticipate <i>wh</i>-in-situ questions? Cross-linguistic reading time evidence from Mandarin Chinese and French

Abstract The understanding of wh -in-situ questions relies naturally on contextual and prosodic information for their early discrimination from declarative sentences. However, there is scarce evidence on the parsing processes involved during the online incremental processing of these questions. In this study, we investigate the incremental reading of wh -in-situ sentences with no prosodic or contextual information available to aid the parser by comparing them to their declarative counterparts. We investigated two wh -in-situ languages: Mandarin Chinese (in-situ only) and French (optionally in situ). This comparison allows us to determine whether wh -in-situ questions are processed similarly across languages and whether the parsing process is related to language-specific question formation strategies. Results of four word-by-word self-paced reading experiments on two types of wh -in-situ phrases (simplex or complex) in Mandarin Chinese and French show an interpretation strategy in which the most frequent structure, declarative, is considered in both languages, independently of the available question formation strategy. Nevertheless, the timing of the online interpretation and the observed effects are affected by the nature of the wh -phrases (simplex or complex) and the definiteness of the noun phrases contained in the declaratives, which confirms that several processes occur concurrently introducing a limit on the capability to extract conclusions on the processes based solely on behavioral measures.