Structural computation and the role of morphological markings in the processing of Japanese

In an SOV, strictly head-final language with otherwise relatively free word-order such as Japanese, disambiguating information including verb information is not available until the end of a clause. In such a language, a sentence theoretically has numerous possible syntactic structures before the verb information becomes available. The current study investigated (1) whether syntactic computation takes place in an ambiguous sentence fragment in Japanese and (2) if and how information from constituents other than the verb is utilized. Two experiments were conducted using automated, word-by-word visual presentations with a lexical decision task on an extrasentential target. The results suggest that syntactic computation takes place in Japanese before verb information becomes available. A strong preference was observed for a simplex clause when three differently Case-marked arguments belonged to the same clause. The findings also indicate that information from surface Case marking is utilized productively, and when there is ambiguity in marking between arguments and adjuncts, the distinction is made before incorporating the constituents into the existing structure.