Cantonese Relative Clause Processing: A Working Memory Study

This dissertation, "Cantonese Relative Clause Processing: a Working Memory Study" by Wai-yan, Sabena, Lam, 林緯欣, was obtained from The University of Hong Kong (Pokfulam, Hong Kong) and is being sold pursuant to Creative Commons: Attribution 3.0 Hong Kong License. The content of this dissertation has not been altered in any way. We have altered the formatting in order to facilitate the ease of printing and reading of the dissertation. All rights not granted by the above license are retained by the author. Abstract:  The processing of relative clauses has been a subject of recurrent interest in psycholinguistics. While a consistent processing preference for subject relative clauses has been found in English (e.g. Just & Carpenter, 1992; King & Just, 1991; King & Kutas, 1995; Traxler, Morris, & Seely, 2002), studies in Mandarin Chinese have not yielded consistent results in the processing asymmetry between subject and object relative clauses (e.g. Hsiao & Gibson, 2003; Lin, 2006). On the other hand, human cognitive preferences have been found to play a role in parsing (Gibson & Pearlmutter, 1998; Tanenhaus & Trueswell, 1995). This dissertation is interested in relating memory with syntactic structures in the course of syntactic parsing, in the hope that it could resolve the conflicting findings in Chinese. We address this by studying Cantonese relative clauses from a processing and working memory perspective. Two experiments were carried out. Experiment 1 involved a self-paced reading task together with an n-back memory task. We found some evidence, although not very conclusive, that object relative clauses were more difficult to comprehend, as reflected in poor comprehension accuracy and response latency. There was significant memory intervention in reading times, though its role in syntactic processing was not clear. Experiment 2 reported a maze task together with an n-back memory task. A processing disadvantage for object relative clauses was evident from a significantly longer reading time on a whole and in the post-relative clause region. A memory effect was found only at the post-relative clause region, possibly indicating that memory capacity as measure by n-back was not influencing the syntactic processing of relative clauses. This dissertation addresses a heavy task effect between Experiment 1 and 2, and the possible discrepancy between processing performance and production preferences. Processing models emphasizing the role of memory and syntactic structures as well as the "cooperation" of the two models were also discussed. DOI: 10.5353/th_b5053414 Subjects: Cantonese dialects - Relative clauses