作者: Shravan Vasishth , Zhong Chen , Qiang Li , Gueilan Guo
DOI: 10.1371/JOURNAL.PONE.0077006
关键词: Syntax 、 Posterior probability 、 Subject (grammar) 、 Biology 、 Relative clause 、 Natural language processing 、 Space (commercial competition) 、 Object (grammar) 、 Grammar 、 Artificial intelligence 、 Psycholinguistics 、 General Biochemistry, Genetics and Molecular Biology 、 General Agricultural and Biological Sciences 、 General Medicine
摘要: A general fact about language is that subject relative clauses are easier to process than object clauses. Recently, several self-paced reading studies have presented surprising evidence relatives in Chinese relatives. We carried out three experiments attempted replicate these results. Two of our found a subject-relative preference, and the third study an object-relative advantage. Using random effects bayesian meta-analysis fifteen (including own), we show overall current for advantage quite strong (approximate posterior probability given data: 78–80%). argue retrieval/integration based accounts would difficulty explaining all experimental These findings important because they narrow theoretical space by limiting role class explanation—retrieval/integration cost—at least clause processing Chinese.