Syntactic Prediction in Sentence Reading: Evidence from Eye Movements

Abstract

The current study was conducted to examine syntactic prediction and structural representation of Chinese coordinate structure during the process of Chinese sentence comprehension. We recorded the eye movements of each participant by using SensoMotoric Instruments (SMI) iView Hi-Speed system while they were seated in front of a monitor (resolution of 1024 × 768 pixel). Sixty native speakers of Chinese were recruited, and were paid for their participation. All of them were native speakers of Mandarin Chinese who had normal or corrected-to-normal vision. The experiment materials were composed by 24 critical sentences and other 60 filler sentences. A 2 × 2 factorial design was adopted, with syntactic prediction and temporal ambiguity as within-participant variables. The results from four reading time measures, including the first fixation duration, first pass reading time, go-past time as well as the percent regressions, showed that syntactic prediction could facilitate the processing of Chinese coordinate sentence with local ambiguity. For both types of ambiguous sentences, the occurrence of the word "either" could greatly reduced the time of first fixation duration, first pass reading, go-past, and percent regressions on the critical region and the spillover region, although there was no significant influence of syntactic prediction on the NP region. As revealed by the statistic analysis, there was a significant interaction between these two experimental variables. Moreover, the manipulation of temporal ambiguity was also significant when syntactic prediction was unavailable but not when syntactic processing could be successfully predicted. In contrast with the results of previous English studies, the present study demonstrates that syntactic prediction can facilitate Chinese sentence comprehension and help people deal with ambiguous syntactic constituents. Therefore, in our study, when readers met the sentences with syntactic prediction, they would adopt the sentence-coordination analysis. However, if the sentences were not predicated, they would prefer the NP-coordination analysis and treated the second NP as part of the direct object rather than as the subject of a clause.