The Effects of Implicit Verb Causality and Accentuation on Pronoun Processing
动词隐含因果性和重读对代词加工的影响

Introduction There are many factors influencing pronoun processing in spoken language comprehension, such as implicit verb causality and accentuation. Numerous studies (McDonald & MacWhinney, 1995; Garnham, Traxler & Oakhill, et al, 1996; Debra & Long, 2000; Sun Y, Shu H, Zheng X, et al 2001) have shown that implicit verb causality influenced pronouns resolution. Focus hypothesis and integration hypothesis(Stewart, 2000)both claim that implicit causality makes some candidate antecedent prominent. However, these studies are not consistent on its time course on pronoun processing: on line or off line. In spoken language domain, researchers(Akmajian, 1970; Cahn, 1995; Kameyama, 1999; Venditti,Stone & Tepper, 2001; Wang D & Yang Y, 2004) have studied the effect of accentuation on pronoun processing but have not come to a conclusion. Some researchers claim that stressed pronoun shifts the center of attention, whereas others argue that it only works in parallel-structure sentences. As mentioned above, different prominence of candidate antecedents arisen by implicit verb causality provides us a natural instance for observing the effect of accentuation. The present study aimed to explore the effects of implicit verb causality and accentuation on pronoun processing during spoken Chinese sentence comprehension as well as their time course, using auditory moving window technique (experiment 1) and cross-modal probe method (experiment 2 and 3), by comparing the comprehension time, percentage of choosing NP1, reaction time to probe words between NP1 verb and NP2 verb level, stressed pronoun and unstressed pronoun level. Method One hundred and one paid volunteers participated in the study (30, 20, and 21 in experiments 1, 2, and 3, respectively). In the self-paced listening paradigm, participants listened to the sentences clause-by-clause and interpreted the pronoun. In the cross-modal probe procedure, sentences were presented by earphone and the probe word (NP1, NP2) was presented on screen. In experiment 2, participants reacted at once when the sentence ended. In experiment 3, participants responded as soon as encountering the pronoun. Results The result of experiment 1 showed that percentage of choosing NP1 in NP1 verb condition was significantly higher than that in NP2 verb condition, and that reading time of the second clause in stressed pronoun condition was significantly longer than that in unstressed pronoun condition. The result of experiment 2 showed that in NP1 verb condition, RT (reaction time) to NP1 was significantly shorter than that to NP2, whereas in NP2 verb condition, the probe time was similar. In NP1 verb condition, RT to NP1 in stressed pronoun condition was significantly longer than that of unstressed condition, whereas RT to NP2, the reverse tendency was observed. The results of experiment 3 were similar to those in experiment 2, except that RT to NP2 did not achievesignificant level. Conclusions This study supported the focus hypothesi