英语论文网

Reflexives and pronouns should have comparable distributions in terms of interpretive preferences, according to a binding theory prediction. In (2a), “himself” can therefore relate to “Ben,” but “him” cannot. In other forms, such as (2b) though, this complementarity seems to fail or is at least less apparent. 

Chapter Three METHODOLOGY .......................... 31

3.1 Research Questions ........................ 31

3.2 Experiment 1: Pronoun Resolution in Coargument  Contexts ...... 32

Chapter Four RESULTS AND DISCUSSION .................... 45

4.1 Results ............................ 41

4.1.1 Results of Experiment 1: Pronoun Resolution in  Coargument Contexts ............................. 45

4.1.2 Results of Experiment 2 ............................ 48

Chapter Five CONCLUSION ....................... 69

5.1 Major Findings ........................ 69

5.2 Implications .................................... 69

5.3 Limitations and Suggestions for Further Studies ............................. 71

Chapter Four RESULTS AND DISCUSSION

4.1 Results

4.1.1 Results of Experiment 1: Pronoun Resolution in Coargument Contexts

The lmerTest package in R and linear mixed-effects models were used for the analysis. In Cunnings and Sturt’s study (2018), they conducted a single analysis for each reading time measure that included both analysis regions but treated region as a fixed effect in order to minimize the number of comparisons across regions. This is in contrast to prior work in which each measure at each region is analysed independently, which may increase the false positive discovery rate (von der Malsburg & Angele, 2017). Comparing this approach to analyzing each region separately, Cunnings and Sturt believe it has a number of advantages. The first benefit is that it increases the power to detect impacts that might be uniform but insignificant across regions. Second, it offers a mechanism to specifically examine time-course effects.Finding a specific reading time pattern at the critical zone and a distinct pattern at the spillover region, for instance, has allowed many earlier studies to draw conclusions regarding the time-course. Generally, no direct statistical testing has been done to determine the reliability of such time-course effects. The ability to explicitly assess potential time-course impacts across regions is made possible by include region as a fixed effect.

Chapter Five CONCLUSION

5.1 Major Findings

We observed that readers preferred to resolve a pronoun as referring to a structurally nonlocal antecedent in coargument and PNP eye-movement trials. Participants did occasionally take into account local antecedents for pronouns in an offline antecedent choice task, particularly when the pronouns were in possessed picture noun phrases. We contend that in a picture noun phrase, our results indicate that pronouns in both coargument and picture noun phrase contexts preferentially cue the retrieval of nonlocal antecedents. However, this 本论文由英语论文网提供整理，提供论文代写，英语论文代写，代写论文，代写英语论文，代写留学生论文，代写英文论文，留学生论文代写相关核心关键词搜索。