英语论文网

5
4.4 Summary.......58
Chapter Five Conclusion....60
5.1 Summary of the Study.... 60
5.2 Limitations of the Study and Suggestions for Further Study........62

Chapter Four Results and Discussion

In this chapter, the collected data including the behavioral data and ERPs datawill be analyzed elaborately. The accuracy rates of both probe-word judgment taskand assertive-sentence judgment task are taken into consideration to ensure thevalidity of the experimental results. A thorough discussion is offered, in which thedifferences in construal processing between rhetoric sentences, literal sentences andnonsensical sentences are analyzed in details. The comparisons are conducted withinthree time windows (50-230ms, 230-550ms and 550-1000ms). Twenty subjectsparticipated in this experiment. However, the results of one subject are excludedsince his data is contaminated by artifacts. Finally, 19 subjects’ results, including sixmales and thirteen females, are analyzed.

4.1 Behavioral Data Results and Analysis

The accuracy rates for probe-word judgment task in rhetorical, literal andnonsensical sentences are 0.995, 0.992, and 0.983 respectively. And the accuracyrates for assertive sentence judgment task in rhetorical, literal and nonsensicalsentences are0.955, 0.949and 0.937 respectively. High accuracy rates suggest thatsubjects fully comprehend the words and sentences in this experiment. The meanreaction time of probe words for rhetorical sentences is 1050.48ms, that for literalsentence is 989.59ms, and that for nonsensical sentences is 1090.13ms. Besides themean reaction time of assertive sentence judgment task in rhetorical, literal andnonsensical sentences are 1364.68ms, 1343.39ms and 1362.26ms respectively. Thesefigures can be seen clearly in Table 4.1and Table 4.2.It is obvious that all of these three kinds of sentences have high accuracy ratesin probe-word judgment task and assertive-sentence judgment task, which meanssubjects fully understand the experiment materials and the results of experiment arevalid. As for the reaction time of probe words, the nonsensical sentences are thelongest, while the literal sentences are the shortest. This finding is consistent withour original expectation. We speculate that subjects may respond to the probe wordsin literal sentences more quickly than in rhetorical sentences; Responding to theprobe words in nonsensical sentence may take the subjects the longest time. Becausebased on the ADRIM Model, when understanding the rhetorical sentences, thesubjects need to spend more time extracting the possible features than in literalsentences. In addition, unlike rhetorical and literal sentences, nonsensical sentencesare more difficult to understand and the process of extracting possible feature isblocked, that is to say, possible feature is nonextractable in nonsensical sentences.Therefore, the longest time is spent in understanding probe words in nonsensicalsentences. For further study, A statistic analysis is made by SPSS16.0, Table 4.3illustrates the results.

.........

Conclusion

In this chapter, we will make a conclusion of the whole study. First, we willrefer to the major findings. Then, the limitations of the present study and suggestionsfor future studies are presented.As the first one to study Chinese personification from empirical perspective, thepresent study use ERPs