CAJ | 학술논문

文本阅读中，读者往往对事件的后续发展进行预期推理。预期推理有两种倾向，要么是倾向于根据客观现实条件进行的现实预期，要么是倾向于根据主观的个人意愿进行的意愿预期。3个实验通过自定步调阅读范式探讨了文本阅读中读者产生的现实预期和意愿预期的保持。结果发现，现实预期和意愿预期在长时记忆中的保持情况差异明显，现实预期不能在长时记忆中保持，而意愿预期则能保持；但意愿预期也不能单独存在，会受到现实条件的制约，在受到现实否定后即时消退，不再影响读者的进一步阅读。
문본열독중，독자왕왕대사건적후속발전진행예기추리。예기추리유량충경향，요요시경향우근거객관현실조건진행적현실예기，요요시경향우근거주관적개인의원진행적의원예기。3개실험통과자정보조열독범식탐토료문본열독중독자산생적현실예기화의원예기적보지。결과발현，현실예기화의원예기재장시기억중적보지정황차이명현，현실예기불능재장시기억중보지，이의원예기칙능보지；단의원예기야불능단독존재，회수도현실조건적제약，재수도현실부정후즉시소퇴，불재영향독자적진일보열독。
Predictive inferences （PI） in text reading can have been defined as estimation and expectation to the upcoming matters in the text based on the current reading information. Research on PI in text reading has been one of the hotspots in the field of narrative reading. Numerous researchers have studied on the issues of activation, factors, and representation of contextual PI, coming up with various achievements. Reality factor andpreference factor are two main factors influencing the process of readers＇ PI, hence PI in text reading can be categorized into two different types： reality-PI and preference-PI. Reality-PI are PI to the results based on the objective situation of contextual information, combined with common experience. The readers will predict and infer based on their own emotion and preference, i.e., preference-PI. Previous studies on PI mainly focused on the processing mechanism of reality-PI, few focus on preference-PI. The aim of the present study was to investigate the difference of preservation of these two types of PI in long term memory. Three experiments were designed, using self-paced reading procedure. Experiment 1 and 2 have investigated the preservation of reality-PI and preference-PI in long term memory. Experiment 3 investigated whether preference-PI will vanish by negating the reality condition in the sentence which may trigger preference. Participants were asked to read short narratives firstly and answer a related question after reading each passage. All materials were presented on a computer-controlled monitor. Participants read the passages in a self-paced manner, advancing the text one line at a time by pressing the space bar. Participants were instructed to read carefully. The reaction time of the prime was recorded and analyzed. The results of the present study have suggested that there is obvious difference between the preservation of reality-PI and that of preference-PI in the long term memory. Preference-PI can be preserved in the long-term memory while reality-PI can not. However, readers can no longer preserve the original preference-PI if it is restricted by real-world condition, which has suggested that preference-PI can not solely exist and is restricted by real-world condition. It will vanish if negated by the reality and will no longer affect readers＇ further reading.