文本阅读中背景信息的加工过程：激活与整合

采用移动窗口技术和眼动技术探讨了文本阅读中当前信息跟背景信息的具体加工过程。被试为华南师范大学本科生92名。实验1使用移动窗口技术，计算机屏幕上每次只呈现一个句子，让被试自己按键逐句进行阅读，通过分析不同实验条件下探测词的再认反应时间和不同实验条件下目标句的阅读时间来探讨文本阅读中信息加工的具体过程。实验2使用眼动技术在一种更自然的情境中通过分析不同实验条件下眼动指标的差异来进一步探讨文本阅读中信息加工的具体过程。实验结果表明，文本阅读中背景信息的加工过程包括激活和整合两个阶段，这两个阶段相互独立但又紧密联系，激活是整合发生的前提，有整合必先发生激活，但激活发生后并不一定会发生整合；文本阅读过程中读者阅读时间的延长主要发生在整合阶段

The Course of Background Information Processing in Text Reading: Activation and Integration

Introduction Text reading is one of the most complex and unique cognitive activities of human beings and is also an important way for us to get information. The process of background information in text reading has been a hot point in experimental psychology for a long time. According to the memory-based text processing view, readers activate the background information and integrate it with the current information by resonance in text reading. However, what is the course of background information processing? In our view, the course of background information processing in text reading includes two phases - activation and integration. The main purpose of the present study was to investigate whether the two phases existed and how they were associated. Method A self-paced, line-by-line, reading paradigm was used in experiment 1. In experiment 1a, 32 university students were asked to read 12 narrative passages and judge whether a probe word appeared in the previous section of the text. The relation between the elaboration on a characteristic of the protagonist and subsequent target action carried out by the protagonist was the first independent variable, which had two levels: consistent version and qualified version. The position of the probe word was the second independent variable, which also had two levels: previous the target sentence and behind the target sentence. The reaction time and the accuracy of the probe word were the dependent variables. In experiment 1b, 30 university students were asked to read 12 narrative passages. The relation between the elaboration on a characteristic of the protagonist and subsequent target action carried out by the protagonist was the independent variable, which had three levels: consistent version, qualified version and control version. The reading time of target sentence which describing the actions of protagonist was the dependent variable. An eye-monitoring procedure was used in experiment 2. Thirty university students were asked to read 12 narrative passages in which, a second target sentence, which was summarily to the first target sentence - the target sentence in experiment 1. was added in the latter part of the passage. Each passage was presented on one screen. The relation between the elaboration on a characteristic of the protagonist and subsequent target action carried out by the protagonist was the independent variable, which had two levels: consistent version and qualified version. The dependent variable was the first-pass reading time, second-pass reading time, total-pass reading time, regressions out and regressions in on the three key regions including the section of the elaboration on a characteristic, the first target sentence and the second target sentence. Results In experiment 1a, the reaction time for the probe word behind target sentence was shorter than that for previous target sentence in both the consistent and qualified condition. There were no differences between the accuracy of the probe word behind target sentence and that for previous target sentence in both the consistent and qualified condition. In experiment 1b, there were no differences between reading time in the consistence condition and that in the control condition, but the reading time in both the consistence and the control condition were shorter than that in the qualified condition. In experiment 2, for the section of the elaboration on a characteristic, there were no differences between the consistence and qualified consistence conditions in the index of first-pass reading time, but the second-pass reading time, total-pass reading time and regressions-in in the qualified condition were longer than those in consistence condition. For the first target sentence, there were no differences between the consistence and qualified conditions in the index of first-pass reading time, but the second-pass reading time, total-pass reading time and regressions out in the qualified condition were longer than those in consistence condition. For the second target sentence, there were no differences between the consistence and qualified condition in all indexes. Conclusions The results showed that background information processing consisted of activation stage and integration stage. Activation occurs before integration. Though integration must take place based on activation, not all information activated leads to integration. Furthermore, the delay of reading time during text reading mainly happened in the integration stage