中文摘要：

采用移动窗口技术和探测词再认技术，探讨文本中出现的身份信息不明确的人物（神秘人物）对读者阅读加工过程的影响。实验1对Love，McKoon和Gerrig（2010）实验中的探测信息的位置进行调整，探测词不与神秘人物出现在同一句中，两者间隔一句填充句，以探讨神秘人物的信息是否在工作记忆中形成了阅读焦点，从而对后续阅读过程产生影响。实验2进一步探讨将神秘人物出现句推进长时记忆后，神秘人物是保持在工作记忆中，还是已被推进了长时记忆，但是保持易感性更容易得到恢复。实验结果表明：（1）神秘人物出现后，在随后的阅读中会在工作记忆中维持较高的关注度，影响对后续材料的加工过程；（2）神秘人物出现句被推进长时记忆后，神秘人物也进入长时记忆，但仍然具有较高的通达性，更容易被相关信息所激活恢复。

英文摘要：

In narratives, authors craft their stories by using the mysteries to attract people who read them. In fact, the information authors choose not to relate is precisely what impels readers to continue reading. Previous researchers defined a type of character appeared in the text comprehension without any connection to the rest of the story as mystery character. The Construction-Integration Model （Kintsch, 1988） specifies how a text moves from short-term memory to long-term memory in cycles. In each cycle, propositions from the text enter short-term memory, which are connected to each other by argument overlap, and then the resulting representation is copied into long-term memory. Nevertheless, some propositions that cannot be connected to any of the propositions from the rest of text or the long-term memory are held over to the next cycle in short-term memory. As a result, the propositions, such as mystery characters, remain more available through subsequent text than characters who are not mysterious. Moreover, recycling mystery characters in short-term memory left diminished resources for the processing of subsequent propositions, which resulted in the weaker encoding of the following propositions. Some researchers had provided evidences that a mystery character introduced without information linking him or her to the story affected readers＇ narrative processing. The systematic analysis of previous theories and evidences about the topic that how mystery characters affected readers＇ text comprehension raised another important question that with new information entering into our brain, whether the mystery characters still held available in short-term memory, or it had been pushed into long-term memory with susceptibility. This research was intended to throw light on the question mentioned above. We conducted two experiments. Experiment 1 was designed in a new condition to explore whether a mystery character still remained more available in short-term memory through subsequent text than character who was not