Abstract

Publisher Summary This chapter focuses on the way in which new information from successive sentences of a text is integrated with the text representation. The fact that integration is necessary for giving the text cohesion is a familiar notion in the study of rhetoric and in composition manuals. Some investigators pointed out that inferences are often necessary for accomplishing textual integration. New information in a sentence must be connected to a unique antecedent in the representation. Integration of new information is a delayed process taking place at specific locations rather than continuously with each new-argument noun. There are characteristic differences between readers in terms of integration processes. Slower readers tend to integrate at linguistic junctions, whereas fast readers rely on both physical layout and linguistic junctions. Different views of text-level integration, including an immediate versus a delayed integration model, are described in the chapter. The integration theory and the buffer mechanisms that support the delay of integration are also discussed in the chapter. The critical empirical support for the buffering of text-level processes is that reading times increase with the cumulative number of new-argument nouns prior to specific locations, including sentence, clause, and line boundaries.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call