| Abstract: | Tested the hypothesis that readers represent a text's topics and their interrelations as they read and then use those representations to access information about each topic. In 2 experiments, 222 undergraduates were required to read and free recall an expository text of approximately 1,100 words in length. Exp I manipulated both the order of topics in the stimulus text and whether the introductory paragraph stated the topics and their organization. It was found that Ss recalled information about fewer topics if the topics were randomly ordered and the introductory paragraph was uninformative than if topics were logically ordered or if the introductory paragraph was informative. Differences in recall of topics accounted for much of the variance in overall recall and recall errors. Exp II examined the effects of the presence or absence of topic sentences and of variations in the physical marking of paragraph boundaries. Findings show that Ss recalled information about more topics if the text contained topic sentences than if it did not. Results are consistent with the hypothesis that readers use a representation of a text's topic structure to guide recall. (40 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved) |
