A new tool for measuring and understanding individual differences in the component processes of reading comprehension. 论文

摘要

This thesis proposes a new measure of individual differences in reading comprehension ability that is theoretically motivated, easy to administer, and has high predictive power. Participants read three-sentence paragraphs that describe the relations among a set of real and artificial terms such as: 'A NORT resembles a JET but is faster and weighs more., A BERL resembles a CAR but is slower and weighs more., A SAMP resembles a BERL but is slower and weighs more'. By using the relations described in the paragraph, participants can construct linear orderings (e.g., for speed linear ordering NORT > JET > CAR > BERL > SAMP); however, because the fact that a ' JET is faster than a car' is not explicitly mentioned, participants need to access their existing world knowledge to construct this ordering. After studying the paragraph, participants respond to true-false statements that assess their abilities on four separate component processes of reading comprehension. Text memory statements (e.g., 'A NORT is faster than a JET'.) test memory for information explicitly mentioned in the paragraph; no prior knowledge is required. Text inferencing statements (e.g., ' A SAMP is slower than a CAR'.) test implicit information that can be inferred by combining information that appears explicitly in the text (i.e., 'A BERL is slower than a CAR; A SAMP is slower than a BERL '). Knowledge access statements (e.g., 'A JET is faster than a CAR'.) can be answered by accessing prior knowledge; no new text information is required. Knowledge integration statements (e.g., ' A NORT weighs more than a CAR'.) test implicit information that require participants to access their prior knowledge (i.e., 'a JET weights more than a CAR') and integrate this knowledge with the text fact (i.e., 'A NORT weighs more than a JET'). The components of the task predict performance on a test of global reading comprehension, and on a range of specific comprehension tests, each of which draws more heavily on one particular component. The component processes task is better at predicting reading comprehension than is a typical working memory span task, and has the potential of advancing the understanding and measurement of a range of linguistic and cognitive tasks.