Metacomprehension monitoring: Cues, criteria, and accuracy.

摘要

Metacomprehension monitoring and control play important roles in reading. Accurate metacomprehension monitoring is crucial to effective metacomprehension control. However, previous research has showed that metacomprehension monitoring is far from accurate. Hence, two key issues were addressed in this thesis: (1) what factors constrain the accuracy of metacomprehension monitoring, and (2) how effective metacomprehension control is.In study one, two experiments were conducted to investigate whether or not the accuracy of metacomprehension monitoring is determined by the interaction among the monitoring judgment, the orienting activity that may provide cues for the judgment, and the criterion tests.The participants first read six expository texts and either generated five keywords or took a pretest on each text. Then they either rated their level of comprehension, or predicted their test scores in a criterion test for each text. Finally, they took a criterion test, by which their monitoring accuracy was measured.The results of the two experiments show that (1) when the criterion test was a multiple-choice test, the readers' monitoring was more accurate if they had predicted their test scores than if they had rated their comprehension. However, when the criterion test was a summary or a concept map task, the readers' monitoring was more accurate if they had rated their comprehension than if they had predicted their test scores. (2) The two types of monitoring judgments were each affected by the two orienting activities in different ways. Before the readers predicted their test scores, their prediction would be more accurate had they taken a pretest than had they generated keywords. But if the readers rated their comprehension, the reverse was found.These results suggest that rating comprehension and predicting test performance may tap different aspects of metacomprehension monitoring. Hence, the view that metacomprehension monitoring is a unitary process is called into question. These results also suggest that concordance among monitoring judgment, monitoring cues, and criterion test is the key for accurate metacomprehension monitoring.In study two, metacomprehension control was compared with regulation based on objective, external feedback. The procedure was the same as study one except that the participants were asked to generate keywords either immediately or at a delay, or not at all. Then, there was a chance to select some of the texts for re-study. The selection was made either by the participants themselves (based on metacomprehension monitoring) or by the computer on the basis of the comprehension test.The results showed that as long as metacomprehension monitoring was accurate (i.e., in the delayed keyword condition), metacomprehension control was as effective as external regulation. However, in the no or immediate keyword conditions in which monitoring accuracy was low, external regulation was more effective. These results suggest that external feedback is a convenient and reliable way to regulate study.The present studies have both theoretical and practical implications. Firstly, they provide an integrated framework for understanding the effects of multiple factors on metacomprehension monitoring accuracy. Secondly, they provide suggestions for metacomprehension research and reading instruction.