抄録
This study examined the effect of working memory on learning from texts. In Experiment 1, participants preformed a word clustering task involving key words from an explanatory text (pretest), and then read the text, which was presented sentence-by-sentence. Next, they performed a second clustering task (post-test), a problem solving task, and a reading span test (RST). The results suggested that the individual differences of the RST scores correlated with the scores for problem solving. In Experiment 2, the results suggested that the individual differences of the RST scores influenced the clustering performance at the level of the situation model when the text was presented all together. Moreover, the result of multiple dimension scaling suggested that the situation models of high-span readers reflected the structure of the text more than those of low-span readers. These results indicate that readers with high reading span scores construct coherent situation models of texts and make use of them for learning from texts.
