| This dissertation study explored the connections between perspective-taking, language, and reading comprehension outcomes. Little previous work has empirically linked perspective-taking, a metacognitive skill that involves understanding others' reasoning and mental states, to literacy, despite the fact that other metacognitive skills have been connected to language and literacy outcomes. In one-on-one sessions, 52 second- and third-grade students completed (a) a perspective-taking measure, (b) an oral narrative, (c) a vocabulary assessment, and (d) factual and inferential reading comprehension questions. The inferential comprehension questions were further divided by whether or not they contained psycho/social content. Correlational analyses were conducted to determine association among all the variables and multivariate regression models were built to test whether different comprehension types could be predicted by perspective-taking skills. To capture students' reasoning about others' perspectives, the Strange Stories vignette measure was used and two aspects were analyzed: accuracy and mental state expression. Narratives were also analyzed for inclusion of mental state information. Overall, vocabulary proved to be highly related to both perspective-taking and comprehension outcomes. Higher vocabulary scores were associated with better performance on the Strange Story accuracy scale, although not with greater use of mental state terms. In the narrative task, children with higher vocabulary skills produced longer utterances but did not necessarily score higher on other measures of length, structure, or perspective-taking content. Performance on the Strange Stories was not related to mental state expression in narrative, suggesting that the two measures may capture different aspects of metacognition. For comprehension, scores on inference questions were not related to scores on fact questions nor were the two types of inference related. As expected, children scored better on the fact questions compared to the inference questions. Contrary to expectations, inference questions that contained psycho/social content proved to be easier for students than those that did not. Across all comprehension types, vocabulary and fluency were the best predictors, however it appears that perspective-taking skill may play a mediating role between vocabulary and comprehension outcomes. Conclusions are presented about perspective-taking and its role in comprehension, children's mental state expression, different types of comprehension, and future work in this area. |
