Elementary school children generally do not revise frequently or skillfully in the classroom. Two studies were conducted to learn if children's ability to revise problematic texts could be facilitated through training in a comprehension monitoring strategy. In the first study, third-and sixthgrade children who were trained in a self-questioning text-evaluation strategy located and revised significantly more target text problems than did control children. The goal of the second study was to compare the effects of prior exposure to problematic texts and self-questioning strategy training. The results showed that a combination of the two approaches was most effective in increasing third graders' revision scores. The results from both studies show that acquiring a strategy for evaluating the comprehensibility of a text can help children make appropriate revisions to improve that text's communicative quality.

No abstract

Two experiments were done to test the hypothesis that beginning readers (firstgrade children) could evaluate the referential-communicative adequacy of simple, two-word messages better if they saw them written out while hearing them spoken than if they only heard them spoken. Oral-plus-written messages did prove significantly easier for the children to evaluate accurately than did oral-only ones. They were also easier to evaluate than control oral-plus-written messages, in which the words were written as two illegible scribbles rather than printed clearly. This facilitation effect was equally strong whether the legible written message remained visible during message evaluation or was erased almost immediately after being written. Reading the message apparently did not improve message evaluation by improving message recall: Message evaluation and message recall were uncorrelated. The results seem consistent with Olson's theory that learning to read and write helps children attend to and analyze the literal meaning of a message.This research was supported by National Institute of Child and Human Development Grant HD 09814.We are grateful to the children, teachers, and parents whose cooperation made these studies possible. We also thank Sophia Cohen and four anonymous reviewers for their helpful criticisms and suggestions.Requests for reprints should be addressed to John H.

Two studies were conducted to examine the effect of children's social cognitions on their ability to monitor their comprehension in the referential communication paradigm. Specifically, it was hypothesized that children's beliefs about the intentions and cooperativeness of speakers prevent them from accurately evaluating messages. In Experiment 1, it was found that children who were made aware of the possibility that speakers themselves (rather than simply their words) might refer to more than one referent, were significantly better able to detect referential ambiguity than were children who, as in traditional referential communication studies, assumed that speakers attempted to describe a single referent. In Experiment 2, it was found that children faced with potentially uncooperative or dishonest speakers, carefully examined the words of referential directions and therefore detected more problems with those directions than children faced with honest, if potentially incompetent speakers. The results are discussed in terms of the interaction between the social and linguistic aspects of communication.A metalinguistic development that has received a good deal of attention recently is the phenomenon of comprehension monitoring. Specifically, between the ages of five and ten children become increasingly aware of how well they have understood verbal input. This awareness is usually assessed in the referential communication paradigm (Bonitatibus 1988a). Children * The authors wish to gratefully acknowledge the ideas, support and technical assistance of Carole Beale. We are also thankful for the assistance of the students, teachers, and parents of the Maple Avenue Elementary School, without whose cooperation this work would have been impossible. Thanks also to Lowell Roberts and Michael Carrier for assistance with data analysis.

No abstract