Reading problems create difficulties for many learning disabled (LD) students. Little is known, however, about the causes of reading failure among LD students. Researchers have recently focused on the comprehension processes of this population, particularly the metacognitive processes involved in reading.

An important area of investigation is how comprehension monitoring can be facilitated during reading. Research in reading and learning disabilities suggests that LD students do not adequately monitor their comprehension during reading. According to schema theory, the activation of readers' prior knowledge should enhance LD students' ability to monitor their comprehension during reading. This study assessed the effects of pre-reading schema activation and schema presentation tasks on the comprehension monitoring of LD adolescents. Specifically, this study investigated the effectiveness of self-questions, structured overviews + self-questions, and a control condition on subjects': 1) detection of passage errors and 2) accuracy on two statement identification tasks.

Fifty-four LD adolescents in grades 9-12 from public schools in Southwest Virginia participated in this study. All subjects were required to meet specific placement, verbal IQ, reading comprehension and exclusionary criteria before they were included in the study.

Subjects were randomly assigned to three experimental conditions (two pre-reading and one control) . Dependent variables were the number of embedded errors detected and accuracy on two statement identification tasks (SIT). The first set of SIT items measured subjects' recognition of information in the passages. The second set was designed to measure subjects' accuracy on determining the plausibility of information in the passages.

Data were analyzed using a repeated measures MANOVA and one-way univariate tests. Post hoc tests were used to determine which means were significantly different from each other. In addition, the number and types of questions generated, probe data and interview responses were analyzed and reviewed.

Major empirical findings include: 1) subjects in the structured overview and self-questioning condition identified significantly more embedded errors than subjects in either the self-questioning or the control condition; 2) subjects in the control condition recognized less passage information (SIT 1) than subjects in either the self-questioning or the structured overview and self-questioning condition; and 3) no significant differences were found among groups in the identification of statement plausibility (SIT 2) .