Monitoring Local Coherence Through Bridging Integration.

Two experiments are reported, which are aimed at verifying a main prediction: since the monitoring of the processes that are necessary for correctly maintaining or restoring text local coherence is a significant feature of both text comprehension and writing, educational treatment can improve this type of monitoring in writing. The treatment works directly on reading comprehension and/or on the writing revision phase that consists in detecting incorrect text gaps. The independent variables in Exp.l were three educational sessions that focused on text gaps requiring inference from both prior knowledge and previous text information in order to restore coherence. In Exp.2 as many sessions focused on a revision task concerning peers' written texts containing impossible-to-bridge gaps. The common dependent variable was a local coherence measure applied to written texts produced by participants (18 approximately-12-yr-old children) These experimental participants were matched, based on their writing pre-tests scores, with 18 controls of the same age, and they outperformed the controls on the written post-test. In Exp.2 a writing revision post-test was also used, where experimentals again outperformed controls, but no correlation between the two dependent measures was found. New hypotheses were drawn from this datum. [ABSTRACT FROM AUTHOR]