Two interrelated studies serve to test the main hypotheses that (a) metacognitions are internalizable during interactions with a guidance-providing computerized tool when mindful abstraction is induced, and that (b) the improved mastery of metacognitions is transferable to novel ill-structured problems under conditions of induced motivation for greater validity (versus motivation to """"""""satisfice""""""""). A computerized, guiding-providing Writing Partner serves as the studies' main tool. In the first experiment, 130 ninth graders are divided into three groups -- one that writes expository essays guided by the Writing Partner under conditions of induced mindfulness (preparing to teach writing to younger peers) while a second group does the same without that inducement; a third group serves as a control group, writing with a regular word processor. Subjects are further divided into two groups for a delayed posttest measuring the use of metacognitions during the solution of an ill-structured problem. Half the subjects are tested under role-playing conditions, assumed to arouse motivation for greater validity. In the second study, half the subjects of the first study actually teach younger peers to write using the Writing Partner. The other half continue to work with the tool. All subjects are tested for improved writing quality, employment of self-guiding metacognitions during writing, and transfer of these to new ill-structured problems immediately following the termination of each study and again three months later. Within and across study comparisons are carried out to test the hypotheses and to answer a number of open-ended questions. These studies can shed new light on the cultural-artifactual factors in metacognitive development and on the psychological conditions that promote their generalizability to new instances.
