Scripting Collaborative Problem Solving with the Cognitive Tutor Algebra: A Way to Promote Learning in Mathematics RummelNikol DiziolDejana SpadaHans McLarenBruce M 2007 Interest in developing improved instructional methods for mathematics has increased since TIMSS and PISA. In our project, we introduce a new way to promote knowledge acquisition in mathematics: we enhanced the Cognitive Tutor Algebra, an intelligent tutoring system for learning mathematics in high school that only has been used for individual learning so far, to a collaborative learning setting. Although the Tutor has shown to increase learning, there also are several shortcomings. For instance, learning with the Tutor places emphasis on improving students’ problem solving skills, yet a deep understanding of underlying mathematical concepts is not necessarily achieved. To reduce these shortcomings, we extended the learning environment to a dyadic setting, adding new learning opportunities such as the possibility to mutually elaborate on the learning content. A script was developed to guide students’ interaction and to ensure that students profit from these new learning opportunities. The script structured students’ collaboration in an individual and a collaborative phase and prompted students to engage in fruitful interaction. An adaptive script component provided hints when the dyad encountered difficulties in their problem solving. Finally, dyads engaged in a reflection activity following each problem to improve their collaboration over subsequent interactions. The scripts’ effect on students’ collaborative learning was tested in a classroom study that took place over the course of one week. We compared the learning of dyads collaborating at the Tutor without script support to the learning of dyads scaffolded by the script. To measure different aspects of learning, students solved several post-tests on the Tutor and with paper and pencil. The results of the paper and pencil test assessing improvement of students’ conceptual understanding show significantly better performances of the script condition. The Tutor log data are still being analyzed and the results will be presented at the conference.