Rosé, C., Jordan, P., Ringenberg, M., Siler, S., VanLehn, K., & Weinstein, A. (2001). Interactive conceptual tutoring in Atlas-Andes. In J. D. Moore, C. L. Redfield, & W. L. Johnson (Eds.). AI in Education: AI-ED in the Wired and Wireless Future (pp. 256-266). Amsterdam: IOS Press.

The goal of the Atlas project is to increase the opportunities for students to construct their own knowledge by conversing (in typed form) with a natural language based ITS.  In this paper we present the results of a comparative evaluation between a model tracing tutor, the Andes system [9], with the otherwise equivalent dialogue enhanced Atlas-Andes [6].  Andes is a model tracing tutor (MTT) that presents quantitative physics problems to students.  The focus of Andes is to help students develop good physics problem solving skills.  While Andes has been successful at this task, nevertheless, there is ample evidence to suggest that teaching students to solve physics problems is not all that is required to provide them with a solid grounding in physics.  While students in elementary mechanics courses have demonstrated an ability to master the skills required to solve quantitative physics problems, a number of studies have revealed that the same students perform very poorly when faced with qualitative physics problems [13, 12, 11].  Atlas provides Andes with the capability of leading students through directed lines of reasoning that teach basic physics conceptual knowledge, such as Newton's Laws.  The purpose of these directed lines of reasoning is to provide a solid foundation in conceptual physics to promote meaningful learning and to enable students to develop meaningful problem solving strategies.  In this study students using the dialogue enhanced version performed significantly better on a conceptual post-test than students using the standard version of Andes.

For a PDF full article version, click here (1.07MB).

Note: This paper won the "2nd best paper" award of the conference.