The application of cognitive diagnosis to the quantitative analysis of simple electrical circuits
The problem of diagnosing performance errors has been of interest to both educationalists and artificial intelligence researchers for many years. Multicolumn subtraction has been a popular domain in which to experiment with the diagnosis of errors. Ohlsson has proposed an approach termed “cognitive diagnosis” which uses a form of heuristic search to find plausible sequences of actions that lead from a multicolumn subtraction task to a known solution.
We have developed an application of cognitive diagnosis to a different domain from the one for which it was originally developed. The system solves simple electrical circuit problems. The reimplementation has raised a number of questions about the cognitive diagnosis technique and its general applicability. We briefly discuss these issues in terms of the future utility of the technique.
Unable to display preview. Download preview PDF.
- 1.P. Brna and J. Howe. Automated diagnosis of misconceptions about electricity: what are the prospects? In Proceedings of the NATO Workshop on Learning Electricity or Electronics with Advanced Educational Technology, Springer-Verlag, forthcoming.Google Scholar
- 2.R.R. Burton. Diagnosing bugs in a simple procedural skill. In D.H. Sleeman and J.S. Brown, (eds), Intelligent Tutoring Systems, pages 157–183, London: Academic Press, 1982.Google Scholar
- 3.A. Caiger. Cognitive Diagnosis of the Solutions to Electrical Circuit Problems. Master's thesis, Department of Artificial Intelligence, University of Edinburgh, 1990.Google Scholar
- 4.M. Caillot. L'Intelligence Artificielle au service de la formation. 1984. Paper presented at FORUM EAO 84.Google Scholar
- 5.S. Hennessy. Why bugs are not enough. In M. Elsom-Cook, (ed.), Guided Discovery Tutoring: A Framework for ICAI Research, chapter 10, pages 173–195, London: Paul Chapman, 1990.Google Scholar
- 6.P. Langley and S. Ohlsson. Automated cognitive modeling. In Proceedings of the 4th National Conference on Artificial Intelligence, pages 193–197, Los Altos, CA: William Kaufman, 1984.Google Scholar
- 7.S. Ohlsson and P. Langley. Psychological Evaluation of Path Hypotheses in Cognitive Diagnosis, chapter 3, pages 42–62. New York: Springer-Verlag, 1988.Google Scholar
- 8.O. Paliès. Diagnostic du fonctionnement cognitif et apprentissage symbolique. In G. Gauthier and C. Frasson, (eds.), Proceedings of Intelligent Tutoring Systems-88, pages 201–206, University of Montreal, 1988.Google Scholar
- 9.O. Paliès, M. Caillot, E. Cauzinille-Marmèche, and J. Mathieu. Student modelling by a knowledge-based system. Computational Intelligence, 2:99–107, 1986.Google Scholar
- 10.R. Plötzner, H. Spada, M. Stumpf, and K. Opwis. Learning Qualitative and Quantitative Reasoning in a Microworld for Elastic Impacts. Research Report 59, Psychological Institute, University of Freiburg, 1990. To be published in the European Journal of Psychology of Education.Google Scholar
- 11.P. Reimann. Towards general knowledge-diagnosis systems for student-and user-modelling. In H. Mandl, E. de Corte, N. Bennett, and H.F. Friedrich, (eds.), Learning and Instruction: European Research in an International Context, pages 349–365, Oxford: Pergamon Press, 1990.Google Scholar
- 12.P.S. Rosenbloom, J.E. Laird, A. Newell, and R. McCarl. A preliminary analysis of the SOAR architecture as a basis for general intelligence. Artificial Intelligence, 47:289–325, 1991.Google Scholar
- 13.K. VanLehn. Student modelling. In M.C. Poison and J.J. Richardson, (eds.), Foundations of Intelligent Tutoring Systems, chapter 3, pages 55–78, Hillsdale, NJ: LEA, 1988.Google Scholar