This paper introduces the concept of Law Encoding Diagrams, LEDs, a novel class of knowledge representations, that may have significant implications for instruction. A LED is a representation that correctly encodes the underlying relations of a law, or a system of simultaneous laws, in the structure of a diagram by the means of geometric and topological constraints. The diagrammatic nature of LEDs permits exploitation of the benefits that diagrammatic representations confer on problem solving, whilst the underling laws ensure that solutions are nevertheless correct. Many different forms of inference are supported by LEDs, ranging from simple reasoning to complex problem solving. The instructional role of LEDs is that of intermediate representations, which may be adopted as the foundation conceptual entities for expertise. Particular pedagogical issues are considered, including the use of LEDs in intelligent tutoring systems.
N.B. For a revised and extended version of this report see/request: Cheng, P. C.-H. (1996). Law encoding diagrams for instructional systems. Journal of Artificial Intelligence in Education, 7(1), 33-74.