An experiment with conversation theory as a candidate for a meta learning theory

This paper argues that educational theories are complex and it is their structural complexity that prevents understanding and easy application to learning design. In order to support e-learning, software tools to aid the understanding and application process are more likely to be successful if there is a formal model for describing educational theories -- a computer readable conceptual model which can act as a reference model and provide visual representations of their inherent complex nature. This paper suggests that before it is possible to describe a meta model for learning theories it is necessary to have a more detailed understanding of the nature and form of a meta learning theory. The paper reports on an experiment on using Gordon Pask's Cybernetics based Conversation Theory as a candidate meta learning theory. The paper contributes a UML representation of Conversation Theory and presents experimental comparison data with other theories using the model as the basis for comparison