Towards an Epistemology of Intelligent Design and Modelling Environments: The Hypothesis Testing Approach

The main purpose of Intelligent Design and Modelling Environments is to offer students the opportunity to acquire knowledge while working on a sequence of given or self-selected problems chosen from the application domain. Earlier, we developed intelligent problem solving environments (IPSEs, Möbus, 1995) for various curricula and applications. Now we extend this approach to intelligent design and modelling environments in the areas of pneumatic circuits, chemical polymerisation reactions, economic simulation games, causal modeling, and diagnosis. On the surface being very different our systems follow a common design theory: the student acquires knowledge by testing his own hypotheses. First we want to show that hypothesis testing plays a fundamental role in a cognitive science orientated theory of knowledge acquisition (ISP-DL-Theory). This theory is the basis of our systems. In a case study three of our most recent systems (PULSE, WULPUS, MEDICUS) and their relationship to hypothesis testing are discussed. Then we define the concept of hypothesis testing in a logic framework. We describe knowledge acquisition events and learning effects. It is argued that knowledge acquisition stimulated by Intelligent Design and Modelling Environments is based fundamentally on self explaining the responses of the system to the student's hypotheses. Furthermore, we will show how the concepts of hypotheses testing and self explanation of the system‘s feedback apply to the four case systems presented. Thus the topic of this paper is the epistemological implication: If there is empirical evidence for the ISP-DL-Knowledge Acquisition Theory which relies on self explanation then Intelligent Design and Modelling Environments should be designed like our systems which stimulate self explanation.