Providing Feedback for common problems in learning by conceptual modeling using expectation-driven consistency maintenance

Learning by building models and reflecting on them by running simulations is an important paradigm in education. New modeling tools are emerging that take this approach, and focus on learning conceptual knowledge explaining how systems behave and why they do so. These tools are promising in the learning effect they bring about, but progress gets hampered when learners want, but are unable to adjust their model so that the simulation results align with their expectations. In the research reported here, automated support is being developed that helps learners identify the problem(s) in their model. To this end, model-based techniques developed for diagnosing devices are adapted and used to automate the conceptual modeling assistance. This paper describes the solutions developed thus far.