Human-in-the-loop: a Feedback-driven Model for Authoring Knowledge-based Interactive Learning Environments*

While computer-based interactive learning environments can provide authentic and supportive settings for problem-based learning, they are very difficult to build. The traditional development model requires significant upfront development and results in systems that are hard to deploy and customize. In this article, we describe a feedback-driven authoring model that aims to reduce the development difficulty by including teachers in the feedback loop to complement system feedback and incrementally author the content in the learning environment during real use. We discuss the design of INDIE, a learning environment authoring toolkit, for supporting the incremental authoring model. We present empirical results obtained from the development and use of Corrosion Investigator, a learning environment delivered by INDIE, as an example to show how the incremental authoring model is implemented in educational settings. *This work was supported primarily by the Engineering Research Centers Program of the National Science Foundation under Award Number EEC-9876363.