Abstract
In this paper, we present an integrated approach to model-based evaluations of interactive prototypes. By combining a state-of-the-art cognitive architecture, Act-R, with an elaborated prototyping tool, Antetype, we enable UX designers without modeling experience to derive quantitative performance predictions for interactive tasks. Using Antetype-Pm, an interface designer creates an interactive prototype and demonstrates the action sequences to complete relevant application scenarios using the monitoring and/or instruction mode of Antetype-Pm. The system learns the interaction paths and predicts the interaction times over trials using Act-R’s symbolic and subsymbolic (i.e. statistical) learning mechanisms. To illustrate the working of Antetype-Pm, an example is provided and contrasted with empirical data.