Tangible Toolkits for Reflective Systems Modeling

Understanding dynamic systems is an important but difficult task across a wide range of disciplines. Building such intuitions is particularly difficult for novices, for whom traditional representations of systems models are often opaque and confusing. Yet these novices often have deep, situated knowledge that is crucial to building powerful models. This thesis leverages work in the field of tangible interfaces to create an infrastructure for building domain-specific toolkits for models of dynamic systems that are transparent and actively support collaboration in their use. We report on two case studies for which we built toolkits based on this infrastructure. Drawing on these studies, we develop criteria for the design of toolkits that provide support for reflection by novices and experts alike. We further outline a framework for the effective use of reflective systems modeling in both academic and corporate settings. Thesis Supervisor: Bakhtiar Mikhak Research Scientist, MIT Media Lab