VIS Arts Program
Authors: Aaron Hill, Clare Churchouse, Michael Schober
Abstract: In data visualization, the representation of uncertainty and error estimation is often difficult to display effectively. Constraints on the number of dimensions that can be expressed visually as well as limitations of statistical graphing software often lead to data visualizations that inadvertently omit and/or poorly convey the uncertainty and vulnerability of the underlying data. This research is based on more than 400 works of fine art from museum collections and galleries across several countries, curated and analyzed for inspiration and information on effective ways to visually communicate uncertainty, ambiguity, and vulnerability. The selected artworks were chosen for their unique ability to convey uncertainty using a range of approaches and techniques. This paper includes key findings from the analysis, examples of compelling works of art from the research, and an exploration of ways these works can inform data visualization practice, specifically for the visual display of uncertainty.