Continuum: A Task-Based Evaluation of Temporal Visualisations

Time is a powerful and common attribute for exploring information, and timelines can be an excellent way to help represent and explore temporal data. Most timelines hit problems when the dataset has a large number of values or attributes, or when the data to be conveyed is multidimensional. We previously introduced Continuum, a tool to support rich temporal visualisations by allowing nested hierarchies, cross-concept relationships and meaningful levels of detail at all levels of zoom. Here we present a qualitative and quantitative study of Continuum in comparison with the current state-of-the-art Simile Timeline. The quantitative study investigates performance via three exemplar kinds of temporally oriented tasks: event finding, counting, and comparison tasks. The qualitative components of the study investigate user satisfaction and affect with the tool. Overall, Continuum demonstrated a significant improvement for user experience and accuracy. We consider the attributes of the tool that account for this success towards a generalized approach for temporal visualizations. Author