Incorporating evolutionary adaptions into the cognitive fit model for data visualization

Abstract

With the proliferation of data has arisen an array of information visualization tools to help humans convert data to information for better decision-making. This ever-expanding visualization toolset has outrun the development of theories of data visualization usage for decision support. In the current research, we contribute to the development of visualization theories in two ways. First, we incorporate our understanding of evolutionary adaptations into the cognitive fit theory to explain why 3-D surface graphs often outperform traditional 2-D graphs in various sensemaking tasks. Drawing on geon theory to interpret the 3-D advantage, we develop the concept of geonic/non-geonic information representations and tasks. We tested and found support for our hypotheses that geonic representations have a speed advantage for the tasks we studied, derived from the human evolved capacity for geographic orienteering, but accuracy depends on the match of representation and task, as geonic or non-geonic. Our experiment compared the geonic representation (3-D surface graph) to non-geonic planar representations (cluster bar, heat map, and line) in geonic (integration, proportion) and non-geonic atomic (acquisition, comparison) tasks. For our second contribution, we build on this result to expand the cognitive fit model to incorporate classic human information processes and evolutionary adaptations. We suggest that the cognitive fit model can expand to include new advances in neuroscience. We discuss the theoretical and practical implications of our experiment and our expanded model.