Art, Affect and Color: Creating Engaging Expressive Scientific Visualization

摘要

As the complexity of scientific data and the needs to communicate the science have grown, the requirements for visualization design and use have become more sophisticated. We increasingly need more effective ways of communicating the science across multiple audiences, including non-experts in the field. The challenges of enriching the representation have moved from the more naive ideas of making it “aesthetically attractive” to more profound constructs of visual language: how to enhance nuances in the data, and how to support more expressive visualizations that elicit different cognitive and communicative affect to tell the science story. In this paper, we describe how artistic color techniques drawn from paintings can be operationally applied to produce more evocative and informative scientific visualization. We illustrate how the color use in a painting can reveal structure and information priority and elicit affect using examples from current work with our scientific visualization colleagues. Our results highlight the value of engaging with artists in long-term, multidisciplinary science teams, but also emphasize the comprehension gaps that exist across the disciplines and the need for methods and techniques that bridge them so they are accessible to a wider range of data scientists. Our color extraction methods and results are a small example of such bridging techniques.