Visualization, Human Systems, Information Systems

We propose to develop and evaluate a social media visualization testbed, using a novel first-principles design approach based on visual thinking algorithms(VTA"s). Much like software design patterns, VTA"s are reusable blueprints for design of new interactive visualizations. VTA's describe the steps in the visual thinking process that capture both operations occurring in a computer and in the human brain. They facilitate design and psychophysical assessment of cognitively effective visualizations that exploit multiple independent visual channels, including size, shape, color, orientation, texture, and motion, to maximize information gain while minimizing cognitive and visual memory workload. Unique features of our approach include 1) extending the VTA methodology to larger workflows that connect multiple smaller VTAs into a dynamic environment for exploratory visual analysis, 2) extending pipeline dataflow models for scientific visualization to social media analysis and 3) applying these techniques to visualize and gain insight into the dynamics of meme propagation and the structure and evolution of social communities within and across multiple social media. This research will result in a flexible visualization framework for rapid implementation of end-user-friendly interactive visual analytic environments from existing analytic and visual components.