Design guidelines for animated data visualization based on perceptual capacity limits

Abstract Data visualizations are used widely to help people see patterns in data across research, policy, education, and business. Computer screens allow these visualizations to become animated, which can effectively show processes of change. While animations can be engaging, ineffective design can also make them confusing or overwhelming. We develop new guidelines for designing effective animated data visualizations by reviewing 40 real-world visualization examples, and categorizing the visual tasks people perform when viewing them. These categories include tracking tasks, holistic judgments, and noticing objects added to or removed from a display. We then evaluate the known capacity limits of each task from human motion processing literature and use these to inform design techniques that enable visualizations to respect these capacity limits. Together, the tasks, limits, and corresponding techniques form new, broadly applicable guidelines that should help designers create effective animated visualizations informed by theory of human perception.