ANALYZING VISUAL REPRESENTATIONS OF CLIMATE CHANGE

Scientists often rely on visual representations to convey complex phenomena such as climate change. Often, these representations are developed first by scientists for academic purposes, and then are modified as visuals to use for instruction and outreach. We tested the hypothesis that careful attention to information design principles would produce visual representations of climate change that were more accessible and better conveyers of information than traditional scientific imagery. Beginning with images developed for the Intergovernmental Panel on Climate Change (IPCC) report, we developed new images, aligned with information design, which we believed would convey important climate change messages more clearly than the original IPCC images. Eye-tracking, think-aloud, and retrospective think-aloud protocols were used to evaluate viewer perceptions and interactions with traditional and redesigned images. We utilized standard eye tracking as well as innovative ArcGIS-based approaches to quantify differences in viewing patterns across image types. We found that: 1) Experts engage with visuals differently than novices, with expert gaze behavior likely driven by underlying conceptual understanding. Novice gaze is often driven by either search efficiency or visually salient features; salient features are not necessarily the conceptually most important features; 2) Redesigned images are more likely to convey messages quickly, as measured by viewing time and statistical comparison of fixation dispersion; and 3) Image preference is strongly dependent upon education and expertise, with experts strongly preferring traditional imagery. These data align with other studies of expert-novice viewing behavior and visual preferences, and suggest that information design should play a larger role in the design of images intended to communicate science to the general public.