From Visual Multiplexing to Zebra Maps

Abstract

Visual multiplexing refers to the simultaneous presentation of multiple pieces of information associated with the same spatial location. By leveraging human perceptual and cognitive capabilities, viewers can often de-multiplex multiple visual signals from a shared representation. Despite its widespread presence in visualization, the deliberate use of multiplexing as a design strategy remains underexplored. This poster presents an ongoing investigation of multiplexing through a case study in machine learning workflows for solving partial differential equations (PDEs). In this application context, researchers frequently compare multiple spatially aligned scalar fields, making it difficult to identify local relationships using conventional juxtaposed visualizations. Building upon a collection of multiplexed representations, we explore layered designs that combine heatmaps, contours, glyphs, and a novel Zebra Map representation. Zebra Maps introduce contour-like value boundaries into dense scalar fields while preserving heatmap patterns, enabling more effective comparison between aligned fields. Through collaboration with PDE experts, we continue to refine multiplexing strategies and are developing an interactive visualization environment for constructing and exploring multiplexed visualizations.