Improving Low-Vision Chart Accessibility via On-Cursor Visual Context
New 'Dynamic Context' method reduces task effort by 40% for low-vision individuals reading charts.
A research team from multiple institutions, including Yotam Sechayk, Mark Colley, and Takeo Igarashi, has published a CHI '26 paper introducing innovative methods to solve a critical accessibility gap: making data charts usable for Low-Vision Individuals (LVI). Their work, 'Improving Low-Vision Chart Accessibility via On-Cursor Visual Context,' addresses the fundamental challenge that reading charts requires viewing data points within a global context—something difficult for LVI who rely on magnification or have a restricted field of view. The team conducted a formative study with five LVI to identify four essential contextual elements (axes, legend, grid lines, overview) common across chart types, which informed their technical solution.
The researchers proposed and evaluated two pointer-based interaction methods: 'Dynamic Context,' a novel focus+context interaction, and 'Mini-map,' an adaptation of overview+detail principles. In a controlled study with 22 LVI participants, Dynamic Context demonstrated a significant positive impact on access, usability, and effort reduction, though it came with the trade-off of increased visual load. The Mini-map method strengthened spatial understanding but was less preferred for the task. The findings offer actionable design insights for developers, highlighting the need to balance providing crucial visual context with managing cognitive and visual load in future accessible systems.
- Dynamic Context method showed significant positive impact on access, usability, and effort reduction for LVI in study with N=22 participants.
- Identified four fundamental contextual elements for charts: axes, legend, grid lines, and the overview, based on formative study with five LVI.
- Research provides concrete design insights to guide future development of systems that support LVI while balancing visual load trade-offs.
Why It Matters
Directly addresses a major data accessibility barrier for millions, providing evidence-based tools for inclusive design in analytics and business intelligence.