Three Modalities, Two Design Probes, One Prototype, and No Vision: Experience-Based Co-Design of a Multi-modal 3D Data Visualization Tool
Summary: arXiv:2604.09426v1 Announce Type: cross
The development of three-dimensional (3D) data visualizations is crucial in various STEM fields, from biomedical imaging to spectroscopy. However, these visualizations are often inaccessible to individuals who are blind or have low vision (BLV). To bridge this gap, we initiated an Experience-Based Co-Design project involving BLV co-designers knowledgeable in non-visual data representations. The goal was to create an accessible, multi-modal, web-native visualization tool that meets the unique needs of BLV users.
Methodology
Our research team comprised five BLV individuals and one sighted researcher. We employed a multi-phase methodology that involved two iterative design sessions. These sessions allowed participants to compare a low-fidelity tactile probe with a high-fidelity digital prototype. The iterative approach was instrumental in refining the tool based on direct feedback from co-designers who provided valuable insights into their experiences and preferences.
Key Features of the Prototype
The co-design process yielded a prototype that integrates several empirically grounded features aimed at enhancing accessibility and usability. These features include:
- Reference Sonification: Provides auditory feedback to help users interpret data points effectively.
- Stereo and Volumetric Audio: Utilizes spatial audio techniques to convey information about data relationships and structures.
- Configurable Buffer Aggregation: Allows users to customize data representation through adjustable parameters, enhancing data analysis capabilities.
Core Analytic Tasks
Our research specifically targeted core analytic tasks essential for non-visual exploration of 3D data, which include:
- Orientation: Understanding the spatial arrangement of data points.
- Landmark and Peak Finding: Identifying significant features within the data set.
- Comparing Local Maxima versus Global Trends: Analyzing localized versus overall patterns in the data.
- Gradient Tracing: Following the direction of change in the data.
- Identifying Occluded or Partially Hidden Features: Detecting elements that may not be immediately visible or accessible.
Implications for Future Research
This study provides accessibility researchers and developers with a comprehensive co-design protocol. It emphasizes the importance of translating tactile knowledge into digital interfaces for improved accessibility. Furthermore, it offers concrete design guidance for future systems aimed at extending accessible 3D visualization into embodied data environments.
Conclusion
By involving BLV co-designers in the development process, we have created a multi-modal 3D data visualization tool that significantly enhances analytic accuracy and learnability. This collaborative effort not only advances accessibility in data visualization but also sets a precedent for future research in creating inclusive technologies.