Glanceable Data Visualizations for Older Adults: Establishing Thresholds and Examining Disparities Between Age Groups

Document Title

Glanceable Data Visualizations for Older Adults: Establishing Thresholds and Examining Disparities Between Age Groups

Document Information

• Subject Area: Data Visualization and User Experience for Older Adults
• Keywords: Data Visualization, Older Adults, Smartwatches, Threshold Time, Comparative Tasks, Technology Adoption, Data Interaction, User Studies, Visualization Strategies

Research Background and Problem

• Problem: With the growing older population and the proliferation of portable devices like smartwatches, older adults face perceptual and cognitive challenges in understanding visualized data on small screens, which remain underexplored.
• Significance: Visualized data is widely used to support personal health monitoring and decision-making. A lack of targeted research on data design for older users affects their user experience, particularly in scenarios requiring quick information extraction, such as smartwatch applications.
• Motivation and Related Work:
  • Current research primarily focuses on visualization design effectiveness for younger users.
  • Older adults may face greater difficulties in data interpretation and comparison due to declines in perceptual abilities (e.g., vision) and cognitive abilities (e.g., short-term memory).
  • This study aims to fill this theoretical gap in the intersectional field by extending existing research on health data and interface design to older populations.

Solution

• Proposed Approach: Conduct a perception study with older adults (aged ≥65) to investigate their response times and performance with three visualization designs: bar charts, donut charts, and radial charts, while testing the impact of varying data point quantities (7, 12, 24).
• Innovations:
  • Quantified, for the first time, the time thresholds for older adults processing visualized data on smartwatch screens.
  • Further segmented the older population into "younger-old" (65-74 years) and "older-old" (75 years and above) groups to explore the effects of aging.
  • Compared the performance of younger and older users to propose potential design optimization recommendations.
• Implementation Steps and Techniques:
  1. Experimental Design: Employed a two-alternative forced-choice (2AFC) task where participants compared the magnitude of two data points in visualizations.
  2. Equipment and Procedure: Used the Sony SmartWatch 3 and dynamically adjusted display time to measure reaction performance.
  3. Data Collection and Grouping: Collected threshold data under experimental conditions and conducted comparative analyses (between younger and older users, as well as within older age groups).
  4. Strategy Analysis: Gathered and categorized the strategies participants used during the experiment.

Research Findings

• Key Findings:
  • The older group exhibited significantly slower response times under all conditions compared to the younger group, with the gap widening as the number of data points increased.
  • Donut charts performed best overall, while radial charts performed the worst, particularly with higher data point counts (e.g., 24 points).
  • As data complexity increased (more data points), response times for older adults, especially the older-old group, increased significantly.
• Advantages:
  • Filled a research gap regarding the perceptual performance of older adults with visualized data, providing benchmark data for the field.
  • Offered clear design preference recommendations that can be applied to optimize smartwatch interfaces.
• Experimental or Evaluation Results:
  • Under the donut chart condition, the older group responded the fastest (312ms), while the radial chart condition yielded the slowest responses (5460ms for the older-old group).
  • Older adults performed worse than younger users, primarily due to significant physiological differences such as vision and memory capacity.
  • Older participants tended to use quick scanning strategies and resorted to guessing more frequently under time pressure.
• Limitations and Future Directions:
  • Limitations: The experimental design did not adequately test more complex tasks (e.g., precise value extraction); participants may not fully represent the general older population.
  • Future Directions:
    • Extend the study to other types of visualization tasks, such as dynamic data interaction or health monitoring.
    • Explore the practical impact of wearing smartwatches, such as hand-eye coordination issues.
    • Develop broader design guidelines for older populations, with a particular focus on optimizing multi-data displays and annotation information.