What's That Shape? Investigating Eyes-Free Recognition of Textile Icons

Document Title

What's That Shape? Investigating Eyes-Free Recognition of Textile Icons

Document Information

• Subject Area: Human-Computer Interaction (HCI), particularly research on tactile interaction and textile interface design.
• Keywords: Textile interface, textile icons, design recommendations, eyes-free interaction, tactile recognition

Research Background and Problem

• Problem or Challenge:

  • Current home device control faces accessibility issues, such as lost or unreachable remote controls, and the awkwardness of using voice assistants in certain social contexts.
  • While significant research has been conducted on the technical implementation of textile interfaces, systematic guidance for tactile cues in design is lacking.
  • Icons are critical for conveying information in visual interfaces, but how to design tactile icons for textile interfaces remains unclear.

• Significance:

  • Integrating textile interfaces into furniture and home environments can enhance user experience, such as controlling a TV via a sofa armrest or setting scenes through curtains.
  • Supporting "eyes-free interaction" improves interface usability, especially when users cannot see the interface.

• Research Motivation and Related Work:

  • The study focuses on how design factors such as texture, shape, and raised height affect users' tactile recognition of icons.
  • Previous research indicates that simple geometric shapes and significant height differences aid shape recognition, but systematic studies integrating these design guidelines are insufficient.

Solution

• Research Method and Experimental Design:

  • Created 14 different icon shapes, each implemented in 6 variations, including raised, recessed, and flat height configurations, as well as filled and outlined designs, resulting in 84 textile icons.
  • Conducted experiments where participants touched textile icons under visual constraints and evaluated recognition accuracy, time, and user preferences.

• Innovative Aspects of the Solution:

  • Developed a process using medium-density fiberboard (MDF) to integrate raised or recessed icons into textile surfaces.
  • Proposed results-driven design guidelines, such as prioritizing raised icons for tactile recognition scenarios.

• Implementation Steps:

  • Used an automated sewing machine to create flat icons and an MDF framework to produce raised or recessed icons.
  • Conducted user experiments to comprehensively evaluate the effects of design variables on different shapes and variations.

Research Findings

• Specific Findings:

  • Raised icons outperformed flat or recessed icons in tactile scenarios, achieving a recognition accuracy of up to 93.71% and an average recognition time of 6.82 seconds.
  • Simple geometric shapes (e.g., circles, squares, triangles) performed well across all variations, while complex shapes (e.g., lightning bolts) were less recognizable.

• Advantages Over Existing Solutions:

  • Provides design recommendations supported by extensive experimental evidence, whereas existing studies are often limited to small-scale or qualitative research.
  • Offers practical selection guidelines for textile interface designers through the creation of a large-scale textile icon set and user testing.

• Experimental or Evaluation Results:

  • Flat icons performed significantly worse in tactile recognition, with a success rate of only 46.43% and the longest recognition time of 15.61 seconds.
  • Participants expressed significantly higher confidence and comfort with raised icons compared to other design variations.
  • Shape confusion primarily occurred with complex shapes, such as hearts being mistaken for arrows, highlighting the need for more distinct design treatments for complex icons.

• Limitations and Future Directions:

  • The current study used a limited set of icon shapes and collections, which could be expanded to include more complex or widely applicable shapes in the future.
  • The research only covered one type of textile material, leaving performance on other textures and elastic textiles to be explored.
  • The overall design language of textile interfaces and the impact of icon combinations on user perception remain unexamined.

Design Recommendations

• Icon Crafting Choices:

  • Prioritize raised filled or raised outlined icons, as both perform similarly well.
  • If recessed designs are necessary, recessed filled icons are recommended.
  • Flat icons are only suitable for visually explicit scenarios with unique stylistic requirements.

• Icon Shape Selection:

  • Simple geometric shapes (e.g., circles, squares, minus signs) have high recognition rates and are easy for users to identify.
  • Avoid shapes prone to confusion or with high user-reported ambiguity (e.g., lightning bolts, arrows).
  • For complex shapes, emphasize distinguishing features (e.g., protruding points or stitching details at corners).

• Application Design Goals and User Interaction:

  • In eyes-free interaction, icons should have prominent tactile characteristics and consider both functionality and semantics.
  • Different crafting variations (e.g., raised and recessed) can be used to differentiate functional modules or sections.

Conclusion

• This study validates the feasibility of textile icons as elements for eyes-free interaction interfaces through large-scale experimentation, providing specific design recommendations for textile interfaces.
• It offers a design framework to improve tactile recognition performance and clarifies the prioritization of icon crafting and shape selection.
• Future research could expand to more shapes, materials, and application domains to optimize the design language and user experience of textile interfaces.