Understanding Wheelchair Users' Preferences for On-Body, In-Air, and On-Wheelchair Gestures

Title of the Paper

Understanding Wheelchair Users’ Preferences for On-Body, In-Air, and On-Wheelchair Gestures

Paper Information

• Research Domain: Assistive technology design, interaction for users with motor impairments, gesture-based interaction
• Keywords: Gesture input, wheelchair users, motor impairments, mobility impairments, gesture elicitation, in-air gestures, on-body gestures, wheelchair gestures, motor impairment studies, assistive technology

Research Background and Problem Statement

• Identified Issues or Challenges:

  • Gesture input typically requires a high level of motor ability, which may not be applicable to users with motor impairments, resulting in significant accessibility challenges when using touchscreens and gesture interfaces.
  • Existing research primarily focuses on able-bodied users and touch-based input, with limited studies on gestures performed by wheelchair users in their personal, in-air, or wheelchair spaces.

• Significance of the Research:

  • Exploring wheelchair users’ preferences for different types of gestures and the relationship between gestures and motor impairments can contribute to designing more user-friendly interaction systems for this group.

• Motivation and Related Work:

  • Wheelchairs are common assistive devices, and wheelchair users may have unique gesture preferences due to their specific motor function levels.
  • Most existing research on in-air and on-body gestures targets able-bodied users, while studies on gesture input within the "chair space" of wheelchair users are extremely limited.

Proposed Solution

• Methodology or Proposed Solution:

  • Conduct a gesture elicitation study by inviting 11 wheelchair users to design suitable gestures for 21 common commands (e.g., turning on the TV, playing music).
  • Analyze the types of gestures proposed by users (on-body gestures, in-air gestures, and wheelchair gestures) and their relationship with self-reported motor impairments.

• Innovative Aspects of the Solution:

  • This study is the first to systematically investigate wheelchair users’ preferences for on-body, in-air, and wheelchair gesture inputs.
  • Proposes ability-based design principles to support these input types, enabling personalized gesture sets tailored to users’ capabilities.

• Implementation Steps and Key Techniques:

  1. Select 21 system function referents covering operational, content access, and navigation commands.
  2. Invite participants to design a suitable gesture for each referent.
  3. Extract gesture characteristics (e.g., gesture location, type, range of motion).
  4. Analyze the relationship between gestures and participants’ self-reported motor impairments (e.g., low strength, rapid fatigue).
  5. Calculate consistency across users’ gestures and perform data modeling.

Research Findings

• Specific Results:

  • Gesture Preferences:
    • Participants showed a clear preference for on-body gestures (47.6%) and in-air gestures (40.7%) over wheelchair gestures (11.7%).
    • Most gestures were performed using one hand (78.3%), with touch-based output being the most common (34.2%).
  • Consistency Analysis:
    • Agreement rates for gestures across users for the same referent were very low (≤5.5%), highlighting highly personalized gesture needs.
  • User Self-Reported Evaluations:
    • Participants rated their gestures as easy to perform (average score 6.83/7), easy to remember (5.48/7), and socially acceptable (6.42/7).

• Advantages and Innovations:

  • This study is the first to systematically reveal the characteristics of gestures performed by wheelchair users within the "chair space," providing empirical support for ability-based gesture design.
  • Offers design recommendations to support the coexistence of on-body, in-air, and wheelchair gestures, enhancing flexibility and adaptability in design practices.

• Limitations and Future Directions:

  • Limitations:
    • Small sample size (11 participants), which may limit the generalizability of the findings.
    • Gesture data was not recorded using computational methods, restricting feature modeling and evaluation.
  • Future Directions:
    • Increase sample diversity and size to study the impact of cultural backgrounds on gesture preferences.
    • Integrate machine learning methods to enable adaptive modeling of gesture recognition systems.
    • Explore complementary use of gesture input with devices such as smartphones to enhance user experience.