Navigating User-System Gaps: Understanding User-Interactions in User-Centric Context-Aware Systems for Digital Well-being Intervention

Title of the Paper

Navigating User-System Gaps: Understanding User-Interactions in User-Centric Context-Aware Systems for Digital Well-being Intervention

Bibliographic Information

• Subject Area: Human-Computer Interaction (HCI) and User-Centric Context-Aware Systems for Digital Health Interventions
• Keywords: User-centric, Context-aware systems, User-system technical gaps, Trigger actions, Digital health intervention, Real-time evaluation, Delayed evaluation, Context uncertainty, User mental model, System design
• DOI: https://doi.org/10.1145/3613904.3641979

Research Background and Issues

• Problems and Challenges:
  • Users often encounter issues where the behavior of context-aware intervention systems does not align with their goals and expectations, a mismatch referred to as the "technical gap."
  • Users must perform multiple complex steps to translate personal goals into system rules, potentially facing uncertainties in context mapping and context recognition.
  • The dynamic nature of user goals increases the complexity of system design, making it difficult for systems to fully capture users' actual needs.
• Significance:
  • Understanding these technical gaps is critical for designing user-centric systems, as such gaps affect user experience and may lead to system abandonment.
  • Particularly in the field of digital health interventions, individualized and diverse needs require flexible and reliable system support.
• Research Motivation and Related Work:
  • Providing empowering interfaces allows users to flexibly define rules and customize system behavior, thereby improving system adaptability.
  • While context-aware technologies have been studied extensively in areas like behavior management and digital health, there is a lack of research exploring the usability gaps and user experience in fully customizable intervention systems.

Solution

• Methods and Innovations:
  • Proposed and developed a simplified context-aware mobile intervention system called FocusAid, which provides an interface for users to define context-action rules.
  • Conducted technical audits and prototype development to ensure that rule-setting technically supports users' goal requirements as much as possible.
  • Analyzed two major uncertainties—context mapping uncertainty and context recognition uncertainty—using real-time and delayed evaluation cycles.
• Implementation Steps and Key Technologies:
  1. Rule Design:
     • Provided three generalized context conditions (location, time, activity).
     • Adopted state-based context conditions instead of event-based conditions to reduce user confusion.
  2. Intervention Types:
     • Supported four main behavioral actions: restricting specific app usage, hiding notifications, changing silent mode, and enabling Do Not Disturb mode.
  3. Management Features:
     • Offered rule review functionality to help users clearly understand rule triggers and release conditions.
     • Supported manual activation or deactivation of rules.
  4. User Study:
     • Conducted laboratory studies and three-week real-world experiments to explore user behaviors and adaptation during real-time and delayed evaluation cycles.

Research Findings

• Specific Outcomes:
  • Prototype Development: The FocusAid system enables users to define, manage, and modify rules to meet digital health needs, with targeted features designed to reduce context uncertainty.
  • Insights into User Behavior:
    • Identified two core uncertainties (context mapping and recognition).
    • Users often employed alternative methods to address difficulties in rule articulation.
    • Many users adapted to system design limitations by adjusting rules to better align with their dynamic personal needs.
• Experimental and Evaluation Results:
  • Laboratory studies showed a high accuracy rate in participants correctly programming rules (17.5/21), though they reported a high task load (32.8/100).
  • In field studies, users adjusted personalized rules over the medium to long term, reducing overall usage time (over three weeks) and, in some cases, decreasing disruptive screen unlock frequency.
• Advantages Over Existing Solutions:
  • Provides a more empowering interface, allowing for personalized interventions tailored to different contexts.
  • Focuses on the specific characteristics of the digital health domain, with enhanced support for addressing uncertainties and user mental models, significantly improving the user experience.
• Limitations and Future Directions:
  • Limitations:
    • The current study primarily targets university students, and the limited scenarios affect the generalizability of the results.
    • The existing framework is restricted to the Android platform, with no evaluation of iOS or cross-platform applicability.
  • Future Directions:
    • Expand research to other user groups (e.g., professional settings).
    • Explore enhanced interfaces based on "seamless design" to improve user transparency and cognition.
    • Conduct longer-term studies to validate users' long-term rule adaptation and behavioral changes.