Trusting Tracking: Perceptions of Non-Verbal Communication Tracking in Videoconferencing

Research Background and Issues

• Issues/Challenges:

  1. While video conferencing improves communication efficiency, its use of cameras raises privacy concerns and social pressures.
  2. Users may unintentionally expose personal living environments and nonverbal reactions during video calls, leading to blurred privacy boundaries.
  3. Prolonged use of cameras can cause "Zoom fatigue," increasing cognitive load.
  4. Current nonverbal tracking tools aim to enhance communication but have yet to fully address users' concerns about privacy and acceptability.

• Significance: Nonverbal communication (e.g., body language, tone of voice, gestures) is crucial for interpersonal understanding and building trust. In the context of the growing prevalence of remote work and online learning, optimizing nonverbal communication tools is of great importance.

• Research Motivation and Related Work:

  1. Various existing systems (e.g., MeetingCoach, CoCo, Emodash) attempt to use nonverbal data to enhance meeting communication but have not thoroughly studied user acceptance of these systems.
  2. Exploring the design of nonverbal communication tracking tools may strike a balance between improving communication and protecting privacy, addressing gaps in current research.

Solution

• Method or Solution: Conduct a global survey (200 participants) and in-depth interviews (20 participants) to evaluate user acceptance of nonverbal data tracking in video conferencing, including which data is acceptable, in what form it should be delivered, and to whom it should be shared.

• Innovations:

  1. A systematic study of global attitudes toward the trade-off between privacy and functionality in nonverbal tracking.
  2. A mixed-methods approach combining quantitative and qualitative methods to guide the design of related tools from the perspective of users across different cultural backgrounds.

• Implementation Steps and Key Techniques:

  1. Survey Design and Distribution: Use Likert scales and other methods to assess the acceptability of nonverbal features and user preferences.
  2. Interview Design and Analysis: Apply open coding and thematic analysis to explore user expectations for designing nonverbal tracking tools.
  3. Scope of Investigation: Define six types of nonverbal features (e.g., collaborative communication, tone of voice, head and body tracking) and four forms of data delivery (real-time feedback, emotion labels, detailed nonverbal cues, and infographics).

Research Findings

• Key Findings:

  1. Users showed higher acceptance of audio data (e.g., tone of voice, background noise) and expressed fewer privacy concerns compared to visual data.
  2. Collaborative communication data (e.g., speaking duration, turn-taking) was the most accepted nonverbal feature, while emotion tracking (especially negative emotions like anger or fear) was widely rejected due to its intrusive nature.
  3. Users were more willing to receive others' nonverbal data than share their own, favoring anonymous and aggregated data formats.
  4. Cross-cultural analysis revealed that while there were minor regional differences, users universally emphasized the need for privacy protection and transparency.

• Advantages: Compared to existing tools (e.g., Emodash), this study focuses on balancing user privacy and needs, providing more specific guidance for the acceptability of nonverbal tools.

• Experiments or Evaluations: Survey data showed that the most acceptable format was anonymous data presented as infographics, particularly when shared with meeting hosts. Some users also expressed a preference for personalized settings to adjust tracking features.

• Limitations and Future Directions:

  1. The study did not fully explore cultural differences in user preferences across all contexts.
  2. The error rate and potential misinterpretations of emotion tracking technologies need further improvement.
  3. Future research could examine the long-term social and psychological impacts of using these tools.

Conclusion and Design Recommendations

• Privacy-First and User-Centric: Nonverbal tools should prioritize privacy protection and enhance user trust through transparent design and user control.
• Simplified Design and Education: Provide clear explanations of tool functionalities and simple tutorials to reduce user apprehension due to unfamiliarity with the technology.
• Contextual Adaptability: Support user customization of features, enabling dynamic activation of different nonverbal tracking functions based on meeting types (e.g., formal meetings, informal discussions).