Join Me Here if You Will: Investigating Embodiment and Politeness Behaviors When Joining Small Groups of Humans, Robots, and Virtual Characters

Title of the Paper

Join Me Here if You Will: Investigating Embodiment and Politeness Behaviors When Joining Small Groups of Humans, Robots, and Virtual Characters

Paper Information

• Research Area: Human-Computer Interaction, Social Behavior, Virtual Reality
• Keywords: Politeness behaviors, free-form conversational groups, humans, robots, virtual characters, trajectory analysis, group dynamics, social norms

Research Background and Problem

• Identified Issues or Challenges:

  • Politeness and embodiment behaviors are core factors in human-computer interaction, yet how these factors influence decisions to join a group remains unclear.
  • Particularly when interacting with robots or virtual characters, human social comfort, trust, and behavioral choices require further exploration.

• Significance:

  • Understanding how individuals choose to join groups in human-AI interaction scenarios can guide the design of more natural and effective interaction systems, improving the social acceptability of robots and virtual characters.

• Research Motivation and Related Work:

  • While previous studies have demonstrated the positive impact of embodiment on human-computer interaction, its significant influence on task performance remains debated.
  • Prior research has explored the impact of politeness behaviors on social interaction and collaboration in physical and virtual environments, but there has been little focus on how politeness behaviors support new members joining groups.
  • This study aims to fill this research gap by investigating the effects of different embodiment forms and politeness strategies on group-joining behaviors.

Solution

• Research Methods or Solutions:

  • An experiment was designed to evaluate user behavior across three embodiment forms (human, robot, virtual character) and three politeness behaviors (no politeness, indirect invitation, direct invitation).
  • Participants were asked to choose paths to join a group, considering adherence to social norms (avoiding the center of the group) or opting for a more convenient but socially less polite path.

• Innovative Aspects:

  • Combining politeness behaviors with embodiment forms to examine their combined impact on group invitation behaviors and adherence to social norms.
  • Adapting Brown and Levinson's politeness theory by integrating linguistic politeness with non-verbal behaviors for experimental design.

• Implementation Steps and Key Techniques:

  1. The experimental design included 9 conditions (3 embodiment forms × 3 politeness strategies), using controlled experiments to evaluate participants' path choices and perceptions of politeness toward agents.
  2. VR devices were used to record participants' trajectory data, enabling fully immersive interactions with virtual characters.
  3. Six behavioral metrics were measured: persuasiveness, adherence to social norms, path length, path time, distance to the primary agent, and distance to secondary agents.
  4. Statistical analysis (ART method) was employed to compare results across different conditions.

Research Outcomes

• Specific Findings:

  • Politeness behaviors had a more significant impact on persuasiveness and perceptions of politeness than embodiment forms.
  • Direct politeness strategies (e.g., explicit invitations to join a distant group) were the most persuasive but could also be perceived as more restrictive.
  • Participants tended to adhere to social norms, avoiding crossing the central space of the group, even at the cost of additional effort.

• Advantages Over Existing Solutions:

  • This study emphasizes the priority of politeness behaviors in designing robots and virtual characters, rather than focusing solely on embodiment forms.
  • Provides empirical evidence to support the application of robots and virtual characters in public scenarios, including directional guidance and group behavior simulation.

• Experimental or Evaluation Results:

  • Experimental findings:
    • Participants completed tasks faster when interacting with humans, indicating humans are more readily accepted.
    • Virtual characters were less attractive and efficient compared to humans and robots.
    • Path analysis revealed that politeness strategies (especially direct behaviors) significantly influenced participants' choices to take longer paths to adhere to social norms.

• Limitations and Future Directions:

  • Limitations include: cultural differences in interpreting politeness may affect experimental results; only a limited set of politeness strategies and embodiment forms were considered; the study focused on short-term interactions.
  • Future research directions:
    • Investigate the impact of long-term interactions on the effectiveness of politeness strategies.
    • Validate findings across different cultural contexts and practical application scenarios (e.g., healthcare, education).
    • Examine the potential influence of gaze interaction levels, agent size, and formations on social behaviors.

Conclusion

This study highlights the critical role of politeness behaviors in facilitating human-computer interaction, which has not been as widely recognized as embodiment forms. The findings contribute to optimizing the behavioral design of robots and virtual characters, enhancing their effectiveness in social interaction scenarios, and underscore the importance of prioritizing politeness strategies when addressing human-AI interaction challenges.