From Tool to Partner: Expressive Behaviors as the Bridge to Human-Robot Creative Collaboration

Paper Title

From Tool to Partner: Expressive Behaviors as the Bridge to Human-Robot Creative Collaboration

Publication Info

• Topic area: Human-robot interaction in creative collaboration.
• Keywords: Human-robot collaboration, expressive robotics, creative partnership, robot behaviors, co-creation, anthropomorphism, intercorporeal communication, partnership formation, creative systems, robot expressivity.

Background and Problem

• Problem / challenge: Current human-robot creative collaboration systems rely on command-response paradigms, treating robots as tools rather than partners. These systems lack adaptive, embodied communication essential for authentic creative partnerships.
• Significance: Transforming robots into creative partners could enhance mutual inspiration, shared agency, and emotional support during collaborative processes, enabling more engaging and effective human-robot interactions.
• Motivation and related work: Previous systems like FRIDA and CoFRIDA demonstrate technical capabilities but fail to address relational mechanisms for partnership formation. Research on expressive robotics has shown promise in social contexts but lacks domain-specific evaluation in creative collaboration.

Solution

• Proposed approach: Investigating how expressive robot behaviors influence human-robot creative collaboration, shifting perception from tool-like interaction to partnership.
• Novelty:
  1. Empirical evidence showing expressive behaviors transform human-robot relationships into creative partnerships.
  2. Identification of user expectations for expressive behaviors in collaborative painting contexts.
  3. Design insights for implementing expressive behaviors that foster partnership formation.
• Procedure and key techniques:
  • Formative study (N=5) to identify user expectations for expressive behaviors.
  • Controlled experiment (N=18) comparing functional and expressive robot conditions in collaborative figure drawing tasks.
  • Implementation of expressive behaviors using Laban Movement Analysis principles and Wizard-of-Oz protocols to ensure contextual appropriateness.

Results

• Concrete findings:
  • Expressive behaviors significantly enhanced partnership perception (r = 0.847), collaborative enjoyment (p < 0.001), and creative motivation (p = 0.011).
  • Participants attributed mental states and intentionality to expressive robots, perceiving them as creative partners rather than tools.
  • Artworks in expressive conditions showed more integrated and intertwined contributions compared to functional conditions.
• Advantage over baselines:
  • Expressive conditions improved collaboration quality across all metrics, including rhythm synchronization (p = 0.003) and team fluency (p = 0.048), while maintaining similar task difficulty and spatial contribution balance.
• Experiments / evaluation:
  • Within-subjects design comparing functional and expressive conditions.
  • Measures included collaboration quality, communication quality, creative experience, and spatial analysis of artworks.
  • Mixed-methods analysis combining quantitative metrics and qualitative insights.
• Limitations and future work:
  • Single-session interactions limit understanding of long-term partnership development.
  • Controlled lab setting may not reflect real-world creative dynamics.
  • Limited expressive repertoire and reliance on Wizard-of-Oz protocols constrain autonomous system scalability.
  • Future work includes multi-session studies, expanded expressive vocabularies, and exploration of negotiated agency in other creative domains.

Summary

This study demonstrates that expressive robot behaviors transform human-robot creative collaboration, enabling the shift from tool-based interaction to genuine creative partnership. Through systematic investigation, expressive behaviors were shown to enhance collaboration quality, creative experience, and emotional engagement, with participants attributing intentionality and agency to the robot. Artworks produced in expressive conditions exhibited more integrated and intertwined contributions, reflecting deeper co-creation dynamics. The findings provide actionable design principles for expressive creative systems, emphasizing transition-focused expression, strategic ambiguity, and maintained user agency. Future research should explore long-term partnership sustainability and expand expressive capabilities across diverse creative contexts.