Feeling with Many: Rethinking Emotion Regulation with Swarm User Interfaces

Paper Title

Feeling with Many: Rethinking Emotion Regulation with Swarm User Interfaces

Publication Info

• Topic area: Emotion regulation through swarm user interfaces in HCI.
• Keywords: Emotion regulation, swarm user interfaces, participatory design, socially assistive robots, tangible user interfaces, speculative design, human-computer interaction, mental health, swarm robotics, interaction patterns.

Background and Problem

• Problem / challenge: Current digital tools for emotion regulation (ER) predominantly rely on single-agent systems, which fail to replicate the complexity of social dynamics inherent in human emotional experiences. The potential of swarm user interfaces for ER remains underexplored.
• Significance: Emotion regulation is critical for mental health and well-being. Swarm user interfaces could offer novel, socially adaptive interventions that better reflect human emotional complexity.
• Motivation and related work: Prior work on tangible user interfaces (TUIs) and socially assistive robots (SARs) has shown promise for ER but is limited to dyadic, single-agent systems. Swarm robotics research highlights the potential of multi-agent systems to simulate social environments and convey emotional qualities, yet their application in ER is underdeveloped.

Solution

• Proposed approach: The study explores swarm user interfaces as a novel medium for emotion regulation through participatory design workshops, where participants envision and prototype swarm-based systems for personal ER needs.
• Novelty:
  1. Exploration of how users conceptualize swarm user interfaces for ER.
  2. Synthesis of eight interaction patterns that translate user-generated metaphors into robotic behaviors.
  3. Identification of design opportunities and challenges for swarm interfaces in ER.
• Procedure and key techniques:
  • Conducted three speculative participatory design workshops with 15 participants using the Magic Machine Workshop method.
  • Participants created and enacted swarm-based artifacts using craft materials.
  • Data collection included written prompts, photos of prototypes, facilitator notes, and video recordings.
  • Thematic analysis identified contexts of use, swarm metaphors, and interaction modes.

Results

• Concrete findings:
  • Participants envisioned six interaction modes for swarm-based ER: Ambient Presence, Empathic Resonance, Proactive Intervention, Reciprocal Caregiving, Playful Soothing, and Collective Expression.
  • Eight interaction patterns were synthesized, including Greeting Swarm, Advice Swarm, Water Sentinel, Robot Tamagochi, Giant Monster, Cheerleading Squad, Calm Waves, and Breathing Flowers.
• Advantage over baselines:
  • Swarm interfaces offer ontological flexibility, supporting diverse metaphors (e.g., pets, tools, siblings).
  • Multiplicity enables distributed social support and dynamic ambient scaffolding, surpassing single-agent systems.
  • Bridges intrapersonal and interpersonal ER by supporting both cognitive processing and social connection.
• Experiments / evaluation:
  • Workshops involved 15 participants (aged 19–53) from diverse backgrounds.
  • Participants prototyped and performed interactions with swarm robots, generating speculative designs and narratives.
  • Findings were analyzed using thematic coding and triangulation across data sources.
• Limitations and future work:
  • Findings are based on speculative designs rather than real-world interactions.
  • Emotional scenarios were limited to successful regulation episodes.
  • Physical materials (e.g., clay cubes) may have biased participants’ designs.
  • Future work should conduct in-situ evaluations and explore larger-scale swarm embodiments.

Summary

This study investigates the potential of swarm user interfaces for emotion regulation through participatory design workshops. Participants envisioned diverse interaction modes and metaphors, highlighting the swarm’s ability to provide distributed, adaptive, and relational emotional support. Eight interaction patterns were synthesized, offering conceptual frameworks for implementing swarm-based ER systems. While the findings demonstrate the promise of swarm interfaces, challenges such as personalization, social acceptability, and balancing agency with control remain. Future research should focus on real-world evaluations and scaling the designs to larger, more expressive swarm systems.