Ready for the Touch: Exploring Users' Perceived Transparency of Robot Pre-Touch Cues

Paper Title

Ready for the Touch: Exploring Users' Perceived Transparency of Robot Pre-Touch Cues

Publication Info

• Topic area: Human-Robot Interaction (HRI), focusing on pre-touch communication in physical interactions.
• Keywords: Human-robot interaction, perceived transparency, robot touch, pre-touch cues, user experience, perceived safety, physical human-robot interaction, multimodal communication, design guidelines, eye-tracking.

Background and Problem

• Problem / challenge: Current research on human-robot interaction (HRI) lacks a systematic understanding of how robots can effectively convey their intent before physical touch. Ambiguous or absent pre-touch cues can lead to user confusion, anxiety, and reduced perceived safety.
• Significance: Ensuring clarity in robot pre-touch communication is critical for improving user trust, safety, and acceptance in physical human-robot interaction (pHRI), which is increasingly relevant in industrial, medical, domestic, and social contexts.
• Motivation and related work: While transparency in HRI has been explored in domains like explainable AI and autonomous vehicles, little attention has been paid to transparency in close-proximity, physical interactions. Existing studies often focus on the touch itself rather than the pre-touch phase, leaving a gap in understanding how to design effective pre-touch cues.

Solution

• Proposed approach: The study investigates the impact of nine robot pre-touch cues on perceived transparency and safety using a mixed-methods approach, including subjective ratings, eye-tracking, and interviews.
• Novelty:
  1. Systematic evaluation of diverse pre-touch cues across multiple contexts.
  2. Identification of the relationship between perceived transparency and perceived safety.
  3. Development of evidence-based design guidelines for context-aware robot touch communication.
  4. Introduction of a dynamic roadmap for adapting pre-touch cues based on user familiarity and interaction phase.
• Procedure and key techniques:
  • Conducted a pre-investigation survey with 340 participants to identify user preferences for pre-touch cues.
  • Designed a video-based experimental study with 41 participants to evaluate nine pre-touch cues (e.g., Video Display, Voice Prompt, Beeping Sound).
  • Measured user experience using subjective questionnaires, eye-tracking data, and semi-structured interviews.
  • Analyzed the effectiveness of cues based on perceived transparency, safety, and context suitability.

Results

• Concrete findings:
  • Semantically explicit cues like Video Display (VD) and Voice Prompt (VP) achieved the highest perceived transparency and safety.
  • Ambiguous cues like Air Puff (AP) and Beeping Sound (BS) performed poorly, often causing confusion and anxiety.
  • A strong positive correlation (r = 0.70, p < 0.001) was found between perceived transparency and perceived safety.
  • Task-oriented cues were rated higher in industrial and medical contexts, while emotion-oriented cues were preferred in social and domestic settings.
• Advantage over baselines:
  • VD and VP significantly outperformed the no-signal control condition in both transparency and safety metrics.
  • Multimodal cues combining visual and auditory elements were more effective than single-modality cues.
• Experiments / evaluation:
  • Participants (n=41) evaluated nine pre-touch cues in a controlled video-based setup.
  • Eye-tracking data revealed that semantically rich cues effectively captured and guided user attention.
  • Semi-structured interviews provided qualitative insights into user preferences and contextual needs.
• Limitations and future work:
  • Limited ecological validity due to video-based methodology; future studies should involve live interactions with physical robots.
  • The industrial appearance of the robot constrained exploration of diverse scenarios.
  • Future research should explore richer sensory feedback (e.g., vibration, temperature) and target diverse user populations.

Summary

This study systematically investigates the role of robot pre-touch cues in enhancing perceived transparency and safety in physical human-robot interaction. Semantically explicit cues like Video Display and Voice Prompt were found to significantly improve user understanding and acceptance, while ambiguous cues like Air Puff caused confusion. The findings emphasize the critical role of perceived transparency in shaping user safety and propose a dynamic roadmap for designing adaptive pre-touch cues based on user familiarity and context. These insights contribute to the development of safer and more trustworthy robotic systems for diverse applications.