Anticipation Without Acceleration: Benefits of Shared Gaze in Collocated Augmented Reality Collaboration

Paper Title

Anticipation Without Acceleration: Benefits of Shared Gaze in Collocated Augmented Reality Collaboration

Publication Info

• Topic area: Effects of shared gaze visualization in collocated AR collaboration.
• Keywords: Shared gaze, augmented reality, collaboration, joint attention, task performance, social connectedness, interaction space, user experience.

Background and Problem

• Problem / challenge: While shared gaze has demonstrated benefits in remote collaboration, its impact in collocated AR settings is less understood. Existing studies suggest potential benefits but lack consistent evidence for objective performance improvements.
• Significance: Understanding the role of shared gaze in collocated AR could refine collaborative tools, enhancing coordination and user experience in professional and educational environments.
• Motivation and related work: Prior research highlights shared gaze as a tool for improving joint attention, coordination, and social connectedness in remote and asymmetric XR settings. However, in collocated AR, natural cues like gestures and speech may already provide sufficient grounding, making the added value of shared gaze unclear. This study aims to address this gap by systematically evaluating shared gaze across tasks and interaction spaces.

Solution

• Proposed approach: A systematic investigation of shared gaze visualization in collocated AR, focusing on its effects on task performance, shared attention, social connectedness, and user experience across three task types (puzzle, negotiation, search) and two interaction spaces (plane, room).
• Novelty:
  1. A controlled study isolating the effects of shared gaze from private gaze in collocated AR.
  2. Evaluation across diverse task types and spatial setups to explore task- and space-dependent effects.
  3. Comprehensive analysis of subjective and objective metrics, including joint attention, task completion time, and user preferences.
  4. Insights into user concerns and design implications for shared gaze systems.
• Procedure and key techniques:
  • Conducted a within-subject study with 24 dyads (n=48), varying gaze visibility (private vs. shared), interaction space (plane vs. room), and task type (puzzle, negotiation, search).
  • Measured task completion time, joint attention (via eye-tracking), social connectedness (IOS, GEQ), workload (RTLX), and user preferences.
  • Analyzed gaze data using fixation detection algorithms and statistical models to quantify joint attention.

Results

• Concrete findings:
  1. Shared gaze increased joint attention ratios but did not reduce task completion times.
  2. Participants rated shared gaze as helpful for anticipating actions and reducing misunderstandings, especially in spatial tasks like search and puzzle.
  3. No significant effects of shared gaze on social connectedness (IOS, GEQ) or perceived workload (RTLX).
  4. Participants preferred shared gaze for spatial problem-solving tasks but not for negotiation.
• Advantage over baselines:
  • Shared gaze improved subjective coordination quality (e.g., anticipation, reduced misunderstandings) compared to private gaze.
  • No measurable efficiency gains over private gaze in task completion times.
• Experiments / evaluation:
  • Tasks: Puzzle (path construction), negotiation (decision-making), and search (visual matching).
  • Interaction spaces: Plane (tabletop) and room (3D distributed layout).
  • Metrics: Task completion time, joint attention, social connectedness (IOS, GEQ), workload (RTLX), and user feedback.
• Limitations and future work:
  • Limited to short, controlled tasks in a laboratory setting; results may not generalize to complex, time-critical, or outdoor scenarios.
  • No comparison to a no-gaze baseline; future studies should include this condition.
  • Longitudinal studies needed to assess sustained utility and user comfort over time.
  • Scalability to larger groups and more diverse spatial configurations remains unexplored.

Summary

This study investigated the effects of shared gaze visualization in collocated AR collaboration across diverse tasks and spatial setups. While shared gaze increased joint attention and was subjectively preferred for spatial tasks, it did not improve task completion times or social connectedness. These findings suggest that shared gaze enhances coordination quality rather than efficiency in collocated AR. Future work should explore its applicability in more complex, time-critical, or large-scale scenarios and investigate adaptive systems for task-specific gaze visualization.