HieraVisVR: Hierarchical Visual Analytics for Motion-Centric VR Playtesting

Paper Title

HieraVisVR: Hierarchical Visual Analytics for Motion-Centric VR Playtesting

Publication Info

• Topic area: Visual analytics for VR playtesting and motion-based game analysis.
• Keywords: VR playtesting, motion-based games, hierarchical analytics, visual analytics, player behavior, game design, eye-tracking, clustering, data visualization, usability evaluation.

Background and Problem

• Problem / challenge: Current tools for VR playtesting lack standardization, efficient methods for analyzing large datasets, and integrated systems for examining gameplay moments and behavioral patterns. Existing solutions are often domain-specific and fail to support group-based analysis or cross-player pattern discovery.
• Significance: Addressing these challenges is critical for improving the efficiency and effectiveness of playtesting workflows, enabling designers to better understand player behavior and refine game design.
• Motivation and related work: Prior research in game analytics and sports visualization has explored static and dynamic data visualizations but has not adequately addressed the unique challenges of motion-based VR games. Existing tools like Caldera and SCOPE.GG are limited in scope and lack features such as motion data visualization and player grouping.

Solution

• Proposed approach: HieraVisVR, a hierarchical visual analytics framework for motion-centric VR playtesting, which organizes analysis into three stages: exploration, grouping, and explanation.
• Novelty:
  1. A three-stage hierarchical workflow for top-down playtesting analysis.
  2. Integration of motion-based visualizations (e.g., walking paths, eye gaze heatmaps) and clustering techniques for player grouping.
  3. Real-time and static visualizations for detailed player behavior analysis.
  4. Demonstration of the framework across multiple VR applications.
• Procedure and key techniques:
  1. Exploration Stage: Provides a high-level overview of gameplay using replayers with anchor modes, motion-based visualizations, and camera controls.
  2. Grouping Stage: Enables clustering of players based on gameplay features or eye gaze data, with visualizations such as heatmaps and trajectory plots.
  3. Explanation Stage: Allows detailed analysis of individual or paired players using synchronized replays, first-person views, and embedded visualizations like gaze balls.

Results

• Concrete findings:
  • Expert participants rated the system highly for assisting in evaluating game quality (M = 4.60, SD = 0.55) and usability (M = 4.40, SD = 0.89).
  • The explanation stage was the most preferred, followed by the exploration and grouping stages.
  • Heatmaps and gaze ball visualizations were particularly effective for identifying player focus and behavior.
• Advantage over baselines:
  • Unlike manual playtesting and existing tools, HieraVisVR supports motion data visualization, cross-player analysis, and a structured workflow for identifying gameplay patterns.
  • Provides broader applicability across different VR games and training applications.
• Experiments / evaluation:
  • Conducted a formative study with 30 VR practitioners to identify challenges.
  • Evaluated the system with 5 expert game testers using a motion-based VR game (Reflex), with tasks designed to mimic real-world playtesting workflows.
  • Demonstrated applicability in two additional VR applications: a fire evacuation training and an escape room game.
• Limitations and future work:
  • Limited support for real-time annotation and intelligent algorithm suggestions.
  • Overlapping player replays and lack of detailed documentation for clustering algorithms were usability concerns.
  • Future work includes integrating emotion-aware metrics, enhancing event anchor functionality, and expanding the participant pool for validation.

Summary

HieraVisVR is a hierarchical visual analytics framework designed to streamline playtesting workflows for motion-based VR games. It organizes analysis into three stages—exploration, grouping, and explanation—enabling designers to visualize motion data, group players, and analyze behaviors in detail. The system was evaluated through expert studies and demonstrated its utility across multiple VR applications. While effective in identifying gameplay patterns and improving data analysis workflows, future enhancements could address usability issues, expand functionality, and incorporate emotion-aware metrics. HieraVisVR represents a significant step forward in VR playtesting and motion-based game analytics.