Modelling Visuo-Haptic Perception Change in Size Estimation Tasks

Paper Title

Modelling Visuo-Haptic Perception Change in Size Estimation Tasks

Publication Info

• Topic area: Visuo-haptic perception dynamics in size estimation tasks, with implications for HCI and VR.
• Keywords: Visuo-haptic perception, size estimation, haptic drift, visual priming, tangible interaction, virtual reality, sensory integration, perceptual adaptation, feedback loop, HCI.

Background and Problem

• Problem / challenge: Existing research on visuo-haptic illusions has largely focused on short-term effects and static conditions, leaving gaps in understanding how perception evolves over time and how visual priming influences this process.
• Significance: Understanding dynamic visuo-haptic perception is critical for designing effective tangible interactions in VR, where mismatched sensory cues can create novel experiences but also introduce perceptual errors.
• Motivation and related work: Prior studies have explored visuo-haptic illusions like the rubber hand illusion and haptic retargeting, but these have primarily examined static or short-duration interactions. There is limited knowledge about how prolonged exposure, sensory realignment, and visual priming affect perception over time.

Solution

• Proposed approach: A user study with 80 participants to investigate visuo-haptic perception dynamics using a custom-designed device capable of changing size (6 cm to 8 cm) and varying visual priming conditions.
• Novelty:
  1. Demonstrating how size perception drifts over time and stabilizes at an asymptote.
  2. Showing that active, size-changing devices influence perception differently than passive objects.
  3. Highlighting the role of visual priming in calibrating and resetting perception.
  4. Proposing a first-order control system model to describe visuo-haptic perception dynamics.
• Procedure and key techniques:
  1. Participants completed forced choice size estimation tasks in VR under four conditions: fixed-size device without priming, size-changing device without priming, fixed-size device with correct visual priming, and fixed-size device with misleading visual priming.
  2. A custom haptic device dynamically altered its size, and participants interacted with it during acclimation games and estimation tasks.
  3. Data were analyzed using sigmoid fitting to calculate perceptual thresholds and points of subjective equality (PSE).
  4. A mathematical model based on a first-order control system was proposed to describe perceptual drift.

Results

• Concrete findings:
  • Perception of fixed-size objects drifted upwards over time, with a 2.1 mm increase in perceived size for a 6 cm object over 57 minutes.
  • Active size-changing devices caused divergent effects: smaller configurations were increasingly underestimated, while larger configurations were increasingly overestimated.
  • Correct visual priming reduced perceptual drift and improved accuracy, while misleading priming skewed perception but had a lesser impact than correct priming.
  • Perceptual drift followed an exponential pattern, stabilizing at an asymptote, and was modeled as a first-order control system.
• Advantage over baselines:
  • The study extended prior work by examining prolonged interactions and introducing dynamic size changes, revealing new insights into how perception evolves over time.
  • Visual priming was shown to act as a calibration mechanism, reducing perceptual errors and resetting drift.
• Experiments / evaluation:
  • Four experimental conditions with 20 participants each, using a between-subjects design.
  • Tasks included forced choice size estimation, acclimation games, and visual priming.
  • Metrics included PSE, just noticeable differences (JND), and NASA-TLX workload scores.
  • Data were analyzed using sigmoid fitting and a proposed mathematical model.
• Limitations and future work:
  • The study focused on size perception; future work should explore other object properties like weight, texture, and stiffness.
  • Individual differences in perception and the effects of grasping behavior and object properties require further investigation.
  • The proposed model needs validation across broader interaction scenarios and sensory modalities.

Summary

This paper investigates how visuo-haptic perception evolves over time, focusing on size estimation tasks in VR. The study reveals that perception drifts in a predictable manner, influenced by interaction duration, device dynamics, and visual priming. A mathematical model based on a first-order control system is proposed to describe this process. Findings highlight the importance of visual priming as a calibration mechanism and show that active haptic devices impact perception differently than passive ones. These insights have significant implications for designing tangible interactions in VR, particularly for applications requiring prolonged or dynamic haptic feedback.