Related Experiment Video
Updated: Oct 11, 2025
06:38
Using a Virtual Reality Walking Simulator to Investigate Pedestrian Behavior
Published on: June 9, 2020
5.0K
Vision-based speedometer regulates human walking
Shinya Takamuku1, Hiroaki Gomi1
1NTT Communication Science Laboratories, Nippon Telegraph and Telephone Corporation, 3-1 Morinosato-Wakamiya, Atsugishi 243-0198, Kanagawa, Japan.
Iscience
|November 29, 2021
Summary
Humans can estimate self-motion speed using vision. This study reveals a visual "speedometer" that integrates depth and motion cues, robustly regulating walking speed despite varying object depths.
Area of Science:
- Neuroscience
- Vision Science
- Biomechanics
Background:
- The human visual system estimates self-motion direction from optic flow.
- It remains unclear if visual perception also estimates self-motion speed.
- Estimating speed requires separating it from object depth information.
Purpose of the Study:
- To investigate if the human visual system can estimate self-motion speed.
- To determine the cues used for visual self-motion speed estimation.
- To test the robustness of visual speed regulation against depth variations.
Main Methods:
- An automatic visual regulator of walking speed was developed.
- The regulator's robustness to changes in object depth was tested.
- Interocular distance and monocular depth cues were manipulated.
Main Results:
- The visual speed regulator demonstrated robustness to changes in object depth.
- This robustness was not explained by temporal-frequency-based speed coding.
- The system's performance degraded with manipulated interocular distance and deceptive monocular cues.
Conclusions:
- The human visuomotor system likely contains a speedometer for self-motion speed.
- This speedometer integrates both monocular and binocular depth with motion cues.
- This mechanism allows for accurate self-motion speed perception independent of object distances.
