Mapping Real World Locomotion Speed to the Virtual World in Large Field of View Virtual Environments

In virtual environments, tracking physical movements in the real world and mapping them to movement in a virtual world increases immersion and the experience of presence. For example, walking on a treadmill in the physical world may be mapped to camera movement in a first-person view of the virtual world. However, due to interrelated factors relating to the field of view and distortion of objects in the virtual environment, matching physical movement speed to virtual world movement speed world so that it 'feels right' to a user can be complex. This perceived mismatch is detrimental as it can induce motion sickness and reduce the experience of presence. Although previously investigated with head-mounted displays, there is little information about how to overcome this mismatch when using large 2D screens that provide a very different viewing environment. To address this gap, we investigate how a 180-degree display that nearly fills the entire human FOV impacts this perceptual mismatch while walking and running on a treadmill. Our results show that people prefer camera speeds that actually exceed their physical movement speed, and increasingly so at higher speeds. Interestingly, though, people's tolerance for deviations from the ideal camera speed mapping does not change with movement speed. We propose a simple personalized linear model that can be quickly calibrated for a user to provide the best match. This work provides important findings to inform and improve the design of virtual environments for an improved user experience.