Physical walking is often considered the gold standard for VR travel whenever feasible. However, limited free-space walking areas in the real-world do not allow exploring larger-scale virtual environments by actual walking. Therefore, users often require handheld controllers for navigation, which can reduce believability, interfere with simultaneous interaction tasks, and exacerbate adverse effects such as motion sickness and disorientation. To investigate alternative locomotion options, we compared handheld Controller (thumbstick-based) and physical walking versus a seated (HeadJoystick) and standing/stepping (NaviBoard) leaning-based locomotion interface, where seated/standing users travel by moving their head toward the target direction. Rotations were always physically performed. To compare these interfaces, we designed a novel simultaneous locomotion and object interaction task, where users needed to keep touching the center of upward moving target balloons with their virtual lightsaber, while simultaneously staying inside a horizontally moving enclosure. Walking resulted in the best locomotion, interaction, and combined performances while the controller performed worst. Leaning-based interfaces improved user experience and performance compared to Controller, especially when standing/stepping using NaviBoard, but did not reach walking performance. That is, leaning-based interfaces HeadJoystick (sitting) and NaviBoard (standing) that provided additional physical self-motion cues compared to controller improved enjoyment, preference, spatial presence, vection intensity, motion sickness, as well as performance for locomotion, object interaction, and combined locomotion and object interaction. Our results also showed that less embodied interfaces (and in particular the controller) caused a more pronounced performance deterioration when increasing locomotion speed. Moreover, observed differences between our interfaces were not affected by repeated interface usage.

Continuous locomotion in VR provides uninterrupted optical flow, which mimics real-world locomotion and supports path integration . However, optical flow limits the maximum speed and acceleration that can be effectively used without inducing cybersickness. In contrast, teleportation provides neither optical flow nor acceleration cues, and users can jump to any length without increasing cybersickness. However, teleportation cannot support continuous spatial updating and can increase disorientation. Thus, we designed 'HyperJump' in an attempt to merge benefits from continuous locomotion and teleportation. HyperJump adds iterative jumps every half a second on top of the continuous movement and was hypothesized to facilitate faster travel without compromising spatial awareness/orientation. In a user study, Participants travelled around a naturalistic virtual city with and without HyperJump (equivalent maximum speed). They followed waypoints to new landmarks, stopped near them and pointed back to all previously visited landmarks in random order. HyperJump was added to two continuous locomotion interfaces (controller- and leaning-based). Participants had better spatial awareness/orientation with leaning-based interfaces compared to controller-based (assessed via rapid pointing). With HyperJump, participants travelled significantly faster, while staying on the desired course without impairing their spatial knowledge. This provides evidence that optical flow can be effectively limited such that it facilitates faster travel without compromising spatial orientation. In future design iterations, we plan to utilize audio-visual effects to support jumping metaphors that help users better anticipate and interpret jumps, and use much larger virtual environments requiring faster speeds, where cybersickness will become increasingly prevalent and thus teleporting will become more important.

Using standard handheld interfaces for VR locomotion may not provide a believable self-motion experience and can contribute to unwanted side effects such as motion sickness, disorientation, or increased cognitive load. This paper demonstrates how using a seated leaning-based locomotion interface -HeadJoystick- in VR ground-based navigation affects user experience, usability, and performance. In three within-subject studies, we compared controller (touchpad/thumbstick) with a more embodied interface ("HeadJoystick") where users moved their head and/or leaned in the direction of desired locomotion. In both conditions, users sat on a regular office chair and used it to control virtual rotations. In the first study, 24 participants used HeadJoystick versus Controller in three complementary tasks including reach-the-target, follow-the-path, and racing (dynamic obstacle avoidance). In the second study, 18 participants repeatedly used HeadJoystick versus Controller (8 one-minute trials each) in a reach-the-target task. To evaluate potential benefits of different brake mechanisms, in the third study 18 participants were asked to stop within each target area for one second. All three studies consistently showed advantages of HeadJoystick over Controller: we observed improved performance in all tasks, as well as higher user ratings for enjoyment, spatial presence, immersion, vection intensity, usability, ease of learning, ease of use, and rated potential for daily and long-term use, while reducing motion sickness and task load. Overall, our results suggest that leaning-based interfaces such as HeadJoystick provide an interesting and more embodied alternative to handheld interfaces in driving, reach-the-target, and follow-the-path tasks, and potentially a wider range of scenarios.

Research on self-motion perception and simulation has traditionally focused on the contribution of physical stimulus properties (“bottom-up factors”) using abstract stimuli. Here, we demonstrate that cognitive (“top-down”) mechanisms like ecological relevance and presence evoked by a virtual environment can also enhance visually induced self-motion illusions (vection). In two experiments, naive observers were asked to rate presence and the onset, intensity, and convincingness of circular vection induced by different rotating visual stimuli presented on a curved projection screen (FOV: 54° × 45°). Globally consistent stimuli depicting a natural 3D scene proved more effective in inducing vection and presence than inconsistent (scrambled) or unnatural (upside-down) stimuli with similar physical stimulus properties. Correlation analyses suggest a direct relationship between spatial presence and vection. We propose that the coherent pictorial depth cues and the spatial reference frame evoked by the naturalistic environment increased the believability of the visual stimulus, such that it was more easily accepted as a stable “scene” with respect to which visual motion is more likely to be judged as self-motion than object motion. This work extends our understanding of mechanisms underlying self-motion perception and might thus help to improve the effectiveness and believability of virtual reality applications.

“Spatial updating” refers to the process that automatically updates our egocentric mental representation of our immediate surround during self-motions, which is essential for quick and robust spatial orientation. To investigate the relative contribution of visual and vestibular cues to spatial updating, two experiments were performed in a high-end Virtual Reality system. Participants were seated on a motion platform and saw either the surrounding room or a photorealistic virtual model presented via head-mounted display or projection screen. After upright rotations, participants had to point “as accurately and quickly as possibl ” to previously learned targets that were outside of the current field of view (FOV). Spatial updating performance, quantified as response time, configuration error, and pointing error, was comparable in the real and virtual reality conditions when the FOV was matched. Two further results challenge the prevailing basic assumptions about spatial updating: First, automatic, reflexlike spatial updating occurred without any physical motion, i.e., visual information from a known scene alone can , indeed, be sufficient, especially for large FOVs. Second, continuous-motion information is not, in fact, mandatory for spatial updating---merely presenting static images of new orientations proved sufficient, which motivated our distinction between continuous and instant-based spatial updating.