Astral body
More actions
| Astral body | |
|---|---|
| ??? | |
| Information | |
| Type | Seated, Standing, Room-scale |
| Required Devices | Gamepad or VR controller with thumbstick |
| Description | First person when you need to interact with the environment, Third person when you need to run around in the environment |
| Pros | Easy to navigate and interact, clean |
| Cons | Immersion breaking for some |
| Creator | ??? |
| Examples | Alice Mystery Garden |
Astral body is a Locomotion Method. In virtual reality (VR), the user's viewpoint leaves their avatar and moves to a third-person perspective for travel, then returns to a first-person view when the user needs to interact with the environment.[1] In the version studied by accessibility researchers, the user steers the avatar with a controller thumbstick while watching it from above.[2]
The technique is one of the 109 records in Locomotion Vault, a database of VR locomotion techniques from academia and industry published at CHI 2021 by Massimiliano Di Luca, Hasti Seifi, Simon Egan and Mar Gonzalez-Franco.[3] The database places it in its "controller" category and gives the 2017 VR puzzle game Alice Mystery Garden as its example.[1] Researchers at the University of Washington later used Locomotion Vault to select Astral Body as one of six seated techniques in a 2023 accessibility study with participants who have upper-body motor impairments, where it was among the three best-performing techniques.[2]
Additional information
First person when you need to interact with the environment. Third person when you need to run around in the environment. Similar to an out-of-body experience, your camera gets pulled out to third person when you need to navigate but you may fly back into your body when you need to interact or socialize.
Locomotion Vault describes the same split: the user, or the experience itself, changes the viewpoint from first person to third person, with first person used for interacting with the environment and third person used for moving the avatar.[1]
How it works
In the implementation tested by Rachel L. Franz, Jinghan Yu and Jacob O. Wobbrock, the participant sees their avatar from a third-person perspective looking down on the environment and pushes the thumbstick in the direction the avatar should travel. Both forward movement and direction come from the thumbstick, so the technique needs one hand and, in the authors' classification, only thumb movement and a low level of effort.[2] The other first-person techniques in that study added a thumbstick-driven stationary turn in 45-degree steps; Astral Body did not need one because the user controls the avatar from outside it.[2]
The authors note that the original implementation, which they identify as Alice Mystery Garden, let the user switch between third-person and first-person perspectives. They disabled that switch in their study, because first-person movement could count as a different technique and would have made trials less comparable between participants.[2]
Use in games
Alice Mystery Garden is a single-player VR action-puzzle game by AMG Games, released on Steam for Windows on 20 December 2017. It is VR-only and supports tracked controllers.[4] The player controls the character Alice while switching between a top-down "player perspective" and Alice's in-world view. From the overhead view the player can check the route to the goal and grab and place blocks to build new paths for Alice, for example a bridge over a gap; Alice's own view reveals caverns that cannot be seen from above. The goal of each of its 17 stages is to collect five looking-glass fragments that unlock a gate.[4] Both Locomotion Vault and the 2023 study by Franz and colleagues point to this game as the source of the technique.[1][2]
Locomotion Vault classification
Locomotion Vault codes each of its 109 techniques with the same set of implementation and user-experience attributes; the user-experience ratings were assigned by the paper's authors, all of whom are experienced in VR.[3] The Astral Body entry records the following values:[1]
| Attribute | Rating |
|---|---|
| Category | Controller |
| Year | 2017 |
| Hardware | Gamepad |
| Posture | Any |
| Degrees of freedom | 3 |
| Speed | High |
| Granularity | High |
| Spatial awareness | High |
| Accessibility | High |
| Energy required | Medium |
| Intuitiveness | Medium |
| Large virtual environments | Medium |
| Vection | High |
| Nausea | Low |
| Embodiment | Low |
| Breaks presence | Yes |
| Magical | Yes |
The entry distinguishes Astral Body from a combination of third-person view and teleportation, which it describes as zooming out and in from an outside perspective and which can include dragging and scaling the world around the user.[1]
Research
Franz, Yu and Wobbrock presented their comparison at the ASSETS 2023 accessibility conference. Nineteen participants with upper-body motor impairments completed a seated navigation task with six techniques: Astral Body, Chicken Acceleration, Grab and Pull, Sliding Looking, Teleport and Throw Teleport. The test used a Meta Quest 2 headset and controllers connected to an Alienware m15 laptop, running a custom testbed built in Unity 2019.4.17.[2]
Astral Body, Sliding Looking and Teleport were the three best techniques on trial time and targets hit. All three needed one controller input and little or no upper-body movement, and they had significantly lower NASA-TLX physical demand scores than Chicken Acceleration, Grab and Pull, and Throw Teleport.[2] In interviews, eight participants named Astral Body the most physically comfortable technique, more than any other, but only two chose it as their preferred technique. Seven said it would be the technique they used second-most in VR. Participants said its controls were easy to use and learn and that it gave them a heightened sense of control over navigation; some felt more oriented in space because they could see the whole environment. The authors suggest that lower presence explains the gap between comfort and preference.[2] A 2024 CHI workshop paper by the same group summarised the result as Teleport, Astral Body and Sliding Looking being the fastest and most accurate of the six techniques.[5]
Two 2025 preprints by Franz and Wobbrock extend the work to 20 participants with and 20 without upper-body impairments. In one, participants with impairments rated Astral Body, along with Grab and Pull, Teleport and Throw Teleport, as significantly more physically demanding than participants without impairments did, and they re-entered the target area more often with it, which the authors read as less control over keeping the avatar inside the target.[6] The other, which models a ranked list of each user's fastest techniques, cites the earlier study's finding that Teleport was the fastest technique for 14 participants with impairments and Astral Body for three. It also found that Astral Body and Chicken Acceleration worked best as "calibration" techniques for the predictive models, and it notes that Astral Body needs only thumb movement.[7]
Problems
Locomotion Vault rates Astral Body as a technique that breaks presence, with low embodiment and high vection.[1] In the 2023 study, a few participants said they felt less present with Astral Body and Teleport and compared the experience to playing a traditional video game.[2] Some participants also found the headset heavy and uncomfortable with Astral Body in particular, because the overhead view made them look down at the environment for long periods.[2] In the 2025 comparison, participants with impairments had more trouble keeping the avatar inside a target than participants without impairments.[6]
Solutions
To address the target-control problem, Franz and Wobbrock propose "gravity wells": subtle attractive forces centred on important objects or targets, so that users need less manual control once the avatar is close to them.[6] In a related design, out-of-body locomotion by Nathan Navarro Griffin and Eelke Folmer, the third-person camera stays still while the user does not move, to avoid generating optical flow, and the view returns to first person automatically when the user loses line of sight to the avatar.[8]
Related techniques
Several techniques also move the user's viewpoint out of the avatar:
- Third Person Teleport is a separate Locomotion Vault entry, credited to Convrge (2019). Moving the joystick locks the viewpoint and detaches the avatar, which the user can move around; a button press then teleports the user to the avatar's position. Its examples include VRChat's Holoport locomotion, and its disambiguation note points to Astral body.[1]
- Out-of-body locomotion (Griffin and Folmer, VRST 2019) lets users switch between first-person and third-person avatar views; from the third-person view they can move the avatar precisely and continuously, and they return to first person when they break line of sight with the avatar or press a button. In a study with 22 participants on an obstacle course, it needed 67% fewer viewpoint transitions than pointer-based teleportation, with no significant difference in performance, and participants judged it faster.[8]
- Outstanding (Sebastian Cmentowski, Andrey Krekhov and Jens Krüger, CHI PLAY 2019) lets players switch on demand from first person to a third-person bird's-eye view, command their avatar from above and start long-distance travel across open worlds. The authors report a significant increase in spatial orientation while avoiding cybersickness and preserving presence, enjoyment and competence.[9]
See also
References
- ↑ 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 "Locomotion Vault dataset: Astral Body entry". Locomotion Vault. https://locomotionvault.github.io/data/locomotionvault-dataset.json. Retrieved 2026-09-27.
- ↑ 2.00 2.01 2.02 2.03 2.04 2.05 2.06 2.07 2.08 2.09 2.10 Rachel L. Franz, Jinghan Yu, Jacob O. Wobbrock (2023-10). "Comparing Locomotion Techniques in Virtual Reality for People with Upper-Body Motor Impairments". Proceedings of the 25th International ACM SIGACCESS Conference on Computers and Accessibility (ASSETS '23). ACM. doi:10.1145/3597638.3608394. https://faculty.washington.edu/wobbrock/pubs/assets-23.02.pdf. Retrieved 2026-09-27.
- ↑ 3.0 3.1 Massimiliano Di Luca, Hasti Seifi, Simon Egan, Mar Gonzalez-Franco (2021). "Locomotion Vault: the Extra Mile in Analyzing VR Locomotion Techniques". Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems (CHI '21). ACM. doi:10.1145/3411764.3445319. https://www.microsoft.com/en-us/research/wp-content/uploads/2021/01/CHI_2021_Locomotion_VR.pdf. Retrieved 2026-09-27.
- ↑ 4.0 4.1 "Alice Mystery Garden on Steam". Steam. Valve. https://store.steampowered.com/app/751300/Alice_Mystery_Garden/. Retrieved 2026-09-27.
- ↑ Jacob O. Wobbrock, Rachel L. Franz, Melanie Kneitmix (2024-05-11). "Improving the accessibility of virtual reality for people with motor and visual impairments". ACM CHI 2024 Workshop on Building a Metaverse for All. https://faculty.washington.edu/wobbrock/pubs/chi-24.02.pdf. Retrieved 2026-09-27.
- ↑ 6.0 6.1 6.2 Rachel L. Franz, Jacob O. Wobbrock (2025-10-09). "Quantifying Locomotion Differences Between Virtual Reality Users With and Without Motor Impairments". arXiv preprint 2510.07987. https://arxiv.org/abs/2510.07987. Retrieved 2026-09-27.
- ↑ Rachel L. Franz, Jacob O. Wobbrock (2025-10-07). "Locability: An Ability-Based Ranking Model for Virtual Reality Locomotion Techniques". arXiv preprint 2510.05679. https://arxiv.org/abs/2510.05679. Retrieved 2026-09-27.
- ↑ 8.0 8.1 Nathan Navarro Griffin, Eelke Folmer (2019). "Out-of-body Locomotion: Vectionless Navigation with a Continuous Avatar Representation". Proceedings of the 25th ACM Symposium on Virtual Reality Software and Technology (VRST '19). ACM. https://doi.org/10.1145/3359996.3364243. Retrieved 2026-09-27.
- ↑ Sebastian Cmentowski, Andrey Krekhov, Jens Krüger (2019). "Outstanding: A Multi-Perspective Travel Approach for Virtual Reality Games". Proceedings of the Annual Symposium on Computer-Human Interaction in Play (CHI PLAY '19), arXiv 1908.00379. doi:10.1145/3311350.3347183. https://arxiv.org/abs/1908.00379. Retrieved 2026-09-27.