Exploring Seated Locomotion Techniques in Virtual Reality for People

with Limited Mobility

Marlene Huber

1,2 a

, Simon Kloiber

3 b

, Annalena Ulschmid

2 c

, Agata Marta Soccini

4 d

Alessandro Clocchiatti

4 e

, Hannes Kaufmann

2 f

and Katharina Kr

osl

2 g

VRVis GmbH, Vienna, Austria

TU Wien, Vienna, Austria

Graz University of Technology, Graz, Austria

University of Torino, Turin, Italy

Keywords:

Virtual Reality, Accessibility, Locomotion, User Study.

Abstract:

Virtual reality (VR) is often designed as a standing experience, excluding individuals with limited mobility.

Given that a signiﬁcant portion of the population experiences lower-body mobility restrictions, accessible VR

locomotion must accommodate users without requiring lower-body movement. To build a comprehensive

understanding of suitable locomotion techniques (LTs) for this demographic, it is crucial to evaluate the fea-

sibility of various approaches in virtual environments (VEs). As a starting point, we present our evaluation

approach and a user study on the feasibility and potential of selected LTs for accessible seated locomotion in

VR. Our ﬁndings indicate that common LTs can be adapted for seated stationary VR. Teleportation-based tech-

niques, in particular, stand out as viable options for accessible locomotion. Although our simulated wheelchair

was less popular with non-disabled participants, it was well-received by wheelchair users and shows promise

as an intuitive LT for this target group.

1 INTRODUCTION

According to the World Health Organization, about

1 billion people (∼ 15% of the global popula-

tion) live with a disability, with 70 million relying

on wheelchairs (World Health Organization (WHO),

2021, 2017). Ensuring accessibility to emerging tech-

nologies like Virtual Reality (VR) for individuals

with limited mobility is of societal importance. De-

signing VR applications for this group requires lo-

comotion techniques (LTs) that accommodate seated

and stationary users. Existing LTs often depend on

costly hardware or physical effort, limiting accessibil-

ity (Vailland et al., 2021; G

otzelmann and Kreimeier,

2020; Brachtendorf et al., 2020). However, research

https://orcid.org/0000-0001-7138-6172

https://orcid.org/0000-0003-1186-7630

https://orcid.org/0000-0002-0539-9378

https://orcid.org/0000-0002-7571-8637

https://orcid.org/0009-0002-1451-7775

https://orcid.org/0000-0002-0322-9869

https://orcid.org/0000-0002-9939-0517

on accessible LTs for VR is sparse. To address this

gap, we pose the following research questions:

Q1: Are commonly used LTs also feasible in a seated

stationary VR setting and therefore in principle

accessible for people with limited lower-body mo-

bility?

Q2: Which LT with only upper-body control is the most

efﬁcient to move through a VE?

Q3: Which LT is best suited for a seated stationary VR

experience overall?

We explore digital LTs for seated, stationary VR,

requiring no additional hardware. This novel ap-

proach evaluates LTs in such settings, emphasiz-

ing techniques potentially enhancing accessibility for

users with limited lower-body mobility. With this

work, we make the following contributions:

• LT Selection. Five LTs were chosen based on

literature and their potential feasibility for users

with limited lower-body mobility.

• Improved LTs. We present modiﬁed versions of

the selected LTs which make them suitable for a

Huber, M., Kloiber, S., Ulschmid, A., Soccini, A. M., Clocchiatti, A., Kaufmann, H. and Krösl, K.

Exploring Seated Locomotion Techniques in Virtual Reality for People with Limited Mobility.

DOI: 10.5220/0013305000003912

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 20th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2025) - Volume 1: GRAPP, HUCAPP

and IVAPP, pages 161-173

ISBN: 978-989-758-728-3; ISSN: 2184-4321

161

(a)

(b)

(c)

(d)

(e)

Figure 1: We compare different locomotion techniques in our evaluation environment: the common Standard Teleport (a), a

teleport version we call Volumetric Teleport (b), a virtual Wheelchair (c), Grab&Pull (d), and the multi-perspective locomotion

technique Outstanding (e).

seated, stationary VR setting.

• Evaluation Environment and Task Design. To

make the selected LTs comparable in a user study,

we developed an evaluation environment, includ-

ing a task design for three tasks that represent dif-

ferent application areas: moving across long dis-

tances, on different elevation levels, or around ob-

stacles.

• User Study. We present the results of a user study,

where all ﬁve locomotion metaphors were exe-

cuted in a seated, stationary position, and give rec-

ommendations for accessible locomotion based

on our ﬁndings.

2 RELATED WORK

Research explores diverse LTs such as natural walk-

ing (Langbehn et al., 2017), treadmills (Cherni et al.,

2021), walking-in-place (Wilson et al., 2016), telepor-

tation (Weißker et al., 2018), joystick steering (Clifton

and Palmisano, 2020), leaning (Buttussi and Chit-

taro, 2021), and arm-movement (Coomer et al., 2018).

Most LTs rely heavily on physical mobility, especially

the legs, limiting accessibility for users with disabil-

ities. Accessibility in VR means ensuring VEs ac-

commodate users with diverse disabilities, a challeng-

ing task given the wide range of needs. While previ-

ous work, such as Soccini (2020); Soccini and Cena

(2021); Soccini et al. (2022), explored VR applica-

tions for users with physical disabilities, our goal is

to enable these users to access experiences designed

for non-disabled users. Mott et al. (2019, 2020) iden-

tiﬁed barriers for users with limited mobility in VR,

including setup and cord management. We address

these challenges in our study (see Section 3).

2.1 Seated Locomotion in VR

When users’ hands are occupied, controller-based LTs

in VR become impractical. Solutions like VR Strider

(Freiwald et al., 2020), a seated LT using an exer-

cise bike with trackers and feedback devices, offer

higher presence, comfort, and better spatial estima-

tion compared to teleportation and joystick locomo-

tion. Similarly, Buttussi and Chittaro (2021) com-

pared teleportation, leaning, and joystick LTs, ﬁnd-

ing leaning uncomfortable and prone to causing neck

and spine fatigue. Alternative methods like pres-

sure sensors on thighs (Ohshima et al., 2016) show

promise for wheelchair users but require special-

ized hardware and lack thorough evaluation. Seated

steering LTs, as explored by Clifton and Palmisano

(2020), reduce disorientation compared to standing

versions and perform similarly in cybersickness met-

rics, suggesting our seated stationary LTs would align

with their standing counterparts. TriggerWalking

(Sarupuri et al., 2017) enables walking-like gestures

via controller buttons, while Zielasko and Riecke

(2020) emphasize that seated users prioritize the sen-

sation of movement in VR over the illusion of walk-

ing. Following this, we avoided simulating a standing

experience or artiﬁcially increasing user height in our

approach.

2.2 Wheelchair Locomotion in VR

A unique form of seated LTs involve virtual

wheelchairs. Abstract LTs are often more acces-

sible than those mimicking real walking (Di Luca

et al., 2021). Studies like Vailland et al. (2021)

and Yang et al. (2021) demonstrate VR’s potential

in training wheelchair users for joystick operation

and mechanical propulsion, but these focus on train-

ing rather than broader VR accessibility. Specialized

hardware is being developed for wheelchair-based VR

locomotion. Whee’llConnect(Hansen et al., 2019)

mounts power wheelchairs onto a stationary setup to

translate propulsion into virtual movement. Physical

wheelchairs combined with ergometers(G

otzelmann

and Kreimeier, 2020; Brachtendorf et al., 2020) have

also been used for urban planning and force feedback.

However, such devices are costly, experimental, and

require switching between controllers and wheelchair

wheels. Other efforts include a virtual wheelchair

simulation using swivel chair rotation (Kr

osl et al.,

2018), which is inaccessible for users with limited

GRAPP 2025 - 20th International Conference on Computer Graphics Theory and Applications

162

Table 1: List of feasible LTs for people with limited lower-

body mobility. The top ﬁve LTs were selected for the study

and modiﬁed for seated stationary locomotion.

Locomotion Based On

Grab&Pull Coomer et al. (2018)

Outstanding Cmentowski et al. (2019)

Standard Teleport No speciﬁc source

Volumetric Teleport No speciﬁc source

Wheelchair Majetich (2021)

Arm-Swinging Wilson et al. (2016)

Leaning (Upper Body) Langbehn et al. (2015)

Swimming Huang et al. (2019)

TriggerWalking Sarupuri et al. (2017)

Node-based Jacob Habgood et al. (2018)

Steering Langbehn et al. (2018)

lower-body mobility. Research by Gerling et al.

(2020) highlighted the importance of adaptable con-

trols, safe interaction, and retaining some physical

movement for VR accessibility. Stress from mov-

ing real wheelchairs in a room while wearing an

HMD is another challenge stationary LTs can ad-

dress. While Majetich (2021) prototyped a mechan-

ical wheelchair simulation, broader research on this

LT remains sparse. We see potential for this approach

as an inexpensive, accessible solution for home use,

eliminating the need for additional hardware or phys-

ical wheelchair movement during VR interaction.

3 LOCOMOTION TECHNIQUES

For our user study, we decided to limit our investi-

gation to LTs for a speciﬁc target audience, namely

wheelchair users that have a full range of motion in

their upper body (hands, arms, head). Based on the

limitations of existing solutions as outlined in the pre-

vious section, we deﬁne the following requirements

for the selection of suitable LTs for our user study:

1. Upper-Body Movement Only. LTs must rely

solely on hand and arm movements, accommodat-

ing users with limited lower-body mobility.

2. Holistic Virtual Movement. LTs must support

seated, stationary 3D navigation, including virtual

rotation and exploration of uneven terrain and al-

titudes.

3. No Additional Hardware. LTs must function us-

ing only standard VR equipment: a HMD and two

controllers.

LTs that meet the above requirements are listed in

Table 1. We selected ﬁve of them and LTs that do not

fulﬁll all three requirements are considered infeasi-

ble. To prevent sickness-inducing movements (Lang-

behn et al., 2018), we excluded LTs with continuous

Figure 2: A comparison of how users move with the differ-

ent LTs. VE colors are simpliﬁed for better visibility.

Steering or joystick control and only included those

enabling unrestricted movement in a VE without pre-

selected nodes (Node-based LTs). Our goal was to

select a range of LTs with different archetypes: Tele-

port is a widely used LT. Outstanding offers a con-

tinuous version for long-range travel. From similar

LTs like Grab&Pull, Swimming, and Arm-Swinging,

we selected Grab&Pull for its independence between

movement direction and camera orientation and its

ability to function with one hand. We included

Wheelchair as an LT that closely mimics real-world

movement for people with limited lower-body mobil-

ity. TriggerWalking was excluded due to complexity,

and Leaning was excluded for comfort issues (But-

tussi and Chittaro, 2021). We adapted the selected

LTs to ensure usability for individuals with limited

mobility (Figure 2 visualizes hand and head move-

ments in the VE, following Kloiber et al. (2020)). The

LTs and evaluation environment were implemented

in Unity 2021.3.0f1 using the XR Interaction Toolkit

2.0.1. To address VR accessibility barriers (Mott

et al., 2020), we used the Oculus Quest 2, a stand-

alone HMD with inside-out tracking requiring only

an HMD and controllers.

Grab&Pull. This continuous LT, based on Point-

Tugging (Coomer et al., 2018), involves users grab-

bing the air to pull or push themselves through the

VE. It relies on upper-body movements and enables

continuous navigation.

Outstanding. A continuous multi-perspective LT

where users control an avatar by selecting target

points (Cmentowski et al., 2019). Users can switch

between ﬁrst- and third-person views to traverse large

distances (Figure 1e).

Teleport. A discontinuous LT where users select a

target point and “jump” there, guided by trajectory or

volumetric indicators. This LT is widely used due to

Exploring Seated Locomotion Techniques in Virtual Reality for People with Limited Mobility

163

its reduced likelihood of motion sickness (Langbehn

et al., 2018).

Wheelchair. Drawing on Majetich (2021), this LT

simulates a mechanical wheelchair. Users push vir-

tual wheels using VR controllers, eliminating the

need for additional equipment or hand tracking (Fig-

ure 1c). Since the wheelchair is physically based, this

LT requires a lot of movement from users, and may

exhaust them. However, since the target audience are

people who use mechanical wheelchairs and perform

this movement daily, this LT might not feel as strenu-

ous to them.

3.1 Modiﬁcations and Implementations

To maximize accessibility for stationary, seated loco-

motion, we adapted all chosen LTs to enable orienta-

tion changes while sitting.

Grab&Pull. This LT builds on the concept by

Coomer et al. (2018), with modiﬁcations informed by

informal user tests. To address the slow pace of move-

ment, we added a multiplier (m in Eq. 3) to enable

covering greater distances efﬁciently. A multiplier

of 2.0 proved effective, avoiding noticeable jitter in

initial testing. Attempts to integrate rotation directly

into the grabbing motion caused jitter and nausea. To

resolve this, we separated position changes from ro-

tation, allowing orientation adjustments via joystick

control. Vertical movement (y-axis) was incorporated

realistically by interpolating between the user’s cam-

era height (cam

) and the controller height (con

) at

the grab point (Eq. 2), with interpolation determined

by hand movement (t, Eq. 1). The updated position

(NewPos) is calculated using Eq. 3, where dist

max

rep-

resents the maximum controller movement distance in

the x-z plane.

t = clamp



dist

current

dist

max



∈ [0, 1] (1)

y = t · con

+ (1 − t)· cam

(2)

NewPos =





current

+ (grab

− con

) · m

current

+ (grab

− con

) · m





(3)

To mitigate potential user fatigue from extended hand

positions, the grabbing mechanism operates at any

point in space. However, this ﬂexibility can inﬂu-

ence the relative movement distance achievable from

a grab point.

Figure 3: Left: Standard Teleport indicator. Right: translu-

cent Volumetric Teleport indicator.

Outstanding. Cmentowski et al. (2019) provided

the source code of their LT Outstanding for our study,

including the basic functionality of zooming out, se-

lecting a target via a trajectory, and allowing the

avatar to follow a path (using Unity AI Navigation)

to reach the target point.

We modiﬁed the ﬁrst-person camera to account for

height differences in the VE and added a continuous

rotation feature, enabling the user to adjust their view-

ing direction. This adjustment replaces the original

LT’s reliance on physical rotation in the real world.

Standard and Volumetric Teleport. Teleport is a

promising LT, but seated users may face occlusions in

VEs with obstacles. We tested two variants: Standard

Teleport (Figure 3, left; Figure 1a) with a ﬂat circle in-

dicator and Volumetric Teleport (Figure 3, right; Fig-

ure 1b) using a translucent cross for better visibility.

Following Weißker et al. (2018), both teleports use a

pointing method, operate in vista space, and provide

visual feedback with instant transitions. Users acti-

vate teleportation with the primary button (A on the

right Oculus Touch Controller) and release to execute.

Discontinuous snap-rotations (45° increments) were

added for in-place orientation using the right joystick.

Wheelchair. Our wheelchair LT builds on Ma-

jetich’s VR wheelchair implementation (Majetich,

2021) and uses a 3D model based on the Meyra Bud-

get 2 (Model 9.050

). Each wheelchair component

exists as both a visual 3D object and a physics in-

stance, synchronized to move the player with the

wheelchair. Users grab the large wheels (60cm di-

ameter) to push or pull, creating a grab point that dis-

appears if they let go or exceed a set distance. When

stopped, braking is simulated by increasing the nega-

tive torque of the wheel based on its angular velocity.

If released without braking, motion gradually slows

due to physics, with inclines requiring more effort

than ﬂat surfaces. We ﬁne-tuned the physics param-

eters (Majetich, 2021) through informal user testing

Mechanical wheelchair: Meyra Budget 2,

Model 9.050, https://www.fruehwald.net/mobilitaet/

rollstuehle-u-elektrorollstuehle/standardrollstuhl/

∼

PRODUKT/42124/meyra-budget-2-modell-9050

GRAPP 2025 - 20th International Conference on Computer Graphics Theory and Applications

164

Figure 4: The ground plane of the VE with all three tasks

active.

for improved usability. Haptic feedback on the con-

trollers indicates successful wheel grabs, reducing the

need to look down while interacting.

4 EVALUATION ENVIRONMENT

AND TASK DESIGN

To test the accessibility of our chosen LTs, we de-

signed a custom VE with three tasks (T1, T2, T3) ad-

dressing key aspects of stationary seated locomotion:

(T1) Differences in Altitude. T1 features a hill with

an ∼ 10

◦

incline to evaluate the usability of LTs

for reaching positions at varying heights.

(T2) Long Distance. T2 includes a 200-meter-long

straight path to assess the performance of LTs

over extended distances.

(T3) Obstacle Course. In T3, users navigate a

course with static obstacles to test precise posi-

tion adjustments without collisions. The course

layout was consistent across all LTs.

They were positioned at the corners of a 200 by

200 meter plane, equidistant from the center (see Fig-

ure 4). The starting point was placed at the fourth cor-

ner, with tasks arranged such that the distances from

the starting point to T1 and T3 are equal, while the

distance to T2 is longer. Each task requires users to

navigate the environment and locate a target marked

by a black pillar with a task-button on top, which must

be pressed to complete the task. A large pink arrow

highlights the target’s location in the VE. To form a

continuous path for three tasks in any order, a replica

of the ground plane is placed so the new starting po-

sition aligns directly beneath the previous target (see

Figure 5). Each added ground plane displays only one

task. In our VE, obstacles like trees and stones (see

Figure 2) have reliable colliders. However, there is

no feedback when a collision occurs except that the

Wheelchair cannot move through it. To enhance the

user’s sense of movement with the continuous LTs,

we use a grass texture on the ground instead of a plain

green color.

Figure 5: One of six possible task orders (2, 3, 1) in the VE.

T1 is marked in magenta, T2 in green, and T3 in blue.

5 USER STUDY

Our goal is to facilitate the design of inclusive LTs

for VR, speciﬁcally for users with limited mobility.

To achieve this, we identiﬁed LTs suitable for sta-

tionary, seated settings through a literature review and

tested them ﬁrst with non-disabled individuals to eval-

uate feasibility. In the second part of the study, qual-

itative feedback was collected from two wheelchair

users. For the ﬁrst part, we employed a within-subject

design where 15 non-disabled participants (seven fe-

male, eight male, aged 26–55, M = 33.5, SD = 4.7)

tested all ﬁve LTs across three tasks. Participants var-

ied in VR experience: one had no prior experience,

four had tried it once, seven multiple times, and three

were regular users. Tasks were completed in a con-

sistent environment using the same hardware and a

stationary non-swivel chair for all LTs. In the sec-

ond part, two wheelchair users provided qualitative

feedback. P1 (male, 24) regularly uses a mechani-

cal wheelchair and has prior VR experience. P2 (fe-

male, 27) alternates between a mechanical wheelchair

at home and an electrical one outside, with limited VR

experience. P1 used his mechanical wheelchair, while

P2 used her electrical wheelchair during the study.

5.1 Experiments with Non-Disabled

Participants

5.1.1 Procedure

Our evaluation environment and tasks were designed

to accommodate an arbitrarily large user study with

an a-priori unknown number of participants. The sys-

tem automatically randomized the order of experi-

mental conditions (LTs) and tasks within each con-

dition. Since counterbalancing would require knowl-

Exploring Seated Locomotion Techniques in Virtual Reality for People with Limited Mobility

165

edge of the expected participant count, it was not im-

plemented. The procedure lasted approximately 45 to

60 minutes and consisted of the following steps:

1. Introduction. Participants were briefed on the

study procedure and signed a consent form.

2. Demographic questionnaire. Participants pro-

vided information on gender, age, wheelchair us-

age, and VR experience.

3. Hardware introduction. Participants were intro-

duced to the Oculus Quest 2 setup and the evalua-

tion environment.

4. Testing phase. Each set of three tasks began

with a test scene where participants were intro-

duced to the current LT while wearing the HMD.

They could try out the controls and were shown

the task-buttons, targets, and indication arrows.

Once comfortable, participants initiated the tasks

by pressing the task-button. After completing the

three tasks described in Section 4, participants

removed the HMD and completed the VR Sick-

ness Questionnaire (VRSQ). If sickness symp-

toms arose, extended breaks were taken, and the

study continued only with the participant’s con-

sent.

5. Questionnaires. After completing all LTs, partic-

ipants ﬁlled out the System Usability Scale (SUS)

questionnaire and answered additional questions

about each LT. This process allowed for compara-

tive feedback across all LTs.

Participants could skip tasks or scenes if they felt

too exhausted, motion-sick, or frustrated. All ques-

tionnaires were completed regardless of task com-

pletion to capture participants’ opinions. Breaks of

any length were allowed, and participants could end

the experiment at any time. The study did not focus

on learning time for operating an LT. During testing,

participants were given unlimited time to familiarize

themselves with the controls. Path memorization was

unnecessary as targets were always marked with a red

arrow, minimizing the impact of potential learning ef-

fects.

5.1.2 Data Collection

We recorded data on task completion time, HMD and

controller positions, collisions with obstacles, and in-

stances where participants skipped tasks.

Virtual Reality Sickness Questionnaire. We eval-

uated the feasibility of the chosen LTs for seated, sta-

tionary VR by assessing nausea potential using the

VRSQ (Kim et al., 2018).

System Usability Scale. To evaluate the usability of

our LTs, we used the adapted SUS from Bangor et al.

(2008). While the NASA-TLX assesses workload, we

chose the SUS as it better distinguishes between us-

able and unusable systems, making it more appropri-

ate for our study. We excluded two statements from

the SUS: statement 4, as traditional technical support

is not applicable in our VR setting, and statement 5, as

our LTs do not have diverse functionalities requiring

integration. The SUS score calculation was adapted

to reﬂect these changes.

Additional Questions. After completing all 15 tri-

als (three tasks per LT), participants rated each LT on

a 7-point Likert scale and indicated their overall pre-

ferred technique, providing reasons for their choice.

Additionally, they were asked to describe any differ-

ences they noticed between tasks when using the same

LT and to provide general feedback not addressed by

other questions. We did not include questions about

presence or immersion, as these topics have been ex-

tensively studied in prior research (Jacob Habgood

et al., 2018; Cmentowski et al., 2019; Wolf et al.,

2020; Buttussi and Chittaro, 2021) and are outside the

scope of our study.

5.2 Experiments with Wheelchair Users

In the second part of the study, feedback was collected

from wheelchair users. The procedure included: (1)

an introduction to the study and hardware, (2) a test-

ing phase with a thinking-aloud protocol, (3) a demo-

graphics questionnaire with a 7-point Likert scale to

rate LT preferences, and (4) a semi-structured inter-

view with the following questions:

• Which was your favorite LT?

• What would be the ideal LT for you or how would

you improve your favorite LT?

• Do you ﬁnd it important that there are adapted

LTs like the ones you just experienced?

• Did you ever have the problem that you were not

able to use a VR app because of your limited mo-

bility?

• Would you describe the wheelchair LT as intuitive

for you? What were the positive/negative aspects

of that LT for you?

6 EVALUATION

To address research question Q1 on the feasibility of

LTs in a seated, stationary VR setting, we surveyed

GRAPP 2025 - 20th International Conference on Computer Graphics Theory and Applications

166

n=14

n=15

n=10

n=5

n=14

n=15

n=7

n=5

n=15

n=13

n=9

n=6

****

***

****

***

****

F(4,12) = 13.79, p = 0.00019, η

= 0.74 F(4,12) = 40.96, p = <0.0001, η

= 0.89 F(1.42,7.08) = 22.78, p = 0.001, η

= 0.72

200

400

600

Task 1 Task 2 Task 3

Time (sec)

a a a a aStandard Teleport Volumetric Teleport Outstanding Grab&Pull Wheelchair

Figure 6: Boxplot of the completion time per task per LT (♦ mean, ∗ weak outlier). Statistically signiﬁcant relations according

to pairwise t-tests are connected on top with the number of stars indicating the signiﬁcance level. The number of participants

per group after outlier-removal is reported bellow each box. One-way repeated-measures ANOVA results are reported bellow

each Task.

Task 1

p=0.0039 p=0.062 p=0.25 p=0.058 p=0.29

Task 2

p=0.083 p=0.99 p=0.68 p=0.76 p=0.36

Standard

Teleport

Task 3

p=0.02

Volumetric

Teleport

p=0.14

Outstanding

p=0.21

Grab&Pull

p=0.071

Wheelchair

p=0.44

Normality of the Duration per Task and Method

Figure 7: QQ-Plots and Shapiro-Wilk test results for the

completion time.

the literature and then tested the usability of our ﬁve

selected LTs, with the following hypothesis:

H1: All LTs reach an adapted SUS score > 50.

To ﬁnd the most efﬁcient LT (Q2), we formulated

these hypotheses:

H2: Continuous LTs cause more collisions.

H3: Discontinuous LTs result in faster task comple-

tion.

H4: Different LTs produce different movement

paths.

Finally, to overall determine the most suitable LT

for seated, stationary VR (Q3), we tested:

H5: Volumetric Teleport is better in cluttered envi-

ronments.

H6: LTs with less sickness symptoms and higher us-

ability scores will be preferred.

H7: The wheelchair LT is more intuitive to use for

regular wheelchair users.

We use the standard α = 0.05 for all statistical

tests and, in case of multiple testing, report adjusted

p-values after applying Bonferroni correction. As ef-

fect sizes, we report Cohen’s d, Wilcoxon’s r and

Kendal’s W (< .5 small, .5 − .8 medium, > .8 large),

and generalized eta squared η

(< .06 small, .06 −.14

medium, > .14 large). As the same participant expe-

rienced every LT, we use paired or repeated measure

methods to address the dependency of the individual

measurements on the subject. An a-priori power anal-

ysis with β = .8 indicates a required sample size of

n = 8 for a one-way ANOVA with a medium effect,

and n = 15 for a paired, two-sided t-test with a large

effect. Greenhouse–Geisser correction is applied, if

the sphericity assumption is violated. Result ﬁgures

use the IBM colorblind-safe palette (IBM, 2018).

6.1 Results from Non-Disabled

Participants

Not all participants completed every task; some

skipped individual tasks, and three users had to abort

testing the Grab&Pull and Wheelchair LTs due to

motion sickness. As a result, the number of measure-

ments per parameter or task varies.

Exploring Seated Locomotion Techniques in Virtual Reality for People with Limited Mobility

167

6.1.1 Objective Data

Collisions. Collisions were recorded only in the ob-

stacle course (T3). Since Outstanding uses Unity’s

AI Navigation pathﬁnding, which does not produce

collisions, it is not evaluated here. Most partici-

pants caused no collisions, but one participant cre-

ated 42 with Grab&Pull, possibly due to not recog-

nizing small plants as obstacles. Grab&Pull had the

most collisions, and Wheelchair the least. Due to the

aforementioned extreme outlier and non-normality of

the data, we applied a non-parametric Friedman test,

which yielded no signiﬁcant result (χ

(3) = 4.53,

p = .21, W = .28). Therefore, we reject H2.

Task Completion Time. We tracked task comple-

tion time as a primary measure of efﬁciency. Two

extreme outliers from the Standard Teleport LT were

removed. Shapiro-Wilk tests revealed violations of

normality for the ﬁrst and third tasks using Stan-

dard Teleport. However, the Quantile-Quantile (QQ)

plots depicted in Figure 7 suggest that the data ap-

proximately follows the linear reference line. Thus,

we consider all times normally distributed. A two-

way ANOVA reveals a signiﬁcant inﬂuence of the LT

used (F(4, 4) = 15.694, p = .01, η

= 0.896), but a

negligible effect of the task (F(2, 2) = 16.665, p =

.057, η

= 0.332). Subsequent one-way ANOVAs

for each task and pairwise comparisons using two-

way t-tests show substantial differences between most

LTs (Figures 6), but no signiﬁcant difference between

Grab&Pull and Wheelchair (T1: p = .54, d = 1.54;

T2: p = 1, d = 1.14; T3: p = 1, d = .044) or between

the Teleport variations (T1: p = .34, d = .63; T2:

p = 1, d = .157; T3: p = 1, d = −.026). Grab&Pull

and Wheelchair were on average the slowest, fol-

lowed by Outstanding, which is restricted by the pre-

set avatar’s speed. Discontinuous Teleport variations

were statistically signiﬁcantly faster than most LTs,

except Wheelchair at T1, probably due to the small

sample size, as the min-max intervals are distinctly

not overlapping. Given the evidence that discontin-

uous LTs lead to faster completion times, we accept

H3. Next to the standard list-wise deletion for miss-

ing data, we additionally ﬁtted a linear mixed-effects

model combined with conditional multiple imputa-

tion (N = 100, MICE based on Bayesion regression),

to predict missing values. The conclusions remained

consistent with the original analysis.

Movement Paths. Most movement paths for the

Teleport variations showed inaccuracies, as partici-

pants prioritized speed over accuracy, resulting in dif-

ferently sized loops around the target. In T3, some

Table 2: Descriptive statistics of the questionnaire results.

VRSQ SUS

Mean SD Median Mean SD Median

Std. Teleport 2.89 6.72 0 92.7 10.1 96.9

Vol. Teleport 3.34 6.99 0 93.3 8.17 96.9

Outstanding 5.56 5.27 3.34 77.1 22.2 87.5

Grab&Pull 21.89 15.88 19.2 66.7 21.7 68.8

Wheelchair 25.00 11.75 22.5 42.9 21.6 37.5

participants found small openings to project their

teleport trajectory through, allowing faster move-

ment through the obstacle course (see Figure 2,

4). Grab&Pull and Wheelchair paths showed more

zigzag patterns, possibily contributing to participants’

aversion to these LTs. Additionally, the Wheelchair

required realistic rotation of the virtual wheelchair,

increasing effort on corners (see Figure 2, 5, lower

right). These ﬁndings demonstrate that different LTs

result in different movement paths, supporting H4.

6.1.2 Questionnaires

Virtual Reality Sickness Questionnaire. The

VRSQ score (0-100) indicates sickness, with higher

scores reﬂecting greater sickness. Volumetric Tele-

port, Standard Teleport, and Outstanding had low

scores, while Grab&Pull and Wheelchair had higher

ones (Table 2). A Friedman test (χ

(4) = 46.3,

p < .001, W = .772) found signiﬁcant differences

between LTs. Pairwise Wilcoxon signed rank tests

showed Grab&Pull and Wheelchair had signiﬁcantly

higher scores than the others (p ≤ .011, r ≥ .851).

Participants reported Discomfort, Fatigue, and Full-

ness of Head with Grab&Pull and Wheelchair, while

Teleport and Outstanding caused fewer symptoms,

mostly General Discomfort and Fullness of Head.

System Usability Scale. Bangor et al. (2008) con-

sider SUS scores above 70 as good and below 50 as

unacceptable. Since we used a modiﬁed SUS with

fewer questions and adjusted calculations, our results

are not directly comparable to standard SUS ratings,

but serve as general usability indicators for compar-

ing our LTs. A Friedman test indicates signiﬁcant

differences between LTs (χ

(4) = 40.6, p < .001,

W = .677). Table 2 shows that both Teleport vari-

ations received high scores (mean > 90), afﬁrming

their popularity. Pairwise Wilcoxon signed rank tests

demonstrate signiﬁcant differences between the Tele-

port variations and all other LTs (p ≤ .046, r ≥ .761).

Outstanding achieved a mean score of 77.08, while

Grab&Pull scored 66.67, which we both interpret

as acceptable despite mixed ratings (SD > 20). The

Wheelchair LT scored lowest on average (42.92),

highlighting the need for rigorous improvement de-

spite some high ratings. Thus, while H1 (SUS > 50

GRAPP 2025 - 20th International Conference on Computer Graphics Theory and Applications

168

for each LT) holds for both Teleport LTs, Outstand-

ing, and Grab&Pull, it does not for the Wheelchair

LT, leading us to reject H1.

Preference and Enjoyment. We asked participants

about their preferred LT and overall enjoyment of

each LT. Contrary to our hypothesis that Volumet-

ric Teleport is better for seated VR in cluttered envi-

ronments, nine out of 15 participants preferred Stan-

dard Teleport, while only ﬁve favored the Volumet-

ric Teleport, rendering H5 false. One participant pre-

ferred Outstanding, and none voted for Grab&Pull or

Wheelchair. Participants stated ease of use, speed,

and intuitiveness as reasons for preferring the tele-

port variations, with Volumetric Teleport users high-

lighting the easier to see volumetric indicator com-

paired to the ﬂat one of Standard Teleport. Fig-

ure 8 illustrates LT enjoyment ratings: Wheelchair

was rated worst, Grab&Pull received mixed feed-

back, Outstanding had mostly positive ratings, and

both Teleport variations scored highest. There is a

signiﬁcant inverse correlation between the VRSQ and

SUS score (Spearman’s ρ = −0.643, p < .0001) and

task completion times (VRSQ: ρ ≥ .6117; SUS: ρ ≤

−.5; p < .0001). Differences between participants’

best-scored LT and their preferred LT (VRSQ: 3;

SUS: 3, no overlaps) showed no signiﬁcant deviation

from zero (VRSQ: p = .1814; SUS: p = .149; one-

sided Wilcoxon signed rank test), indicating align-

ment between ratings and preferences. Thus, LTs

with the lowest sickness symptoms and highest us-

ability scores are preferred, supporting H6.

6.1.3 Additional Questions

Finally, participants provided free-text feedback, not-

ing task-speciﬁc differences. Opinions on Outstand-

ing varied: Some appreciated the third-person view

for the obstacle task T3, while others found the view-

switching tedious. Grab&Pull was criticized for

its slow speed and therefore limited suitability for

long distances, but praised for precise navigation in

T3. Speed was also a drawback when using the

Wheelchair, with difﬁculties reported when navigat-

ing uphill (T2) or on uneven terrain.

6.2 Results from Wheelchair Users

Figure 9 shows responses on a 7-point Likert scale

from P1 and P2 to the question “Did you like/enjoy

this Locomotion Technique?” and the time they spent

with each LT.

Teleport. P1 preferred Teleport LTs, despite spend-

ing the least time with them, due to their speed, ease

of use, and familiarity with gaming. In contrast, P2

also used to fast gaming interactions, rated them as

neutral, stating, “I didn’t enjoy the teleport, none

of them, because it was easy and you didn’t have

to move”. P1 generally prefers joystick movement

in VR, self-reporting no motion sickness, and noted,

“It’s relatively easy and it is equally usable for ev-

eryone [non-disabled people and individuals using

wheelchairs], there are no pros or cons when every-

one can use the same [LT].”.

Outstanding. P1 found Outstanding slower than

Teleport LTs, although otherwise preferring walking

characters in games. P2 rated Outstanding slightly

above neutral, noting, “Outstanding works for me just

because I am a very patient person. It wouldn’t work

for people with no attention or no patience.”

Grab&Pull. P1 found Grab&Pull unsuitable in his

wheelchair, stating, “Since I don’t have trunk stabil-

ity, I fall forward the moment my arms are extended.”,

and explaining that sitting on a stable chair or couch

would improve this by allowing to lean back without

losing stability. Finding the LT too exhausting, P1

considered combining it with Outstanding a suitable

solution, stating,“[This] would deﬁnitely be the ideal

[LT] from this selection [of LTs].”.

Wheelchair. Both participants liked the Wheelchair

LT and spent similar amounts of time using it. P2

rated it the highest and selected it as her favorite, valu-

ing its visual representation over other, more abstract

LTs (e.g., pulling an invisible rope). She found the

representation crucial for immersion. Conversely, P1

became deeply immersed, noting, “[...] I wouldn’t

even be able to tell, if the breaks of [his] wheelchair

are on.” P1 initially struggled with the movement,

explaining, “I intuitively do the upper-body move-

ments as I would have to, when moving in my real

wheelchair to not fall off. This means I move my up-

per body forward and backward. [...] So, the intu-

itive movements do not work, because I just sit there

[while moving the virtual wheelchair].” However, he

eventually found that the virtual wheelchair’s move-

ment closely matched that of a physical wheelchair.

With time, he realized that reaching further back to

grab the virtual wheels, as in real life, allowed him

to generate more movement. Similarly, P2 noted that

users might require time to understand the LT’s me-

chanics. Both participants enjoyed the Wheelchair LT

but found it physically demanding. P2 compared the

upper-limb exhaustion to playing sports. While she

Exploring Seated Locomotion Techniques in Virtual Reality for People with Limited Mobility

169

Strongly

disagree

Moderately

disagree

Slightly

disagree

Neutral Slightly

agree

Moderately

Agree

Strongly

agree

Standard Teleport Volumetric Teleport Outstanding Grab&Pull Wheelchair

Figure 8: Participants’ responses to the question “Did you like/enjoy this locomotion technique?”.

Figure 9: Preference ratings and time spent exploring an LT of the wheelchair users.

appreciated the controls, she suggested less strenuous

arm movements could improve the interaction design,

remarking, “It would be nice for example to put my

arms up and down to move.” P1 acknowledged sim-

pler alternatives for moving through a VE but noted

he might prefer using his physical wheelchair for such

tasks, provided the controllers were integrated effec-

tively. P1 concluded, “It’s something that needs time

getting used to. In real life, I also need time to get

used to a wheelchair, since every wheelchair is dif-

ferent and this [virtual] one is also manageable after

some time.” Feedback from both participants inspired

potential improvements for the Wheelchair LT. For in-

stance, P2 suggested, “I would change the look of the

wheelchair. I would add some colors or make it look

like a spacecraft or a motorbike.” Their positive re-

ception, compared to non-disabled participants, sup-

ports H7, though a larger study with more wheelchair

users is required for a deﬁnitive evaluation.

7 DISCUSSION

By looking at our quantitative and qualitative data and

observations made during the experiments, we gained

the following insights.

Speed. Discontinuous LTs were generally faster

than continuous ones (H3 = true). Some users strug-

gled with height differences using the Wheelchair,

as reﬂected in the outliers for T1 (involving altitude

changes) in Figure 6. While expected, these ﬁndings

highlight considerations for designing accessible LTs.

Accuracy. Collisions were observed with the Tele-

port variations during rapid movement, potentially

impacting precision when aiming between obstacles.

Unexpectedly, the Wheelchair LT resulted in few col-

lisions (H2 = false). This may be due to our instruc-

tions encouraging conscious movement or because

users were fully immersed in the VE and maintained

a natural distance from obstacles, as they would in the

real world.

Motion Sickness. As expected, the Teleport vari-

ations resulted in low sickness levels (assessed with

VRSQ). In contrast, Outstanding caused nausea in

a few participants, the same individuals who strug-

gled with Grab&Pull and Wheelchair due to sickness.

Grab&Pull showed the highest Fatigue scores on the

VRSQ, aligning with participants’ qualitative feed-

back labeling it as too exhausting. Some participants

reported HMD jitter during movement, which may

explain the elevated scores in other VRSQ segments

for this LT. The poor VRSQ scores for Wheelchair

and Grab&Pull likely stem from a mismatch be-

tween visual and vestibular information, consistent

with prior ﬁndings on continuous joystick move-

GRAPP 2025 - 20th International Conference on Computer Graphics Theory and Applications

170

ments (Langbehn et al., 2018; Clifton and Palmisano,

2020).

Usability. The Volumetric Teleport, Standard Tele-

port, Outstanding, and Grab&Pull all received SUS

scores above 50, with the Teleport variations scoring

above 90. This high usability may be attributed to

the familiarity of Teleport as one of the best-known

LTs and its simplicity, requiring only pointing, press-

ing one button, and joystick rotation. Conversely, the

Wheelchair LT scored below 50 among non-disabled

participants, likely because they were not regular

wheelchair users and required time to adjust to its me-

chanics. However, both wheelchair users favored the

Wheelchair LT despite its physical demands. They

found it accurate, usable, and enjoyable (H7 = true),

experiencing no motion sickness or tracking issues.

P1 even experimented with faster movement by ex-

tending further back with the controllers, akin to his

real-life wheelchair use, without encountering HMD

tracking problems. Since Outstanding and Teleport

also performed well with wheelchair users, combin-

ing Wheelchair for precise navigation with Teleport

or Outstanding for efﬁcient long-distance movement

could be a promising avenue for future research.

7.1 Research Questions

Q1. Informed by our literature survey, we proposed

three requirements for LTs to be accessible to indi-

viduals with limited lower-body mobility: (1) only

upper-body movement, (2) holistic virtual movement

(including virtual rotation), and (3) no additional

hardware beyond standard VR equipment. Based on

these requirements, we identiﬁed all LTs listed in Ta-

ble 1 as suitable for accessible locomotion when mod-

iﬁed to include virtual rotation. Our quantitative eval-

uation of ﬁve modiﬁed LTs showed acceptable us-

ability (SUS scores) for all but low usability for the

Wheelchair(H1). As discussed in Section 2, the use

of a mechanical wheelchair simulation as an LT in VE

remains uncommon and relatively unexplored in re-

search. While improvements are likely needed, our

wheelchair users generally appreciated this LT. To ad-

dress Q1, many commonly used LTs are feasible in

a seated, stationary VR setting if modiﬁed to account

for the lack of physical rotation, making them, in prin-

ciple, accessible for people with limited lower-body

mobility.

Q2. Since locomotion should be fast, easy to use,

and maneuverable, we evaluated the efﬁciency of LTs

primarily by task completion times, as well as mo-

tion paths and the number of collisions as measures

of accuracy. Given that we accepted H3, the Teleport

variations emerged as the most efﬁcient LTs, with the

lowest task completion times. Although the Teleport

variations recorded a slightly higher absolute number

of collisions, these results were not statistically sig-

niﬁcant (H2 = false). Regarding movement paths (ac-

cepted H4), the Teleport variations appeared subjec-

tively more efﬁcient for navigating the VE due to their

discontinuous movement between target points. To

address Q2, our ﬁndings suggest that among LTs with

only upper-body control, a Teleport LT is the most ef-

ﬁcient for navigating a VE.

paragraphQ3. As shown, all our modiﬁed LTs are

suitable for seated stationary VR (Q1). The Teleport

variations were the most efﬁcient and preferred LTs

in our study (H6 = true), achieving the highest SUS

and lowest VRSQ scores. We hypothesized that Volu-

metric Teleport would perform better in cluttered VEs

(H5). While this was true for ﬁve participants who

preferred Volumetric Teleport, the majority favored

Standard Teleport, leading us to reject this hypoth-

esis. Outstanding combines some beneﬁts of Tele-

port while remaining continuous. Users rated it well

across categories but found precise movements chal-

lenging. In its original version, Cmentowski et al.

(2019) paired it with real walking for ﬁner adjust-

ments. Future research could explore combining Out-

standing with Grab&Pull, as the latter excels at pre-

cise positioning but is unsuitable for long distances.

One wheelchair user also considered this combina-

tion ideal for accessible LTs. While the Wheelchair

LT was poorly received by non-disabled participants,

wheelchair users found it potentially effective for

their needs (H7). To answer Q3, the optimal LT for

a seated stationary VR experience depends on the tar-

get group. However, Teleport variations are generally

well-received by both non-disabled participants and

wheelchair users.

8 CONCLUSION AND FUTURE

WORK

While many LTs exist, their suitability for seated, sta-

tionary VR settings remains underexplored. Based on

a literature review, we proposed a set of requirements

for accessible LTs for individuals with limited mo-

bility and identiﬁed potentially viable LTs for seated,

stationary locomotion (Table 1). To evaluate these, we

developed an evaluation environment and task design,

conducting a user study with 15 non-disabled partic-

ipants and two wheelchair users. The Teleport varia-

tions outperformed other LTs in efﬁciency, usability,

and user preference among non-disabled participants.

Exploring Seated Locomotion Techniques in Virtual Reality for People with Limited Mobility

171

For wheelchair users, the Wheelchair LT was equally

favored, alongside the Teleport LTs. Outstanding

received good ratings among continuous LTs from

both groups. Grab&Pull caused exhaustion, motion

sickness, and low usability, while the Wheelchair LT

had similar results for non-disabled users. However,

wheelchair users were less affected and generally ap-

preciated this LT. Based on current ﬁndings, we rec-

ommend a Teleport variation for an inclusive design-

for-all approach, supporting both non-disabled users

and those with limited lower-body mobility. Future

work aims to reﬁne the Wheelchair LT based on feed-

back from wheelchair users, as it showed promise for

this target group. Additionally, we tested Teleport

LTs with implicit orientations. Given their versatil-

ity, follow-up studies could explore specialized tasks

and alternative orientation methods (e.g., Mori et al.

(2023)) to better understand their advantages.

ACKNOWLEDGEMENTS

This work was enabled by the Competence Cen-

tre VRVis. The VRVis GmbH is funded by BMK,

BMAW, Styria, SFG, Tyrol, Vorarlberg and Vienna

Business Agency in the scope of COMET - Com-

petence Centers for Excellent Technologies (879730,

911654) which is managed by FFG. Furthermore, we

would like to thank Sebastian Cmentowski

for pro-

viding an adapted implementation of his project Out-

standing: A Multi-Perspective Travel Approach for

Virtual Reality Games (Cmentowski et al., 2019).

REFERENCES

Bangor, A., Kortum, P. T., and Miller, J. T. (2008). An

Empirical Evaluation of the System Usability Scale. In-

ternational Journal of Human–Computer Interaction,

24(6):574–594.

Brachtendorf, K., Weyers, B., and Zielasko, D. (2020). To-

wards Accessibility in VR - Development of an Afford-

able Motion Platform for Wheelchairs. In 2020 IEEE

Conference on Virtual Reality and 3D User Interfaces

Abstracts and Workshops (VRW), pages 291–292, At-

lanta, GA, USA. IEEE.

Buttussi, F. and Chittaro, L. (2021). Locomotion in Place in

Virtual Reality: A Comparative Evaluation of Joystick,

Teleport, and Leaning. IEEE Transactions on Visualiza-

tion and Computer Graphics, 27(1):125–136.

Cherni, H., Nicolas, S., and M

etayer, N. (2021). Using

virtual reality treadmill as a locomotion technique in a

navigation task: Impact on user experience – case of

sebastian.cmentowski@uni-due.de

the KatWalk. International Journal of Virtual Reality,

21(1):1–14.

Clifton, J. and Palmisano, S. (2020). Effects of steering lo-

comotion and teleporting on cybersickness and presence

in HMD-based virtual reality. Virtual Reality, 24(3):453–

468.

Cmentowski, S., Krekhov, A., and Kr

uger, J. (2019). Out-

standing: A Multi-Perspective Travel Approach for Vir-

tual Reality Games. In Proceedings of the Annual Sym-

posium on Computer-Human Interaction in Play, CHI

PLAY ’19, page 287–299, New York, NY, USA. Asso-

ciation for Computing Machinery.

Coomer, N., Bullard, S., Clinton, W., and Williams-

Sanders, B. (2018). Evaluating the Effects of Four VR

Locomotion Methods: Joystick, Arm-Cycling, Point-

Tugging, and Teleporting. In Proceedings of the 15th

ACM Symposium on Applied Perception, SAP ’18, New

York, NY, USA. Association for Computing Machinery.

Di Luca, M., Seiﬁ, H., Egan, S., and Gonzalez-Franco, M.

(2021). Locomotion Vault: The Extra Mile in Analyz-

ing VR Locomotion Techniques. In Proceedings of the

2021 CHI Conference on Human Factors in Computing

Systems, CHI ’21, New York, NY, USA. Association for

Computing Machinery.

Freiwald, J. P., Ariza, O., Janeh, O., and Steinicke, F.

(2020). Walking by Cycling: A Novel In-Place Loco-

motion User Interface for Seated Virtual Reality Experi-

ences, page 1–12. Association for Computing Machin-

ery, New York, NY, USA.

Gerling, K., Dickinson, P., Hicks, K., Mason, L., Simeone,

A. L., and Spiel, K. (2020). Virtual Reality Games for

People Using Wheelchairs, page 1–11. Association for

Computing Machinery, New York, NY, USA.

otzelmann, T. and Kreimeier, J. (2020). Towards the

Inclusion of Wheelchair Users in Smart City Planning

through Virtual Reality Simulation. In Proceedings of the

13th ACM International Conference on PErvasive Tech-

nologies Related to Assistive Environments, PETRA ’20,

New York, NY, USA. Association for Computing Ma-

chinery.

Hansen, J. P., Trudslev, A. K., Harild, S. A., Alapetite,

A., and Minakata, K. (2019). Providing Access to VR

Through a Wheelchair. In Extended Abstracts of the

2019 CHI Conference on Human Factors in Computing

Systems, CHI EA ’19, page 1–8, New York, NY, USA.

Association for Computing Machinery.

Huang, Z., Zhang, Y., Quigley, K. C., Sankar, R., Wormser,

C., Mo, X., and Yang, A. Y. (2019). Accessibility of Vir-

tual Reality Locomotion Modalities to Adults and Mi-

nors.

IBM (2018). IBM COLOR BLIND SAFE PALETTE. https:

//www.ibm.com.

Jacob Habgood, M. P., Moore, D., Wilson, D., and Alapont,

S. (2018). Rapid, Continuous Movement Between Nodes

as an Accessible Virtual Reality Locomotion Technique.

In 2018 IEEE Conference on Virtual Reality and 3D User

Interfaces (VR), pages 371–378, Tuebingen/Reutlingen,

Germany. IEEE.

Kim, H. K., Park, J., Choi, Y., and Choe, M. (2018). Virtual

GRAPP 2025 - 20th International Conference on Computer Graphics Theory and Applications

172

reality sickness questionnaire (VRSQ): Motion sickness

measurement index in a virtual reality environment. Ap-

plied Ergonomics, 69:66–73.

Kloiber, S., Settgast, V., Schinko, C., Weinzerl, M., Fritz, J.,

Schreck, T., and Preiner, R. (2020). Immersive analysis

of user motion in VR applications. The Visual Computer,

36(10-12):1937–1949.

osl, K., Bauer, D., Schw

arzler, M., Fuchs, H., Suter, G.,

and Wimmer, M. (2018). A VR-based user study on the

effects of vision impairments on recognition distances of

escape-route signs in buildings. The Visual Computer,

34:911–923.

Langbehn, E., Eichler, T., Ghose, S., von Luck, K., Bruder,

G., and Steinicke, F. (2015). Evaluation of an omni-

directional walking-in-place user interface with virtual

locomotion speed scaled by forward leaning angle. In

Proceedings of the GI Workshop on Virtual and Aug-

mented Reality (GI VR/AR), pages 149–160, Sankt Au-

gustin, Germany. GI-Group VR/AR.

Langbehn, E., Lubos, P., Bruder, G., and Steinicke, F.

(2017). Application of redirected walking in room-scale

VR. In 2017 IEEE Virtual Reality (VR), pages 449–450.

Langbehn, E., Lubos, P., and Steinicke, F. (2018). Eval-

uation of Locomotion Techniques for Room-Scale VR:

Joystick, Teleportation, and Redirected Walking. In Pro-

ceedings of the Virtual Reality International Conference

- Laval Virtual, VRIC ’18, New York, NY, USA. Associ-

ation for Computing Machinery.

Majetich, J. (2021). VR Wheelchair. https://github.com/

justinmajetich/vr-wheelchair.

Mori, S., Hashiguchi, S., Shibata, F., and Kimura, A.

(2023). Point & teleport with orientation speciﬁcation,

revisited: Is natural turning always superior? Journal of

Information Processing, 31:392–403.

Mott, M., Cutrell, E., Gonzalez Franco, M., Holz, C., Ofek,

E., Stoakley, R., and Ringel Morris, M. (2019). Ac-

cessible by Design: An Opportunity for Virtual Real-

ity. In 2019 IEEE International Symposium on Mixed

and Augmented Reality Adjunct (ISMAR-Adjunct), pages

451–454, Beijing, China. IEEE.

Mott, M., Tang, J., Kane, S., Cutrell, E., and Ringel Mor-

ris, M. (2020). “I Just Went into It Assuming That I

Wouldn’t Be Able to Have the Full Experience”: Under-

standing the Accessibility of Virtual Reality for People

with Limited Mobility. In The 22nd International ACM

SIGACCESS Conference on Computers and Accessibil-

ity, ASSETS ’20, New York, NY, USA. Association for

Computing Machinery.

Ohshima, T., Ishihara, H., and Shibata, R. (2016). Virtual

ISU: Locomotion Interface for Immersive VR Gaming

in Seated Position. In Proceedings of the 2016 Virtual

Reality International Conference, VRIC ’16, New York,

NY, USA. Association for Computing Machinery.

Sarupuri, B., Hoermann, S., Steinicke, F., and Lindeman,

R. W. (2017). Triggerwalking: A Biomechanically-

Inspired Locomotion User Interface for Efﬁcient Realis-

tic Virtual Walking. In Proceedings of the 5th Symposium

on Spatial User Interaction, SUI ’17, page 138–147,

New York, NY, USA. Association for Computing Ma-

chinery.

Soccini, A. M. (2020). The induced ﬁnger movements ef-

fect. In SIGGRAPH Asia 2020, pages 1–2, New York,

NY, USA. ACM.

Soccini, A. M. and Cena, F. (2021). The ethics of rehabili-

tation in virtual reality: the role of self-avatars and deep

learning. In 2021 IEEE International Conference on

Artiﬁcial Intelligence and Virtual Reality (AIVR), pages

324–328, New York, NY, USA. IEEE, IEEE.

Soccini, A. M., Clocchiatti, A., and Inamura, T. (2022). Ef-

fects of frequent changes in extended self-avatar move-

ments on adaptation performance. Journal of Robotics

and Mechatronics, 34(4):756–766.

Vailland, G., Devigne, L., Pasteau, F., Nouviale, F., Fraudet,

B., Leblong,

E., Babel, M., and Gouranton, V. (2021).

VR based Power Wheelchair Simulator: Usability Eval-

uation through a Clinically Validated Task with Regular

Users. In 2021 IEEE Virtual Reality and 3D User Inter-

faces (VR), pages 420–427, Lisboa, Portugal. IEEE.

Weißker, T., Kunert, A., Fr

ohlich, B., and Kulik, A. (2018).

Spatial Updating and Simulator Sickness During Steer-

ing and Jumping in Immersive Virtual Environments. In

2018 IEEE Conference on Virtual Reality and 3D User

Interfaces (VR), pages 97–104, Lisboa, Portugal. IEEE.

Wilson, P. T., Kalescky, W., MacLaughlin, A., and

Williams, B. (2016). VR Locomotion: Walking > Walk-

ing in Place > Arm Swinging. In Proceedings of the 15th

ACM SIGGRAPH Conference on Virtual-Reality Contin-

uum and Its Applications in Industry - Volume 1, VRCAI

’16, page 243–249, New York, NY, USA. Association for

Computing Machinery.

Wolf, D., Rogers, K., Kunder, C., and Rukzio, E. (2020).

JumpVR: Jump-Based Locomotion Augmentation for Vir-

tual Reality, page 1–12. Association for Computing Ma-

chinery, New York, NY, USA.

World Health Organization (WHO) (2017). Wheelchair

publications – an overview. https://www.who.int/

news-room/articles-detail/wheelchair\ publications\

%E2\%80\%93an\ overview.

World Health Organization (WHO) (2021). Disability

and health. https://www.who.int/news-room/fact-sheets/

detail/disability-and-health.

Yang, Y.-S., Koontz, A. M., Hsiao, Y.-H., Pan, C.-T., and

Chang, J.-J. (2021). Assessment of Wheelchair Propul-

sion Performance in an Immersive Virtual Reality Sim-

ulator. International Journal of Environmental Research

and Public Health, 18(15):8016.

Zielasko, D. and Riecke, B. E. (2020). Can We Give Seated

Users in Virtual Reality the Sensation of Standing or

Even Walking? Do We Want To? In 2020 IEEE Con-

ference on Virtual Reality and 3D User Interfaces Ab-

stracts and Workshops (VRW), pages 281–282, Atlanta,

GA, USA. IEEE.

Exploring Seated Locomotion Techniques in Virtual Reality for People with Limited Mobility

173