A Unique Way to Increase Presence of Mobility Impaired Users - Increasing Confidence in Balance

Rongkai Guo, Gayani Samaraweera, John Quarles The University of Texas at San Antonio

ABSTRACT Previous research on healthy subjects showed that a higher sense of presence can be elicited through full body avatars versus no avatar. However, minimal avatar research has been conducted with persons with mobility impairments. For these users, Virtual Environments (VEs) and avatars are becoming more common as tools for rehabilitation. If we can maximize presence in these VEs, users may be more effectively distracted from the pain and repetitiveness of rehabilitation, thereby increasing users’ motivation. To investigate this we replicated the classic virtual pit experiment and included a responsive full body avatar (or lack thereof) as a 3D user interface. We recruited from two different populations: mobility impaired persons and healthy persons as a control. Results give insight into many other differences between healthy and mobility impaired users’ experience of presence in VEs.

Keywords: Virtual Reality, Presence, Mobility Impairments, Avatar, Rehabilitation, User Studies

Index Terms: H.5.1 [Information Interfaces and Presentation]: Multimedia Information Systems—Artificial, augmented and virtual realities;

1 INTRODUCTION Avatars have been shown to be one of the main factors that can affect presence [1-6] in Virtual Environments (VEs). Although results suggest that avatars can significantly increase presence for healthy people, their effects on Mobility Impaired (MI) persons have been minimally studied. If avatars can increase presence of MI persons, then it may help immerse them in a VE and potentially distract them from the pain and repetition of rehabilitation. Thus, since this could have a potential impact on the effectiveness of VR rehabilitation, it is important to understand how avatars affect MI persons’ presence.

Specifically, in this paper we present results of a study that investigated how a responsive, full-body avatar can affect the presence of MI persons. The study utilized a virtual pit [6-8] (Figure 1). In contrast to previous studies, our study investigates MI persons. We recruited MI participants who require the use of a cane. Results give insight into how avatars (or lack thereof) can differently affect the presence of MI persons and suggest how to increase presence specifically for MI users.

2 BACKGROUND AND RELATED WORK Researchers have often found that avatars can affect presence. However, almost all of the previous avatar studies have been conducted only with persons without disabilities. For example, in 2010, Gonzalez-Franco et al. suggested that synchronous mirror reflection was able to give participants a significantly greater body ownership illusion (i.e., embodiment) than an asynchronous condition [2]. Betty et al. found that a fully registered avatar could help participants make more accurate distance estimation [4]. Several other studies have shown that a first person experienced, registered avatar could temporarily affect participants’ presence [5, 9]. Researchers also found interesting correlations between participants behavior and avatars [1, 10]. Recent research conducted by Konstantina et al. [3] suggested that participants behaved differently when they had avatars with different appearances for the same task. However, this study was conducted with only healthy participants.

3 MATERIALS

Figure 1: Participant’s view of the virtual pit.

We designed a task specifically to keep the users within a narrow walkway, walking back and forth in straight lines to facilitate quantitative gait analysis with our GAITRite mat. The task is to pick up a virtual ball from a dispenser on one side of a narrow walkway over an 11ft pit (Figure 1), carry the ball to the other side of the walkway, and drop the ball into a box. The participant can only pick up one ball at a time from the dispenser. At both ends, the participant hears a sound to indicate they reached the end of the walkway (i.e., to prevent tracking errors and make sure that they walked far enough for adequate gait analysis). To increase usability and immersion, the participant also hears 3D enabled sounds when: the dispenser generates a ball, the drop box opens, a ball is successfully dropped into the drop box, or a ball falls into the water in the pit. They need to successfully drop five balls into the drop box to finish the study.

4 METHODS We conducted a study to investigate whether an avatar could affect the presence differently for MI persons than for healthy

* Rongkai Guo, rguo@cs.utsa.edu

* Gayani Samaraweera, gsamaraw@cs.utsa.edu

* John Quarles, jpq@cs.utsa.edu

77

IEEE Virtual Reality 201429 March - 2 April, Minneapolis, Minnesota, USA978-1-4799-2871-2/14/$31.00 ©2014 IEEE

persons in a stressful VE. In our study, there were two populations being compared 1) 11 MI persons and 2) 11 healthy persons who were of similar demographics to the MI population.

We hypothesize: MI participants will find the VE more realistic than the healthy participants, but only in the avatar condition.

Pay

Pre-Study

Study Procedure

1. Introduction2. Informed Consent

Study Session

Rate Fatigue

Rate Fatigue

Questionnaire

Break

Repeat 4 times

Repeat Twice

Rate Fatigue

Gait Analysis

Rate Fatigue

Human Calibration

Training Session

Figure 2: The flowchart of the study procedure.

5 RESULTS We ran Wilcoxon signed ranks tests with the questionnaire between MI and healthy pariticipants. In the AVATAR condition, MI participants found the VE to be more realistic than the Healthy participants did (Figure 3). This is based on the following presence question:

Q5: “How much did your experiences in the virtual environment seem consistent with your real-world experiences?” That is, we found a significant difference between MI (mean =

6.27, median = 6) and Healthy (mean = 5.36, median = 6) in the AVATAR condition (Wilcoxon Signed Ranks: Z = -2.45, p = .016, r = .52). Thus, our hypothesis cannot be rejected.

Figure 3: The mean values of Q5.

5.1 Interview From observation, we noticed that participants did not look down at the avatar very often. The interview confirmed this. Of note, many MI participants said that if they knew there was an avatar, it gave them more confidence in their balance while they were walking. Most of them thought that this increased confidence helped them more than the visual feedback from the avatar itself.

6 CONCLUSION Differences in presence between healthy and MI showed that avatars make the VE seem more realistic to MI users than to healthy users. Thus, MI presence is actually higher than that of healthy users when there is an avatar, based on realism factors.

Avatars influence both healthy and MI users. However, the influence is different. We should consider using a different design strategy for MI users: 1) no avatar seems better for MI users to build up confidence in their own sense of proprioception, since they cannot rely on visual feedback as much. 2) An avatar may be

able to help their concentration on training and improve their sense of realism but may counteract their sense of control as it relates to their gait.

7 FUTURE WORK We expect that there are more other unique ways to increase presence of MI users through manipulating core VE system aspects, such as latency, Field of View, 3D stereoscopic vision and so on. These all could have an impact on motivation in rehabilitation. To investigate this, we aim to recruit additional MI participants, people with different disabilities, and elderly people to generally investigate how to maximize presence and motivation of persons with disabilities,

ACKNOWLEDGMENTS The authors wish to thank the study participants. This work was supported by grants from NSF (IIS-1153229 and IIS-1218283).

REFERENCES [1] H. Farmer, A. Tajadura-Jiménez, and M. Tsakiris, "Beyond the

colour of my skin: How skin colour affects the sense of body-

ownership," Consciousness and cognition, vol. 21, pp. 1242-1256,

2012.

[2] M. González-Franco, D. Pérez-Marcos, B. Spanlang, and M. Slater,

"The contribution of real-time mirror reflections of motor actions on

virtual body ownership in an immersive virtual environment," in

Virtual Reality Conference (VR), 2010 IEEE, 2010, pp. 111-114.

[3] K. Kilteni, I. Bergstrom, and M. Slater, "Drumming in immersive

virtual reality: the body shapes the way we play," Visualization and Computer Graphics, IEEE Transactions on, vol. 19, pp. 597-605,

2013.

[4] B. J. Mohler, S. H. Creem-Regehr, W. B. Thompson, and H. H.

Bülthoff, "The effect of viewing a self-avatar on distance judgments

in an HMD-based virtual environment," Presence: Teleoperators and Virtual Environments, vol. 19, pp. 230-242, 2010.

[5] M. Slater, B. Spanlang, M. V. Sanchez-Vives, and O. Blanke, "First

person experience of body transfer in virtual reality," PloS one, vol.

5, p. e10564, 2010.

[6] M. Meehan, B. Insko, M. Whitton, and F. P. Brooks Jr,

"Physiological measures of presence in stressful virtual

environments," in ACM Transactions on Graphics (TOG), 2002, pp.

645-652.

[7] R. Guo and J. Quarles, "Differences in presence between healthy

users and users with multiple sclerosis," in Perceptual Illusions in Virtual Environments (PIVE), 2012 IEEE VR Workshop on, 2012, pp.

1-6.

[8] P. Khanna, I. Yu, J. Mortensen, and M. Slater, "Presence in response

to dynamic visual realism: a preliminary report of an experiment

study," in Proceedings of the ACM symposium on Virtual reality software and technology, 2006, pp. 364-367.

[9] K. Kilteni, J.-M. Normand, M. V. Sanchez-Vives, and M. Slater,

"Extending body space in immersive virtual reality: a very long arm

illusion," PloS one, vol. 7, p. e40867, 2012.

[10] V. Groom, J. N. Bailenson, and C. Nass, "The influence of racial

embodiment on racial bias in immersive virtual environments,"

Social Influence, vol. 4, pp. 231-248, 2009.

6.273 5.364 5.909 5.909

0

2

4

6

8

Mea

ns o

f Q5

MI Healthy MI Healthy

Avatar Avatar No Avatar No Avatar

78