Elsevier

Displays

Volume 29, Issue 2, March 2008, Pages 58-69
Displays

Virtual reality induced symptoms and effects (VRISE): Comparison of head mounted display (HMD), desktop and projection display systems

https://doi.org/10.1016/j.displa.2007.09.005Get rights and content

Abstract

Virtual reality (VR) systems are used in a variety of applications within industry, education, public and domestic settings. Research assessing reported symptoms and side effects of using VR systems indicates that these factors combine to influence user experiences of virtual reality induced symptoms and effects (VRISE). Three experiments were conducted to assess prevalence and severity of sickness symptoms experienced in each of four VR display conditions; head mounted display (HMD), desktop, projection screen and reality theatre, with controlled examination of two additional aspects of viewing (active vs. passive viewing and light vs. dark conditions). Results indicate 60–70% participants experience an increase in symptoms pre–post exposure for HMD, projection screen and reality theatre viewing and found higher reported symptoms in HMD compared with desktop viewing (nausea symptoms) and in HMD compared with reality theatre viewing (nausea, oculomotor and disorientation symptoms). No effect of lighting condition was found. Higher levels of symptoms were reported in passive viewing compared to active control over movement in the VE. However, the most notable finding was that of high inter- and intra-participant variability. As this supports other findings of individual susceptibility to VRISE, recommendations are offered concerning design and use of VR systems in order to minimise VRISE.

Introduction

In the early 1990s there was a rapid increase in the development of commercial virtual reality (VR) systems and expectation of widespread application of the technology in industrial, public and domestic environments. At that time the interest was mostly in systems using head mounted displays (HMDs) and datagloves for personal viewing and interaction with a virtual environment (VE). So-called non-immersive systems, which display the VE on a standard PC monitor, were often not regarded as ‘true’ VR. Interest in VR applications led to speculation of potential side-effects from using these systems, ranging from anecdotal reports of flash-backs producing driving difficulties post-exposure to scientific reports of ‘simulator sickness’ following participation in VR [18], [30], [39].

However, in recent years the focus of display technologies has moved from HMD based systems to projection displays. Projection displays have the advantage of the potential for collaborative viewing and interaction and are an attractive financial option as the technologies can often have multiple uses rather than requiring expensive purchase of dedicated VR displays. This paper presents a series of studies examining projection based VR display systems.

Early work into VR health and safety established a number of findings relating to the health and safety implications of virtual reality [26], [6]. Firstly, a framework of factors influencing the production of VR induced symptoms and effects was developed, with four main factor groups identified as; VR technical system, Virtual Environment (the content of the virtual “world”) design, circumstances of use and individual participant characteristics [26]. Secondly, although the symptoms and effects identified were similar to those found with other simulators and in transportation, the causes and symptom patterns were considered to be sufficiently different to justify a new term: virtual reality induced symptoms and effects (VRISE) [6]. The term “cybersickness” has also been used to describe the sickness element of this symptom set. Thirdly, a wide variety of individual differences in susceptibility to, and experience of, effects was observed. From data obtained from over 200 participants, 80% of participants across all experiments reported some experience of VR induced symptoms. For most people these were mild and short-lived but 5% of participants experienced symptoms so severe that they had to end their period of VR exposure [6].

One of the advantages of desktop viewing is that it allows several users to view a VE at the same time. Whilst two or three viewers may sit comfortably at a PC workstation, the potential for additional VE viewers may be facilitated by projecting the VE onto a larger screen using a standard PC-compatible projector. Such displays are commonly used for slide presentations and can be used to display a VE for viewing by a group or with which the participants can interact in real-time using the PC input devices, although some degradation in display quality is experienced. In addition, a number of higher resolution methods of displaying a VE in stereo or monoscopic modes are available. These technologies include CAVES, passive and active stereo systems with magnetic or infra red tracking and horizontally curved screen displays (sometimes termed “reality theatre). A curved display set-up typically consists of a room containing a 7.5 m diameter screen over 150° of arc across the room, and from floor-to-ceiling. Three colour projectors are used to display computer generated images on this screen, and an advanced audio sound system enhances the impression of immersion in the VE. Although these display systems require expensive computing resources they can be used to display VEs created on standard PC-based systems and therefore may be used for final presentation of designs, layouts or training to a group of users. These displays are not stereoscopic but do attempt to promote a sense of immersion (physical enclosure in the display, thought to be associated with a sense of presence) via the size of the display.

It is feasible to envisage that VE applications in the workplace may use standard PC displays for single users (e.g. in the development of product design, architecture, or training applications) and projection screen displays for meetings and presentations. This has been seen by the authors in a number of industrial contexts (including aerospace and automotive) over recent years as a useful tool for communicating rationale behind design decisions that have been made using virtual prototype models or for supporting the design decision making process.

Each of these different types of viewing conditions produce varying combinations of sensory input to the participant. In all of the conditions there is a basic difference between the information received by the visual system and the vestibular or non-vestibular proprioceptive system during movement around a VE, where the visual information indicates that the participant is moving, but the vestibular and proprioceptive informs the participant that they are stationary. Sensory conflict theory [29] uses this difference as the basis for the causative theory of motion sickness. This theory also states that where unexpected conflict occurs between sensory inputs, the participant is more likely to experience sickness. There are differences in the extent to which this conflict occurs, and the degree to which this conflict is expected, in different VE viewing conditions. For example, in the desktop viewing condition, the participant usually has control over their movement within the VE, whereas in a curved large screen display, the most likely scenario is that the movement around the VE is governed by an independent controller. The participants in the desktop viewing condition will have a higher degree of expectation about the direction in which they are likely to travel, and the interactions with the VE that might be performed. Therefore it is necessary to consider not only how participants’ experiences of sickness differ in the different viewing conditions, but what the underlying differences between these conditions are, and therefore what might cause the differences in VRISE experienced. In addition, the lighting conditions in the viewing room may vary from light to dark, and this may affect the symptoms experienced by the participants. In this paper, both a general examination of the prevalence and severity of sickness in four display conditions, and a controlled examination of the role of two aspects of VE display conditions (active vs. passive viewing, light vs. dark viewing) are presented.

One of the main aims of this research project was to complete a controlled assessment of the influence of different VE display types on VRISE. This section of the literature review summarises previous work that has examined the effects of different display media.

The range of commercially available HMDs and the variety of conditions under which they are used (e.g. different virtual environments, users completing different tasks under different constraints and over differing time periods) make comparisons in symptom profiles between headsets difficult. The following section summarises the prevalence and severity of symptoms observed by the dominant researchers in the field. Mon-Williams et al. [22] state

“a number of factors have the potential to cause such stress to the visual system. These factors include poor illumination, poor contrast, and an unusually close working distance. All of these factors diminish as the quality of design improves and the technology of the components increases” [22, p. 207].

The dangers of such visual stresses following immersion in the workplace are summarised by

“The possibility of causing long- term problems through repeated immersions appears remote, although no studies have directly addressed this issue yet…of greater concern is the potential for a VR user to drive, or operate machinery, with unstable binocular vision and a decrease in visual acuity following immersion in a virtual world” [22, p. 210].

In a group of 20 adult volunteers with normal vision, a ten minute immersion in a virtual environment wearing a stereoscopic Head Mounted Display (Eyephone LX) resulted in 60% of participants reporting symptoms of eyestrain, headache and nausea, and 20% reporting a reduction in binocular visual acuity. In addition, 95% of participants demonstrated significant shifts in heterophoria towards asophoria and the number of subjects with associated heterophoria increased from 5% pre to 55% of participants post immersion [23].

Regan and Price [30] reported that 61% of participants immersed in a VE for a 20 min period reported symptoms, of which the greatest were reported at the end of the immersive period, with 45% of participants reporting some symptoms (whilst wearing a DVisor HMD). Similarly, Lampton et al. [20] investigated symptom severity during and following immersion in a range of HMDs. They reported that 4–16% of participants dropped out of the immersion due to adverse symptoms before their allotted time was over. However, they stated that most participants enjoyed the experience but reported some discomfort.

The prevalence and severity of side-effects associated with standard PC use were investigated in a review of the current literature by Dillon and Emurian [8]. They found that self report of visual fatigue may be evoked by: viewing distance, time on task, glare and lighting. They concluded a viewing distance of 65 cm was optimum [34], task time on extended periods of work (3–4 h) can cause temporary discomfort and should be avoided or include rest breaks. Workstations should be glare free and contain stable character presentation and proper lighting should be available [37].

Few papers have directly compared the presentation of similar stimuli on different display mediums. Burns [3] reviewed the differences between viewing electronic print (displayed on a VDU) vs. print (displayed on paper). He identified certain characteristics of CRT displays that may decrease the eyes comfort and efficiency as: screen flicker (disrupts fine eye movements affecting perception); spatial regularity of the display (minor effects on contrast sensitivity and post-task colour vision); screen declination (might necessitate separate spectacles for comfortable use of VDU). He concluded VDU and paperwork viewing of print possess different visual ergonomic attributes based on the characteristics identified above.

Deisinger et al. [7] investigated the differences between viewing a personal computer monitor screen vs. an HMD and screen based projection compared in an interactive environment. They found screen based projection gave inexperienced users the best feeling of immersion, and it was preferred to that of the HMD. The monitor display had much more acceptance with participants as they were familiar with using it in a normal working environment. Within the HMD, subjects had difficulty reading numbers on far away objects and hence task completion time was greater for this condition.

Garris-Reif and Franz [11] conducted a comparison in simulator sickness scores following viewing a HMD and personal computer screen system. Twenty-four participants, completed the Simulator Sickness Questionnaire (SSQ, developed by Kennedy et al. [17]) after completing a target location task viewed on a conventional table-top computer screen or a head slaved HMD. They found that average sickness scores were higher in the HMD group than the personal computer screen group (p < 0.026). Howarth [14] investigated symptom changes over 20 min immersions in 41 participants across five experimental conditions, two of which involved using a VDU (playing solitaire or DOOM), and the remaining three involved different head mounted displays (I-glasses, Virtuality, Visette and Division, PV100). He found that

“HMDs used to view environments, cannot be considered as a homogeneous group  symptom patterns are idiosyncratic to each HMD, or system” [14].

Howarth points out that many studies to date have examined the consequences of immersion in a single virtual environment and HMD and side effects reported cannot readily be attributed to any specific aspect of the environment, HMD, or hardware and software and such findings can therefore relate only to the consequences of immersion applicable to their specific equipment and not HMDs generally. His main results were found on visual, sickness and dizziness symptom subscales of the SSQ. Visually, blurred vision was significantly greater with the I-glasses than in any of the other displays, and general visual discomfort was significantly greater with both the I-glasses and Division conditions. Sickness increased with the I-glasses and Division HMDs compared to no changes in virtuality or either VDU conditions; for dizziness increases were observed for the Division HMD compared to VDU conditions, but not between the other HMD conditions.

Peli [27] measured the visual effects of completing a thirty minute task wearing I-glasses (in mono and stereo mode) and using a conventional desk-top computer display. No functional visual differences were found between the HMD and desk-top computer, however subjective data concerning comfort of the display types revealed using the HMD in stereo mode was significantly less comfortable than viewing the desk-top computer screen.

Despite an extensive literature search on projection screen technology and curved large screen displays (“reality theatres”), almost all published articles describe development of the technology itself rather than evaluating the effect of viewing displays on this technology on the participant. However, anecdotal reports indicate that some people, perhaps 0.5–1%, have found viewing VEs on a large display uncomfortable, in some cases distressing. Objective data is however difficult to obtain since the severity of the VR experience, whether it be simple viewing or purposefully dramatic, such as during simulated exposure to falling from a great height, appear to influence the effects (Rhodes, personal communication).

Interaction within virtual environments is achieved by using various input devices from keyboards, joysticks and three dimensional mice to data gloves. Using such inputs the user can determine the speed and direction of movement and the manipulation of objects. Such manipulations are responsible for the action occurring within the environment and so the user who generates such actions has ‘control’.

The other aspect of user control refers to ‘perceived control’ [31] based on the subjective cognitive appraisal of a situation, where the presence of perceived control can reduce the aversiveness of a situation [1], [38]. Unfortunately the amount of control a user has within a virtual environment varies between different virtual systems/environments. Some systems/environments allow user control over the speed and direction of travel and the ability to pick up and manipulate objects within the environment. However, other systems/environments might restrict user movements by slaving movement to a set path that cannot be deviated from. Such differences in user control may affect the quality of the users immersive experience, which might, in turn influence the incidence and severity of VRISE experienced.

Rolnick and Lubow [31] investigated the role of controllability in motion sickness by exposing 22 pairs of yolked subjects to nauseogenic rotation. Only one of each pair had any control over the rotation and head movements and the other was passively exposed to the same stimulus. They found that the subject in control reported fewer motion sickness symptoms than their passive yolked partner. Several subjects (15.9%) dropped out of the experiment due to severe malaise, of these five (71.4%) were from the passive group and two (28.6%) were from the control group. Similarly, when asked if they would like to participate in any future experiments eighteen (81.8%) of passive subjects declined compared to only nine (40.9%) of control subjects.

Hash and Stanney [12] highlighted research into the impact of a number of technical factors that may contribute to VRISE. However they state that the actual impact of these factors is not clear as issues of user control within the different experimental scenarios were not the same. The importance of user control is summarised:

“control may provide users with a means of adapting or to accommodate cue conflicts by building conditioned expectations through repeated interactions with a virtual world (e.g., when a users head turns the user learns to expect the world to follow milliseconds behind). Lack of control would not allow such expectations to be established since users would not be aware of which way they were turning at any particular moment (i.e., the course would be determined by the system)” [12].

Stanney and Hash [35] tested influence of user-initiated control on the level of cybersickness experienced by 24 college students participants immersed in a virtual environment. They found that the level of control a participant had over their immersive experience was highly significant for all factors on the SSQ with passive conditions leading to more severe symptom reports than active which led to more severe symptom reports than passive–active. Overall, the levels of symptoms reported and the percentages of subjects affected adversely were greater in this experiment than had been reported previously. Participants (95%) reported some adverse effects due to immersion compared to 60% of aircraft pilots in comparable studies [19]. In their conclusions Stanney and Hash [35] state that: firstly, passive observers can be expected to experience a high rate of incidence and severity of symptoms. Secondly, active participants may experience less symptoms but may have problems updating their neural stores due to the amount of sensory information they will be exposed to due to their unrestricted movement and thirdly, coupled control is an effective method of minimising symptoms by allowing participants to have task-orientated control within a VE.

The aim of the experiments presented in this paper was to compare the effects of four different VR viewing conditions on VR induced sickness symptoms. The four conditions compared were; head mounted display, desktop, standard projection screen and reality theatre (horizontally curved large screen display). In addition, two specific variables of interest, lighting conditions and user control, were examined.

Section snippets

Experimental VE

The VE used throughout the experimental programme was a Virtual Factory. This environment was originally developed as a VE to teach adults with learning disabilities about potential health and safety hazards that might be encountered in the workplace [5]. It was chosen as the experimental environment as it was quite complex, and thus included enough potential interactivity and tasks to keep experimental participants occupied for the chosen length of VE exposure of 30 min. It was also quite

Experimental conditions

Experiment 1 was a comparison of VRISE in the four display conditions of interest, with lighting conditions configured as they would normally be expected to be used in the workplace. Table 2 describes the experimental conditions used.

Participants

Seventy-one participants completed the experiment, 38 male and 33 female. The sample consisted mainly of students from the University of Nottingham. The volunteers were paid for their participation in the experiment.

Symptom reports

SSQ scores reported before and after exposure to

Experimental conditions

Two experimental conditions were run to compare the symptoms experienced in light and dark viewing conditions. For dark conditions the windows in the experimental room were blacked out completely with fixed plywood screens that covered the windows preventing any natural light from entering the room. For light conditions – the windows in the experimental rooms were uncovered and natural light could enter the room. In addition, artificial light, which came from fluorescent lights present in the

Experimental conditions

Two experimental conditions were run to compare the symptoms experienced in active and passive control conditions. These conditions used a projection screen display in dark conditions, and the VE was the Virtual Factory. Table 6 describes the experimental conditions used.

Participants

Thirty-one participants completed the experiment, 18 male and 13 female. The participants were mainly students from the University of Nottingham, and were paid for their participation in the experiment.

Symptom reports

Table 7 shows the SSQ

Overview of symptom profiles

Fig. 2 shows the post-exposure symptom profiles for all the experimental conditions run throughout the experimental programme. It can be seen that there is a general predominance of oculomotor symptoms, particularly in the projection screen conditions, and the majority of the Reality Theatre conditions. This is interesting, as it is a symptom profile more familiar in simulator sickness than VR-induced sickness [16]. This may be due to a number of differences between the HMD and projection-based

Discussion of findings

This study provides the first direct comparison of viewing the same virtual environment under alternative display configurations. A series of experiments were conducted to examine VR-induced sickness reported under the following conditions: HMD, desktop, standard projection screen and Reality Theatre under normal viewing conditions; active desktop viewing under light versus darkened room conditions; active versus passive control of a VE viewed on a standard projection screen in a darkened room;

Recommendations and conclusions

The different designs of VE used in this project did provide an indication of the implications of VE design on VRISE, but as they were not manipulated systematically it was not possible to make inferences regarding particular VE design characteristics and VRISE. So et al. [33] have attempted to quantify visual scene movement within a VE with the potential to link this to levels of sickness experience, and the authors have developed a prototype classification – Virtual Environment Description

Acknowledgements

This work was conducted under funding support from the UK Health and Safety Executive and we thank Trevor Shaw for his support and guidance throughout the project. Virtual environments were built by the VIRART at the University of Nottingham and the authors thank Richard Eastgate, Steven Kerr, Rick Barnes and Victor Bayon-Molino. We would also like to thank Tim Whitehouse and David Rhodes from the Centre for Industrial and Medical Informatics (CIMI), University of Nottingham, for providing

References (39)

  • J. Deisinger et al.

    The effect of different viewing devices for the sense of presence and immersion in virtual environments: a comparison of stereoprojections based on monitors, HMDs and screen

  • H. B-L. Duh et al.

    Effects of characteristics of image quality in an immersive environment

    Presence: Teleoperators and Virtual Environments

    (2002)
  • H.B-L. Duh et al.

    An independent visual background reduced simulator sickness in a driving simulator

    Presence: Teleoperators and Virtual Environments

    (2004)
  • R. Garris-Reif et al.

    Simulator sickness and human task performance in conventional and virtual environments

  • P.A.K. Hash et al.

    Control: a primary driver of cybersickness

  • P.A. Howarth, The multi-factorial visual causes of virtual simulation sickness, VISERG Report,...
  • W. Ijseelsteijn et al.

    Effects of stereoscopic presentation, image motion and screen size on subjective and objective corroborative measures of presence

    Presence: Teleoperators and Virtual Environments

    (2001)
  • R.S. Kennedy et al.

    Profile analysis of simulator sickness symptoms: application to virtual environment systems

    Presence

    (1992)
  • R.S. Kennedy et al.

    Simulator sickness questionnaire: an enhanced method for quantifying simulator sickness

    The International Journal of Aviation Psychology

    (1993)
  • Cited by (0)

    1

    Née Nichols.

    View full text