Effects of virtual environment platforms on emotional responses☆
Introduction
Over the past two decades, virtual environments (VEs) have been developed for use as a human–computer interaction platform across a wide variety of applications, including education, skill training (e.g., flight simulator), gaming, and the scientific study of complex human behavior [1]. Within the fields of psychiatry and psychology, VEs broadly have been used both to characterize basic processes in healthy and psychiatric samples and in behavioral interventions (e.g., exposure therapies for anxiety disorders) [2]. More specifically, this growing body of research includes studies across a wide range of samples, including individuals with anxiety disorders [3], addiction [4], eating disorders [5], obsessive–compulsive disorder [6], pain [7], [8], and stroke [9], [10], [11]. In healthy samples, for example, VEs have been used to study basic processes of learning and emotion [12]. In psychiatric samples, VEs have been used to identify candidate behavioral markers of psychiatric problems, in order to ultimately identify objective assessment measures and predictors of treatment response (e.g., compulsive behaviors in OCD; [6]; cue reactivity in addiction [4]). Clinical uses of VEs have been reviewed extensively elsewhere (e.g., [7], [13], [14], [15]), with interventions that have shown promise in, for example, reducing anxiety, coping with acute pain, improving body image, reducing compulsive behavior.
Because VEs permit the scientific study of human behavioral responses in a controlled manner using ecologically valid contextual cues, these platforms have the capability of enhancing both internally and externally valid experimental designs. However, despite the promise of VEs in the experimental investigation of neurobehavioral processes (e.g., emotion, attention, behavioral disinhibition) and interventions for psychiatric disorders, many questions remain about how to optimize the use of VEs.
A key consideration for researchers and clinicians using VEs is the selection of VE platform. For example, researchers or clinicians interested in using VEs to investigate emotional processes (e.g., emotional reactivity) must choose a particular method of presenting VEs to participants. The selection of VE platform (e.g., desktop computer alone, desktop with head mounted technologies, fully immersive technologies, etc.) could influence the emotional responses of participants engaging the VEs, and, accordingly, may influence the primary dependent variables of the research. Although higher emotional arousal is associated with an increased experience of immersion in VEs [16], it is unknown whether fully immersive VE platforms are associated with different emotional responses in emotional stressors tasks compared to less immersive technologies. As such, the primary purpose of the present study was to investigate whether VE platforms elicit differential emotional responses to a stressor task. The aim of this study was to provide empirical evidence capable of helping researchers and clinicians select an appropriate VE platform when using VEs to elicit emotional responses for either basic or applied purposes.
Studies with VEs have utilized a range of technological platforms, including: (1) a simple desktop personal computer (PC) with a standard monitor, (2) a desktop PC or laptop computer with a head mounted display system (HMD; i.e., goggles with limited peripheral vision), and (3) more immersive virtual environments where the participant stands or walks inside a room with a projected computer display across up to six walls, including the floor and ceiling (i.e., cave automatic virtual environment; CAVE). Although the choice of technological platform for any given experiment may be influenced by a number of pragmatic and logistical factors (e.g., cost, space, expertise available, etc.), this decision also could be informed by differential response patterns expected across platforms.
Previous studies have begun to investigate this issue. For example, researchers have evaluated projection-based dome, HMD, desktop, or CAVE systems for navigation tasks, such as way-finding tasks and path-finding tasks [17], [18]. In one of these studies, Swindells et al. [17] found evidence to suggest that large screens optimize task performance for navigation and way-finding tasks. In addition, to identify the appropriate technological platform to use for visualization-based tasks, researchers have compared various platforms to each other, including: (1) CAVE v. a fish tank system (i.e., a monitor-based desktop system with head position tracker), (2) HMD v. fish tank system, and (3) desktop v. fish tank and CAVE systems [19], [20], [21]. In one of these studies, Ware and Franck [22] found that a platform using stereoscopic vision combined with a motion tracker enhanced performance in a three-dimensional (3D) visualization task compared to a traditional desktop PC.
Most previous studies investigating differential responses across VE platforms have been designed to address this issue in tasks involving spatial navigation or visualization. In contrast, considerably less is known about whether basic processes of emotional arousal and the regulation of emotions are differentially influenced by the type of VE technological platform. Juan and Pérez [23] reported that the CAVE provokes more anxiety than the HMD platform under acrophobic VE based on one-item anxiety measures (from 0 = not anxious at all, to 10 = very anxious). However, we are not aware of other studies that have compared VE platforms to study emotional responses and task performance in healthy samples. In the present study, we examined the role of different VE technologies on performance both in high-stress and low-stress tasks, as well as on subjective and psychophysiological dependent measures of stress reactivity. These measures of stress reactivity were obtained in the most representative VE systems spanning low to high-end technological platforms: a simple PC with monitor, a standard PC plus inexpensive commercially available head-mounted display (HMD), and the six-wall fully immersive CAVE system (Duke immersive Virtual Environment; DiVE).
Other well studied factors on selection of a VE technological platform in clinical [24] and non-clinical [23] settings were presence and simulator sickness. In the present study, we also examined effects of different VE systems on presence and simulator sickness, in order to replicate previous findings. The sense of presence is defined as the psychological perception of being in or existing in the virtual environment in which one is immersed [25]. Sanchez-Vives and Slater [26] detailed the importance of presence in a review of VE studies. Indeed, across studies using VE, presence is widely thought to have an important role in the elicitation of emotional arousal and sense of realism in VEs [27]. Because presence is a subjective state, it is typically measured using self-report inventories [28], but also has been investigated with behavioral [29] or psychophysiological measures [30]. Technologies designed to enhance immersion and multi-sensory stimulation in VEs may be likely to increase presence. In line with this, previous studies have found that a CAVE system induces a higher level of presence than HMD [23]. Similarly, Ijsselsteijn et al. [31] suggested that stereoscopy and tactile augmentation significantly increased presence, Hendrix and Barfield [32] showed that head tracking contributed to higher presence, and Welch et al. [33] found that pictorial realism and interactivity increased the level of presence.
The severity of simulator sickness (i.e., motion sickness experienced in response to a VE) has been commonly measured with simulator sickness questionnaires [34]. Using this methodology, researchers have reported that the severity of simulator sickness was correlated with psychophysiological responses, including gastric tachyarrhythmia, eye-blink rate, and cardiovascular functioning [35]. Previous studies also have compared the effects of different VE systems on simulator sickness. Specifically, studies have found that both desktop and CAVE systems induce less simulator sickness than HMD systems [36], [37]. Recently, Sharples et al. [38] also found higher simulator sickness using the HMD compared to a standard desktop monitor or a curved monitor. Based on such findings in previous studies, it can be expected that technological platforms using multi-sensory and fully immersive platforms may increase presence, and that HMD-based VEs may induce the highest level of simulator sickness. However, to the best of our knowledge, there has been no study that directly compares these three different VE systems (i.e., desktop, HMD, and CAVE systems) on presence and simulator sickness in the context of an emotionally evocative stressor task. As such, in the current study, we directly compared these three common VE systems on presence and simulator sickness in response to a low- and high-stress behavioral task.
We examined emotional reactivity and behavioral performance using a modified 3D Stroop task presented in low and high stress contexts across three representative VE systems. The Stroop task [39] is a measure of information processing that has been modified for use to study a wide variety of basic neurobehavioral processes, including attention [39], sensory processing [40], and emotion [41], [42], [43]. Traditionally, this task measures attentional interference by presenting color names in congruent and incongruent colors, with participants instructed to identify the color name as quickly as possible. Conceptually, in the present study we were interested in using a behavioral task that did not directly measure emotional reactivity, but could be used in emotionally evocative contexts. More specifically, this task was chosen because: (a) it is a measure of basic attentional processes with a rich empirical history [41], (b) could be easily modified for use in a low- and high-stress paradigm with non-emotional words, and (c) could be used in future studies investigating emotional reactivity in VEs by using emotionally-relevant words in low- and high-stress VEs to examine the interplay between information processing and emotional reactivity.
The experiment was a 3 (CAVE, HMD and Desktop PC) × 2 (high-stress VE and low-stress VE) within subject design, with the primary dependent variables of: (a) psychophysiological and self-reported emotional reactivity, (b) behavioral performance on the task, (c) presence, and (d) simulator sickness. Generally, we expected more immersive VEs to be associated with higher emotional arousal, presence, and enhanced task performance. More specifically, it was hypothesized that: (1) emotional arousal (i.e., psychophysiological and self-reported) would be significantly higher in the CAVE system compared to the HMD or desktop PC systems; (2) performance on the modified 3D Stroop task would be significantly better in the CAVE compared to either HMD or desktop PC platforms; (3) presence would be significantly higher in the CAVE compared to either HMD or desktop PC; (4) the HMD platform would elicit significantly higher simulator sickness and negative emotion compared to either the CAVE or desktop PC and (5) there would be significant main effect for stress (high v. low) across VE platforms, with high-stress VEs eliciting greater emotional arousal and lower valence than low-stress VEs.
Section snippets
Participants
Prior to the beginning of recruitment, this protocol was approved by the Duke University Institutional Review Board. Participants were students (Age: M = 21.6, SD = 3.9) at Duke University recruited through the Department of Psychology and Neuroscience Subject Pool, a student e-mail list-serve, and through flyers posted on campus. All subjects received $15 for participating. Out of 55 participants who consented to participate, one was excluded due to significant self-reported anxiety (i.e., ≥16 on
Results
The first step of the data analytic process included screening the data for missing entries and data entry errors. Next, we calculated the values of each variable in each condition. As described above, we obtained baseline values of resting state for arousal, valence, and SCR. To calculate the effect of each VE system on these measures, we subtracted the baseline values from each value obtained in response to each VE system. Therefore, our dependent variables of arousal, valence, and SCR were
Discussion
The primary aim of the current study was to investigate whether different VE platforms elicit unique patterns of emotional responses and behavioral performance in both high- and low-stress tasks contexts. We conducted an experiment with three representative kinds of VE systems: A Desktop PC without HMD, Desktop PC with HMD, and a 6-wall fully immersive VE (i.e., DiVE). Dependent variables were obtained using self-report (i.e., sense of presence, simulator sickness, emotional arousal, and
Conflicts of interest
None declared.
Acknowledgements
We are grateful to Billy Dwight and Holton Thompson who helped with data collection and graphic design. K. Kim was supported by the research fund of Hanyang University (HY-2013).
References (62)
- et al.
Using virtual reality to investigate complex and contextual cue reactivity in nicotine dependent
Addictive Behaviors
(2011) - et al.
Development of a computer-based behavioral assessment of checking behavior in obsessive–compulsive disorder
Comprehensive Psychiatry
(2010) - et al.
Virtual reality for persistent pain: a new direction for behavioral pain management
Pain
(2012) - et al.
An interactive 3-D application for pain management: results from a pilot study in spinal cord injury rehabilitation
Computer Methods and Programs in Biomedicine
(2012) - et al.
Development of a system based on 3D vision, interactive virtual environments, ergonometric signals and a humanoid for stroke rehabilitation
Computer Methods and Programs in Biomedicine
(2013) - et al.
Design of an efficient framework for fast prototyping of customized human–computer interfaces and virtual environments for rehabilitation
Computer Methods and Programs in Biomedicine
(2013) - et al.
Affective outcomes of virtual reality exposure therapy for anxiety and specific phobias: a meta-analysis
Journal of Behavior Therapy and Experimental Psychiatry
(2008) - et al.
Virtual reality exposure therapy of anxiety disorders: a review
Clinical Psychology Review
(2004) Viability of virtual reality exposure therapy as a treatment alternative
Computers in Human Behavior
(2008)- et al.
The role of presence in virtual reality exposure therapy
Journal of Anxiety Disorders
(2007)
Virtual reality induced symptoms and effects (VRISSE): comparison of head mounted display (HMD), desktop and projection display systems
Displays
Generalizability of carry-over effects in the emotional Stroop task
Behaviour Research and Therapy
Does the modified Stroop effect exist in PTSD? Evidence from dissertation abstracts and the peer reviewed literature
Journal of Anxiety Disorders
Feasibility and potential effect of a low-cost virtual reality system on reducing pain and anxiety in adult burn injury patients during physiotherapy in a developing country
Burns
Comparison of checking behavior in adults with or without checking symptom of obsessive–compulsive disorder using a novel computer-based measure
Computer Methods and Programs in Biomedicine
Measuring emotion: the self-assessment manikin and the semantic differential
Journal of Behavior Therapy and Experimental Psychiatry
Handbook of Virtual Environments: Design, Implementation and applications
Virtual Reality Therapy for Anxiety Disorders
Effectiveness of computer-generated (virtual reality) graded exposure in the treatment of acrophobia
American Journal of Psychiatry
Virtual reality in eating disorders
European Eating Disorders Review
Development of virtual reality proprioceptive rehabilitation system for stroke patients
Computer Methods and Programs in Biomedicine
The validity of virtual environments for eliciting emotional responses in patients with eating disorders and in controls
Behavior Modification
Affective interactions using virtual reality: the link between presence and emotions
Cyberpsychology & Behavior
Comparing CAVE, wall, and desktop displays for navigation and wayfinding in complex 3D models
Proceedings of Computer Graphics International
Head-mounted display versus desktop for 3D navigation in virtual reality: a user study
Multimedia Tools and Applications
CAVE and fishtank virtual-reality displays: a qualitative and quantitative comparison
IEEE Transaction on Visualization and Computer Graphics
A comparative study of desktop, fishtank, and cave systems for the exploration of volume rendered confocal data sets
IEEE Transaction on Visualization and Computer Graphics
A comparison of immersive HMD, fish tank VR and fish tank with haptics displays for volume visualization
Evaluating stereo and motion cues for visualizing information nets in three dimensions
ACM Transaction on Graphics
Comparison of the levels of presence and anxiety in an acrophobic environment viewed via HMD or CAVE
Presence: Teleoperators and Virtual Environments
Comparison of two VR platforms for rehabilitation: video capture versus HMD
Presence: Teleoperators and Virtual Environments
Cited by (0)
- ☆
These data were presented as a poster at the IEEE VR conference on March 7th, 2012 in Orange County, CA.