Potential of virtual reality as a diagnostic tool for social anxiety: A pilot study

Highlights

• Virtual reality is a potential diagnostic tool for social anxiety disorder.

• High-socially-anxious participants fixated on facial regions significantly shorter.

• Analysis of gaze behavior is a suitable measure for social anxiety.

Abstract

The potential of virtual reality for diagnosing anxiety disorders has been explored to a lesser extent than its use in psychotherapy so far. The current study applied virtual environments as an innovative diagnostic tool for social anxiety and social anxiety disorder and examined differences between low- and high-socially-anxious participants on the basis of specific psychological parameters and recordings of gaze behavior. Out of 119 subjects, 19 low- and 18 high-socially-anxious participants were selected by a social anxiety questionnaire. During the completion of eight tasks in a virtual train and waiting room scenario, the skin conductance response and gaze behavior was monitored. The findings indicate that analyzing fixation durations of faces in a virtual social situation is even more suitable for distinguishing low- and high-social-anxious participants than investigations of skin conductance responses.

Introduction

Virtual reality (VR) is a promising tool, which can simulate the complexity of experiences of the real world in a controllable and reproducible manner in the laboratory and activate fear networks in anxious subjects (Mühlberger et al., 2008, Shiban et al., 2015, Shiban et al., 2013). While the advantages of the application of VR in psychotherapy have already been illustrated (for a review see Freeman et al., 2017, Krijn et al., 2004, Meyerbröker and Emmelkamp, 2010, Opriş et al., 2012), clinical-psychological diagnostics in virtual surroundings are still in the early stages of development. Virtual social scenarios have mainly been generated for therapeutic purposes thus far.

Typically, participants had to give speeches in front of a virtual audience (Harris et al., 2002, Slater et al., 2006). However, additional social settings exist: e.g. a virtual clothing shop (Brinkman, Van der Mast, & de Vliegher, 2008, pp. 85–88), a virtual elevator (Mühlberger et al., 2008) or a virtual application interview (Grillon, Riquier, Herbelin, & Thalmann, 2006).

The current study assessed the application of virtual technology for psychological diagnostics in the case of social anxiety disorder. To simplify the analysis of eye movements, the VR was viewed over a desktop system as opposed to head-mounted displays (HMDs). Hence, problems such as motion sickness or distortion of data due to prevalent gesture-based interactions did not emerge (Kim, Kim, & Nam, 2016).

Train and waiting room scenarios were established in the framework of a behavior test to track the gaze behavior and skin conductance response (SCR) and to expand the results of commonly used virtual social paradigms. The essential feature of social phobia is an intense fear of social situations in which the individual is exposed to scrutiny by others such as social interactions, being observed and performing in front of others (American Psychiatric Association, 2013). In this experiment, we aimed at testing these main criteria. The train scenario includes triggering elements of social interaction and performance as the participants have to make eye contact with the porter and to complete a calculation task, and the waiting room scenario, which requested participants to make eye contact, to ask for a seat and resulted in being rejected, targeted fear of social interactions, observance and rejection. Furthermore, situations which include performance in public and meeting new people represent social situations that are feared most in social phobics (Crome & Baillie, 2014).

There is evidence that the measurement of the SCR can help distinguish between high- and low-socially-anxious people. A slower habituation and more spontaneous fluctuations of palmar skin conductance responses towards social stimuli were shown for socially-anxious individuals compared to controls (Lader, 1967). Dimberg, Fredrikson, and Lundquist (1986) revealed that the slower habituation and the amplitude of the SCR were positively correlated with the severity of social anxiety. Furthermore, a higher skin conductance level was reported for high-socially-anxious individuals during the imagination of socially-threatening stimuli, which were described by an auditory script (McTeague et al., 2009). Nonetheless, others reported results contradictory to the above (Edelmann and Baker, 2002, Mauss et al., 2003, Mauss et al., 2004).

Gaze behavior examinations revealed that, when observing faces, healthy individuals show a consistent pattern of eye movements. Characteristically, attention is paid to salient elements of faces, like eyes, nose and mouth; thus, the gaze movements represent a reversed triangle (Mertens et al., 1993, Walker-Smith et al., 1977). The eyes are especially focused on as they can provide crucial information about the emotions of an interaction partner (Lundqvist, Esteves, & Ohman, 1999). Earlier studies found a deviating behavior in the case of social anxiety disorder. Affected individuals avoided the eye regions of faces (Hofmann et al., 1997, Wieser et al., 2009). Chen, Thomas, Clarke, Hickie, and Guastella (2015) reported that socially-anxious individuals giving speeches fixated longer on non-social regions, defined as regions around the faces of people in a video-taped audience.

Consequently, we assumed that high-socially-anxious individuals have a higher SCR in the behavior test in the VR than low-socially anxious controls.

Furthermore, it was expected that high-socially-anxious individuals exhibit shorter fixation durations on the faces of virtual agents.