Abstract

Virtual reality exposure therapy (VRET) is a commonly used treatment for anxiety disorders. It is often used to treat those with public speaking anxiety (PSA). Yet qualitative research examining users views and experiences of using virtual reality (VR) as a treatment for PSA, in general, is scarce. This study investigates users’ experiences and views of using virtual reality as a treatment for PSA from a qualitative perspective using semi-structured interviews. Participants included 1 male and 5 females all students of Nottingham Trent University. All had previous experience of a VR intervention, in which they had control over their environment. All suffer from PSA. 6 interviews were carried out and data was analysed using thematic analysis. 4 main themes with adjoining sub-themes were identified: “Anticipation of presenting”, “Immersion”, “Usefulness” and “Recommendations”. Views and experiences of using VR were generally positive with participants finding the overall experience realistic and useful. Participants suggested VR to have useful application for reducing PSA, yet suggestions were made for improvement. This included improving the detail of the virtual audience and environment.

Introduction

Public speaking anxiety (PSA)

The most prevalent form of social anxiety is public speaking anxiety (Manuzza et al 1995). As of today, it affects approximately one in five people (Bartholomay and Houlihan., 2016). This can be very debilitating for the sufferer as presentation giving is crucial and a regularity to many jobs, meaning PSA can have a serious impact on one’s career prospects (Pertaub et al., 2001). Those who suffer from PSA in professions that require them to speak, can experience it so intensely that it affects their ability to perform at all (Katz et al., 2000). Those with PSA may suffer mediocre academic performances, enhanced feelings of loneliness or social isolation, and a general lower quality of life (Beidel, Turner, and Dancu, 1985).

Students are not an exception to this as presentations are regularly expected at degree level. Many students fear public speaking and often report the need for classes on how to improve their presentation performance (Marihno et al., 2017). Fear of public speaking can affect one’s personal and social life which could lead to a negative university experience.

Existing treatments for PSA

The preferred treatments for PSA are cognitive behavioural therapies (Wallach et al., 2009). Behavioural therapy techniques for treating phobias and anxiety disorders often involve the use of graded exposure or systematic desensitization (North et al., 1998). These techniques often involve the use of self-flooding or visualization techniques. As some are more capable of producing vivid images of anxiety-provoking scenarios than others, visualization is more effective in those who can create vivid mental images (Ayres and Hopf., 2009). Thus, the use of Virtual reality could potentially overcome some of the difficulties inherent in traditional

treatment of anxiety disorders and phobias (North et al., 1998).

The use of Virtual Reality as a treatment for anxiety disorders and phobias

VR as a therapeutic intervention (sometimes called virtual reality exposure therapy – VRET), creates an immersive virtual world that integrates real-time graphics, visual displays, and body trackers and other sensory inputs (Krijin et al., 2004).

Virtual Reality enables one to enter a computer generated virtual environment in which they can be exposed to feared stimuli within a context relevant setting (Parsons and Rizzo., 2007). Today, Virtual Reality (VR) is often used over other techniques as virtual environments are easier to control than the real world (Kang et al., 2016). Though the focus of this study is based on Virtual reality’s application to Public Speaking anxiety, it is helpful to highlight a few complimentary uses of this technology in psychology.

Useful applications of Virtual reality are well known (Pull et al., 2005). Successful treatments of phobias include: Fear of flying, fear of heights and arachnophobia (Opdyke et al., 1995; Rothbaum et al., 1996; Hoffman and Weghorst., 1997). VR has also been prevalent in the aiding of cognitive rehabilitation in schizophrenia sufferers, stroke patients, autistic disorder and PTSD (Lee et al., 2003; Bojan et al., 2001; Strickland et al., 1998). It can enhance performances in, psychology, medicine and in sporting fields (Aufegger et al., 2015).

Immersion/presence

The main advantage of using VR is that individuals can be exposed to an anxiety provoking situation in a controlled manner, so they can learn to deal with rejection and criticism becoming comfortable in situations that would usually trouble them (Botella et al., 2005). However, efficacious VR therapies must meet specific underlying needs. One being the extent to which the environment emerges the participant so they are immersed and feel present (Owens & Beidel, 2014). Presence is defined as the interpretation of an artificial environment as though it is real (Lee et al., 2004). It has been discussed as the mechanism by which VR may be an effective tool for exposure therapy (Wiederold et al., 2005A). It is a fundamental concept for understanding the effectiveness of virtual environments (Macintyre et al., 2004). The concept is key in defining virtual reality in the context of human experience rather than its technological hardware (Steur et al., 1992).

Owens and Beidel (2014) proposed that a VRET can only be considered effective if it is in line with Foa and Kozak’s (1986) theory of emotional processing. This theory proposes a successful exposure therapy leads to new memory structures being created

Which ‘overrule’ the previous anxiety provoking one’s. Thus, in the context of VR, it could be an effective replacement for normal exposure therapies if it could elicit fear and activate the anxiety-provoking structure (Owens & Beidel., 2014).

It has been argued that humans tend to treat virtual humans as ‘social actors’ instead of a virtual image, so computer-generated virtual audiences can elicit realistic responses (Nass et al., 1996). Furthermore, some studies have detailed that even when participants are aware audiences are virtual, physiological arousal is still displayed in response to different audience types (Rosenblum & Macedonia., 1999; North et al., 1998).

Virtual reality as a treatment for Public Speaking Anxiety

Virtual reality (VR) technology has been successfully applied to help individuals whom are anxious in certain social situations (North et al., 1998). A well-known application of VR is in the treatment of public speaking apprehension. VR is seen to reduce public speaking anxiety and increase performances in reality. (Anderson & Hodges, 2005; Pertaub & Slater, 2002; Wallach et al., 2009). Anderson (2005) discovered lower self-report measures of PSA, and better public speaking performances after a VRET.

The use of VR as a treatment for PSA, involves creating VR components, interwoven with key features of real public speaking situations (Aufegger et al., 2015). This usually consists of the creation of virtual auditoriums and audiences (Slater et al., 1998). There is research which supports the notion that VR environments created for public speaking situations are effective and immersive. For example, North et al (1998) created a virtual auditorium consisting of 3 sections of chairs containing virtual characters up to the hundreds. They confirmed their VRET was an effective treatment of PSA, reducing participant’s anxieties and improving real world performance. Pertaub et al, (2002) created a virtual seminar room consisting of 8 virtual audience members. The virtual audience displayed characteristics which would be observed in a real-life scenario, giving the illusion of life (Kwon et al., 2013). They examined participant’s responses in comparison to 3 different audience attitudes and discovered participants responded appropriately to each audience (i.e. anxiety increased with a negative attitude and decreased with a positive audience attitude. Roy et al, (2003) similarly used a virtual audience to provide situations for four different subtypes of social anxiety, including performance anxiety, assertiveness anxiety, observation anxiety and informal speaking anxiety. All situations successfully provoked anxiety responses and Roy et al, concluded that VR was effective in eliciting an emotional response from participants and in reducing anxiety levels after exposure.

Multiple studies have assumed VRET to be effective in treating PSA. Public speaking performances often improve after virtual exposure. However, observations of VR effectiveness often rely on the use of self-report measures and recordings of physiological arousal (Morina et al., 2015). Thus, there is a lack of research examining, in detail, how and why VR is effective, as user perspective is rarely gained. Furthermore, the application of VR to students, is rare.

Research aim

It is noticeable there is a lack of research which use interviews to gain a first-person perspective into user experience of virtual reality. Thus, potentially limiting ways of improving virtual environments and exposure therapies. To truly understand how effective and immersive virtual reality is, more research must be done in an in-depth qualitative manner. This further enables an understanding of what specific aspects of a virtual environment or audience are effective in producing a realistic experience.

Furthermore, to the knowledge of this study, no research examining VR as treatment for PSA used a virtual environment where participants were given control over their environments.

The aim of this research, therefore, was to use a qualitative approach to gain a holistic understanding of the use of a virtual reality intervention as a treatment for students who suffer from PSA. To gain an in-depth understanding into participants’ views and experiences of using virtual reality to aid Public Speaking Anxiety. The question being addressed in this study was: What were participants views and experiences of using a virtual reality exposure as a treatment for public speaking anxiety? How realistic they found it, how it could be improved, and how controlling their environments affected them were other areas addressed.

Method

 

Description of the VRET

Participants had previously taken part in a VRET which enabled them to enter a virtual lecture theatre where they gave two 20-minute presentations on their experiences of university using a VR headset. Participants were given control over their environments. They could change conditions to their preference. Changeable conditions were: audience size (6/12/20 people), audience mood (agreeable, neutral, disagreeable), and poster type (poster with silhouette of name/ poster with photo of participant and silhouette of name/ no poster). They were also able to decide the number of prompts which were available to help prompt them to remember information (low no. prompts, medium no. prompts, high no. prompts). Participants did this in a private lab with an experimenter present in the room with them.

Procedure

Participants in the study were recruited with the help of the researchers who conducted the previous VRET experiment. Researchers sent out an email inviting those who had taken part in the VRET to be interviewed about their experiences. Following the gaining of consent, participants were invited to a one on one, hour long, semi-structured interview. Interviews were conducted in a quiet experimental cubicle, booked through the Nottingham Trent social science room booking webpage. Participants were notified that they would be required to answer 30 questions based on their previous experience of presenting to a virtual audience. Questions were asked by the researcher. Data was recorded on an audio tape recorder. All questions were open-ended. Participants were encouraged to expand on answers when possible.

Participants

Participants consisted of 5 females and one male who had all taken part in the previous experiment using VRET. All 6 were students of Nottingham Trent University. All suffered from PSA. Participants had a mean age of 18.7 years.

Materials

A schedule consisting of 30 items was used to guide the interview. The schedule consisted of open ended questions and took the priority of attempting to understand the views and experiences of VRET. Responses were recorded using an audio recorder and later typed up as a transcript onto a computer.

Analysis

Thematic analysis was chosen for analysing the interviews as it was believed to be the best method for being able to describe and explore the views and experiences of participants using the VR. The reasoning behind the choice was that it would allow an in-depth description of patterns in the datasets to be found and thus be suitable for the exploratory aims of the research. An essentialist/realist method was adopted to report the experiences, meanings and the reality of participants (Braun & Clarke, 2006).

Braun and Clarke’s (2006) guidelines on how to conduct a thematic analysis were followed closely when analysing the data from the interviews. This included reading and re-reading interviews, coding the interviews, finding and reviewing themes, and then developing and expanding on themes.

Results

Table 1. Themes and emergent subthemes

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

The thematic analysis revealed 4 main themes with their respective sub-themes listed beneath them.

The interview number was placed in brackets beside each quote and ‘P’ and ‘I’ were abbreviated for participant and interviewer.

The analysis focused on participants views and experiences of using virtual reality to give a presentation inside a virtual environment.

Theme 1 – Anticipation

All participants were asked on how they felt before giving a presentation in the virtual environment. Most participants suggested they felt some form of anxiety or apprehension before presenting in front of the virtual audience.

Anxiety

When asked about how they felt before taking part in the virtual exposure, most participants, similar to a real public speech, were apprehensive about presenting in the virtual environment.

I: “How did you feel before presenting in front of the virtual audience?”

P: “I felt like before I started I felt more nervous than an actual presentation because I didn’t know like what to expect in the like virtual thing” (James, interview 2)

Similarly, another participant stated their fears before presenting:

P: “I was so scared, I couldn’t work out why I was so scared because I knew it wasn’t real but I didn’t want to come, I dint want to come like 2 days before this was happening, I was really, really nervous about it” (Rachel, interview 4)

 

One participant detailed that it took them a while before they could become confident enough to speak:

P: “I was very nervous it actually took me a long time before I could start to speak” (Katie, interview 6)

Another participant mentioned:

P: “Um, I don’t know, it was a bit weird it sort of felt like quite real so I was still a bit like oh no I have to present now” (Charlotte, interview 3)

Feelings of anxiousness and nervousness are common feelings experienced prior to public speaking. Here the feelings were seen to be a result of the novelty of the virtual exposure, students were unsure of what to expect from the virtual audience as none had taken part in virtual reality before. Research has proposed the highest levels of anxiety experienced in people with PSA occurs just before speaking (Behnke & Sawyer, 2009). This was no different for a presentation inside a virtual environment.

 

 

 

Theme 2 – Immersion

 

Most participants mentioned during their interviews that they felt immersed within the virtual environment during the exposure. However, participants were aware the environment was artificial, yet they still found the experience immersive and realistic to an extent.

Some aspects like the audience movement were detailed as being unrealistic and prevented the participants from being fully immersed for the whole presentation. Therefore, the extent to which participants were immersed within the virtual environment differed.

Translation to real life – “Realistic but I knew it wasn’t reality”

When asked how realistic the experience was, one participant mentioned that they experienced physiological arousal within the virtual environment.

P: “definitely yeah I felt like I was really there um it was weird because it did steam up because I was sweating so much um so It was a little bit blurry at times but it was quite immersive” (Sarah, interview 4)

 

When asked the same question, another participant similarly stated:

P: “I kept going in and out of reality like I was told I would have 4 minutes until the end of the presentation and that brought me out a bit, but then the 4 minutes went so quickly because I was so engrossed in the presentation” (Rachel, interview 5)

 

Another participant detailed how involved they were with the virtual environment:

IQ: “How real did the experience feel?”

P: “When you are inside it you literally felt as if you are giving a real presentation but once I took it off and came back into reality it was a massive difference but once I was inside it, it felt quite real I got really sucked into it” (Katie, interview 6)

 

Some participants detailed that they were immersed, but at the same time, they were constantly aware that it was an artificial situation:

P: “I was aware that it was virtual reality but It did feel quite real like I had the same reactions as I would do doing a normal speech” (James, interview 2)

 

Similarly, another participant stated:

P: “um it was quite real um apart from obviously the pause breaks when you’re like mid-sentence and you just go back and are like where was I which I wouldn’t be able to do in reality so I did know that it was virtual” (Charlotte, interview 3)

However, one participant, in contrast, mentioned that she was not immersed within the virtual environment at all and was aware the situation was artificial for the entire presentation.

I: “How real did the experience feel?”

P: “Not real at all I knew It wasn’t real the whole time” (Olivia, interview 1)

5 out of 6 participants found the overall experience of presenting in front of the virtual audience realistic and immersive. However, participants who did find the experience immersive, were still consciously aware that the situation was artificial. Furthermore, it was detailed by some that once they removed the virtual headset, there was a clear contrast between the virtual environment and reality. Thus, suggesting that despite how immersive VR is capable of being, an experience which entirely represents reality is unlikely. However, it has been proposed that the extent to which one is immersed within virtual reality can be determined how open-minded they are towards the experience (Murray et al., 2007). Therefore, some participants may have been less immersed within the virtual environment due to their attitudes beforehand.

One participant detailed that she experienced physiological symptoms of anxiety whilst in the VR headset. “It did steam up because I was sweating so much”. This demonstrates that the VR exposure was in-fact capable of eliciting anxiety responses, providing support for the VR exposure as being immersive. This is further demonstrated through the following subtheme which highlights how the audience were causal in eliciting anxiety responses.

A disagreeable audience increased anxiety

 

A subtheme which emerged upon analysis of the data transcripts, was that the disagreeable audience were anxiety inducing. All participants attempted presenting with the audience mood set to disagreeable. Most participants reported some form of apprehension towards the disagreeable audience type.

 

P: “I felt like I was actually more nervous than in an actual presentation because I didn’t know what to like expect in the virtual thing and I didn’t know how the audience were going to react when they changed to disagreeable and when they did I was quite scared because they were moving around loads like I didn’t know what this meant like I didn’t know what they were doing”. (Sarah, interview 4)

When asked about how the audience mood affected how they felt, one participant stated:

I: “How did the audience mood affect how you felt?”

P: “Before they changed to disagreeable I was quite nervous cause when they were neutral agreeable they didn’t really do anything different, but when it was disagreeable they got quite agitated and were moving everywhere so this made me a bit distracted from what I was saying I got a little tongue tied” (James, interview 2)

 

Like this, another participant noted that it was the anticipation of the change which made them anxious:

P: “I would say I was equally as scared as I would be for a normal presentation purely because in real life like the audience aren’t all going to be like agreeing or disagreeing they’ll probably stay the same but like knowing that they were going to be disagreeable I started to panic a bit while I was speaking, it felt really like personal” (Katie, interview 6)

The extracts presented demonstrate support for the idea of virtual audiences and situations of being able to evoke anxiety responses. It was seen that the anticipation of change in audience mood was anxiety provoking, and when the mood did change, this had an effect on participant’s anxiety levels. However, participants specifically mentioned they were “unsure” about what the audience were doing when they turned disagreeable as they were “moving around loads”. Thus, the extent to which anxiety responses were seen to be due to the realistic characteristics of the virtual audience is questionable. It was not the specific detail of the virtual audience that was capable of eliciting the anxiety response, yet it was the anticipation of what the disagreeable audience might be like beforehand which increased participant’s anxiety responses. As participants all suffer from PSA, their fear of negative appraisal was made evident through their anxiety responses to the disagreeable audience. Participants were experiencing feelings similar to those in a real public speaking scenario, providing n increased view of authenticity in the virtual environment.

 

 

Theme 3 – Usefulness

 

The third main theme which emerged from the datasets was “usefulness”.

After taking part in the VR intervention, some participants detailed that it had successfully helped improve their public speaking performances in reality. They believed VR to be a useful tool to aid PSA.

Application in reality – “It has helped me already”

IQ: Do you think using virtual reality to create public speaking situations could be beneficial in helping reduce PSA?

P: “Um I’ve actually done 2 presentations since, one in the lab and one in tutorial and I’ve actually been a lot calmer so yeah, I didn’t think it would work straight away but it probably has helped me already” (Katie, interview 6)

 

Another participant also detailed the positive effect the virtual exposure had on them in reality:

P: “I did a presentation the other day and it actually went really well so it wasn’t until then that I realised it actually helped” (Charlotte, interview 3)

 

One participant mentioned they did not find the experience realistic but still found it useful:

“P: I didn’t find it realistic but I had to speak for my professional practice presentation and I think it helped probably because it’s a good practice I guess” (Olivia, interview 1)

 

Participants detailed the VR intervention to be helpful in improving their presentation performances throughout various modules of their university degree’s. However, one participant mentioned that she did not find the experience realistic, but it still helped her presentation in reality, suggesting useful implications for VR as a practice technique. However, it seems that not all participants were not entirely aware the VR intervention was helpful until a public speaking situation occurred.

“A good practice” – could be used to help others

Participants believed the VR exposure to be useful for both themselves and others. Participants emphasised the importance of having a safe place to practice their public speaking in which the audience were not real.

P: “It does seem quite representative of actual public speaking so it would be a good practice for other people I think” (Katie, interview 6)

P: “I think with anything practice is probably the way to become confident at something um so I definitely think it could be used sort of like immersion as well I think people may be more willing to do it as well because it’s not real so you could build up the confidence to doing the real thing” (Sarah, interview 4)

P: “I think it’s safer to practice in the virtual thing because if I was practicing in real life even with people I knew I still would have struggled” (James, interview 2)

 

P: “It would be a lot calmer for someone presenting in front of a virtual audience

Participants proposed the VR intervention to be useful in the sense that it enabled them and others to gain some experience of practicing public speaking in an artificial situation. One participant proposed that the virtual environment was a “safe” setting to practice in. Simulated performance spaces have been classified as ‘safe’ environments. They are potentially less harmful than typical vivo desensitization exposure (Aufegger et al., 2015). Participants confirmed this with their suggestions of the VR intervention being a safe environment free from real factors such as negative appraisal.

Control

Participants emphasised that being able to control their conditions inside the virtual environment was a useful aspect of the VR intervention. The ability to change the audience mood and size, number of prompts was detailed as a positive aspect of the VR exposure.

P: “I think when I had more prompts and also knowing that the audience were agreeable made me feel more comfortable purely because I could control how they felt” (Katie, interview 6)

P: “Being able to like choose what conditions you wanted was good because at the start I wasn’t going to do a disagreeable audience but then I changed my mind and wanted to test myself so I ended up doing disagreeable with less prompts” (James, interview 2)

P: “being able to control prompts is definitely something I would like to be able to have in real life I think that was the best condition out of them to use” (Charlotte, interview 3)

 

Participants being able to control conditions in virtual reality exposures is not something regularly observed in research. Yet the ability to control this was viewed as a significantly useful aspect of the virtual intervention. It was seen to lower participant’s anxiety knowing that they had control over their environment enabling them to feel more at comfort during the virtual exposure.

 

 

 

Theme 4 – Recommendations

 

The final theme to emerge from the datasets was recommendations. All participants provided some feedback for the virtual exposure in which they took part in. Though they found the overall experience to be realistic, there were some aspects of the virtual environment which prevented them from being immersed fully and made the interaction between themselves and the environment feel artificial. One of the main recommendations suggested was to improve the detail of the audiences faces, more specifically, the direction in which they were looking. They all viewed this as something important, suggesting that it is eye contact which induces their anxiety in real public speaking scenarios.

When asked as to whether participants had any improvements to the VR intervention participants

P: “it’s kind of like a bit not blurry but like the detail isn’t as much as that you can see who’s looking at you and who isn’t, it was hard to see what their facial expressions meant so apart from their moving around it was hard to see what mood they were actually in” (James, interview 2)

 

P: “The only thing with the audience is that I couldn’t actually see where they were looking like the eye contact wasn’t clear the whole time so when it was a small audience it was a bit weird so I would make this better but overall I found it pretty realistic yeah” (Rachel, interview 5)

 

P: “I wasn’t sure if some of the virtual people were looking at me, sometimes when the mood changed it didn’t feel like they were actually looking right at me so I think to make the whole thing more realistic they should look at you in the eye” (Olivia, interview 1)

 

The lack of eye contact was seen to make the experience unrealistic. However, participants acknowledged that the overall experience was realistic. Eye contact is one of the many factors which influence PSA (Daly et al., 1989). The fact that the VR intervention confused participants as to where the audience were looking suggests that improvements must be made of the application in order for it to be deemed an effective exposure.

The experimenter’s presence was felt throughout

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Discussion

 

In respect to the study’s research question, the results showed that while students did not find the VR intervention entirely realistic, the context of the virtual environment, more specifically the audience, were able to contribute to an overall immersive experience. The results also found the views of using the VR intervention as positive. Results of improved performance in reality were detailed and participants believed it to be useful in the treatment of others whom may experience worse anxiety than themselves. However, all participants provided feedback for the intervention, with suggestions of improvement including the detail of the audiences faces and movement and having less interaction with the experimenter throughout.

Effective VR simulations should provide adequate graphical reality and sensory feedback to enable the experience to be immersive (Gervaultz, 1996). The VR simulation in this study was effective at producing realistic visual cues to the extent where participants detailed feeling a sense of immersion. Immersion experiences were heightened through interaction with the virtual audience, and the different audience moods which were seen to influence anxiety levels. In line with Foa and Kozak’s (1986) theory of emotional processing, the VR intervention examined in this research was capable of activating the anxiety-provoking structure.

Furthermore, detailed success of the VR intervention suggests that it was capable of ‘overruling’ the previous anxiety provoking structure leading to new memories being created and improving participant’s presentation performance in reality.

While the improving of graphical features on the audience’s faces was suggested, the VR intervention was still able to succeed in eliciting anxiety. Thus, the results are seen to be in line with research by North et al, (1998) whom proposed one can experience a sense of presence in virtual reality even when the virtual environment does not completely or accurately represent the real-world situation. This supports the assumption that anxiety may be induced more by the thought of being in a stressful situation than the actual graphic realism itself (Kwon et al., 2013; Klinger et al., 2005). However, in contrast, Kwon et al (2013) proposed that the level of graphic realism in virtual reality is essential in sustaining the state of anxiety over the course of the VR exposure. Therefore, the quality of graphic realism used in the VR intervention may have been influential on preventing participants from being emerged for the whole presentation.

The disagreeable audience mood was able to elicit anxiety responses in participants. Though the extent to which the details of audience mood was causal in this is questionable, a disagreeable audience was able to induce anxiety responses.

This is in line with previous research which found participants with PSA to benefit from VR (Pertaub et al., 2002; Slater et al., 2006; Roy et al., 2003), this study provides significant support for the notion that VR is effective in reducing PSA. Participants deemed the VR exposure to be significantly helpful, providing them with real life application. Participants proposed the VR intervention to be useful for themselves and for others with potentially worse PSA than themselves.

Research examining virtual reality with user control over virtual settings is not something that has been previously observed. Thus, the findings that show participant’s appreciation of this trait is a new finding not previously reported. This is a proposed idea for future research aiming to examine the effectiveness of virtual reality. For example, this provides a possible marker for improving virtual exposure therapies. Typical exposure therapies use graded exposure in the treatment of anxiety disorders or phobias, yet this finding supports the notion that providing sufferers with control over their environments may be a more useful technique in exposure therapies.

The importance of sensory feedback from virtual audiences and environments has been established (Kwon et al., 2013). Yet little research has examined the role of the physical surroundings in which one uses VR equipment (E.g. the environment in which one uses to interact with virtual reality). Participants suggested that the role of the experimenter, whom partially interacted with participants by setting up the equipment and guiding them when questions were asked, was felt throughout. Research has hinted at the usage and development of mobile virtual reality apps that could be set up and used, by the user as being an effective self-treatment for PSA (Kim et al., 2016). As the findings from the present study hint that the physical environment in which one engages with VR as being influential in withdrawing them from the virtual environments, future research examining the use of self-training with virtual mobile apps to reduce PSA should be conducted.

• Mention prompts in discussion.

Limitations

The benchmark for thematic analysis usually lies within recruiting 6-10 participants (Braun & Clarke., 2006). Thus, the sample of participants used in this study was sufficient, yet it could have been more representative. A sample of 6 could be seen as unrepresentative of the sample who took part in the VR intervention.

As the sample of participants were all sufferers of PSA, a further limitation could be the responses they gave. Responses were limited and not as in depth as of, perhaps, what another sample of participants might have offered. So, though the depth of responses might not have improved, it would have been wiser to obtain a stronger sample of participants to gain a better insight into the experiences of users of the VR. Themes could have therefore been different across a larger sample size.

The recruitment of participants was another limitation of this research. As the original researchers of the VR intervention emailed those who had taken part, it is possible not all participants would have noticed or seen invites to take part. Furthermore, fatigue from previous intervention could have deferred participants from taking part as they had already given their time to taking part in the VR intervention.

Additionally, a further limitation could be the execution of the interviews and whether this effected what the participants were willing to share with the interviewer. As all participants volunteered to do the interview, they should have had a positive attitude to revealing information. The interview guide only used open-ended questions and the main focus of these was the virtual environment which should have limited anxiety and any feelings of a negative presentation performance. The questions did not focus on the performance of participants which should have restricted any feelings participants had about negatively performing in the presentation. These factors should have produced a non-judgemental interview atmosphere where participants should not have felt any need to restrict the information that they disclosed. The questions asked during the interview were of quite a broad nature and in turn of this the data received were also of quite a broad scope. If the questions asked were more specific questions, it is possible that more in depth data could have been obtained.

Another limitation would be that some of the interviews took place after a considerable amount of time since participants used the VR. This could mean that some of the answers were unreliable and participants may have over or underestimated the extent of anxiety to which they felt during the VR intervention. Furthermore, it is possible they may have forgotten some detail of the virtual environment since usage. Recall bias is often a problem with qualitative research. It is often that if the participant cannot remember something when asked, they will often make up what they believe the interviewer would like to hear. For example, they would tell the interviewer what they wanted to hear about VR acting in a way similar to social desirability which is also a common shortfall of the qualitative research method.

There seemed to be a paradox between the themes discovered. Though participants suggested the VR was realistic and translated well to real life. Most participants also suggested that there was a lack of some important detail on the virtual audience making the experience somewhat unrealistic for them. Yet, anxiety was seen to occur in every participant suggesting that the VR must have been realistic and immersive enough to produce anxiety responses.

Conclusion