Introduction

Young adults experience life stressors which contribute to poor mental health and well-being (Truskauskaite-Kuneviciene et al., 2022). COVID-19 exacerbated psychological distress, causing increased anxiety and depression (Emery et al., 2021; Kwong et al., 2021). Studies indicate that younger people are more vulnerable to mental health problems and adjusted poorly to the pandemic (Varma et al., 2021). Stress management practices, including meditation, mindfulness, and yoga, have been shown to increase relaxation and reduce stress (Varvogli & Darviri 2011), but adherence to these practices is inconsistent, especially among young people (Danilewitz et al., 2018; Erogul et al., 2014; Gard et al., 2012). This highlights the need for acceptable, engaging, and effective self-management interventions that improve well-being for young adults.

Virtual reality (VR) is a simulated digital three-dimensional experience that can be accessed using a head-mounted display (HMD). VR shows promising results for stress management and well-being (Best et al., 2022; Freeman et al., 2017; Imperatori et al., 2020). Unlike traditional relaxation techniques involving memory and imagination, VR immerses users into virtual environments, facilitating a more realistic experience with less effortful engagement (Bohil et al., 2011). Studies show that nature-based virtual environments, including beaches, forests, and water scenes, increase relaxation and decrease stress (Anderson et al., 2017; Soyka et al., 2016; Van Houwelingen-Snippe et al., 2021). Systematic reviews indicate that VR relaxation has high user satisfaction and decreases stress for both clinical and non-clinical samples (Riches et al., 2021; Riches et al., 2023a, b, c). A recent study shows that exposure to a virtual forest reduces stress and negative affect in young adults after two supervised laboratory sessions (Chan et al., 2021). Decrease in heart rate and reduced psychological stress are reported in another study where young adults are exposed to virtual environment of a beach scene during a single 15-min laboratory-based VR session (McGarry et al., 2023). The incorporation of waterfall sounds into a virtual forest was also found to relieve anxiety states in university students (Hsieh et al., 2023). This body of research indicates that VR relaxation has the potential to support young adults coping with stress; however, existing VR relaxation studies are limited by lack of unassisted VR use in naturalistic, non-laboratory-based settings and limited qualitative data on participants’ subjective experience of VR.

This mixed-methods study investigated VR relaxation for young adults during COVID-19. A remotely delivered pilot study of home-based VR relaxation was designed, which adhered to social distancing and infection control protocols for COVID-19 and contrasted with laboratory-based VR studies (Mottelson et al., 2021). The aims were to test feasibility and acceptability of delivering HMDs to participants’ homes, of participants using HMDs unsupervised for relaxation and self-management of stress, and of remote data collection (Ma et al., 2018; Mottelson & Hornbæk, 2017; Steed et al., 2016).

Materials and Methods

The study gained ethical approval from King’s College London (HR-19/20–19650). Young adults, aged 18–30, based in London, UK, with no reported mental health or neurological conditions, were recruited using social media to participate in a VR relaxation study. Participants were assigned to a VR relaxation group, who received an HMD for daily self-use, or a control group, who received no intervention. The aim was to recruit 10–12 participants for each group, as recommended for pilot studies (Julious, 2005; Lewis et al., 2021). Group allocation was based on geographical feasibility of researchers delivering HMDs to participants’ homes. It was not possible for researchers to match groups for demographic characteristics. Participants provided informed consent via an online form. No financial incentives were offered.

Oculus Go wireless HMDs programmed with the VRelax app (https://vrelax.com/en/home-us/) were delivered by researchers to VR participants’ homes in London from May to July 2021, when the UK was coming out of a lockdown but when there were still significant restrictions on travel and socialising. VR participants were loaned HMDs for a 14-day intervention period. Participants were encouraged to carry out at least one VR session per day. There was no requirement to complete a certain number of sessions because the study aimed to investigate the feasibility and acceptability of participants engaging in this intervention. In line with COVID-19 safety measures, HMDs were thoroughly cleaned, re-boxed, and quarantined for a week between deliveries to participants. VR sessions were unsupervised and conducted by participants in their own homes. Participants chose the time and duration of each session. Data collection for the control group occurred concurrently with VR group data collection to reduce risk of confounders.

Contacts between researchers and participants were conducted remotely. VR participants received a pre-intervention video call briefing and a VR training session from a researcher. Written instructions on using HMDs were emailed to VR participants. The VR environment comprised 360° nature scenes with audio, including beaches, swimming with dolphins, meadows, mountains, meditation exercises, and interactive games (Table 1 and Fig. 1) (Veling et al., 2021). On day 14, VR participants were debriefed by video call.

Fig. 1
figure 1

Impression of virtual reality relaxation environments from VRelax app

Full size image

Measures

All data were collected anonymously using Qualtrics online surveys that participants accessed remotely using weblinks emailed to them by researchers.

Pre- and Post-Intervention

On day 1 of the intervention period, all participants provided demographic information on age, gender, ethnicity, occupation, and impact of COVID-19 (Table 1), and completed the following psychological well-being measures: Generalised Anxiety Disorder scale (Spitzer et al., 2006); Patient Health Questionnaire (Kroenke et al., 2003), which measures depression symptoms; Trauma Screening Questionnaire (Brewin et al., 2002), which measures posttraumatic stress disorder symptoms; Perceived Stress Scale (Cohen et al., 1983); and Interpersonal Sensitivity Measure (Boyce & Parker, 1989). All scales have high validity and reliability for adults (Dekkers et al., 2010; Denovan et al., 2019; Kroenke et al., 2003; Spitzer et al., 2006). The day after the intervention period ended, participants repeated all well-being measures (Table 2). VR participants also completed the Sense of Presence Scale (Slater et al., 1994), which measures immersion in VR; and a video call-based semi-structured interview investigated their experience of the VR (Fig. 2).

Fig. 2
figure 2

Post-intervention interview questions assessing virtual reality participants’ experience of virtual reality relaxation

Full size image

Pre- and Post-VR Sessions

VR participants completed visual analogue scales (VAS) evaluating their mood states, relaxation, and connectedness to nature on a scale of 0 (“not at all”) to 10 (“very”) immediately pre- and post-VR sessions. VAS of helpfulness of, and immersion in, the VR were completed post-sessions. See Table 3 for full VAS. Participants could also provide written feedback post-sessions.

Analysis

SPSSv27 was used to analyse quantitative data. Demographic data was reported to convey participant characteristics and contextual stress levels associated with COVID-19. Fisher’s exact test was used to determine differences between VR and control groups in demographic characteristics and experiences of COVID-19.

Feasibility was measured by HMD delivery and collection, remote data collection, completion rate of outcome measures, intervention adherence, number of daily VR sessions, dropout rates, and participants’ feedback. Post hoc power analysis using G*Power showed insufficient power to compare pre- and post-intervention well-being outcomes between the two groups due to small sample sizes, so they were not carried out. Instead, data from these measures was reported solely to evaluate feasibility of participants completing measures remotely. Control participants’ demographic characteristics were reported and compared to the VR group to investigate the possibility of recruiting and collecting data from appropriate controls.

Acceptability was measured by calculating changes in pre- and post-session VAS scores and investigating any adverse VR effects. Wilcoxon matched-pairs signed-rank tests compared mood, relaxation, and connectedness to nature VAS scores pre- and post-VR sessions, given small samples and normality violations. For each VAS, data were pooled for all sessions of all VR participants.

VR participants were divided into two subgroups to investigate novelty bias: participants who completed ≤ 7 sessions and participants who completed ≥ 8 sessions. This cutoff was selected to determine whether participants used VR more or less than half the time of the intervention. Wilcoxon matched-pairs signed-rank tests compared pre- and post-session VAS for subgroups.

Post-session and post-intervention qualitative data were pooled for all VR participants, and thematic analysis was conducted in NVivo12 to assess acceptability and user experience. One researcher (IK) carried out the analysis including data familiarisation, coding, and theme development. A second researcher (SN) subsequently reviewed the data and the identified themes and provided suggestions for refinement. The overall thematic analysis was regularly discussed with the lead researcher (SR) throughout the study. Thematic analysis was organised by researchers into categories of Feedback on VR Experience, Recommendations, and Future Applications. Themes listed under each of these categories were included if they were endorsed by ≥ 4 participants.

Results

There were 23 participants (11 VR, 12 controls). Table 1 reports full participant demographic characteristics and impact of COVID-19. Participants were mostly female, of White ethnicity, and in their early 20s. Approximately half were full-time employed, and a quarter were students. Nearly half had experienced stress or adverse life events due to COVID-19. Most had self-quarantined, a fifth had been physically ill (although fully recovered), a third reported reduced ability to earn money, and a fifth had lost their jobs. More than half had a family member or friend who had been ill and nearly a third had one who had died. Although participants acknowledged that they were a lower risk group, most reported feeling tense, anxious, sad, and depressed during COVID-19. There were no significant differences between the two groups on demographic factors or impact of COVID-19, apart from a significant difference in ethnicity (Table 1), with greater ethnic diversity in the control group.

Table 1 Demographic characteristics and impact of COVID-19 pandemic on participants (N = 23)
Full size table

All participants completed all pre- and post-intervention well-being measures (Table 2). All HMDs were delivered to VR participants’ homes and collected as scheduled. VR participants successfully learned to use HMDs with remote training. There was a technical issue with one HMD, but this was quickly resolved by a technician.

Table 2 Pre- and post-intervention outcome measures for virtual reality and control groups (N = 23)
Full size table

Participants carried out a total of 84 sessions. Mean number of sessions during the intervention period was 7.64 (SD = 2.16, range 3–11). Mean VR session duration was 22.89 min (SD = 9.83, range 11–59). Table 1 reports the VR environments visited by participants. Post-VR sessions, there were significant increases in relaxation, happiness, and connectedness to nature, and significant decreases in anxiety, stress, and sadness (medium effects—Table 3).

Three participants carried out ≤ 7 VR sessions, with a total of 15 sessions between them. Eight participants carried out ≥ 8 VR sessions, with a total of 69 sessions between them. Post-VR sessions, there were the same VAS findings for participants who completed ≥ 8 sessions as for the whole sample. For participants who completed ≤ 7 sessions, relaxation and connectedness to nature significantly increased, and stress significantly decreased. There were no significant changes on other VAS. Table 3 reports VAS scores and tests.

Table 3 Pre- and post-session measures of mood, relaxation, and virtual reality experience (N = 11)
Full size table

Table 4 reports full themes, explanations, and illustrative quotes from VR participants. Feedback on VR experience themes was that participants found it a relaxing experience, particularly when stressed, due to the calming sounds and images. There were technical issues, such as glitches and poor resolution, which hindered participants’ experience. A novelty effect meant participants expressed initial excitement and curiosity about VR and its potential benefits; however, participants reported diminished interest over time, due to repetitiveness, limited interactivity, and monotony of the virtual environments. Participants reported an enjoyment of the beach and sea because they found these environments calming and entertaining. Participants also reported a positive impact on sleep. Recommendations were to increase interactivity, through more diverse environments and gamification, and improve user experience, by reducing technical problems. Future applications were to introduce VR relaxation in workplaces and stressful environments.

Table 4 Thematic analysis of participants’ experience of using virtual reality relaxation (N = 11)
Full size table

Discussion

This study investigated feasibility and acceptability of remotely delivered VR relaxation for stress in young adults. Remote delivery of HMDs to participants’ homes, unsupervised VR relaxation, and data collection were feasible, suggesting that VR relaxation can be an accessible self-help intervention for young adults at home. Quantitative and qualitative findings, including lengthy duration of sessions, indicate participants found VR acceptable, with short-term benefits for relaxation and positive mood at a time of heightened stress.

Findings are consistent with research indicating VR relaxation can decrease stress among the general population (Riches et al., 2021) and young adults in isolated environments (Anderson et al., 2017), and suggest applications for young adults in other stressful or restrictive environments, such as in prison or in psychiatric hospitals, or where individuals’ mobility issues hinder their ability to access real-world nature (Riches et al., 2023a, b, c). Positive feedback on sleep and relaxation was consistent with studies showing that VR can promote nighttime relaxation and improve sleep in young people (Yüksel et al., 2020). However, frequency of participants’ VR sessions was lower than suggested by researchers, indicating daily use may be impractical or require further support. Although participants’ interest diminished and technical problems hindered user experience, appraisals of these issues may have been exacerbated due to limited technical assistance, compared to that which they might receive in laboratory-based studies or through greater online support. Nevertheless, participants felt there were positive future applications for VR relaxation, including in the workplace and specific stressful environments (Nijland et al., 2021; Riches et al., 2023b, c; Riches & Smith, 2022).

This study reports an innovative methodology using entirely remote recruitment, participant consultation, and data collection for a VR-based psychosocial intervention. Previous “remote” VR studies have provided preliminary evidence, but have tended to either not be entirely remote, with participants collecting HMDs in-person (Mottelson & Hornbæk, 2017), or only recruiting participants who already owned an HMD (Ma et al., 2018). Further strengths of this study include use of a control group, qualitative feedback on the subjective experience of VR relaxation, and the naturalistic setting of participants’ homes, which increased ecological validity. Limitations include the small homogenous sample and lack of physiological measures of stress. The fact that participant characteristics, number of sessions, and duration of sessions were self-reported is a potential limitation because there is no way to verify this data. It is also unclear how representative the demographic characteristics of the sample are compared to the general population, especially those that use VR interventions. Investigating such issues of representativeness fell outside of the remit of the present study.

Future research could investigate longer-term engagement and targeted use of VR relaxation for stressful circumstances with more representative and diverse samples (Riches et al., 2023a). Studies should also be adequately powered and compare intervention groups with control groups on the main outcome measures. To improve engagement and user experience, VR developers could design more diverse, interactive, gamified environments (Kaleva & Riches, 2023). VR relaxation could also be incorporated into videogames, social media, and virtual online communities in psychologically informed ways that target user well-being. Avatar-based technical support in the VR could provide instruction and troubleshooting to reduce technical problems.

In conclusion, remotely delivered VR relaxation in young adults’ homes is feasible and appears to provide short-term stress relief, but further research should investigate optimal frequency of use with a larger sample to assess longer-term impact on psychological well-being.