Virtual Reality as an Automated Stress-Reduction Therapy Tool - Case Study on War Refugees

3.1 VR application

As we mentioned in the Introduction, this is not our first approach to VR-based stress reduction. Our first application, focused on stress reduction among office workers, was in the form of a walk in the woods, where the user performed exercises at successive stops. Although the stress level was reduced among most participants, 7 participants reacted negatively to the session or fell asleep during the experience. Therefore, we decided to conduct an extensive interview with those participants and consult them during the development of the new release of our application. We intended to make improvements that wouldn’t significantly affect the methodology used in the application but would allow for more promising results for different users. The most significant changes (determined during the interviews) introduced concerning the previous version of the application are presented in Tab. 1.

The proposed version of the VR application is in the form of a cozy apartment in the mountains (see Fig. 2). During the session, participants relax on a comfortable chair (the same is mapped in the VR environment) in any convenient position (see Fig. 3 a). The main scene is calibrated automatically (based on the user’s head position) - the main element (a window with the mountain view) of the scene is centered vertically and horizontally. The narrator assists throughout the session, acting as a therapist.

Figure 2:

View Figure

A short adaptation scene precedes the therapy scene. First, the user is asked to look around the apartment and get comfortable in the chair. Then, the user moves seamlessly into the therapy sessions, which involve various relaxation techniques such as breathing methods, visualization, and bilateral stimulation.

Figure 3:

View Figure

3.2 Stress monitoring methods

The sessions were monitored using EEG and GSR sensors. For this specific group, we eliminated the cumbersome EEG wet cap (gel or saline liquid is used to increase the conductivity value) by replacing it with a discreet, HDM-integrated, dry EEG Looxid Link (see Fig. 3) at the expense of reducing the number of channels - 6 channels with sensors mounted on the user’s frontal lobe. The EEG system remains almost unnoticeable. Users did not complain about the inconvenience (preparation for the session, soaking the hair, and the need to wash the hair afterward). We used also a GSR sensor with convenient mounting on the user’s fingers. Those sensors allow the recording of raw measurement data and real-time monitoring of interpretation of the user’s involvement and stress level. Before and after the session, users fill out General Health Status Questionnaires (GHQ12 [7] and GMS [12]) so that we know their subjective feelings about their mental state at that moment, the intensity of stress and the emotions accompanying the moment.

The system’s unique feature is the capacity to compare data recorded during the user’s subsequent therapy sessions. Thus, the therapist can easily monitor the progress and achieve effects. These tools in the pilot study phase facilitated us in objectively evaluating the effectiveness of the application.

3.3 Course of the experiment

At the end of 2021, we began work on improving the system based on the measurement experience we gathered and user interviews. Our interventions were evolutionary: each implemented improvement was tested by a group of volunteers who took part in our previous research. Once the implementation was verified and accepted, we implemented the next one. The final version coincided in time with the start of the war on Ukrainian territory. Local media (newspapers, online portals, radio and television) played a vital dissemination role.

Both the media and our observations of refugees arriving in Poland inspired us to adapt the developed application for such users. We initiated extensive cooperation with local city authorities as well as non-governmental organizations involved in assisting refugees from Ukraine. We modified the content of the narratives - the narrator’s voice was lent by a professional native Ukrainian speaker. When the application was ready for use in March 2022, we launched an information campaign together with Lodz city council. To increase the level of trust, we hired a Ukrainian refugee who was in charge of recruiting subjects for VR sessions. With limited equipment resources (one complete station) and limited availability of the premises, during three months, we managed to organize more than 250 sessions, which lasted from 45 to 60 minutes each.

The study protocol was approved by the local Ethical Committee (resolution no. 3/2021 dated October 21, 2021). Before each session, users were informed about the course of the VR experience and the exclusionary conditions, and provided informed consent. The entire course of the session was monitored by the supervisory staff. We assumed that each participant would be able to benefit from five therapeutic sessions (one session per week). Unfortunately, due to personal reasons and relocation, not everyone has completed the session sequence.

Finally, we recruited n = 55 participants, 52 female, 3 male, aged M = 35, SD = 10,93. The study was conducted in a period when the university was closed (there were no students), in a quiet, air-conditioned room. We provided a shopping voucher for the equivalent of USD 20 as remuneration for participating in one session.

3.4 Results

To verify the effectiveness of the presented application, we used pre and post-tests (subjective measurement) and the GSR and EEG signal (objective measurement). The results obtained from pre and post-test, in the form of mean stress level, are presented in Fig. 4. First, we smoothed GSR signal with a moving average (30 samples) to reduce measurement noise. Next, to determine the change in signal waveforms at the beginning and end of the session, the group analysis (average across all subjects for each session) in the form of mean and standard deviation values of resistance gain over time is presented in Fig. 4 a. Since the Looxid Link provides analytical information such as relaxation level (in a range of 0 to 1) over time, similarly as in the case of GSR, first we smoothed those information with a moving average (30 samples). Than, we determined the change at the beginning and end of the session (average across all subjects for each session) in the form of mean and standard deviation values of relaxation level gain over time. The results are presented in Fig. 4 b

Figure 4:

View Figure

4.1 Reflections on the recruitment process

In order to recruit refugees, we faced a significant challenge in effectively engaging potential participants. When we began the recruitment process, we initially assumed it would be relatively easy, thinking that extending a helping hand and offering support to address apparent mental health issues using proven therapeutic methods in a more user-friendly VR environment would be well received. However, this assumption turned out to be far from reality. We underestimated the deep distrust and skepticism of refugees who found themselves in a foreign country with limited resources and often incomplete family units. Despite the warm initial welcome by Polish society, Ukrainian refugee communities were initially closed, suspicious, and paralyzed by fear of meeting their basic existential needs. In the early stages of their stay, refugees frequently changed their place of residence in order to find one that would guarantee the satisfaction of their basic needs. Other priorities and concerns overshadowed therapeutic considerations. Various channels were used to reach our target group, including city authorities, official government refugee registration offices, non-governmental organizations providing refugee assistance, local media and social media platforms. Direct access to refugee communities proved challenging. These groups were tight-knit and suspicious of strangers, often using communication channels that were inaccessible to us. Surprisingly, we were bombarded with numerous questions, suggesting that even the subject of therapeutic assistance at a national university (sessions were organized at the Lodz University of Technology) had to be sufficiently credible. Potential therapy participants had doubts about whether it was "human experimentation", they knew nothing about VR technology and considered it a dubious machine that could harm them. Overcoming these barriers to recruiting early participants became a formidable challenge. One of our team members invited refugees into their home to build trust. As a result, they agreed to attend the first sessions. The idea of formally hiring one of them to help with the project emerged and became a turning point. Someone with a background in war refugee communities found it much easier to connect with the target groups, largely due to direct access through social media and the common language, eliminating language and cultural barriers. The city authorities played a crucial role in organizing an information campaign using local media and providing us with municipal office space for two weeks, which greatly enhanced the credibility of the project. From that point on, the recruitment process became much more effective. The hired Ukrainian effectively handled new enrollments by fully explaining the project. Still, some uncertainty remained among the target group consisting mainly of women. They preferred to attend the VR sessions in pairs or bigger groups to have the sense of security and companionship. This experience highlighted the importance of a rational approach to the existing hierarchy of needs within the target group when planning outreach programs. It was also a reminder that providing the subjects with a sense of security and safety should be a priority. Another conclusion is that finding effective ways to reach the desired community directly is essential to the success of any relief program.

4.2 Reflections on the course of the study

Some of the refugees, for reasons beyond our control, did not complete the full series of tests. Only 30 subjects participated in all offered sessions. 6 relocated and were not able to continue participation due to a very inconvenient commute. 8 people returned to their country of origin. 11 people, without giving a reason, did not report back to participate in further sessions. Based on behavioral observation and analyzing data from VR therapy sessions, we noticed that virtually 100% of the subjects participating in the first study showed signs of heightened distrust, skepticism and critical attitude towards our initiative. These perceptions changed after subsequent sessions - with each visit we could observe much more engagement and willingness to participate in the study. In comparison to Polish office workers who did not suffer any trauma, we recognized the refugees as more willing to continue the sessions because they clearly provided them with comfort and much needed stress reduction. Only two of the recruited subjects had any previous exposure to a VR experience. Participants of the first sessions in the conducted interview made it clear that they were astonished by the level of immersion and very keen on exploring the presented VR environment. It is likely that for this reason, the expected measurable objective therapeutic effect was obtained only during the second and subsequent sessions (89% of cases).

The recording of biomedical signals did not cause any critical problems. Thanks to the adopted procedure of placing the GSR sensors on the index and ring finger of the left hand, we did not have any case of shortened contacts or discontinuity of measured quantity. Similarly, we did not encounter any issues with recording eye tracking signals and HMD localization (6 degrees of freedom - the object can move forward/backward, up/down, and left/right; it can also rotate around three different axes: yaw, pitch, and roll). In contrast, monitoring EEG signals proved to be more troublesome. Since the vast majority of participants were women, applied face makeup was a source of some problems, i.e. dry Looxid electrodes sometimes lost good connection with the skin. In one case, getting a stable EEG signal was impossible due to the very “low” forehead of the participant. It is worth mentioning that the HMD mounting procedure included the need to expose the forehead by moving away the fringe as contact of hair with EEG electrodes virtually always leads to temporary or permanent loss of signal.

We would also like to discuss some organizational issues that we faced. First of all, we were initially too optimistic about scheduling therapy sessions. In spite of our best efforts, it was challenging to keep the sessions on track. Being in an unfamiliar city, suffering from communication problems, and having no experience in using local public transport caused trouble when reaching our University, and thus significant delays in our schedule. Since we often had to plan the session cycle in the evenings, there was a rather unfortunate situation of accidentally setting off the Institute’s monitoring system, triggering a loud acoustic alarm and, more concerning for participants, activating the intervention procedure for an "unauthorized intruder in the building." This experience was particularly unpleasant for individuals with a history of war refugee trauma.

What is more, the rise of national threat level connected with the situation in Ukraine, caused increased security measures to be introduced in public institutions. The participants had to go through security and identify themselves, which often caused additional setbacks. We have noticed a lack of empathy from the facility’s security personnel. We mention these incidents to underscore the diverse situations we may encounter in practice. Following each occurrence, we implemented interventions to adjust our organizational procedures, ensuring avoiding similar issues in the future. This approach proved effective in resolving and mitigating the following challenges.

Another unforeseen logistic problem was that women often attended the sessions accompanied by their young children. Providing the children with care and entertainment during their guardian’s session was an additional challenge. Fortunately, we swiftly organized a set of building blocks and provided colorful crayons and markers.

4.3 Reflections on the study results

For the vast majority of measurements, skin resistance increased steadily during each session. The intensity of this increase decreased with each successive session. This may indicate that the initial stress level for subsequent sessions is lower. In addition, we noticed that the trends among all participants were more consistent with each session. Similarly, measurements from the Looxid device indicate a state of meditation of drowsiness among participants, which was especially noticeable in the period of BLS (which shares many features with a meditative state). This positive condition deepens from one session to another session.

The results were promising and showed that VR-based BLS is efficient in mood improvement and stress reduction. These effects are also related to a decrease in the subjective perception of stress levels. Data analyses showed that the repetition of VR training does positively affect its efficacy in symptom reduction which is relatively stable from one session to another. Taking into account the special sample we worked with, our results are more than promising. However we need to mention some limitations of our pilot study. The first group of limitations concerns the sample. Firstly, it was a sample of random volunteers who responded to media announcements. Secondly, participants were under the constant pressure of uncontrollable events and sudden demands and possibilities which were impossible to control in the study but they might affect our results. The third issue is the fact that the group was relatively small, with a high dropout rate. Another group of limitations concerns the way we measured mental health status and study design. Although GHQ-12 is a well recognized and reliable measure, possessing an independent clinical diagnosis of mental health would be an additional advantage. Moreover, taking into account the complexity of uncontrollable problems, challenges, stress and trauma faced by our participants – case-control design of study might increase the reliability of our result. However, despite the limitations mentioned above, the results are very promising and guide our imagination to the era of affordable, accessible, home-based treatment of adverse mental health symptoms.

We did not find any negative effects of VR stress management sessions in terms of VR sickness symptoms. Experimentally we asked the participants to fill in VRSQ [11] before each session to check if they experienced any oculomotor or disorientation symptoms. It appeared that the number of negative symptoms of VR sickness reported before the session is significantly higher than that reported after experienced immersion.

4.4 Reflections on the consequences associated with the project

Thanks to the press conference and various media publications in newspapers, the Internet, radio and television, we had no problems recruiting volunteers for our study. Additionally, we were contacted by many NGOs interested in implementing the application at their spot. Since our solution was designed for research, it was rather unsuitable for such applications: the hardware is expensive and additional trained personnel is required.Thus, we prepared a simplified version dedicated to the Oculus Quest 2 HMD, supplemented with a self-service IU. This version has been designed for individual users or patients receiving therapy sessions at a therapy clinic, at home, or refugee assistance center. Since standard Oculus Quest HMD is not equipped with additional EEG, eye tracking, and GSR sensors, this version supports only one selected scenario and the amount of recorded data has been naturally limited. We believe that the therapist will be able to conduct a control session and adjust home therapy. Despite significant functional limitations, the application still fulfills its primary mission of stress reduction.

Despite the great initial interest in the software, ultimately only two foundations decided to implement it at their sites: Masz Wszystko w Sobie and Together-Razem Centre. The first one, with the help of donors, purchased 5 Oculus Quest sets and have been working with since then. The other one wishes to use the expanded version of the application and is waiting for financial support. The most obvious parties interested in the application - refugee aid centers - unfortunately did not have facilities or personnel to make use of the software. Despite initial intentions, when conversation reached the issue of financial and human resources, the organizations began to withdraw their interest. We have also received information about the need to implement the system in elementary and secondary schools due to the fact that many children have problems with adaptation. Unfortunately, after consultations with a child psychologist, it turned out that the created scenario will not fulfill its purpose. We are in the process of preparing an application tailored for children in school age, which will be focused on stress reduction, but in a more dynamic, task-oriented environment.

After reading one of article about our project, the board of a significant VR goggles distribution company in the US decided to provide us with VR kits in any quantity to work with refugees. Unfortunately, due to procedural reasons, the implementation of this initiative faced considerable delays. Local legal and tax regulations further posed challenges to this endeavors. Thus, this valuable initiative eventually met an unfortunate end.