Desensitizing Anxiety Through Imperceptible Change: Feasibility Study on a Paradigm for Single-Session Exposure Therapy for Fear of Public Speaking.

BACKGROUND: Exposure therapy (ET) for anxiety disorders involves introducing the participant to an anxiety-provoking situation over several treatment sessions. Each time, the participant is exposed to a higher anxiety-provoking stimulus; for example, in the case of fear of heights, the participant would successively experience being at a greater height. ET is effective, and its counterpart, virtual reality (VR) exposure therapy (VRET), where VR substitutes real-world exposure, is equally so. However, ET is time-consuming, requiring several sessions. OBJECTIVE: This study aimed to compare the results of single-session exposure with those of traditional VRET with regard to reducing public speaking anxiety. METHODS: We introduced a paradigm concerned with public speaking anxiety where the VR exposure occurred in a single session while the participant interacted with a virtual therapist. Over time, the therapist transformed into an entire audience with almost imperceptible changes. We carried out a feasibility study with 45 participants, comparing 3 conditions: single-session exposure (n=16, 36%); conventional multiple-session exposure (n=14, 31%), where the same content was delivered in successive segments over 5 sessions; and a control group (n=15, 33%), who interacted with a single virtual character to talk about everyday matters. A week later, the participants were required to speak on a stage in front of a large audience in VR. RESULTS: Across most of the series of conventional public speaking anxiety measures, the single-session exposure was at least as effective in reducing anxiety as the multiple-session exposure, and both these conditions were better than the control condition. The 12-item Personal Report of Confidence as a Speaker was used to measure public speaking anxiety levels, where higher values indicated more anxiety. Using a Bayesian model, the posterior probabilities of improvement compared to a high baseline were at least 1.7 times greater for single- and multiple-session exposures compared to the control group. The State Perceived Index of Competence was used as a measure of anticipatory anxiety for speaking on a stage in front of a large audience, where lower values indicated higher anxiety. The probabilities of improvement were just over 4 times greater for single- and multiple-session exposures compared to the control group for a low baseline and 489 (single) and 53 (multiple) times greater for a middle baseline. CONCLUSIONS: Overall, the results of this feasibility study show that for moderate public speaking anxiety, the paradigm of gradual change in a single session is worth following up with further studies with more severe levels of anxiety and a larger sample size, first with a randomized controlled trial with nonpatients and subsequently, if the outcomes follow those that we have found, with a full clinical trial with patients.

Imperceptible body transformation in virtual reality: Saliency of self representation.

Change blindness (CB) is the perceptual phenomenon whereby people are blind to dramatic changes in their visual environment. In virtual reality (VR) a person's body can be substituted by a life-sized virtual one that moves synchronously with their real body movements as their self-representation. We consider whether CB occurs in VR, and whether there are differences in the case of changes to their own virtual body compared with the body of another. Forty people took part in a Qi Gong lesson in VR led by a virtual instructor. During the lesson both their own and the instructor's face dramatically changed in appearance. Overall, 73% and 85% did not notice the changes to their own and instructor's face respectively. People make iconic inferences about their visual surroundings without sampling detail, and reduced CB in the case of their own body may be a marker for self-representation.

The sentiment of a virtual rock concert.

We created a virtual reality version of a 1983 performance by Dire Straits, this being a highly complex scenario consisting of both the virtual band performance and the appearance and behaviour of the virtual audience surrounding the participants. Our goal was to understand the responses of participants, and to learn how this type of scenario might be improved for later reconstructions of other concerts. To understand the responses of participants we carried out two studies which used sentiment analysis of texts written by the participants. Study 1 (n = 25) (Beacco et al. in IEEE Virtual Reality: 538-545, 2021) had the unexpected finding that negative sentiment was caused by the virtual audience, where e.g. some participants were fearful of being harassed by audience members. In Study 2 (n = 26) notwithstanding some changes, the audience again led to negative sentiment-e.g. a feeling of being stared at. For Study 2 we compared sentiment with questionnaire scores, finding that the illusion of being at the concert was associated with positive sentiment for males but negative for females. Overall, we found sentiment was dominated by responses to the audience rather than the band. Participants had been placed in an unusual situation, being alone at a concert, surrounded by strangers, who seemed to pose a social threat for some  of them. We relate our findings to the concept of Plausibility, the illusion that events and situations in the VR are really happening. The results indicate high Plausibility, since the negative sentiment, for example in response to being started at, only makes sense if the events are experienced as actually happening. We conclude with the need for co-design of VR scenarios, and the use of sentiment analysis in this process, rather than sole reliance on concepts proposed by researchers, typically expressed through questionnaires, which may not reflect the experiences of participants. Supplementary Information: The online version contains supplementary material available at 10.1007/s10055-022-00685-9.

Self-observation of a virtual body-double engaged in social interaction reduces persecutory thoughts.

The proportion of the population who experience persecutory thoughts is 10-15%. People then engage in safety-seeking behaviours, typically avoiding social interactions, which prevents disconfirmatory experiences and hence paranoia persists. Here we show that persecutory thoughts can be reduced if prior to engaging in social interaction in VR participants first see their virtual body-double doing so. Thirty non-clinical participants were recruited to take part in a study, where they were embodied in a virtual body that closely resembled themselves, and asked to interact with members of a crowd. In the Random condition (n = 15) they observed their body-double wandering around but not engaging with the crowd. In the Targeted condition the body-double correctly interacted with members of the crowd. The Green Paranoid Thoughts Scale was measured 1 week before and 1 week after the exposure and decreased only for those in the Targeted condition. The results suggest that the observation of the body-double correctly carrying out a social interaction task in VR may lead to anxiety-reducing mental rehearsal for interaction thus overcoming safety behaviours. The results also extend knowledge of the effects of vicarious agency, suggesting that identification with the actions of body-double can influence subsequent psychological state.

Evaluating participant responses to a virtual reality experience using reinforcement learning.

Virtual reality applications depend on multiple factors, for example, quality of rendering, responsiveness, and interfaces. In order to evaluate the relative contributions of different factors to quality of experience, post-exposure questionnaires are typically used. Questionnaires are problematic as the questions can frame how participants think about their experience and cannot easily take account of non-additivity among the various factors. Traditional experimental design can incorporate non-additivity but with a large factorial design table beyond two factors. Here, we extend a previous method by introducing a reinforcement learning (RL) agent that proposes possible changes to factor levels during the exposure and requires the participant to either accept these or not. Eventually, the RL converges on a policy where no further proposed changes are accepted. An experiment was carried out with 20 participants where four binary factors were considered. A consistent configuration of factors emerged where participants preferred to use a teleportation technique for navigation (compared to walking-in-place), a full-body representation (rather than hands only), the responsiveness of virtual human characters (compared to being ignored) and realistic compared to cartoon rendering. We propose this new method to evaluate participant choices and discuss various extensions.