Table 1.

Details of included studies.

Author Country/Setting	Design/ Participant	Intervention	Control	Outcome Measures	Knowledge	Attitudes	Empathy	Stigma
[51] United States	2 arm quasi-experimental Patients with schizophrenia and healthy control	G1: Patients with schizophrenia, n = 25 - Participants interact with surroundings, objects in a virtual environment using mouse. They were asked to take pills with virtual distractions (siren, ringing doorbells).	G2: Healthy control, n = 16 - same intervention	Pre - The Medication Management Ability Assessment (MMAA) - Continuous Performance Test	↑ Better pill selection, less time discrepancy when taking medications and greater MMAA score with reminder notes, clock	Nil	Nil	Nil
[47] United States	Pre, post single group quasi-experimental Nursing under-graduates from 3 classes	G1: Virtual Dementia Tour, n = 163 - equipment that altered participants’ vision, hearing and touch - complete 5 tasks - 10 min	Pre group	Pre, post - Dementia Attitudes Scale - Knowledge about Memory Loss and Care - Healthcare tour survey Post only - Students’ reflection	↑ Improvement noted post-test although it did not reach significant levels	↑ Significant improvement at post-intervention	↑ Students’ reflection highlighted that patients with dementia required empathy and sensitivity. Students affirmed the need to support caregivers and families.	Nil
[49] Spain	Post only single group quasi- experimental Psychology under-graduates aged 18 to 28	G1: Stigma-Stop, n = 26	Nil	Post only - Open ended questions and the following questions rated ‘yes’ and ‘no’: (1) Whether character is emotionally well? (2) Whether participant can help the character? (3) Whether participant had similar experience?	-- Most participants rated characters as emotionally unwell: a. 96%- panic disorder with agoraphobia b. 96% for schizophrenia c. 73% for bipolar disorder d. 100% for depression	-- Most participants felt they could help the character: a. 100%- panic disorder with agoraphobia b. 88% for schizophrenia c. 79% for bipolar disorder d. 92% for depression	Nil	↓ Participants’ open-ended answers showed reduction of stigma.
[36] Spain	2 arm pre post RCT High school students aged 14–18, from 24 classes	G1: Stigma-Stop, n = 484 from 21 classes - Non immersive virtual reality game presenting four characters experiencing depression, bipolar disorder, schizophrenia, panic disorder with agoraphobia.	G2: n = 68 from 3 classes	Pre, post - Student Attitudes toward Schizophrenia measures stereotypes and other aggressiveness	-- Most participants rated characters as emotionally unwell: a. 96.8%- panic disorder with agoraphobia b. 86.8% for schizophrenia c. 61.8% for bipolar disorder d. 96.1% for depression	-- About half to most of the participants felt they could help the character: a. 82.4%- panic disorder with agoraphobia b. 62.5% for schizophrenia c. 53% for bipolar disorder d. 90% for depression	Nil	↓ Stigma-Stop group had significantly lower stigma
[34] Australia	Pre-post single group quasi- experimental Public and psychology under-graduates	G1: VR, n = 50 - with head mounted display, headphones and Xbox controller, - virtual environment with VH environmental sounds and voices suggestive of danger - 10–15 min	Pre group	Pre, post - Knowledge * about psychosis - Attitude ^ by Reavley and Jorm (2011) - Empathy by clinical empathy scale	↑ Significant improvement at post-test	↑ Significant improvement in attitude scores at post-test	↑ Significant improvement in empathy scores at post-test	Nil
[35] Australia	Pre-post 2 arm quasi-experimental Medicine and pharmacy under-graduates	G1: Australia Vic Virtual Dementia Experience, n= 80 - multisensory, virtual simulation to allow participants to experience perceptual and cognitive difficulties by patients - 1.5 h	G2: Waitlist control group, n = 198 - curriculum as usual	Pre post - Dementia Attitudes Scale	Nil	↑ Intervention group had significantly better attitude scores at post-test	Nil	Nil
[44] United States	Descriptive study Nursing students, n = 126	G1: VR condition, n = 126 - Students viewed virtual neighbourhood with two houses: one belonged to someone with schizophrenia, another with depression. Students enter house and interact with patient. - Students were given case studies on the patients. - 45–60 min	Nil	Post only - * 35-item Second Life (SL) Simulation Evaluation Survey - Two open-ended questions regarding feedback about stimulation	↑ Second Life Simulation, as a teaching modality, was moderately effective.	Nil	Nil	Nil
[45] United States	Pre- and post-intervention study 4th year nursing undergraduates	G1: VR, n = 149 - one virtual simulation case study weekly on depression, bipolar disorder, anxiety, alcohol withdrawal and schizophrenia - 30 min, student can repeat simulation	G2: Non-simulation group, n = 150	Post only - Two vignettes, schizophrenia and depression. Participants answer about their perception of helpfulness of certain people (including healthcare workers, traditional healers, family and friends), specific medications and interventions (e.g., physical activity, massage, relaxation, specific therapies).	↑ For perceived helpfulness of pharmacological interventions, intervention group was less likely to rate antipsychotics and sleeping pills as ‘do not know’ for the depression vignette. Control group was more likely than G1 to rate psychiatric hospital admission and electroconvulsive therapy as ‘do not know’.	-- Both groups agreed solving the problem by self as unhelpful.	Nil	Nil
[50] Australia	Pre post single group quasi-experimental Students and public	G1: Visit with Viv, n = 35: - VR about Viv, who recounts her experiences of dementia > life-size in art gallery, > Occulus Quest VR headset in university, n = 36 - 15–20 min	Pre group	Pre post - comprehensive state empathy scale - Change in emotional distance scale	Nil	Nil	↑ Significant improvement in empathy scores from pre-test	↓ Significant reduction in stigma levels at post-test
[46] Brazil	Pre post single group quasi-experimental Medical students from 3 universities	G1: AR, n = 21 - figures and voices from narratives of three patients with schizophrenia - voices included whispers, commanding and threatening speech - 3 min	Pre group	Pre post - Schizophrenia stigma * Post only - Evaluate environment simulation	Nil	↑ Significant increase in the average score of help-giving at post-test	↑ Increase in empathy	↑ Increase in stigma especially in fear, pity and segregation.
[48] Ireland	Pre post repeated single group quasi-experimental Health professionals, voluntary groups and public	G1: Virtual Dementia Tour + watching another group doing distortion session, n = 240 - 2 h - 10 min of sensory distortion - 30 min debriefing	Pre group	Pre post, follow up at 3 months * Tool that measures empathy, understanding of behaviours and role of the person in care decisions.	↑ Significant improvement in understanding of behavioural impact of dementia across time points	Nil	↑ Significant improvement in empathy across time	Nil
[41] United States	4 arm post only RCT Psychology research participant pool and university community	G1: VR condition, n = 26 - Participant plays the character of someone experiencing schizophrenia, visiting a pharmacist asking for prescription refill. - 4.5 min G2: Empathy condition, n = 26 - Participants were asked to pen their thoughts about experiencing VH and AH while getting prescription - 1 min G3: VR+ empathy condition, n = 26 - Empathy before VR condition	G4: control, n = 26 No intervention control group	Post only - Empathy 12-items - Social Distance Scale - Attitudes Toward People with Schizophrenia, 7 items - Evaluation of simulation - Pre-existing attitudes towards people with schizophrenia, 8 items	Nil	↑ Intervention group had better attitudes but this did not reach significant levels	↑ Significant improvement in empathy in intervention groups using VR	↓ VR group had significantly lower stigma
[42] Germany	3 arm post only RCT Majority are students from university, n = 114	G1: VR, n = 31 - Young male actor speaking about his experiences with schizophrenia, including how his loved ones cope with it. G2: regular video, n = 45 - Similar as the VR but fixed perspective	G3: No intervention control group, n = 38	Post only - Stigma with four related constructs: anxiety *, social proximity ^, empathy (by Kinnebrock et al., 2010, - Benevolence (using Community-Attitudes-Toward-the-Mentally Ill Inventory)	Nil	Nil	Nil	↑ VR contact did not decrease stigmatization compared with control but had increased stigmatization compared with video.
[53] The Netherlands Alzheimer’s Society	Pre post single group quasi-experimental Informal caregivers caring for those with dementia	G1: VR, n = 35 360-degree simulation movie on virtual reality (first person view) and e-course, n = 42- different scenes of interactions e.g., confronted by daughter about remote control in cupboard and she talks to people on the phone about you - 13 min	Pre group	Pre, post - Empathy measured by Person-centeredness subscale of Approach to Dementia questionnaire and ‘perspective-taking’ subscale of Interpersonal Reactivity Index	Nil	Nil	↑ Significant improvement using the perspective-taking subscale	Nil
[52] United States	Post only single group quasi- experimental Second Life Users	G1: Second Life VR, n = 579 - Character toured environment, experiencing hallucinations including voices, posters changing text to profanities, floor that fall away, TV that encourage suicide and gun with voices telling character to commit suicide, and own reflection with bleeding eyes.	Nil	Post only - Questions about understanding of hallucinations	-- Intervention group had improvements of understanding of (1) AH- 76.86% (2) VH- 69.91% (3) Schizophrenia- 73.9%	Nil	Nil	Nil
[43] Hong Kong	3 arm pre post RCT University students 18 years old and above	G1: Immersive animation, n = 82 Participants played a character, Yan, who had mixed anxiety and depression, and who was speaking with an uncle. Pop up messages illustrated problems in Uncle’s communication. G2: Text condition, n = 80 Participants read same story in 2D effect, without immersive experience. All 3 groups: 10 min, had VR headset, Oculus Go.	G3: Control, n = 82 Exoplanet VR video, 360°	Pre, post, 1 week follow up: - Stigma by 21-item Public Stigma and Acceptance Scale - 7-item Sense of Embodiment Scale * - Story Transportation	Nil	Nil	Nil	↓ Immersive animation and text condition had significantly lower public stigma at post-test and follow up compared with the control group. Immersive animation vs. text condition did not have significantly different stigma levels between them.

AH = auditory hallucinations; AR = augmented reality; diff = difference; G = group; M = Mean; MMAA = Medication Management Ability Assessment’ min = minutes; sig = significantly; VH = visual hallucinations; * constructed by authors; ^ modified by authors, ↑ = increased, ↓ = decreased, -- = non-quantitative measurement.