Virtual reality in assessment and treatment of schizophrenia : a systematic review

Objective: To conduct a systematic review about the use of virtual reality (VR) for evaluation, treatment and/or rehabilitation of patients with schizophrenia, focused on: areas, fields and objectives; methodological issues; features of the VR used; viability and efficiency of this resource. Methods: Searches were performed about schizophrenia and virtual reality in PsycINFO, Academic Search Complete, MEDLINE Complete, CINAHL with Full Text, Web of Science and Business Source Premier databases, using the following keywords: [“schizophrenia”] AND [“virtual reality” OR “serious game”] AND [“treatment” OR “therapy” OR “rehabilitation”]. The search was carried out between November 2013 and June 2014 without using any search limiters. results: A total of 101 papers were identified, and after the application of exclusion criteria, 33 papers remained. The studies analysed focused on the use of VR for the evaluation of cognitive, social, perceptual and sensory skills, and the vast majority were experimental studies, with virtual reality specifically created for them. All the reviewed papers point towards a reliable and safe use of VR for evaluating and treating cognitive and social deficits in patients with schizophrenia, with different results in terms of generalisation, motivation, assertiveness and task participation rate. Some problems were highlighted, such as its high cost and a constant need for software maintenance. Conclusion: The studies show that using the virtual reality may streamline traditional evaluation/rehabilitation programmes, allowing to enhance the results achieved, both in the cognitive and in the social field, helping for the legitimisation of this population’s psycho-social inclusion.


Palavras-chave
Esquizofrenia, realidade virtual, avaliação, reabilitação, revisão.iNtrODuCtiON Sixty years ago, the only treatment alternative for patients with schizophrenia was confinement in mental hospitals, where most of them would remain for the rest of their lives.In 1952, the experimental use of chlorpromazine in psychiatric patients had excellent results in the remission of the psychotic symptoms of those patients, leading to the introduction of pharmacology applied to schizophrenia.Currently, second-generation or atypical anti-psychotic drugs are available which, apart from acting on the positive effects of schizophrenia, also have positive results in decreasing the negative symptoms of this pathology.This result had a direct impact on the optimisation of other interventions (e.g.psycho-social, cognitive), as well as on the reduction of the number of hospitalisations.However, for the inclusion and maintenance of this population in society, there was a need to improve the treatment techniques in order to solve some of the remaining negative symptoms (avolia, apathy, social isolation, disorganisation) which hinder the integration of patients with schizophrenia into the different occupational performance contexts (e.g.family, study, work) [1][2][3][4] .Therefore, in addition to symptom control, the treatment of schizophrenia started focusing on the improvement of this population's social participation through competent performance in daily activities.
Studies point out that the low occupational and social performance of these people is related to deficits in their cognitive (e.g.executive functions) and social skills (e.g.emotional recognition) 5,6 .Currently, cognitive impairment has been pointed out as a core characteristic of schizophrenia, which may be related to poor social relationships and to the inability to manage an independent life.Treatment and rehabilitation programmes for these skills have been developed, preferably through tasks in a lab environment, focusing on attaining some improvement in terms of psycho-social functioning and mood, and, to a larger extend, on the optimisation of psychiatric treatment 7,8 .Meanwhile, the therapeutic approach to schizophrenia has evolved in terms of the shape and contents of evaluation, treatment and rehabilitation programmes, as new techniques and technologies emerge.With the advent of information technology, software programs have been created and adapted to the evaluation and treatment of these people's symptoms and skills.Recently, studies reported the inclusion of virtual reality (VR) in programmes using for evaluating and training cognitive and social skills with promising results [9][10][11][12][13] .
VR is a computer technology which enables immersion in complex environments that refer to reality through devices that transmit different types of stimuli, which are perceived by one or more senses of the user 14 .VR has three basic features: sense of presence, immersion and interaction.The sense of presence is the feeling of being present, as be a part of the virtual environment.Integrating the sense of presence, the immersion is the existence of sensory context, meaning that reality is experienced also by the sensory system (e.g.vision and hearing), creating the illusion of being in another place.VR can be classified as immersive or nonimmersive due the sense of presence that provides.Multisensory devices like head-mounted display (HMD), trackers or electronic gloves are needed to capture the movements and behaviours to transport the user to the application domain, in an immersive experience.When shipping to the virtual world is in part provided through a window (e.g. a monitor), set up a non-immersive experience.This set generates the third central aspect of the RV, the interaction.It is related to the computer's ability to detect user actions and react in real time, responding to commands given by the user and altering the landscape.The creative possibilities are limitless, similar to the real world or merely synthetics, the virtual environments provide situations that are handled and safe for achieving results often impossible in a real context 14,15 .
Since the first description of the use of VR applied to psychiatric disorders (acrophobia) 16 , the results of the use of this resource have proven favourable and efficient for the treatment of the symptoms of some pathologies, such as anxiety disorders 17,18 , eating disorders 19,20 and attention deficit disorder 21 .Regarding schizophrenia, published studies hi-ghlight the efficacy of this resource, showing positive results in the evaluation and training of cognitive and social skills [22][23][24][25][26][27][28][29] .
Recently, two literature review studies were published 30,31 regarding research performed using VR for evaluating and treating patients with schizophrenia, and the results point towards the use of VR in schizophrenia in three areas of intervention: cognitive, social, and perception/sensory deficits.They also reported the use of VR for creating symptom models (e.g.paranoia) with the aim of understanding the mechanisms of such manifestations in schizophrenia and also of contributing to creating safer and more efficient virtual realities specifically for this population.As a limitation of this resource, La Barbera et al. 31 point out the high cost of the technology and suggest that this barrier will be overcome as the price of materials and products for the use of VR tends to decrease with the growth of the information technology (IT) market.
When faced with a promising tool for the treatment of schizophrenia deficits, there is a need to summarise the know ledge obtained on the viability and reliability of the use of VR for the treatment of this population.This paper presents a systematic review on the use of VR for evaluating, treating and/or rehabilitating patients with schizophrenia, analysing: (1) areas, fields and objectives; (2) methodological issues; (3) characteristics of the VR used; (4) viability and efficiency of this tool.

MetHODS
This study follows the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) 32 .

Inclusion criteria
The inclusion criteria were the following: (1) the use of virtual reality for evaluating or treating schizophrenia symptoms, (2)  with adult participants, (3) diagnosed with schizophrenia or schizoaffective disorder, (4) published in English.

Search method
The search was conducted on the papers indexed in the PsycINFO, Academic Search Complete, MEDLINE Complete, CINAHL with Full Text, Web of Science and Business Source Premier databases, using the following key words: ["schizophrenia"] AND ["virtual reality" OR "serious game"] AND ["treatment" OR "therapy" OR "rehabilitation"].The search was carried out between November 2013 and June 2014, without using any search limiters.The references to each selected paper were checked for additional studies that could have been overlooked during the database search.
The eligibility of the studies was analysed independently by two reviewers and, whenever there was lack of consensus, a third author was consulted.

Data extraction
A table was created in order to extract the data, and a pilot test was conducted with 10 randomly selected studies, for stablish common criteria among reviewers.One author extracted the data included in the form: nature of the study; area, fields, objectives and focus; country of origin; demographic data, sample size, control group and evaluation tools; VR characteristics; evidence of VR viability and efficiency.The second author verified the extracted data.Potential discrepancies were discussed among the authors, who reviewed the papers together, and whenever consensus could not be reached, a third reviewer was consulted to analyse and clarify the issue.

Database search
A total of 101 papers were identified and, after excluding duplicate studies, 70 potentially relevant papers remained.Upon analysing the abstracts, 28 did not meet the inclusion criteria.After reading the 42 selected papers in full, nine studies were excluded, as shown in figure 1.One study 33 , which used a sample of patients with early psychosis (50% diagnosed with schizophrenia and 50% with non-specific psychosis) was included by the reviewers since it covered important aspects related to the acceptability and safety of the use of VR.In total, 33 papers were analysed in this systematic review.

Country of origin and number of papers per year
Regarding the country of origin, from the 33 studies analysed, 19 came from Asia, eight from Europe, five from North America and one from South America (Figure 2).The temporal distribution of papers is summarised in figure 3, where it can be seen that the first reported study dates back to 2003.

Areas, fields and objectives
The studies reveal three areas (Table 1): cognitive, social, and perceptual/sensory areas, classified into two main fields, namely evaluation of skills and symptoms (n = 27) and skill training (n = 6).

Characteristics of the sample and procedures
Participants were recruited at psychiatric units (outpatient and inpatient units), and the average number of participants in the experiments is 22 (SD = 10).A total of 24 studies included a control group, usually using a paired group with corresponding genders and ages.Considering all the studies, the age of the participants ranged from 28 to 66 years, which resulted in an average age of 31.5 years.
Regarding procedures, 26 were experimental studies and two were randomised controlled trials (RCT).

Evaluations and measurements
Positive and negative symptoms of schizophrenia were evaluated in 29 studies, and in 23 studies, these were evaluated with The Positive and Negative Syndrome Scale (PANSS), as listed in table 2. The feeling of presence was verified in five papers, using the Presence Questionnaire, and side effects related to the use of VR were evaluated in three studies using the Simulator Sickness Questionnaire (SSQ).

Characteristics of the virtual reality used
Most studies (62%) used virtual realities specifically designed for the experiments.The remaining studies resorted to commercially available programs or programs developed originally for other studies.Regarding the feeling of presence, 55% of the studies used a non-immersive interface.

VR viability and efficiency
The conclusions of all (100%) studies indicate that VR is a useful tool to evaluate and treat patients with schizophrenia, as shown in table 2. The results from the evaluations of cognitive, social, and perceptual and sensory skills supported by VR show significant correlations with the results obtained from evaluations with classic tools (e.g.WCST, SBS).In the six studies that used VR for training cognitive and social skills, there was an improvement in results and performance, when compared to training with classic programs (e.g.CRTT, RBANS, SSKT).Out of these six studies, four measured, from a subjective perspective, the usability of VR in the training program (Table 3).
The results of the two randomised trials are consistent, demonstrating an increase in the level of motivation, assertiveness and participation rate of individuals during the performance of the tasks proposed.In the vast majority of the studies (97%), subjects did not have side effects caused by the use of the technology.Three studies reported a possible interference of positive symptoms in task performance and two main limitations were reported regarding the technology: the high cost of software development and the need for constant technical support for the set up and maintenance of the process (n = 1).

DiSCuSSiON
A variety of programs was found for evaluating and treating schizophrenia symptoms (e.g.paranoia, cognitive and social deficits) which used VR with different features and specificities.There seems to be a consensus regarding the viability of applying this technology in patients with schizophrenia, and that confirms Freeman's 53 prognosis, which highlighted this resource as a new paradigm for the study and treatment of schizophrenia.VR tends to overcome some of the limitations in terms of evaluation and treatment of cognitive and social skills, providing a natural environment which is very close to the real context.This seems to be the major difference of VR -the feeling of presence (immersion) -that allows subjects to perceive the environment as if it were real.However, it appears to allow a safer experience than in a real setting, namely regarding emotional issues, such as frustration when faced with failures and mistakes 24,26,[41][42][43][44][45][46] .This tool also allows Virtual reality in schizophrenia: review using and monitoring the performance of complex tasks in a dynamic manner, enabling real-time changing of parameters by therapist.Assessments or programs based on virtual environments overcome the limitations of data obtained in laboratory settings, allowing real-time interventions in a simulated context, with strong ecological validity inasmuch as they greatly resemble everyday functioning 54,55  Participants of VRG and TAG showed a significant improvement in cognition, although VRG showed better performance in Wisconsin Card Sorting Test (WCST) than TAG.This comfortable and motivating scenario, in which patients may face challenges as if they were actually in the real context, thus feeling safe, can also explain the promising result generalisation rates, revealing the potentialities of this promising ecological tool for the evaluating and rehabilitating of patients with schizophrenia 11 .Twelve years have passed since the first experiment with the use of VR in schizophrenia.Despite all the positive facts concerning the use of this technology in the evaluation and treatment of schizophrenia, there are a few number of publications per year.It can be explained by the high cost and complexity of the tool 24,26 , however, as mentioned by La Barbera et al. 31 , this is a market in highly development, which tends to substantiate the facilitation in the access to inputs and technology, and, consequently, enable further researches in this area.
The first researchers group that published studies about VR environments addressed to people with schizophrenia is from Korea 34 .They have published the largest number of studies to date, particularly related to social skills.Nonetheless, is possible to notice an increasing interest from Spanish, Chinese and Canadian groups, to develop pilot studies and randomized trials using VR on training social and cognitive skills programs, contributing to the importance of this tool in the treatment of schizophrenia.
Regarding methodology, the analysis of the selected studies revealed that most of them are experimental studies, with small samples, basically focused on the usability of VR for the evaluation of symptoms and skills (cognitive, social, and perceptual and sensory) of patients with schizophrenia.Studies 35,42 that proposed adapting classic evaluation tasks such as WCST and RBANS to the virtual platform obtained results which correlated positively with those obtained using the classic method.Meanwhile, the technology allowed going beyond the fields established in conventional evaluations, and the immersive environment seems to provide a multidimensional response to the tasks proposed, allowing to observe behaviours that are intrinsic to the pathology.The small number of controlled trials covering the efficiency of the VR programme in the training of cognitive and social skills made impossible to statistically compare the results obtained.It is expected that, in the near future, a larger number of randomised trials with more robust samples is conducted, because the results of the experimental studies so far advocate and encourage this new approach to treatment.
Regarding the virtual environments used in the studies under analysis, a wide diversity of scenarios, specifically designed for each study, was observed, which justifies an experimental stage in the introduction of the VR tool in the treatment of schizophrenia.This situation may interfere in the reproducibility of the studies, due to the complexity of the technology, and this may precisely be the explanation for the difficulties found in the development, replication and validation of VR-based programs.However, studies were found regarding the design, development and application of seven virtual reality programmes applied to schizophrenia: five programs designed for skill training (AVIRC 14 , VRAMMA 38 , VAP-S 40 , VRVTS 26 , SOSKITRAIN-VR 8 ) and two for skill evaluation (VEBFDB 43 , VEFSA 49 ).All authors reported an improvement in the trained skills, as well as greater efficiency and user interest and motivation in the use VR, although these studies reveal obstacles to the technology, such as its high cost and a frequent need for maintenance.Results also show that VR computer programmes can outperform classic computer programs 7 and increase the gains obtained from cognitive and social remediation, by allowing to train skills in environments that can be customised to each user 56 .Therefore, it can be assumed that VR-based skill rehabilitation can also entail gains in the psycho-social performance of this population.
Despite the highlighted potential, experiments with this technology in schizophrenia are recent, when compared to those conducted for other psychiatric disorders.Perhaps the uniqueness of the positive symptoms of this disorder may have been a cause for doubts on the safe use of this tool, leading to the postponement of experimental studies with this population.However, studies 33,[50][51][52] show that VR does not interfere in the worsening of paranoid delusions and that, quite the contrary, it may be a useful tool for understanding and treating those symptoms.Therefore, results from experiments point towards the possibility of setting parameters related to behaviours that are specific to patients with schizophrenia 37 , which may be important for a better understanding of the behavioural responses of these patients and, consequently, for the development of specific programmes for the treatment of schizophrenia.
Even though problems such as its high cost and the constant need for maintenance were highlighted in this review, other pathologies (e.g.phobias) have been widely treated with this technology [57][58][59] , showing that its benefits may overcome such potential obstacles.This paper stresses the need for strengthening partnerships between psychological and technological research teams, in order to solve problems, such as technical support, software cost and maintenance, which could speed up the development of new programmes, possibly more efficient, due to the potential they present for treatment teams, allowing to streamline the attainment of intervention goals in schizophrenia.These programmes may contribute to the inclusion of this population into its social and labour contexts in a competent and significant manner.

CONCLuSiON
This study highlights the benefits of VR for the evaluation, treatment and skill training of patients with schizophrenia, showing that this tool has the potential to become a new effective way of treating the symptoms and deficits experienced by this population.However, it is important to highlight that VR is a complementary tool, which can integrate programs consisting of other therapeutic approaches.This device gives the therapist not only a new instrument, but ascribes an active role in prescription and customization of approaches tailored to the user's needs and context of daily living, as well as in terms of stimulus presentation, monitoring and adjusting the process according to need and developments in.The results suggested to conduct randomised and follow-up multicentre trials, comprising larger samples that will allow to obtain more consistent data regarding the improvement of the skills trained, as well as to validate and perform the meta-analysis of the results, confirming the evidence on the use of VR to assess and treat schizophrenia.This technology also appears to allow a more efficient generalisation of the gains achieved, which may enable a better perception of their improvement by the patients themselves, a factor which is intrinsically related to quality of life 60 .It is important that efforts are made so that research continues in this area, namely through consortia between psychology labs and information technology labs, as VR seems to allow patients with schizophrenia to be exposed to challenges in a safe and efficient way, so that they can overcome their difficulties and thus have an efficient social participation.
iNDiviDuaL CONtriButiONS Mônica Macedo -Contributed in the conception and design of the study, in the analysis and interpretation of the data, in the drafting of the article, and completed and approved the written version for publication.
antónio Marques -Contributed in the conception and design of the study, in the analysis and interpretation of the data, in the revision of the article, and approved the final version for publication.
Cristina Queirós -Contributed in the analysis and interpretation of the data, in the revision of the article, and approved the final version for publication.

autHOr DiSCLOSure StateMeNt
No competing financial interests exist.

aCKNOwLeDGeMeNtS
This research was supported by grants from the Coordenação de Aperfeiçoamento de Pessoal de Ensino Superior (Capes), Ministério da Educação, Brasil.

Figure 3 .
Figure 3. Number of publications by year.

table 1 .
Summary of areas, domains and objectives of selected studies

table 3 .
VR Usability in training abilities (subjective perspective)
18Those vir tual environments features allow to reproduce the real-world challenges in VR.In agreement with this, Emmelkamp et al.18developed a study with 33 acrophobic participants (16 exposures in vivo and 17 with VR), aiming to evaluate the effectiveness of virtual reality (VR) exposure versus exposure in vivo, using virtual environments exactly copied from the real settings used in the exposure in vivo platform.The authors concluded that the VR treatment was as effective as exposure in vivo in combating anxiety and avoidance.Additionally, Tsang and Man 26 reported similar results in a randomized study aiming to investigate the efficacy and effectiveness of VR as a cognitive intervention for enhancing vocational outcomes.Seventy five schizophrenia patients [25 VR group (VRG), 25 therapist administered group (TAG) and 25 conventional group] participated in vocational training, where contents of VRG and TAG were the same, just differing in training mode.