Users’ role in co-designing products has changed: from influencing outcomes to influencing development/design; from standardizing to customising products/outcomes; from participating to engaging designers/developers. Although this participatory design (PD) approach makes users’ role more prominent it has been under-utilised for the technological development of products for people with neurodevelopmental disabilities (NDD). The purpose of this paper is to present a responsible research and innovation example, in conversation skills training for people with autism, using virtual reality (VR).
The PD approach was adopted during the iterative development of the virtual world and training materials. Multiple baseline design was utilised consisting of three participants on the mild/moderate end of the autism spectrum. Participants joined 15–16 sessions over four phases of structured conversations, delivered both face-to-face and virtually.
The feedback sessions revealed that the participants felt VR has the potential in providing training for people with autism spectrum disorders. Moreover, they thought delivering the training in three formats could enhance their learning, since PowerPoints, videos and chatbot would represent teaching, showing and practicing, respectively.
PD promotes a “one-size-fits-one approach”, cultivating agile, inclusive, responsive design approaches for people with NDDs, so that outcome meets their needs and preferences, while VR training allows for a wider implementation, benefiting a wider range of learners.
The RRI approach increases the inclusion of people with disabilities in the decision-making process through dialogue with “experts”, making their role more visible, fostering an ethical and sustainable innovation process, leading to more desirable outcomes.
Politis, Y., Sung, C., Goodman, L. and Leahy, M. (2019), "Conversation skills training for people with autism through virtual reality: using responsible research and innovation approach", Advances in Autism, Vol. 6 No. 1, pp. 3-16. https://doi.org/10.1108/AIA-05-2018-0017Download as .RIS
Emerald Publishing Limited
Copyright © 2019, Emerald Publishing Limited
Autism research made a big leap forward when Baron-Cohen et al. (1985) used the “theory of mind” hypothesis to explain the social interaction and communication impairments associated with the condition. Even though many studies since then have produced similar results and the evidence is robust, more recently there is scepticism about this hypothesis due to the fact that it does not explain the restricted or repetitive interests and behaviours (RRIB), i.e. the earliest manifestations of autism (Tager-Flusberg, 2001). It is not surprising, therefore, that autistic people believe that non-autistic people fail to recognise the importance of RRIB aspects of autism (Bakare et al., 2009; Campbell and Barger, 2014). Studies have shown that according to autistic people it is in fact sensorimotor difficulties that contribute to their socio-communicative challenges (Chamak et al., 2008; Robledo et al., 2012).
A new way of thinking about autism has recently emerged, where autistic people’s lack of interest in socialising is attributed, according to the “expertise” hypothesis, to the fact that they have not developed these skills at an expert level like most neuro-typical people (Warland, 2015). There are those in the autistic community that would reject the idea they need to change to become more neuro-typical because they do not feel autism is a disability and they value the ways that autism makes them unique, in spite of the stigma attached (Jones et al., 2015). They thus see autism as an identity that has both strengths and weaknesses that contribute valuably to the human neurodiversity (Kapp et al., 2013; Pellicano et al., 2014).
However, others see their condition as a barrier to social inclusion. This segment of the autistic community would welcome ways to bring down those barriers and be able to “fit in” to the society (Humphrey and Lewis, 2008). People with autism spectrum disorders (ASD) typically have difficulties with communication and socialisation skills, and they may exhibit repetitive behaviours and/or have restricted interests (Fombonne, 2005; Levy and Perry, 2011; Rivet and Matson, 2011; Suzuki, 2011). Such difficulties have been shown to be linked to the development of problem behaviours, and thus people with ASD do struggle to keep or find employment, become part of the community, and experience low quality of life (National Research Council, 2001). Assistive technology’s role for people with autism is to primarily improve their quality of life and help them in their participation in education, employment and community engagement.
For years interventions are happening through instructional video modelling (Bellini and Akullian, 2007), which is an effective tool. Technological advancements though have added new intervention options with great potential, such as through computerised technologies and mixed reality – a spectrum from augmented reality to virtual reality (VR). VR in recent years has seen a great deal of technological advancement that have made it more affordable, and thus has made VR more appealing to the public. This technology has found application in the socialisation and training of people with neurodevelopmental disabilities (NDDs), because it offers safe and less anxiety-provoking environments, where training and learning exercises and practice can be embedded in an inclusive manner that meets the needs and preferences of each individual. The literature on digital approaches to offering training on communication skills through VW suggests that adolescents with ASD can show improvement in initiating a conversation in a VR café training setting (Parsons et al., 2004; Mitchell et al., 2007). Moreover, social training through a VR platform can enhance the social understanding of people with ASD (Cheng and Ye, 2010; Herrera et al., 2008).
In order for such interventions to gain broad acceptance from the autistic community, they would need to feel ownership of the process and the resulting product. Technological development has traditionally depended on user involvement in the design and testing phases so that the process culminated with a highly usable and desirable product (Coleman et al., 2012). That has been true even more so in recent years in the era of digital technologies and social media. Societies have embraced a shift to a “participatory culture”; consumers create, share and respond to media (Jenkins, 2006). A diverse group of people with varied roles in the process are coordinated towards a shared goal and can have a successful outcome (Lankshear and Knobel, 2010). End-users ceased to be just the recipients and testers of products and became equal partners in the design process (Scaife and Rogers, 1999). Instead of influencing the outcome, they had a relationship with the designers; from helping with the standardisation of products they became engaged in their customization. There has therefore been an evolution in the roles that the different stakeholders, be it researchers, designers, technologists, practitioners and the general public have adopted the development of scientific and technological products and services. That is in essence responsible research and innovation (RRI).
The authors have adopted RRI while developing the intervention that will be presented in this paper. They tried to cultivate an inclusive, responsive approach to designing the virtual world (VW) environment. In this initial, pilot phase of the project, people with autism were involved at every stage of the process, from inception of the VW environment, to testing of the VW and of the training material and finally the intervention itself, followed by feedback of the intervention. By being a part of the design process of the training, autistic people had the opportunity to provide their perspectives and influence the outcome so that it better meets their needs and preferences.
This paper will give a detailed account of the steps the authors took to develop the intervention on conversation skills training through VR for individuals on the autism spectrum. However, before we do that, it would be useful and informative to give the readers definitions of RRI and a deeper understanding of the term, especially in the European context, followed by a list of examples of applications and projects that have adopted the RRI principles of diversity and inclusion, anticipation and reflection, openness and transparency, and responsiveness and adaptive change. We will then present the RRI tool we chose to use for our project, a 3DNovations and Hive-RD VW.
Responsible research and innovation explained in a nutshell
We can all agree that technological advances have been very rapid over recent decades. Society has mostly been the beneficiary of the progress achieved; however, in some instances the outcomes were misaligned with the values of society (e.g. misuse of nuclear energy, satellites. internet). In those instances, the consequences of the new concepts/approaches/products devised were not fully appreciated, nor were the potential ethical conflicts. In response to that, over the last decade or so, there has been more willingness to re-imagine “science and innovation policy and the social contract for science in which scientiﬁc freedom is exchanged for the promise or expectation of socially-beneﬁcial impacts” (Owen et al., 2012; p. 752). These impacts form the basis of “responsible innovation”.
It was recently (2011) that the European Commission explored the idea of RRI. RRI involves all stakeholders from the early stages, which encourages and empowers innovation to be developed collaboratively, ensures co-responsibility and changes the association of the terms “science” and “society” from science in society to “Science with and for Society (SwafS)” (Laroche, 2011). SwafS has increasingly been highlighted as a best practice for research and innovation.
The 2014 Rome Declaration on RRI in Europe stated that benefits of RRI ensure “research and innovation (R&I) deliver smart, inclusive and sustainable solutions to societal challenges by engaging new perspectives, new innovators and new talent” (EC, 2016, p. 6). In 2015, Commissioner Moedas proposed three strategic priorities under the themes of open innovation, open science and open to the world. An important aspect of the open science priority is citizen science, which is an avenue to promote RRI through outreach activities and public engagement. The R&I Commissioners meeting on 25 May 2016 highlighted the need to promote RRI throughout the SwafS Work Programme, with special attention being placed on co-design with stakeholders and end-users.
In essence RRI is about solving real-world problems collaboratively, between researchers, policymakers, other organisations and bodies, industry and concerned citizens, at all stages of the research process. In doing so, it would be more likely that consequences of research outcomes and methods would be anticipated (unethical behaviours, questionable practices), and by involving society in the process, it would help shape a more desirable and functional world for future generations (European Commission, HORIZON 2020, available at: https://ec.europa.eu/programmes/horizon2020/en/h2020-section/responsible-research-innovation).
Responsible research and innovation in practice
The RRI movement follows in the footsteps of increased inclusion, over the past couple of decades, of the public in the governance of science and innovation with the establishment of forums on related issues (Stirling, 2006; MacNaghten and Chilvers, 2014). This more bottom–up approach came as a response to a need for more legitimacy for those areas (Irwin, 2006; Hajer, 2009). The public involvement comes in the form of multi-stakeholder partnerships, inclusion on scientific boards or other modes that encourage debate and discussion between the researchers, scientists, government organisations, NGOs and the public. This trend seems to be quite prevalent in Europe, where we can find many examples of application of the RRI principles.
The RRI approach can be found in a science setting such as the Science Centre Experimentarium in Copenhagen that tried to improve the visitor experience, more specifically create a better flow and navigation throughout the exhibits, through dialogue and participatory planning with all relevant stakeholders but primarily the visiting public. Science cafés are another good example of engaging all relevant stakeholders in an informative but at the same time fun way. The Copernicus Science Centre (Denmark) has adopted the “Reversed Science Café” format, where the public are the experts and the experts are asking the questions. This is a great way of challenging current ideas and ways of thinking, of involving new blood in the conversation, thinking outside the box and gaining a fresh perspective.
An RRI approach can also involve public engagement with ongoing projects such as the example of Atelier Arts Sciences, This endeavour meets a key aspect of the RRI approach, which is helping projects to contact end-users from the very early stages of the development of the project. The process culminates with the Experimenta event, which is the venue where the finalised prototypes are showcased. RRI has also found application in the field of Education. For instance, the EnRRICH (Enhancing RRI through curricula in higher education) tool’s main goal is to help academics re-design their curricular in order to meet the RRI principles.
The examples, which were presented above, are present in the RRI Tools toolkit (www.rri-tools.eu), which is a database of hundreds of examples of tools, inspiring practices, projects and library elements of RRI application across many fields, which has been built with and for the community of practice. The RRI product from the field of education (more specifically training) that the authors’ work was based on will be described below.
3DNovations and Hive-RD
Another very important entry in the education section is that of 3DNovations, which comprises of immersive 3D virtual reality applications for primarily training and research purposes, but which also has a strong community/customer services aspect, targeted at people with autism and other complex needs. It follows the RRI principles because it is designed at every stage with and for people with autism. In fact, 3DNovations has demonstrated that “industry led RRI tools and methodologies can create a more inclusive society by unlocking ideas and talents of people with autism, increasing their employment and improving industrial competitiveness” (European Foundations Award for Responsible Research and Innovation 2016, available at: http://efarri.eu/finalist/3dnovations/).
The 3DNovations products have been developed by Hao2, a company that strives to participate in and contribute to research and innovation. Its workforce comprises 80 per cent of people on the autism spectrum. They are, therefore, part of the research and development teams that create VWs, which enables all relevant stakeholders (people with NDDs, technologists, researchers and crucially, employers) to work together to find ways to include people with complex needs – with the diversity, talents and fresh ideas that they can bring to the table – in the labour market.
Research has shown the breadth of affordances and benefits of VR to the mainstream population; however, very little research has been carried out with people with NDDs (Freina and Ott, 2015). The authors’ project, “Virtual Learning for people with Autism Spectrum Disorder (VL4ASD)”, is addressing that gap in the literature, and they decided to link up with Hao2 and tap into their experience and knowledge of this technology and their understanding of the target population.
Hive-RD provided to the VL4ASD project a VW that was a first iteration, early alpha stage of development, Hive-RD construct powered 3DNovations product. This VW was developed, using RRI principles, in the Unity platform, whose main advantage over other platforms is its high degree of customization that offers the opportunity to adapt training to better address individual needs. A truly participatory design (PD) process was adopted for the development of the VW, where people with autism were involved at every stage, from the inception of the VW, to stress and user testing. An intervention on conversation skills training for people with NDDs (to address the social skills difficulties of the people with autism) was built based on this VW.
The fact that Hao2, as a social business, uses PD approaches (co-design, with and for people with autism) in the development of 3D environments, meant that there was the potential for the real social impact of their developed products and services. Some of Hao2’s 3DNovations products are powered by Hive-RD, a company who is seen as “exemplar in terms of its commitment to inclusive innovation as well as its founders’ technical and design expertise […] (Their work follows the) Responsible Research and Innovation (principles) ie. everything they do at every stage of a project […] is smart, sustainable, collaborative and inclusive” (Nicola Herbertson, Founder of Hao2). The next section will give a detailed account of the development of each element of that intervention.
“Making” the intervention
Virtual world development: early stages
It is made clear then that the creation of the VW was an iterative process that involved a number of development phases. In the first phase, the two developers, who have a great deal of experience and expertise with autism, created a minimally viable version of the VW. This was followed by two software test sessions with 15-user stress test group (8 of whom were people with ASD), which purpose was to identify anomalies or features that were difficult to use. The focus of this phase was mainly on usability issues because the code was under development; there was no noticeable difference in performance between the two groups. The server CPU never showed a reading above 25 per cent at maximum stress on the second test, vs server failure at maximum load (99 per cent+) on the first test. Therefore, based on the feedback from those stress tests, a revised version of the VW was created that showed significant improvements on performance (Politis, Olivia, Olivia and Sung, 2017).
Training material development
The next step involved identifying material that would be appropriate and useful for conversation skills training. The process begun with a thorough review of literature and resources freely available online. The search first looked at social skills training more generally, which was followed by a more focussed search on communication skills training for both mainstream and people with NDDs. Traditional training approaches for the general population included list of ways (with animations) to, for instance, identify good conversation topics, such as conversation starters and how to maintain a conversation (wikiHow website, available at: www.wikihow.com/Come-Up-with-Good-Conversation-Topics). Traditional training approaches for individuals with autism included visual strategies on, for instance, the topic of turn-taking (Autism Teaching Strategies website, available at: http://autismteachingstrategies.com/autism-strategies/turn-taking-in-children-with-asd-visual-based-social-skills-strategies/) and social stories about how to respond to questions (Do2Learn website, available at: www.do2learn.com/organizationtools/SocialSkillsToolbox/RespondingToQuestions.htm).
The examination literature and resources (mentioned in the previous section) illustrated to the authors what are the key components of the “art of the conversation” that would need to be included in the training intervention. The training would consist of six sessions in the VW and would cover the following categories:
What is conversation and how does it work?
Why is conversation useful? And approaching someone to have a conversation.
Starting a conversation and why are topics appropriate and inappropriate?
Finding common interests.
Taking turns and answering questions.
Switching topics and ending a conversation.
The training material was then developed in Opensimulator, which took the form of a brief PowerPoint presentation for each session, containing illustrative photos, images and sketches, and ended with a brief multiple answer quiz (three questions at the end of each presentation); a list of two to three YouTube videos that would provide either role-playing examples or real-life examples of the elements of conversation mentioned in the presentations; and a non-playable character (any character that is not controlled by a player) who was narrating the text of the PowerPoint presentations, so that the participants could read and listen to that material. The developers then transferred these materials to the Unity-based VW.
While this process was ongoing, the authors also worked closely with the developers to explore the feasibility of introducing a “chatbot” (non-playable character), as an additional training element. The chatbot would offer a chance to the participants to test their conversation skills based on their knowledge acquisition from the training with the PowerPoints and videos. The chatbot was programmed to have basic conversational interactions with the users, based on a script that “tried to anticipate a participant’s possible answers to a question or a statement, so it was kept as simple as possible, to avoid complications arising from unforeseen responses” (Politis, Olivia, Olivia and Sung, 2017, p. 7). When all the training elements were completed, the next development phase could commence, namely, feasibility, usability and accessibility testing with people with ASD and with practitioners.
User feedback of training material in the virtual world
This testing phase included a feedback session with six people with ASD, and a feedback session with two practitioners who have had experience working with the above-mentioned population (and those six individuals in particular). The feedback sessions revealed that the participants felt VR has potential in providing training for people with NDDs (ASD in this case) and saw the VW environment in a positive light. Moreover, they thought that delivering the training in three formats (i.e. PowerPoints, videos and chatbot) could contribute to the learning process in different ways, since they felt that PowerPoints would represent teaching, videos would represent showing and chatbot would represent practicing.
The participants of the two feedback sessions (both individuals with ASD and practitioners) offered their opinion on issues regarding “the visual presentation, pacing and organisation of the content” (Politis, Olivia, Olivia and Sung, 2017, p. 9) and on more subjective issues such as the colour scheme of the VW, which are important because they address the customization aspect of the project. Although, as mentioned above, the concept of a chatbot was really appealing to both people with ASD and the practitioners, they both felt that in its current stage of development the chatbot was not very functional due to lack of interactivity. The reason was that when the chatbot could not understand an answer by the participant, it would repeatedly ask them to rephrase until the answer registered as one of the possible answers it was programmed with, which would cause annoyance and make the conversation feel forced (unnatural). Finally, it was interesting to notice that during the feedback sessions, participants focussed more “on design elements (narrative, sounds, characters, etc.), and somewhat neglected the educational content (the training materials in all three formats)” (Politis, Olivia, Olivia and Sung, 2017, p. 10).
At that point, both the VW and the training material were revised to address the feedback provided by the participants. The most significant change was that the chatbot was going to be removed from the intervention tentatively, but that its functionality issues would be addressed again in the future. Identifying a methodology that would offer the optimum way of assessing the effectiveness of the conversation skills training for people with NDDs through a VW was the last piece of the intervention puzzle.
Identifying the most appropriate methodology
Following in-depth analysis of possible delivery methods for the intervention, the authors settled on structured conversations over the four following phases.
Comparisons between Phases 1 and 2 would determine whether VR on itself has a positive impact on the conversation skills of people with NDDs. Phase 3 would determine whether the training instruction on conversation skills in VR is beneficial. Finally, Phase 4 (a repeat of Phase 1) would determine whether the participants can transfer and generalise any acquired skills from the virtual to the real world (Figure 1).
With regards to the mechanics of the intervention, the authors assessed different methodological approaches before concluding that a single case experimental design (SCED) would be most suitable for this project. An SCED affords the opportunity (to the researcher) to identify differences that are specific to each participant, identify methods and techniques that work but more importantly changes that can be made to an intervention in order to enhance it (Plavnick and Ferreri, 2013). SCEDs do not necessarily require a hypothesis; they rather most commonly involve an inductive approach where you “[…] experiment first and let the theories emerge inductively from the data” (Sidman, 1960, p. 14). In such an approach, relations between an independent variable and the behaviour of an individual are identified (Johnston and Pennypacker, 2008). Multiple measurements of that behaviour can lead to conclusions about the aforementioned relationship between the two variables (Neef, 2006). The SCEDs are particularly relevant to an educational study about people with disabilities (Kratochwill et al., 2010), because they allow “for causal inferences in situations where large numbers of sufficiently homogenous participants cannot be recruited (e.g. lower incidence populations)” (Plavnick and Ferreri, 2013, p. 567).
The starting point for each SCED is making many observations over an extended period of time that establishes a baseline condition (Sidman, 1960). The SCED that was decided as most appropriate to employ in this study is the multiple baseline design (MBD), which demonstrates the reliability of an intervention in altering behaviour (Baer et al., 1968; Hersen and Barlow, 1976). More specifically, MBD across subjects (participants), whose purpose is to affect the same behaviour across two or more participants taking part in the intervention in a successive manner.
Conducting the study using single case experimental design
The study would consist of three participants on the mild/moderate end of the autism spectrum. All three participants would participate in 15-16 sessions as indicated in Figure 1, which also indicates the number of sessions per week for each participant.
How often do you use a computer/tablet?
How good are your computer skills?
What do you use it mainly for?
How useful is the internet for you?
How safe do you feel when you are on the internet?
The three subjects will initially all be untreated (at baseline condition). If the measured behaviour remains at a steady baseline prior to the subjects undergoing the intervention, then experimental control is established. The first subject then undergoes the intervention while the other two will remain untreated. If the second and third subject continue to demonstrate the steady baseline condition, while there is an observable and sudden change to the first subject’s behaviour then this infers a functional relation (Cooper et al., 2007). The intervention is then applied to one subject at a time.
Based on the experimental process, the third participant would remain on baseline condition until the other two would have finished with the work in the VW. Each session would be assessed through a conversation the researcher (the lead author in this case) would have with the participants. In Phases 1 and 4, the conversation would be face-to-face; in Phase 2 the conversation would be in the VW; in Phase 3, the conversation would follow the training component and would also be in the VW (see the list “Session 8” above). The topics for Phase 1 were related to school life, likes/dislikes, subjects, teachers, hobbies, etc.; in Phases 2 and 3 the topics covered TV, films, music, travel, animals, computers, etc., so that the participants would have to face both topics they like and dislike, feel comfortable and uncomfortable with. The question sets were used in the same order for all participants. Phase 4 was an open conversation covering topics discussed in the previous sessions. The sessions would last between 10 and 30 min (depending on participant engagement and taking into account that Phase 3 has the training component).
The conversations were semi-scripted for the first three phases in the sense that each session had a specific topic and there were five questions (question sets) that the researcher would try to ask to all participants so that there was a common base of interaction for all three participants. The list below is an example of such a script (Figure 2).
The intervention that was described in this paper has been delivered to three people with ASD in their early 20 s as described in the list “Session 8” above. The sessions were videotaped with their permission. The Conversation Skills Rating Scale was adapted and used for evaluating the participants’ conversation skills. In addition, several pre/post self-reported assessments were used to measure changes in self-efficacy and anxiety before and after intervention, which include the following:
The Perceived Empathic Self-Efficacy Scale (PESE) (Caprara and Steca, 2005), a six-item self-report inventory, was used to measure the participants’ perceived ability to exercise theory of mind in understanding the feelings of others and to be both sensitive and responsive to the emotional states of others. Each item was rated on a five-point Likert-type scale, ranging from 1 (not well at all) to 5 (very well).
The Perceived Social Self-Efficacy Scale (PSSE) (Caprara and Steca, 2005), a five-item inventory, was used to measure participants’ self-perceived ability to express opinions, share personal experiences, work cooperatively and manage interpersonal conflict. Each item was rated on a five-point scale from 1 (not well at all) to 5 (very well).
The General Anxiety Disorder Questionnaire-7 (GAD-7) (Spitzer et al., 2006), a seven-item self-report measure, was used to measure anxiety. Each item was rated on a four-point Likert-type scale to assess frequency of symptoms, ranging from 0 (not at all) to 3 (nearly every day) with total scores range from 0 to 21 (10 being the threshold indicating clinical significance).
According to the preliminary results (see Table I), Participant 2 did not perceive any changes in their self-efficacy, social self-efficacy or anxiety state. Participant 3 reported lower anxiety levels, whereas Participant 1 indicated improvements in empathic self-efficacy and social self-efficacy.
This was a rather complex project that involved simultaneous development of a VW environment fit for participants on the autism spectrum (by engaging with that population in every stage of the development), development of the training material on conversation skills in different formats and development of an intervention that could be effectively assessed. Therefore, there were many “moving parts” and it was not always possible to anticipate and/or address some of the development aspects or issues that arose.
The benefits of using VR have been touched upon in this paper. Literature suggests that most serious games developed in the 2004–2014 period for communication and social skill training for people with ASD were delivered on desktop screens or tablets (Zakari et al., 2014). However, carrying out the intervention using a head mounted display, would have made the experience more immersive, by creating a feeling of presence within the environment (Bimber, 2014), which has the potential to increase users’ motivation and control of the interaction. There were financial constraints to take into account for this decision as well as the added time needed and potential complications for the ethics approval process.
The VW did not have the level of customization initially envisaged. The participants could pick among nine avatar characters (male and female; younger and older – but not one resembling themselves), they could choose their viewing perspective (eye level – top of head – or through a drone); however, they could not change the colour scheme or more importantly the training material could not be adjusted to each individual’s competence level.
The PowerPoint presentation materials were revised several times, initially by the lead researcher and then by other colleagues, so that the language of the text was simplified and mistakes were corrected. The intervention was going to be deployed in the USA (but the text was originally written in UK/Ireland English language), so the text was also checked that it made sense in context to the specific environment. Finding appropriate videos for the training sessions proved more difficult, so a couple of the videos were of British people speaking, which may be difficult to understand for some participants (although the characters’ accents would objectively be considered clear and neutral) or distracting from the task.
There may also be a need to re-examine the question set format, because it may prove difficult to seamlessly include the five scripted questions into the conversation without it appearing as being forced. Moreover, using the question sets in the same order may not provide a clear indication of whether a participant’s like or dislike of a topic impacted on their conversation performance.
Communication problems (e.g. conversation) are still seen by autism research and treatment as impairments within the autistic person (APA, 2013), which has been criticised by both autistic people (e.g. Yergeau, 2013) and non-autistic researchers and advocates (e.g. Dinishak and Akhtar, 2013). That conception of autism lends itself to misconceptions of the condition and can stigmatise people with autism. Nicolaidis’ (2012) hypothesis that an emphasis on normalising and curing autistic people leads to stigmatisation towards the condition among both autistic and non-autistic populations. Findings from a study by Gillespie-Lynch et al. (2017) support that hypothesis. People with autism were asked for their perceived importance of finding a cure for autism; a huge majority (87 per cent) said that it was not important and that the funding would be better used to develop supports for them (views in support of the neurodiversity paradigm).
The project presented in this paper had as its primary aim to create supports for autistic individuals in the form of conversation skills training, using technology (VR) that several studies, which have already been mentioned, have shown to be accepted by people with autism due to their benefits and affordances. Moreover, the development of the VW used for the delivery of the intervention was participatory in nature, meaning that people with autism were a part of each phase of its development. That kind of involvement in the research process not only empowers the autistic population, it also has the potential of improving the understanding of autism, which in turn could lead to greater acceptance and reduced stigma.
The preliminary results presented above indicate some possible positive effects from the intervention; however, its overall efficacy is inconclusive solely based on the self-reported scales.
What this paper has shown is that to better meet the training/education needs of people with NDDs delivered through technological means (e.g. VR, augmented reality, serious games, apps), there is a need to adopt an RRI approach. A closer collaboration between researchers, practitioners, technologists, designers, developers and other professionals relevant to the field of PD would help expand the state of the art of this field, raise awareness of, and knowledge about, the many roles people with NDDs can play in design and stimulate debate on ethical issues, policy and practice relating to the nature of participation.
By conducting research and innovation in a responsible way, we engage with people with NDDs at all stages of technological development for training/education purposes, which can lead to the creation of material that are accessible to a wider range of learners (even neuro-typical). The authors envisage applying PD in future research in the creation of VR environments and conducting longitudinal studies on the effectiveness of PD with people with NDDs. Longitudinal studies would also be a more dynamic approach to collection of research data, because one-off studies cannot keep up with the rapid pace of technological evolution, irrespective of how well they are designed. This could be alleviated if the technology is developed with the end user (people with NDDs).
This proof of concept can be the blueprint for the development of training for other social-communication, life, academic and vocational skills. This type of training through VR/VW not only helps the neurodiverse population employment-related skills, it also has the potential broader implementation and wider access in terms of distance learning that would benefit people living in relatively remote locations or who have mobility issues. Moreover, even though VR technology has high upfront costs (developing the training scenarios, hardware and technical support), it has relatively low maintenance costs, and thus would financially compare favourably with traditional treatments (Politis, Robb, Yakkundi, Dillenburger, Herbertson, Charlesworth and Goodman, 2017).
Future work should focus on ways to better tailor the training in the VW so that it meets everyone’s individual needs and preferences, in a cost-effective way. A PD approach was adopted for the development of the VW for the VL4ASD project, however, the same cannot be said about the training material; this needs to be addressed in a future iteration of this project.
A challenge going forward is to find ways of adapting the training for other user groups. The training material that was developed during the pilot study is for the mild end of the autism spectrum. We would need to create new material in order to cater for the more severe end of the autism spectrum and for possible co-morbid intellectual disabilities. Customising the training material to meet the needs of these users would entail, for example, the use of pictograms or other alternative means of communication in replacement of text. Generally speaking the training material – and thus the target audience – would benefit from less text, more images and better role-playing videos, that demonstrate the situations described in a real-world scenario.
When there is a significant body of research on social skills interventions through VR, that have tested their effectiveness and usefulness for autistic people with a spectrum of severity of the condition, then the next logical step would be to employ more rigorous research design and methodology.
The pre/post results from the PESE, PSSE and GAD scales
APA (2013), Diagnostic and Statistical Manual of Mental Disorders (DSM-5), 5th ed., American Psychiatric Association, Washington, DC.
Baer, D.M., Wolf, M.M. and Risley, T.R. (1968), “Some current dimensions of applied behaviour analysis”, Journal of Applied Behavior Analysis, Vol. 1 No. 1, pp. 91-7.
Bakare, M.O., Ebigbo, P.O., Agomoh, A.O., Eaton, J., Onwukwe, J.U., Onyeama, G.M., Okonkwo, K.O., Igwe, M.N., Orovighwo, A.O. and Aguocha, C.M. (2009), “Knowledge about childhood autism and opinion among healthcare workers on availability of facilities and law caring for the needs and rights of children with childhood autism and other developmental disorders in Nigeria”, BMC Pediatrics, Vol. 9 No. 12, available at: https://doi.org/10.1186/1471-2431-9-12
Baron-Cohen, S., Leslie, A.M. and Frith, U. (1985), “Does the autistic child have a ‘theory of mind?’”, Cognition, Vol. 21 No. 1, pp. 37-46.
Bellini, S. and Akullian, J. (2007), “A meta-analysis of video modeling and video self-modeling interventions of children and adolescents with autism spectrum disorders”, Exceptional Children, Vol. 73 No. 3, pp. 264-87.
Bimber, O. (2014), “Thinking virtual”, Computer, Vol. 47 No. 7, pp. 22-3.
Campbell, J.M. and Barger, B.D. (2014), “Peers’ knowledge about and attitudes towards students with autism spectrum disorders”, in Patel, V.B., Preedy, V.R. and Martin, C.R. (Eds), Comprehensive Guide to Autism, Springer, New York, NY, pp. 2471-611.
Caprara, G.V. and Steca, P. (2005), “Affective and social self-regulatory efficacy beliefs as determinants of positive thinking and happiness”, European Psychologist, Vol. 10 No. 4, pp. 275-86.
Chamak, B., Bonniau, B., Jaunay, E. and Cohen, D. (2008), “What can we learn about autism from autistic persons?”, Psychotherapy and Psychosomatics, Vol. 77, pp. 271-9, doi: 10.1159/000140086.
Cheng, Y. and Ye, J. (2010), “Exploring the social competence of students with autism spectrum conditions in a collaborative virtual learning environment: the pilot study”, Computers and Education, Vol. 54 No. 4, pp. 1068-77.
Coleman, R., Clarkson, J., Dong, H. and Cassim., J. (2012), Design for Inclusivity: A Practical Guide to Accessible, Innovative and User-Centred Design, Gower Publishing Limited, Hampshire.
Cooper, J.O., Herron, T.E. and Heward, W.L. (2007), Applied Behavior Analysis, 2nd ed., Prentice Hall, Upper Saddle River, NJ.
Dinishak, J. and Akhtar, N. (2013), “A critical examination of mindblindness as a metaphor for autism”, Child Development Perspectives, Vol. 7 No. 2, pp. 110-4.
EC (2016), Open Public Online Consultation on the Science with and for Society Work Programme 2018–2020: Description and Analysis of Results, European Commission Directorate-General for Research & Innovation, Brussels.
Fombonne, E. (2005), “The changing epidemiology of autism”, Journal of Applied Research in Intellectual Disabilities, Vol. 18 No. 4, pp. 281-94.
Freina, L. and Ott, M. (2015), “A literature review on immersive virtual reality in education: state of the art and perspectives”, eLearning & Software for Education, Vol. 1, pp. 133-41.
Gillespie-Lynch, K., Kapp, S.K., Brooks, P.J., Pickens, J. and Schwartzman, B. (2017), “Whose expertise is it? Evidence for autistic adults as critical autism experts”, Frontiers in Psychology, Vol. 8 No. 438, available at: https://psycnet.apa.org/record/2017-14941-001
Hajer, M. (2009), Authoritative Governance: Policy Making in the Age of Mediatization, Oxford University Press, Oxford.
Herrera, G., Alcantud, F., Jordan, R., Blanquer, A., Labajo, G. and De Pablo, C. (2008), “Development of symbolic play through the use of virtual reality tools in children with autistic spectrum disorders: two case studies”, Autism, Vol. 12 No. 2, pp. 143-57.
Hersen, M. and Barlow, D.H. (1976), Single Case Experimental Design: Strategies for Studying Behaviour Change, Pergamon Press, New York, NY.
Humphrey, N. and Lewis, S. (2008), “‘Make me normal’: the views and experiences of pupils on the autistic spectrum in mainstream secondary schools”, Autism, Vol. 12 No. 1, pp. 23-46.
Irwin, A. (2006), “The politics of talk: coming to terms with the ‘new’ scientific governance”, Social Studies of Science, Vol. 36 No. 2, pp. 299-320.
Jenkins, H. (2006), Fans, Bloggers and Games: Exploring Participatory Culture, New York University Press, New York, NY.
Johnston, J.M. and Pennypacker, H.S. (2008), Strategies and Tactics for Behavioral Research, 3rd ed., Erlbaum, Hillsdale, NJ.
Jones, J.L., Gallus, K.L., Viering, K.L. and Oseland, L.M. (2015), “Are you by chance on the spectrum? Adolescents with autism spectrum disorder making sense of their diagnoses”, Disability & Society, Vol. 30 No. 10, pp. 1490-504.
Kapp, S.K., Gillespie-Lynch, K., Sherman, L.E. and Hutman, T. (2013), “Deficit, difference, or both? Autism and neurodiversity”, Developmental Psychology, Vol. 49 No. 1, pp. 59-71.
Kratochwill, T.R., Hitchcock, J., Horner, R.H., Levin, J.R., Odom, S.L., Rindskopf, D.M. and Shadish, W.R. (2010), “Single-case designs technical documentation”, available at: http://ies.ed.gov/ncee/wwc/pdf/wwc_scd.pdf (accessed 10 May 2018).
Lankshear, C. and Knobel, M. (2010), DIY Media: Creating, Sharing and Learning with New Technologies, Peter Lang, New York, NY.
Laroche, G. (2011), Presentation at the Responsible Innovation Workshop, French Embassy, London, available at: www.ambafrance-uk.org/IMG/pdf/Gilles_LAROCHE.pdf (accessed 10 May 2018).
Levy, A. and Perry, A. (2011), “Outcomes in adolescents and adults with autism: a review of the literature”, Research in Autism Spectrum Disorders, Vol. 5 No. 4, pp. 1413-20.
MacNaghten, P. and Chilvers, J. (2014), “The future of science governance: publics, policies, practices”, Environment and Planning C: Government and Policy, Vol. 32 No. 3, pp. 530-48.
Mitchell, P., Parsons, S. and Leonard, A. (2007), “Using virtual environments for teaching social understanding to 6 adolescents with autistic spectrum disorders”, Journal of Autism and Developmental Disorders, Vol. 37 No. 3, pp. 589-600.
National Research Council (2001), Educating Children with Autism, National Academy Press, Washington, DC.
Neef, N.A. (2006), “Advances in single subject design”, in Reynolds, C. and Gutkin, T. (Eds), The Handbook of School Psychology, 4th ed., Wiley, New York, NY, pp. 54-64.
Nicolaidis, C. (2012), “What can physicians learn from the neurodiversity movement?”, Virtual Mentor, Vol. 14 No. 6, pp. 503-10.
Owen, R., MacNaghten, P. and Stilgoe, J. (2012), “Responsible research and innovation: from science in society to science for society, with society”, Science and Public Policy, Vol. 39 No. 6, pp. 751-60.
Parsons, S., Mitchell, P. and Leonard, A. (2004), “The use and understanding of virtual environments by adolescents with autistic spectrum disorders”, Journal of Autism and Developmental Disorders, Vol. 34 No. 4, pp. 449-66.
Pellicano, E., Dinsmore, A. and Charman, T. (2014), “What should autism research focus upon? Community views and priorities from the United Kingdom”, Autism, Vol. 18, pp. 756-70.
Plavnick, J.B. and Ferreri, S.J. (2013), “Single-case experimental designs in educational research: a methodology for causal analyses in teaching and learning”, Educational Psychology Review, Vol. 25 No. 4, pp. 549-69.
Politis, Y., Olivia, L., Olivia, T. and Sung, C. (2017), “Involving people with autism in development of virtual world for provision of skills training”, International Journal of E-learning and Distance Education, Vol. 32 No. 2, available at: www.ijede.ca/index.php/jde/article/view/1032/1683
Politis, Y., Robb, N., Yakkundi, A., Dillenburger, K., Herbertson, N., Charlesworth, B. and Goodman, L. (2017), “People with disabilities leading the design of serious games and virtual worlds”, International Journal of Serious Games, Vol. 4 No. 2, pp. 63-73.
Rivet, T.T. and Matson, J.L. (2011), “Review of gender differences in core symptomatology in autism spectrum disorders”, Research in Autism Spectrum Disorders, Vol. 5 No. 3, pp. 957-76.
Robledo, J., Donnellan, A.M. and Strandt-Conroy, K. (2012), “An exploration of sensory and movement differences from the perspective of individuals with autism”, Frontiers in Integrative Neuroscience, Vol. 6 No. 107, available at: https://doi.org/10.3389/fnint.2012.00107
Scaife, M. and Rogers, Y. (1999), “Kids as informants: telling us what we didn’t know or confirming what we knew already”, in Druin, A. (Ed.), The Design of Children’s Technology, Kaufmann, San Francisco, CA, pp. 27-50.
Sidman, M. (1960), Tactics of Scientific Research, Basic Books, New York, NY.
Spitzer, R.L., Kroenke, K., Williams, J.B.W. and Lowe, B. (2006), “A brief measure for assessing generalized anxiety disorder”, JAMA Internal Medicine, Vol. 166 No. 10, pp. 1092-7.
Stirling, A. (2006), “Analysis, participation and power: justification and closure in participatory multi-criteria analysis”, Land Use Policy, Vol. 23 No. 1, pp. 95-107.
Suzuki, M. (2011), “Mental development and autistic behavior in children with pervasive developmental disorders”, Research in Autism Spectrum Disorders, Vol. 5 No. 4, pp. 1517-25.
Tager-Flusberg, H. (2001), “A re-examination of the theory of mind hypothesis of autism”, in Burack, J., Charman, T., Yirmiya, N. and Zelazo, P.R. (Eds), Development in Autism: Perspectives from Theory and Research, Erlbaum, Hillsdale, NJ, pp. 173-93.
Warland, H. (2015), “Children with autism shouldn’t be forced to socialise”, The Conversation, July 23, available at: https://theconversation.com/children-with-autism-shouldnt-be-forced-to-socialise-44585 (accessed 12 October 2018).
Yergeau, M. (2013), “Clinically significant disturbance: on theorists who theorize theory of mind”, Disability Studies Quarterly, Vol. 33 No. 4, available at: http://dx.doi.org/10.18061/dsq.v33i4.3876
Zakari, H., Ma, M. and Simmons, D. (2014), “A review of serious games for children with autism spectrum disorders (ASD)”, in Ma, M., Oliveira, M.F. and Baalsrud Hauge, J. (Eds), Serious Games Development and Applications, Vol. 8778, SGDA, Lecture Notes in Computer Science, Springer, Dordrecht, pp. 93-106.
Benton, L. and Johnson, H. (2015), “Widening participation in technology design: a review of the involvement of children with special educational needs and disabilities”, International Journal of Child-Computer Interaction, Vol. 3 No. 4, pp. 23-40.
Frauenberger, C. (2015), “Disability and technology: a critical realist perspective”, Proceedings of the 17th International ACM SIGACCESS Conference on Computers & Accessibility, ACM, Lisbon and New York, NY, pp. 89-96.
This research was supported by funding from the charity RESPECT and the People Programme (Marie Curie Actions) of the European Union’s Seventh Framework Programme (FP7/2007-2013) under REA Grant Agreement No. PCOFUND-GA-2013-608728.
About the authors
Yurgos Politis holds a Physics Degree from the University of Athens, a Master’s Degree in Education and a PhD Degree in Physics Education from University College Dublin. He is currently ASSISTID Marie Curie Research Fellow. His research interests include social skills training with people with neurodevelopmental disabilities (NDDs) through assistive technologies (ATs), and exploring participatory design approaches with NDD populations for the development of ATs. He has co-founded the Neurodiversity in Design Network, which advocates for access to services, products and technologies for people with NDDs through participatory design. He also co-founded the Early Career Higher Education Researchers’ network.
Connie Sung is Assistant Professor of Rehabilitation Counselling and Co-director of Spartan Project SEARCH at Michigan State University. She holds a PhD Degree in Rehabilitation Psychology from the University of Wisconsin-Madison. Her research interests focus on the biopsychosocial factors associated with successful transition, psychosocial and employment outcomes, and quality of life of individuals with neurodevelopmental disabilities. She is a Mary Switzer Fellow awarded by the National Institute on Disability, Independent Living, and Rehabilitation Research and Associate Editor of the Rehabilitation Counselling Bulletin. She is also Director of Neurodevelopmental Disability and Transition Research Lab and has led multiple community-based intervention projects.
Lizbeth Goodman is Chair of Creative Technology Innovation and Full Professor of Inclusive Design at UCD and Senior Advisor to the Innovation Academy and the Leonardo Panel of the Science Gallery (for TCD). She is Founder/Director of the SMARTlab and the Multimedia and Games Innovation Centre (MAGIC), and of the industry funded INCLUDE project addressing Inclusive Design Engineering using AR/VR/AI and cloud personalisation platforms. She is Founder/Director of UCD’s Inclusive Design Research Centre of Ireland, supporting Thematic PhDs in Inclusive Design and Creative Technology Innovation as the Director of Studies, and mentoring fully funded Post-Docs through the DOCTRID ASSISTID CoFund Project (Marie Curie), as a PI and Academic Chair.
Michael Leahy is University Distinguished Professor of Rehabilitation Counselling and Director of the Office of Rehabilitation and Disability Studies. His research interests include professional competency development and education, professionalisation, regulation of practice, vocational assessment, disability and rehabilitation policy, case management practices, outcomes and evidenced-based practices in rehabilitation counselling. He is Past-President of the National Council on Rehabilitation Education, a Past Chair of the Alliance for Rehabilitation Counselling and a Past-President of the American Rehabilitation Counselling Association. He has led nearly 30 large-scale research, training and service delivery grant projects and has received over 20 national research, teaching and service awards.