From

Time -

Interactions with Human, Thesis

This thesis focuses on the design and use of Extended Reality (XR) environments for supporting autistic children regarding sensory perception.

Speaker : Valentin Bauer

The jury of this thesis is composed of:

  • Sarah Parsons (Rapporteure) – Professor, University of Southampton
  • Giuseppe Riva (Rapporteur) – Full Professor, Università Cattolica; Head Researcher, Istituto Auxologico Italian
  • Ouriel Grynszpan (Examinateur) – Professor, Université Paris-Saclay
  • Daniel Mestre (Examinateur) – CNRS Research Director, Université Aix-Marseille
  • Stefania Serafin (Examinatrice) – Professor, Aalborg University Copenhagen
  • Isabelle Viaud-Delmon (Examinatrice) – CNRS Research Director
  • Patrick Bourdot (Directeur de thèse) – CNRS Research Director, Université Paris-Saclay
  • Tifanie Bouchara (Co-encadrante de thèse) – Associate Professor, Université Paris-Saclay

This thesis focuses on the design and use of Extended Reality (XR) environments for supporting autistic children regarding sensory perception.

It is common today to refer to Extended Reality (XR) as all technologies ranging from Augmented Reality (AR) to Virtual Reality (VR). While autism XR research is promising to extend practitioners’ interventions, it mainly addresses the socio-emotional abilities of autistic children with mild learning disabilities. Yet, common interventions address the entire spectrum by targeting a range of abilities, including sensory perception. Based on these observations, I conducted 34 interviews with autism stakeholders to compare their practical needs with the literature XR uses and designs. Findings confirmed the presence of a research gap, provided a set of XR guidelines, and called for further exploration of XR sensory and mediation approaches. Furthermore, to better consider autistic children with severe learning disabilities and more complex needs, AR seemed more suited than VR, as allowing children to keep contact with the real environment and their usual practitioner. To examine this objective, two research endeavors were conducted within clinical settings in collaboration with practitioners.

In the first part of this thesis, I investigated the possibility of using sensory and mediation XR approaches to support reassurance and to reinforce the child-practitioner relationship. To that end, an AR application called Magic Bubbles was created to complement usual sensory interventions, such as Snoezelen and Sensory Integration Therapy, which are sometimes limited in terms of flexibility or access. A user-centered design process was conducted for a day hospital setting with two psychologists and one psychiatrist, and validated by a clinical team of eleven practitioners. After acceptability and usability testing with ten children with neurodevelopmental conditions, a long-term field study with seven autistic children confirmed its potential for reassurance and social interaction. Moreover, using a grounded theory approach, I built a categorization of children’s experiences, which could inform future autism XR research. To be able to use Magic Bubbles with a larger number of autistic children, the application then had to be further individualized based on an ecological assessment of sensory stimuli inducing negative or positive experiences among these children.

In the second part of this thesis, I thus focused on using AR to assess the Atypical Auditory Functioning (AAF) of autistic children, in collaboration with a psychomotor therapist. Indeed, although AAF largely impacts the everyday life of these children, current sensory profiles and auditory exams prevent practitioners from testing various auditory stimuli in ecological settings. This makes it hard to differentiate between emotional and physical reactions towards sounds, and thus to develop appropriate sensory strategies. To address this issue, I started by developing a sound taxonomy being representative of autistic AAF through a systematic literature review. The items of the taxonomy were then validated, ranked, and enhanced, through an online questionnaire answered by 68 stakeholders. At last, an AR application and experimental protocol were designed to perform such assessments. Following a participative process involving autistic individuals and practitioners, first assessments of the AAF of autistic children are planned at the therapist’s office in the Ile de France area.

The thesis concludes by drawing research perspectives based on the main research findings.