Developing for Autism with User-Centred Design
Rachel Menzies · 2011 · Proceedings of the 13th International ACM SIGACCESS Conference on Computers and Accessibility (ASSETS 2011) · doi:10.1145/2049536.2049624
Summary
This short paper describes the development of software designed to help children with Autism Spectrum Disorders (ASD) explore social situations, specifically the concept of sharing. The author identifies a gap in the use of User-Centred Design (UCD) methods with children with ASD, noting that while technology shows promise for social skills intervention, most existing tools focus on emotion recognition from facial expressions rather than broader social competencies like sharing. The research began with ethnographic observations in naturalistic settings at special schools, using unstructured observation to identify specific problems children with ASD face. Video recording was deliberately avoided as the participant group was unfamiliar with the observer and cameras could be intrusive and distracting. These observations revealed that teachers spent considerable time customising therapeutic efforts for individual needs and that sharing was recognised as a particularly difficult skill. Design workshops were then conducted where children designed characters, participated in drama and role-play sessions, and created fun locations to play in. The resulting designs were analysed using grounded theory, revealing four engagement themes: space, fantasy, animals, and garden.
Key findings
A key methodological finding is the distinction between involving children with ASD as "Design Informants" rather than "Design Partners." Because participants with ASD experience difficulties imagining situations and contexts outside their immediate reality, with potential for increased anxiety and social stress, the researcher takes responsibility for the design by making inferences from workshop data rather than expecting children to participate as equal design partners. The resulting software allows children aged 5-8 (developmental age) to explore sharing scenarios at their own pace. Children select a theme, encounter a sharing problem presented by a character, and manipulate objects by dragging them to appropriate locations. The system provides visual and auditory confirmation of success, and after three unsuccessful attempts, the character provides an explanation rather than a directive correction. Green-coloured buttons allow children to control the pace of interaction, a design choice reflecting the need for predictability. A planned evaluation with six participants over four weeks will use pre- and post-use video recordings in classroom and unstructured situations, along with British Picture Vocabulary Scale testing and interviews with staff and parents.
Relevance
This paper offers practical guidance for accessibility practitioners who need to involve users with cognitive or developmental disabilities in the design process. The adapted UCD methodology — using Design Informants rather than Design Partners — provides a useful model for situations where traditional participatory design methods may cause anxiety or stress for participants. The ethnographic approach of naturalistic observation without video recording demonstrates sensitivity to the needs of autistic participants. For developers building educational or therapeutic software, the design principles are instructive: giving users control over pacing, using non-confrontational feedback mechanisms, incorporating preferred themes identified through user research, and providing predictable interaction patterns. While the paper is a short doctoral consortium contribution and the evaluation was still planned at time of publication, the methodological adaptations it describes remain relevant for anyone designing technology with and for autistic users.
Tags: autism · user-centred design · social skills · inclusive design · ethnography · participatory design · educational technology