PROBLEM FRAMING
How might we know whether an educational game actually works for a child with autism?
Children with Autism Spectrum Disorder face learning barriers rooted in communication and social interaction, alongside restricted and repetitive patterns of behaviour and interest. Digital games — serious games, in this context — have been proposed to reduce those barriers, and several studies report positive results.
The same literature, however, warns that these games are rarely evaluated with methodological rigour, and that the instruments used to collect data are seldom adapted to how these children actually interact. Without a usability evaluation that accounts for those particularities, there is no reliable way to tell whether a game supports learning or gets in its way. That gap is what this project addresses.
RESEARCH
Autism spectrum disorder
• Understanding unique needs: Children with ASD have distinct cognitive, emotional, and sensory profiles, so it’s essential to understand their specific learning styles and challenges.
• Tailored interaction strategies: Research provides insights into communication preferences, attention span, and behavioral patterns, guiding the design of accessible and engaging interactions.
• Avoiding overstimulation: Knowing common sensory sensitivities helps in avoiding overwhelming visual or auditory stimuli that may hinder learning.
Assistive technologies
• Enhanced accessibility: These technologies provide tools and features that help remove barriers to learning, such as alternative input methods or communication aids.
• Support for autonomy: Assistive tools empower children with ASD to engage more independently and confidently in educational activities.
• Evidence-based practices: Research helps identify which technologies have been proven effective in supporting educational development in neurodivergent learners.
Serious digital games
Serious games are games built around a purpose beyond entertainment — training, health or education. For children with ASD they are attractive because they offer predictable rules, a controlled sensory load and immediate feedback, which fit structured teaching. The risk sits on the other side: enthusiasm for the format has run ahead of evidence, and most studies do not document how the game was evaluated or with whom.
GAME DESIGN
The methodology chosen for this project is based on the TEACCH program (Treatment and Education of Autistic and Communication-Handicapped Children), which views autism not just as a disorder, but as a unique way of interacting with the world—marked by specific communication styles, behaviors, and individual strengths and challenges. TEACCH promotes “Structured Teaching,” which uses visual supports and clear routines to help individuals with autism better understand and engage with their environment.
In the game, this method was adapted into a system of progressive levels, each designed to match different stages of cognitive development — from basic motor coordination to more complex tasks like reading and math. By translating these educational strategies into interactive digital elements, the game aims to support learning in a way that is both accessible and effective for children with autism.
Initial Analysis of TEACCH Activities
Activities commonly used in the TEACCH intervention program, in real world environment, were reviewed and cataloged according to the different stages of cognitive development.
Selection criteria for game content
A selection of activities was made based on two criteria:
• Alignment with Brazil’s National Common Curriculum Base (BNCC) for early childhood and primary education (up to 5th grade).
• Feasibility of adapting them into a 2D game environment.
Structured progression by work levels
The game was organized into 5 progressive phases, each corresponding to a TEACCH work level. Tasks gradually increase in complexity to support skill development.
Work system design (Task Layout)
Visual organization followed TEACCH recommendations to support independent task completion:
• Left/top area stores movable items (stimuli).
• Right/bottom area is where tasks are executed.
Use of visual instructions and errorless learning
• Visual clarity was prioritized in all interface elements.
• A color-coded feedback system (green for correct, red for incorrect) was implemented, aligned to behavior analysis techniques such as “errorless learning.”
Feedback and accessibility features
• Pop-up windows provide visual charts indicating skill stimulation per level (motor, cognitive and academic).
• A settings menu includes a performance dashboard and options to mute sound — important for users with auditory hypersensitivity.
In the end, a flowchart with game screens and inputs was designed to guide the development team on the game navigation on the Unity engine.
VALIDATION AND TESTING
To validate the first version, four methods were combined as part of my Master's thesis:
• Heuristic evaluation of the interface, using a set of usability heuristics adapted and translated for this context.
• Group-based expert walkthrough, in which evaluators work through task scenarios and agree on a severity rating for each problem.
• Pre-test questionnaire with the children's legal guardians.
• Usability test with the children.
The study involved 20 participants — children with ASD, their legal guardians, and specialists in usability and in ASD. Data were analysed by identifying patterns and through descriptive statistics.
This evaluation was part of my Master's research; the thesis is not published yet. The game's first version appeared earlier in Educational digital game for children with autism spectrum disorder (Revista Educação Gráfica · 9th Information Design International Conference).
One thing I would change: the method in the first paper is its weak point, and I would not run it that way again. The Master's study was built to fix that, with a defined protocol for each of the four methods and 20 participants. It is the version I stand behind.