GeniAuti: Tracking Challenging Behaviors of Children with Autism for Data-Driven Interventions
Kiwon Choi, Dongho Jang, Dasol Lee, Seoyoung Park · 2018 · Proceedings of the 20th International ACM SIGACCESS Conference on Computers and Accessibility (ASSETS '18) · doi:10.1145/3234695.3240987
Summary
This student research competition paper presents GeniAuti, a mobile application designed to help caregivers of children with autism track challenging behaviors using the ABC (Antecedent-Behavior-Consequence) framework. Challenging behaviors in autism — classified as self-injury, assault, destructive, sexual, and escape behaviors — are complex, vary between children, and are difficult for caregivers to record in the moment because they must simultaneously manage the behavior itself. The researchers conducted formative interviews with three clinical experts (two psychiatrists and a therapist) to understand caregiver difficulties, establishing three design goals: recording should require minimal hand use, data should be visualized as charts for expert review, and caregivers should be able to search reference cases of other children's behaviors. GeniAuti addresses these needs through voice recognition for hands-free ABC recording, data visualization showing behavior frequency, duration, and intensity over time, and a reference database of 18 cases of children with challenging behaviors that caregivers can search by behavior type or function.
Key findings
The app's workflow is designed around the reality that caregivers cannot type while managing a challenging behavior. When a behavior begins, the caregiver taps a timer. When the child calms down, they stop the timer and use voice recording to describe the ABC context. They then categorize the behavior by type (six categories including self-injury, assault, destructive, sexual, escape) and perceived function (attention-seeking, task avoidance, self-stimulatory, tangible-maintained), and rate intensity on a 1-5 scale. The visualization feature shows behavior data on daily, weekly, monthly, and yearly views, displaying total count, average duration, and maximum intensity — information that helps clinicians identify patterns and design personalized interventions. The reference feature provides 18 real cases of children's challenging behaviors searchable by type and function, helping caregivers find relevant coping strategies from similar situations. Initial feedback from one caregiver found voice notes convenient but indicated that the clinical terminology used in the app was difficult to understand, highlighting the need for plain language in caregiver-facing tools.
Relevance
GeniAuti addresses a practical gap in autism care: the need for systematic, low-burden behavior tracking that connects caregivers with clinical experts. The ABC framework is well-established in behavioral psychology but is typically recorded on paper forms that are difficult to complete during or immediately after a challenging behavior episode. By using voice recognition and automated timestamping, the app reduces recording friction at the most critical and stressful moments. The visualization component transforms raw behavioral data into actionable insights that can improve the quality of caregiver-clinician communication. The reference case feature acknowledges that caregivers often feel isolated and need practical examples of how others have managed similar behaviors. The early finding that clinical terminology was confusing to caregivers is an important design lesson — accessibility tools must be linguistically accessible to their intended users, not just technically functional.
Tags: autism · challenging behavior · mobile application · data visualization · voice recognition · caregiving · behavior tracking