Towards Understanding the Effect of Serious Games on Attention, Adherence, and Behavior for Children with ADHD
Jonathan Wang Liu, Mahmood Jasim, Jeong-Heon Song, So-Hwi Ha, Jun-Su Kim, Dayoung Kim, Hyunsuk Lee, Byeong Il Kim, Hee-Tae Jung · 2026 · ACM Transactions on Computer-Human Interaction · doi:10.1145/3772068
Summary
This paper investigates how serious games designed to help children with ADHD impact their attention, adherence, and daily behaviour in real-world home settings. Conducted in two stages, the study first involved nine educators working with children in clinical ADHD settings, then deployed a suite of five mobile games — collectively called NeuroWorld DTx v1.0 — to seven child-caregiver dyads for one month in their homes. NeuroWorld includes four attention-focused games using matching and evasion tasks and one working memory game, each featuring 20 stages, adaptive difficulty, and audiovisual feedback, with sessions designed to last approximately 25 minutes daily. The preliminary educator study found that while NeuroWorld was seen as more engaging than traditional paper-and-pencil attention interventions, educators were concerned about children's ability to self-administer the games without supervision, anticipating distraction and disengagement. The one-month home deployment directly tested these assumptions. Outcome measures included the Child Behavior Checklist (CBCL), ADHD Rating Scale (ADHD-RS), and Comprehensive Attention Test (CAT), alongside game usage data (play time, success rate) and post-deployment semi-structured interviews with both children and parents. All seven children maintained consistent engagement and adherence throughout the month, playing every session independently at home. The research is grounded in HCI methodology, prioritising end-user experience and qualitative findings alongside clinical outcome measures, and draws on self-determination theory and self-efficacy research to contextualise its results.
Key findings
Contrary to educators' expectations, all seven children self-administered NeuroWorld consistently and independently throughout the month, exceeding the 50% compliance threshold used in EndeavorRx clinical trials. Every child played at least 240 seconds (4 minutes) per session, with average success rates above 65% across all 20 sessions. CBCL and ADHD-RS scores showed a positive trend: 13 of 14 assessments across both scales showed reductions, averaging −14 points on the CBCL and −8 points on the ADHD-RS. Four of seven participants met the Minimum Clinically Important Difference (MCID) thresholds for both scales. Cohen's d effect sizes were large (−1.49 for ADHD-RS; −1.48 for CBCL), though changes narrowly missed statistical significance (p=0.0526 and p=0.0536) due to the small sample size. The Comprehensive Attention Test (CAT) showed mixed and inconsistent results with no statistically significant improvements, revealing a meaningful discrepancy between behavioural self-report measures (positive trend) and the computerised attention test (no clear pattern). Qualitative interview analysis identified five major themes: the tension between "serious" and "game" framing, improvement as motivation, the impact of reinforcement strategies, physician-driven participation rationale among parents, and strong self-efficacy and goal-setting behaviour demonstrated by children independently at home.
Relevance
This paper is directly relevant to practitioners and researchers working at the intersection of accessibility, HCI, and neurodevelopmental disabilities. It challenges the assumption that children with ADHD require external supervision to maintain engagement with digital therapeutic tools, with implications for designing home-based accessibility and assistive technology interventions more broadly. The finding that self-report behavioural measures diverged from standardised computerised attention tests highlights a critical measurement challenge relevant to any population using technology-mediated interventions: objective performance metrics may not capture real-world daily functioning improvements. For accessibility practitioners, the paper underscores the value of designing for self-efficacy and autonomy rather than deficit remediation. Limitations include a small sample (n=7), no control group, and a South Korean cultural context. Future work should pursue larger randomised trials and explore AI-enabled adaptive difficulty systems.
Tags: ADHD · serious games · digital therapeutics · children · attention · cognitive training · home-based intervention · self-efficacy · adaptive difficulty · caregiver involvement · HCI · neurodevelopmental disabilities