JMIR Publications

ABSTRACT

Background:

Attention Deficit Hyperactivity Disorder (ADHD) is a common neurodevelopmental disorder in children, characterized by inattention, hyperactivity, and impulsivity that significantly disrupt daily functioning. Traditional diagnostic methods are often lengthy and resource intensive. Recent research suggests that serious games can improve engagement and provide valuable data for ADHD assessment and therapy.

Objective:

This study aimed to develop and validate Attention Robots, a serious game designed to assess attention levels in children and adolescents. The goal was to compare its effectiveness with the widely used D2-R Test of Attention, while also establishing a foundation for future use of biometric data in developing Deep Learning models for ADHD detection.

Methods:

Fifty-eight participants aged 6–18, including children diagnosed with ADHD and neurotypical controls, completed two sessions: playing Attention Robots while their in-game performance and biometric data (EEG, EDA, eye-tracking) were recorded, and performing the D2-R Test of Attention. Correlation analyses using Spearman and Pearson coefficients compared key attention indicators across both tests, including working speed, omissions, commissions, concentration, and precision.

Results:

Statistical analyses showed strong positive correlations between Attention Robots and the D2 Test in key metrics: Spearman: working speed (ρ=0.905, p<.001), omissions (ρ=0.333, p=.011), commissions (ρ=0.496, p<.001), concentration (ρ=0.926, p<.001), and precision (ρ=0.607, p<.001). and Pearson: working speed (ρ=0.911, p<.001), omissions (ρ=0.278, p=.0035), commissions (ρ=0.396, p=.0002), concentration (ρ=0.922, p<.001), and precision (ρ=0.574, p<.001) These findings suggest that Attention Robots is a valid tool for measuring attention performance, with the added advantage of providing a more engaging assessment environment and collecting rich biometric data for future AI model development.

Conclusions:

Attention Robots offers an innovative and effective approach to assessing attention in children and adolescents. By combining serious gaming with biometric data collection, it opens new avenues for ADHD diagnosis and therapy. Future work will focus on extending analysis by participant subgroups, integrating additional biometric signals, and developing Deep Learning models for personalized ADHD detection and monitoring. Clinical Trial: Hospital Ethics Committee of the Alicante Institute for Health and Biomedical Research (ISABIAL; PI2022-139) and the Ethics Committee of the University of Alicante (UA-2023-05-17_1)