Gamified Digital Phenotyping for Micro-Kinematic Cognitive Profiling: Bridging the Explainability and Synthetic Data Gaps
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
The paradigm of cognitive wellness and behavioral performance monitoring is undergoing a critical transformation, moving from episodic evaluations toward continuous, data-driven digital phenotyping. Traditional cognitive assessments provide robust psychometric validation but are primarily administered in isolated, sterile environments, leading to profound user disengagement. Furthermore, developing these tests as strict medical diagnostic tools introduces massive regulatory hurdles and limits everyday accessibility. To resolve these intersecting deficits, this research presents the engineering and validation of a fully deployed, game-focused AI solution. The proposed system is not a medical labeler; rather, it functions as a personal baseline engine built to detect micro-level cognitive drift and deviations from a user's normal state.17 The system translates standardized psychological evaluations into a suite of six high-engagement web modules. Migrated to the Unity Game Engine and exported via WebGL, the architecture leverages WebAssembly (WASM) to natively capture sub-millisecond micro-kinematic telemetry—such as cursor trajectories and task-switching latency—directly within the browser. A Scikit-Learn Random Forest Classifier then predicts cognitive states (e.g., High Cognitive Load, Psychomotor Fatigue) while calculating Feature Importance via Mean Decrease in Impurity (MDI). This native Explainable AI (XAI) mechanism isolates the specific behavioral biomarkers driving the algorithmic prediction, providing transparent insights into user wellness while establishing an empirical roadmap for domain adaptation to overcome the newly identified "Synthetic Data Gap."
Fields of Study & Subjects
Emotion and Mood Recognition
Educational Games and Gamification
Human-Automation Interaction and Safety
Publication details
- Journal / source
- International Journal For Multidisciplinary Research
- Volume
- 8
- Issue
- 2
- Publication date
- 31 March 2026
- Language
- EN