Emotionally Adaptive IoT Ecosystems for Neurodivergent Cognitive Regulation
Abstract
Neurodivergent factors that commonly have their effect on learning, behavior and social participation are sensory hypersensitivity, emotional dysregulation and cognitive overload. It is on this chapter, where the Emotionally Adaptive IoT Ecosystem is proposed and it takes a step beyond the simple identification of emotions, to environmental control, on a closed-loop fashion. It is a framework that integrates multimodal affect sensing, predictive sensory load model, adaptation based on Reinforcement Learning and privacy preserving edge computing into one architecture. The system anticipates the occurrence of overloads and helps to achieve cognitive stability via real-time regulation of the lighting, acoustic and contextual assistance. The factors that are attained through embedded governance mechanisms are transparency, consent and ethical adherence. The model presents the intelligent and human oriented model of neurocognitive support, which has been scaled and provided at homes, classes, and therapeutic locations.
