SCH: A Data-driven Classroom Intervention Framework for Children with Social Peer Engagement Deficits

患有自闭症和其他发育障碍的人在建立和维持社会同伴交往方面表现出明显的困难。这些困难导致友谊有限，社会排斥和孤立率高，焦虑、抑郁和自杀念头的风险增加。与同龄人相比，自闭症和发育障碍患者的就业率、高等教育参与率和独立生活率也较低。尽管自闭症和发育障碍儿童面临社会挑战，但他们渴望参与社会活动，但需要支持。早期社会技能干预，如近距离训练，可以帮助这些儿童提高社会参与度和心理健康。该项目开发了一种辅助技术，以帮助教师在课堂环境中进行这种纠正性社会技能干预。该项目开发人类感知和人工智能处理基础设施，用于儿童同伴参与行为监测。教师使用感知行为和人工智能推断的情绪信息，对被诊断患有自闭症和其他相关行为障碍的儿童进行有针对性的行为干预。该项目的一项关键技术是可穿戴式人类互动追踪器（WHET）标签，它可以测量个人之间的成对互动。当两个人，每个人都戴着名字标签大小的标签，互动时，他们的互动/参与动态被跟踪和定量捕获，由WHET传感器随后无线上传到连接到云存储的接入点。这种教师辅助技术有助于提高干预的质量，以及教师可以同时有效地为儿童提供干预的数量。新开发的技术和算法还将改善对患有抑郁症、焦虑症和多动症等许多其他疾病的儿童的支持。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Individuals with autism and other developmental disabilities show marked difficulty establishing and maintaining social peer engagements. These difficulties result in limited friendships, high rates of social rejection and isolation, and increased risk of anxiety, depression, and suicidal ideation. Individuals with autism and developmental disabilities also experience lower rates of employment, post-secondary participation, and independent living in comparison to their peers. Despite their social challenges, children with autism and developmental disabilities desire to engage socially but require support. Early social skills interventions such as proximity training can help such children improve social engagement and mental health. This project develops an assistive technology for aiding such corrective social skills interventions by teachers in classroom settings.This project develops human sensing and AI-enabled processing infrastructure for children’s peer-engagement behavior monitoring. Teachers use the sensed behavior and AI-inferred mood information to administer targeted behavioral interventions to children diagnosed with autism and other related behavioral disorders. A key technology enabler in this project is a Wearable Human Interaction Tracker (WHET) tag, which measures pair-wise human interaction between individuals. When two people, each wearing the nametag-sized tag, interact, their interaction/engagement dynamics are tracked and quantitatively captured by the WHET sensors for subsequent wireless upload to an access point connected to cloud storage. This assistive technology for teachers helps improve the quality of intervention, as well as the number of children that a teacher can simultaneously and effectively provide interventions to. The newly-developed technology and algorithms will also improve support for children with many other disorders including depression, anxiety, and ADHD.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.