I-Corps: Machine Learning and Bio-Signal Processing for Enhancing Empathy Training

I-Corps：用于增强同理心训练的机器学习和生物信号处理

• 批准号: 2237325
• 负责人: Pattie Maes
• 金额: $ 5万
• 依托单位: Massachusetts Institute of Technology
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-09-15 至 2024-12-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2237325&HistoricalAwards=false
• 关键词: Corps; Machine; Learning; Bio; Signal

The broader impact/commercial potential of this I-Corps project is the development of learning programs and tools to develop empathetic and inclusive behaviors. The proposed technology may offer an alternative to current solutions, such as coaching (typically only accessible to top executives due to high-cost implications) and didactic training (e.g., often mandatory online tutorials or workshops, mostly disconnected from employees’ day to day needs). The goal is to provide a tool to be used during day-to-day work, rather than only in training or coaching sessions. The proposed technology may have broad applications, ranging from the early therapeutic support of children with Autism Spectrum Disorders and their families, enhancing medical staff’s ability to empathize with patients in the context of telemedicine, managers’ inclusive leadership behaviors, military leaders’ ability to empathize and resolve conflicts, and even assisting couples and families with their personal relationships. The proposed technology may be an effective, user-driven, ethical, and accessible way to generate insights into the emotional dynamics of people’s interactions, while generating useful metrics to track and steer progress.This I-Corps project is based on the development of an artificial intelligence (AI)-based bio-signal analysis system aiming to help people enhance their ability to empathize. The proposed technology analyzes a set of bio-signals from people engaging in video calls (e.g., tone of voice, facial expressions, use of language, breathing rate, etc.) with the help of machine learning and a recently developed ultrasound signal processing algorithm. The system then creates a continuous feedback loop making each interlocutor aware of their own and their conversational partner’s evolving emotions. The laboratory experiments performed with early prototypes have provided an initial proof-of-concept - participants have experienced the emergence of empathy-related and inclusive behaviors including a higher awareness of their own and other’s emotions (pausing, reflecting, inquiry, and using pro-social language), as well as a positive learner experience. These insights form the basis of actionable feedback for users to develop more inclusive behaviors. The high accuracy of evaluating emotions based on multi-modal biosignal fusion and distributed AI, and its ability to provide rich feedback in real-time with the help of custom-made biosignal processing algorithms provides opportunities for further development and commercial application.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.

这个I-Corps项目的更广泛的影响/商业潜力是开发学习计划和工具，以发展同情和包容的行为。所提出的技术可以提供当前解决方案的替代方案，例如辅导（由于高成本影响，通常只有高层管理人员才能访问）和教学培训（例如，通常是强制性的在线教程或研讨会，大多与员工的日常需求无关）。 我们的目标是提供一个在日常工作中使用的工具，而不仅仅是在培训或辅导课程中使用。 拟议的技术可能具有广泛的应用，从自闭症谱系障碍儿童及其家庭的早期治疗支持，增强医务人员在远程医疗背景下同情患者的能力，管理人员的包容性领导行为，军事领导人的同情和解决冲突的能力，甚至协助夫妇和家庭的个人关系。 这项技术可能是一种有效的、用户驱动的、符合道德规范的、可访问的方式，可以深入了解人们互动的情感动态，同时生成有用的指标来跟踪和指导进展。这个I-Corps项目是基于开发一个基于人工智能（AI）的生物信号分析系统，旨在帮助人们增强他们的同理心能力。 所提出的技术分析来自参与视频呼叫的人的一组生物信号（例如，语调、面部表情、语言使用、呼吸频率等）借助机器学习和最近开发的超声信号处理算法。然后，该系统创建一个连续的反馈循环，使每个对话者意识到他们自己和他们的对话伙伴的不断变化的情绪。使用早期原型进行的实验室实验提供了初步的概念验证-参与者经历了与移情相关的包容性行为的出现，包括对自己和他人情绪的更高认识（暂停，反思，询问和使用亲社会语言），以及积极的学习者体验。这些见解形成了用户可操作反馈的基础，以开发更具包容性的行为。基于多模态生物信号融合和分布式人工智能的情感评估的高准确性，以及在定制生物信号处理算法的帮助下实时提供丰富反馈的能力，为进一步开发和商业应用提供了机会。该奖项反映了NSF的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。