CAREER: Facilitating Intergroup Communication and Cooperation

该项目促进了促进不同群体之间交流与合作的理论理解和技术发展。有效的群体间沟通为社会带来了许多好处，包括更强的公民参与、更好的合作和整体资源的增加。然而，群体在集体行动中面临着相互信任的挑战；众所周知，群体内与群体外效应会导致负面的刻板印象和对其他群体的排斥，从而减少个人和社会层面上有益的沟通与合作的机会。该项目旨在利用对沟通行为的新兴理解，以及来自网络科学、经济学和自然语言处理的技术，增加人们成功弥合这些差异并形成更强合作和公民关系的机会。为实现其目标，该项目将开发群体间沟通和社交网络合作影响的模型，为改善在线环境中的沟通与合作的干预措施提供信息。研究人员将首先开发一个实验系统，以检查在人类网络中存在群体身份的情况下合作的通信和连接模式。然后，该团队将迭代地设计算法，使用自然语言处理和会话人工智能技术，以及网络科学和用户建模方法，来建议对话伙伴、话题和风格，从而帮助人们更有效地建立跨群体的联系。这些将通过实验室和实地研究相结合的方式进行测试，以衡量具有强大因果关系和外部有效性的已开发干预措施的有效性。总之，该项目将增进对技术如何帮助像我们这样的多元化社会更有效地运作的理解。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。

This project promotes theoretical both understanding and technology development on the topic of facilitating communication and cooperation between distinct groups. Effective intergroup communication provides a number of benefits to society, including stronger civic engagement, better cooperation, and increased resources for the whole. However, groups face challenges trusting each other in collective action; in-group versus out-group effects are known to lead to negative stereotyping and exclusion of other groups, reducing the chances of beneficial communication and cooperation at both the individual and the social level. This project aims to leverage an emerging understanding of communication behavior, along with techniques from network science, economics, and natural language processing, to increase the chance that people successfully bridge these differences and form stronger cooperative and civic ties.To achieve its goals, the project will develop models of intergroup communication and the influence of cooperation in social networks to inform interventions that improve communication and cooperation in online contexts. The investigators will first develop an experimental system to examine communication and connection patterns for cooperation under circumstances where there is group identity in human networks. Then, the team will iteratively design algorithms that use natural language processing and conversational AI techniques, along with network science and user modeling methods, to suggest conversation partners, topics, and styles that may help people make connections across groups more effectively. These will be tested with a combination of laboratory and field studies to measure the effectiveness of the developed interventions with robust causality and external validity. Together, the project will improve understanding of how technology might help a diverse society such as ours operate more effectively.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.