SoCS: Achieving the Interpersonal Function of Affect in Human-Machine Collaboration

SoCS：实现人机协作中的情感人际功能

• 批准号: 1211064
• 负责人: Jonathan Gratch
• 金额: $ 70万
• 依托单位: University of Southern California
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2012
• 资助国家: 美国
• 起止时间: 2012-08-01 至 2017-01-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1211064&HistoricalAwards=false
• 关键词: SoCS; Achieving; Interpersonal; Function; Affect

Displays of anger or joy can profoundly shape the behavior of those around us. This project brings a functionalist perspective to the study of such influences on human decision-making. Rather than treating affective displays as mere manifestation of internal experience, they can be viewed as strategic other-directed signals that facilitate social goals, for example by conveying specific internal beliefs, desires and future intentions. The project explores the theory of "reverse appraisal" as a framework for explaining and computationally modeling this social process. Specifically, this theory will be used to unpack the mechanisms by which affective displays influence outcomes in social dilemmas and negotiations. This functional analysis also informs the design of human-computer and computer-mediated collaboration systems. The project uses a framework based on "virtual confederates" - i.e., anthropomorphic computer characters - to study social affect. It builds on reverse appraisal's insight that what matters is the information (function) not the display in itself (form) and studies how the social functions of affect apply to systems that extend beyond virtual humans.Broader Impact. The project promotes synergies between social and computational sciences: it contributes to understanding the psychology of how affective expressions impact people's decisions in social, economic and daily life; and, it advances the state of human-computer systems for collaboration in practical domains such as negotiation, and has implications for designing collaborative computer-mediated systems. The research results in new tools -- virtual confederates -- or studying social processes, which will be offered freely to the scientific community. Finally, the project supports research and mentoring activities for graduate students and postdoctoral scholars towards a multidisciplinary specialization in affect and decision making.

愤怒或喜悦的表现可以深刻地塑造我们周围人的行为。这个项目带来了一个功能主义的角度来研究人类决策的影响。与其将情感展示视为仅仅是内部体验的表现，不如将其视为促进社会目标的战略性他人导向信号，例如通过传达特定的内部信念，欲望和未来意图。该项目探讨了“反向评估”理论，作为解释和计算建模这一社会过程的框架。具体来说，这一理论将被用来解开情感展示影响社会困境和谈判结果的机制。这种功能分析也通知人机和计算机介导的协作系统的设计。该项目使用了一个基于“虚拟联盟”的框架，拟人化的计算机角色-研究社会影响。它建立在反向评估的洞察力之上，即重要的是信息（功能）而不是显示本身（形式），并研究了情感的社会功能如何应用于超越虚拟人类的系统。该项目促进了社会科学和计算科学之间的协同作用：它有助于理解情感表达如何影响人们在社会，经济和日常生活中的决定的心理学;并且，它推进了人机系统在谈判等实际领域的协作状态，并对设计协作计算机介导系统产生了影响。这项研究产生了新的工具-虚拟联盟-或研究社会进程，将免费提供给科学界。最后，该项目支持研究生和博士后学者的研究和指导活动，以实现影响和决策的多学科专业化。