CAREER: Fostering Prosocial Behavior and Well-Being in Online Communities

Regents Of The University Of Michigan - Ann Arbor, Ann Arbor MI

Investigators

Abstract

This award is funded in part under the American Rescue Plan Act of 2021 (Public Law 117-2). Social media platforms are a common part of daily life that connect people and aid in sharing information. While society has learned how these platforms can negatively impact individuals, little is known comparatively about how our interactions on the platforms benefit us and how to encourage positive behaviors of people on these platforms. Being able to measure the potential benefits of social media can help understand its cumulative effects on people and help identify both ways individuals can change their own behavior and inform the design of social media platforms themselves. This project aims to develop computational tools to identify and measure a variety of ways in which positive behaviors such as compassion or empathy occur online. This research will study the effects of experiencing positive and negative behaviors in social media on people’s psychological well-being to create a holistic picture of social media’s impact. Furthermore, the project will identify strategies for how to phrase positive reinforcement and develop tools to support individuals in behaving positively, both of which should encourage future positive behavior in social media. The project team will also release educational material, courses, and videos for the public and students at multiple levels, as well as provide practical and informative public-facing tools that educate people about the impact of social media on well-being at a personal level. This project studies prosocial behavior in social media and quantifies the impact of interactions, content, and interventions there on specific dimensions of psychological well-being. To accomplish this goal, the project will develop datasets, models, and technologies to identify and encourage positive behaviors. The computational techniques will combine insights from natural language processing and computational social science to recognize subtle social signals in language, which will enable moving from just understanding the meaning of a text to understanding the text’s effects on the audience. The project will address three core technical challenges: (1) recognizing prosocial behavior in online spaces in ways that are sensitive to the norms of the community; (2) quantifying the impact of social media interactions and content on dimensions of psychological well-being and developing computational methods for analyzing this impact at scale; and (3) creating proactive technologies that promote prosocial behavior via behavioral nudges and tools to help people write more prosocially. 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.