On January 6, 2021, the public watched in disbelief as the Capitol building was stormed by hundreds of protestors. Most spectators at home didn't know violence at the Capitol building was already circulating through far-right social media channels for months.
Social media, for better or worse, play a large role in how we consume information – as well as spreading misinformation and conspiratorial propaganda.
Researchers at the University of Pittsburgh received $100,000 as part of the Meta Foundational Integrity Research Award to create a population-level sampling and modeling framework to trace and measure the influence of online conspiratorial content.
“One of the key challenges in stopping the influence of cyber conspiratorial propaganda is the lack of a coherent and generalizable framework to understand the relationship between individuals’ characteristics and spread of this propaganda,” said Amin Rahimian, an assistant professor of industrial engineering at the University of Pittsburgh Swanson School of Engineering and co-investigator on the project.
The project, which is led by Associate Professor at Pitt’s School of Computing and Information Yu-Ru Lin, will pilot a survey to collect representative population and network characteristics across the United States and employ mechanism discovery methods to understand the relationship between conspiratorial propaganda on social media and its users.
Researchers are mostly interested in conspiratorial cascades – the phenomenon in which a large portion of people interact with the same conspiratorial content. Using their sampling methods, they want to determine who is likely to be a believer in or spreader of conspiratorial content and under what conditions do these individuals create a large cascade online.
“Ultimately, this is a social cyber security concern,” Rahimian said. “We’re looking to stop violence at the source.”
Lin added that this study is about understanding social fabrics.
“We need to look at which communities are more vulnerable to the spread of conspiratorial messages and how we can best safeguard those communities,” she explained.
Outcomes of this study, “A Multi-Resolution, Population-Scale Framework to Identify Sociodemogrpahic and Psychometric Factors for Network Influence of Online Conspiratorial Content,” will inform the design of large-scale, population statistics to examine other information integrity issues both nationally and globally and create mathematical tools to discover the causal and social structures of such diffusions.