Hate Speech on Social Media Fueled By Users’ Shared Values, Moral Concerns
Working together, scientists at USC and other institutions a few years ago developed a model for detecting moralized language. It’s based on a prior, deep-learning framework for a computer program that can reliably identify text evoking moral concerns associated with different types of moral values and their opposites. The values, as defined by the Moral Foundations Theory, are focused on care/harm, fairness/cheating, loyalty/betrayal, authority/subversion and purity/degradation.
Moral Foundations Theory is a social and cultural psychology theory that explains the evolutionary origins of human moral intuitions based on innate, gut feelings rather than logical reasoning.
“Morality binds us together and gives our society structure and direction for taking care of those in need, and a vision for a just and prosperous future for the group. But morality also has a dark side, in that extreme forms of it can lead to the opposite of a lot of these positive principles,” said Morteza Dehghani, an associate professor of psychology and computer science. He leads USC Dornsife’s Computational Social Science Lab, where he and others investigate how morality intertwines with prejudice and hate.
Social media platforms help engender extremism and allow extremists to find each other and, as Dehghani describes, “feed each other’s visions of the world and anger towards the outgroup.”
Experimental Studies Further Revealed the Role of Morality in Online Extremism
In three controlled experimental studies, the research team further demonstrated that leading people to believe that others in their hypothetical or real group shared their views on moral issues increases their radical intentions to protect the group at any cost, even by resorting to violent means. When U.S. study participants were led to believe that other Americans shared their moral views, they became more willing to “fight and die” for their country and the values it stands for.
“These findings highlight the role of moral convergence and familial-like bonds in radicalization, emphasizing the need for diversity of moral worldviews within social networks,” said Atari.
But, he acknowledged, that’s easier said than done. More study is needed to determine the most effective interventions for online communities to introduce different views, which may hold the key to stopping radicalization.
#StoptheSteal Had Roots in Online Radicalization
The real-world threats posed by online radicalization were recently illustrated by the Jan. 6 storming of the U.S. Capitol. Those who were convinced the 2020 presidential election had been stolen from former President Donald Trump organized online under the hashtag #StoptheSteal on Facebook and on Gab, which served as a hub for organizing the insurrection.
These radicalization studies were already well underway before the Jan. 6 insurrection. Even so, Atari said the events of Jan. 6 further motivated the research team that had been trying to understand online radicalization.
He added that identifying as conservative or liberal does not necessarily predict who is predisposed to radicalization. “When people are motivated by morality, regardless of their political affiliation, it clouds their judgement,” Atari said.
USC notes that USC researchers in many disciplines are studying political polarization and radicalization — how it starts and how it can be mitigated.
