Social media platforms witnessed a surge in intense political debate following the 2020 United States presidential election. New research indicates that a specific psychological trait known as collective narcissism played a primary role in fueling the “Stop the Steal” movement on Twitter. The study finds that messages expressing an exaggerated sense of group importance combined with victimhood were more likely to go viral. These findings regarding online political behavior appeared in the journal New Media & Society.
Social psychologists describe collective narcissism as a belief system where individuals view their own group as exceptional. This belief is not merely about pride. It comes with a deep conviction that the group is not receiving the recognition or privilege it deserves from others. When the group faces a perceived threat, such as an election loss, this psychological trait can drive intense hostility toward outsiders.
Liwei Shen, a researcher at the University of Wisconsin-Madison, led the investigation into how this dynamic functions online. The research team included Yibing Sun, Luhang Sun, Yun-Shiuan Chuang, and Kaiping Chen. They sought to understand how feelings of group superiority shape digital discourse. They specifically wanted to see if these sentiments helped spread narratives about voter fraud.
The researchers based their work on a framework called the Social Identity Model of Deindividuation Effects. This psychological theory suggests that anonymous online environments change how people relate to their social groups. When individuals lose their personal visibility online, they often identify more strongly with their group. They also feel a stronger pressure to conform to that group’s norms and language.
The team collected a massive dataset from Twitter to test their ideas. They gathered nearly 12,000 original posts and over 23,000 replies related to the “Stop the Steal” movement. The data covered the critical period from January 4 to January 7, 2021. This timeframe included the day of the Capitol attack, marking the peak of the movement’s activity.
Analyzing such a large volume of text required advanced computational methods. The authors utilized a form of artificial intelligence called the BERT model to read the tweets. This model was trained to classify posts based on whether they expressed collective narcissism. The researchers defined this expression as a combination of collective identity and a grandiose view of the in-group that is under threat.
The team also employed a technique called dependency parsing. This method breaks down the grammatical structure of sentences to find the subjects and objects. It allowed the researchers to see exactly which groups were being talked about. They wanted to know who the users viewed as the heroes and villains of their narrative.
The investigation revealed that national identity was the central theme in these expressions. Tweets characterized by collective narcissism frequently used the pronoun “we.” This usage created a sharp linguistic boundary between the in-group and everyone else. Terms like “America,” “Americans,” and “country” appeared often in these posts.
Liwei Shen and the team found that users did not just express support for a candidate. They framed the political contest as a battle for the nation’s existence. The language used often equated loyalty to Donald Trump with patriotism itself. Conversely, any opposition to the movement was framed as a betrayal of the country.
The researchers identified seven recurring themes in the posts labeled as collectively narcissistic. One major theme was “heightened patriotism,” where users claimed the title of “true patriots.” Another common theme involved religious language. Users frequently invoked God to suggest their political stance had divine approval.
A particularly aggressive theme focused on “internal traitors.” The analysis showed that users often directed hostility toward members of their own political party. They accused Republicans who did not support the movement of betraying the group. This policing of the in-group is a hallmark of collective narcissism.
The study also highlighted a focus on external enemies. Many posts contained unsubstantiated claims about foreign interference. Users frequently mentioned China or communism as existential threats to the United States. This focus helped reinforce the group’s sense of being a victim of a grand conspiracy.
One of the primary goals of the study was to see if this type of language increased user engagement. The results from statistical models were clear. Tweets that contained expressions of collective narcissism received substantially more attention than those that did not. They garnered higher numbers of likes, retweets, and quotes.
The researchers found that this rhetoric was highly contagious. When an original post used collective narcissistic language, the replies were likely to use it as well. The online environment appeared to encourage users to mimic the intense, identity-based language of the original poster. This created a cycle where users reinforced each other’s grievances.
This finding supports the idea that social media acts as an echo chamber for group-based emotions. The platform’s design allows users to find validation for their grandiose group image. When they see others expressing similar outrage and superiority, they feel emboldened to do the same. This leads to a rapid diffusion of these sentiments across the network.
The study also looked at the profiles of the users who posted this content. The team used a method called structural topic modeling to analyze user bios. They found that collective narcissistic expressions came from specific subsets of supporters. The most prominent group consisted of users who explicitly identified as Trump supporters.
Another group frequently using this language included conservatives who emphasized faith and country in their profiles. A third group consisted of female users who identified as patriots and Christians. The data showed that not all conservative profiles engaged in this rhetoric. Profiles that focused on family or general religious faith without the nationalist element were less likely to post collective narcissistic content.
These distinctions matter because they show that collective narcissism is not universal among all supporters of a movement. It is driven by specific sub-identities that view the nation and their group as synonymous. The “Stop the Steal” movement successfully tapped into these specific identities. It provided a narrative that validated their feelings of being exceptional but under attack.
The authors noted that the design of social media platforms facilitates this behavior. Platforms like Twitter allow for rapid, widespread broadcasting of information. They also enable users to easily find and connect with like-minded individuals. This structure is ideal for cultivating collective narcissism.
There are limitations to this research that the authors acknowledged. The study focused on a single platform during a unique historical event. Twitter has specific features that might shape discourse differently than Facebook or Reddit. The dynamics of the “Stop the Steal” movement might also be distinct from other political movements.
The researchers could not determine if these patterns hold true for different political orientations. Future research would need to examine if left-leaning movements exhibit similar patterns of collective narcissism. Comparing data across multiple social media platforms would also provide a more complete picture. It remains to be seen how different content moderation policies might affect the spread of such discourse.
Additionally, the study analyzed text but did not measure behavioral outcomes. It is unclear if engaging in this online discourse leads to real-world actions. Future studies could investigate whether online collective narcissism fosters stronger in-group solidarity offline. Understanding the link between digital words and physical actions remains a priority for social scientists.
The study, “How collective narcissism became contagious in public conversations of ‘Stop the Steal’ on Twitter,” was authored by Liwei Shen, Yibing Sun, Luhang Sun, Yun-Shiuan Chuang, and Kaiping Chen.
