![[Photo credit: Gage Skidmore]](https://sp-ao.shortpixel.ai/client/to_webp,q_glossy,ret_img,w_750,h_375/https://www.psypost.org/wp-content/uploads/2025/08/donald-trump-750x375.jpg)
A new study published in PNAS Nexus suggests that people are more likely to believe conspiracy theories about the 2024 attempted assassination of Donald Trump if they heard them from people they know rather than from social media or news outlets. While conspiracy theories spread widely across platforms like Facebook and X (formerly Twitter), the researchers found that interpersonal networks played a stronger role in shaping belief.
The study was motivated by a growing need to better understand how conspiracy theories take hold and spread, particularly in the aftermath of high-profile, emotionally charged events. Previous research has often centered on individual psychological traits, such as a tendency toward suspicion or political extremism, to explain why some people are more prone to conspiratorial beliefs. While these factors remain important, the current study aimed to explore a less studied domain: the role of communication networks and interpersonal influence.
The authors used the July 2024 assassination attempt on Donald Trump as a test case because it quickly became a flashpoint for politically charged misinformation. As with past events like the assassination of John F. Kennedy, the suddenness and ambiguity of the Trump shooting generated intense speculation and competing narratives.
Almost immediately, people began sharing theories across the political spectrum—some suggesting Democratic operatives were behind the attack, others alleging the whole event was staged to benefit Trump politically. This combination of salience, controversy, and rapid information flow created an ideal environment to examine how beliefs about conspiracy theories form and spread.
The research team collected survey data from a nationally diverse sample of 2,765 U.S. adults between July 17 and 21, just days after the shooting took place. The survey was conducted online through a non-probability sample, with efforts made to balance demographics like age, race, gender, and geographic region. To further improve representativeness, the researchers applied post-stratification weights based on census and voting data.
Participants were first asked if they had heard about the Trump assassination attempt. Those who were aware of the incident were then asked where they had received their information—options included television, radio, newspapers, social media, news websites, podcasts, or from people they know. Respondents could select multiple sources. The researchers used this information to examine how different channels of communication were linked to awareness and belief in two specific conspiracy theories: one claiming that Democratic operatives orchestrated the shooting, and another alleging that the event was staged altogether.
To assess belief in the theories, participants who had heard them were asked how likely they thought each was to be true. Responses ranged from “very unlikely” to “very likely” on a five-point scale. The researchers also collected demographic data, political orientation, approval of Trump, general interest in politics, and a standard measure of conspiratorial thinking known as the American Conspiracy Thinking Scale.
Statistical analyses included logistic regression to determine predictors of exposure to conspiracy theories and linear regression to assess what factors were associated with belief.
Nearly all respondents (95%) reported being aware of the assassination attempt. The majority of those who were informed said they got their information from television (64%), followed by social media (43%), and personal contacts (30%).
Of those surveyed, 41% had heard the theory that Democratic operatives were behind the attack. Among this group, 53% had seen the claim on social media, 28% had seen it on television, and 32% had heard it from people they knew. Roughly 29% of those exposed to this theory believed it was likely to be true.
The second theory—that the event was staged—was even more widely circulated, with 53% of participants reporting they had encountered it. Of these, 52% saw it on social media, 34% heard it from personal contacts, and 21% saw it on television. About 29% of those who had heard this theory said they believed it was likely.
Social media appeared to be the main vector for initial exposure to both theories. However, when the researchers examined what influenced actual belief, a different picture emerged. People who heard the conspiracy theories through interpersonal networks were significantly more likely to believe them. This pattern held across both left-leaning and right-leaning narratives.
In contrast, social media use was not strongly linked to belief in the theories once exposure was accounted for. While it increased the likelihood of encountering conspiracy content, it did not appear to increase the chance that people would believe it. This finding runs counter to the often-repeated assumption that social media is the primary engine of conspiracy belief.
Other factors associated with belief included approval of Trump, political partisanship, and a higher score on the conspiratorial thinking scale. These variables tended to predict belief in expected ways. For instance, Republicans and Trump supporters were more likely to believe the Democratic operative theory, while Democrats were somewhat more open to the idea that the event was staged.
Notably, among all the information sources analyzed, interpersonal communication was the only one consistently and positively associated with belief in both conspiracy theories. Hearing about a conspiracy from someone personally known increased the perceived likelihood that the theory was true by 0.2 to 0.4 points on a 0–4 scale.
These findings indicate that conspiratorial thinking is not just a function of individual psychology or online media exposure, but also a social process embedded in everyday relationships. However, there are some limitations to consider. The survey used a non-probability sample, which may not fully represent the U.S. population despite statistical adjustments.
In addition, the study was observational, so it cannot determine causation. It remains unclear whether people adopt conspiracy beliefs because of their social networks or whether they seek out relationships with people who already share those beliefs. Longitudinal or experimental research would be needed to untangle these possibilities.
Future research could explore how strong or confident people are in their beliefs, not just whether they accept a theory. Another promising direction would be to examine the structure of conspiratorial social networks: are there particular patterns of relationship or communication styles that make belief more likely to take hold? Understanding these dynamics could help in designing more effective interventions aimed at reducing the social spread of misinformation.
The researchers also suggest that future studies should consider how interpersonal influence interacts with other media, such as algorithm-driven news feeds, and how conversations about conspiracy theories evolve over time.
The study, “Information from social ties predicts conspiracy beliefs: Evidence from the attempted assassination of Donald Trump,” was authored by Katherine Ognyanova, James N. Druckman, Jonathan Schulman, Matthew A. Baum, Roy H. Perlis, and David Lazer.
