Who would Iron Man vote for? (Miguel Lagoa/Shutterstock)
In a nutshell
- People tend to project their own political views onto figures they admire (heroes) and assign opposing views to those they dislike (villains), even when these figures are fictional.
- The stronger someone identifies with their political party, the more likely they are to engage in this projection, with individuals more inclined to “counter-project” opposing views onto villains rather than project their own views onto heroes.
- This mental habit of associating good with “our side” and bad with “the other side” contributes to increasing political polarization, making it harder to find common ground and understand those with opposing political beliefs.
SOUTHAMPTON, England — In the heated battle between political parties, there’s a psychological trick our brains play that might explain why political divisions feel impossible to overcome. We automatically assume good people share our political views and bad people oppose them. New international research reveals this happens even with fictional characters like Iron Man and Thanos.
Research published in Political Science Research and Methods revealed what researchers call “motivated political projection.” This means we tend to think people we like share our political views, while those we dislike must belong to the opposing political camp. This mental habit might help explain why political divisions feel so hard to bridge today.
Heroes Vote Like Me, Villains Don’t
To investigate this phenomenon, the research team designed two experiments with participants from the United States and Britain. For their first study, which involved 3,200 participants, they showed people images of fictional characters from popular franchises like Marvel, Disney, Harry Potter, and Star Wars and asked them to guess these characters’ political affiliations.
Regardless of a character’s other traits (gender, age, background), participants consistently projected their own partisan identities onto the heroes while assuming villains supported the opposing party. For example, a Democratic voter would be much more likely to think Iron Man votes Democrat while imagining Thanos supports Republicans. Conservative participants showed the same pattern in reverse.
This pattern appeared in both American and British participants, suggesting this mental habit crosses political systems. The effect was stronger among those with stronger partisan identities. The more strongly someone identified with their political party, the more likely they were to make these assumptions.
“Counter-projection,” assigning opposing political views to villains, was actually stronger than the projection of one’s own views onto heroes. In other words, people were more driven to distance themselves politically from characters they disliked than to claim kinship with characters they admired.
Beyond Fiction: Testing with “Real” Politicians
To ensure these findings weren’t just about fictional characters, the researchers conducted a second study with 1,617 British participants using a more realistic scenario. Participants read about a local politician who had either done something virtuous (donated money to charity) or something villainous (stolen money from charity). The politician’s party affiliation was never mentioned, but participants were later asked to recall which party the politician belonged to.
Despite being given easy ways to say “I don’t remember seeing this information,” about one in six participants confidently “remembered” a party affiliation that was never actually stated. Just like in the first experiment, participants were much more likely to “recall” that virtuous politicians belonged to their preferred party, while villainous politicians belonged to the opposing party.
This false memory effect only appeared for political identities; when asked about other details like the politician’s number of children, no such projection effect occurred.
Why This Matters: The Polarization Cycle
These findings extend beyond fictional characters or hypothetical politicians. This tendency to assume “good people” share our politics while “bad people” belong to the other side could feed a dangerous cycle that worsens polarization. If we automatically connect negative traits to our political opponents and positive traits to our allies, we might become increasingly unwilling to find a middle ground or see nuance.
Liberals and those who identified with left-leaning parties (Democrats in the U.S., Labour in the UK) showed stronger projection effects than conservatives and right-leaning party supporters. This difference appeared consistently across both countries and experiments. The researchers propose this might be because those on the political left tend to have stronger negative feelings toward those on the right than vice versa, though they note this finding needs more investigation.
We’re all guilty of this mental shortcut, assuming the “good guys” think like us and the “bad guys” don’t. Breaking free from this pattern requires conscious effort: acknowledging that admirable people exist across the political spectrum, and that villainy plays no favorites when it comes to party membership. Heroes and villains exist in the eye of the beholder, and in our polarized world, so too do Democrats and Republicans, liberals and conservatives.
Paper Summary
Methodology
The researchers used two complementary experiments. First, a visual conjoint experiment with 3,200 participants (half American, half British), who viewed pairs of fictional characters and guessed their political affiliation. Second, a vignette experiment with 1,617 British participants who read about a local politician who either donated to or stole from charity, with no mention of party affiliation. Participants were later asked to “recall” the politician’s party, allowing researchers to measure false recall as evidence of projection.
Results
In the fictional character experiment, participants were 20 percentage points more likely to project their own partisan identity onto heroes than villains. Americans were 21 percentage points more likely to see heroes as sharing their ideology, while British participants showed a 12-point difference. In the politician experiment, about 15% of participants falsely “remembered” a party affiliation. Those reading about a virtuous politician projected their own party 60% of the time, compared to just 26% for those reading about a villainous politician—a 131% increase.
Limitations
The study’s findings are limited to two-party political systems like the US and UK, and may not apply to countries with multiple major parties. The use of fictional characters, while innovative, might not perfectly capture real-world political projection. The finding that left-leaning participants showed stronger projection effects than right-leaning ones was part of the researchers’ pre-registered exploratory analysis, meaning it was identified as an area to explore but wasn’t their primary hypothesis, so additional research would be valuable to confirm this pattern.
Discussion and Takeaways
The tendency to project our politics onto people based on whether we like them—and to project opposing politics onto those we dislike—could create a reinforcing cycle that deepens partisan divides. When we automatically link negative traits with the opposing party, we become less willing to engage across party lines. However, since projection often exaggerates differences, becoming aware of this tendency might help reduce polarization. Recognizing that there’s more diversity within both political camps than we assume could reduce hostility toward political opponents.
Funding and Disclosures
The research was funded by the University of Southampton and the European Research Council (grant 101044069). No conflicts of interest were noted in the paper.
Publication Information
The study, “Heroes and villains: motivated projection of political identities,” appeared in Political Science Research and Methods in 2025 by Stuart J. Turnbull-Dugarte (University of Southampton) and Markus Wagner (University of Vienna). The article is available as Open Access under the Creative Commons Attribution license.
