Linking personal identity to political issues predicts a preference for extreme candidates

A recent study published in the Journal of Experimental Social Psychology suggests that the rising popularity of extreme political candidates may be driven by how voters link their personal identities to their political opinions. The research provides evidence that when people feel an issue defines who they are as individuals, they tend to adopt more radical positions and favor politicians who do the same.

The researchers conducted this series of investigations to explore the psychological reasons why voters might prefer extreme candidates over moderate ones from their own party. Previous explanations have focused on structural factors like the way primary elections are organized or changes in the pool of people running for office.

But the authors behind the new research sought to better understand whether a voter’s internal connection to an issue is a significant factor. They focused on a concept called identity relevance, which is the degree to which an attitude signals to others and to oneself the kind of person someone is or aspires to be.

“Elected officials in the United States are increasingly extreme. The ideological extremity of members of Congress from both parties has steadily grown since the 1970s, reaching a 50-year high in 2022,” said study author Mohamed Hussein of Columbia University.

“State legislatures show similar trends. A recent analysis of more than 84,000 candidates running for state office revealed that extreme candidates are winning at higher rates than at any time in the last 30 years. We were interested in understanding why extreme candidates are increasingly elected.”

Free daily newsletter

“So far, research in this area has focused on structural factors (e.g., the structure of primary elections),” Hussein explained. “In our work, we wanted to pivot the conversation to more psychological factors. Specifically, we tested if the identity relevance of people’s attitudes causes them to be drawn to extreme candidates. ”

The researchers conducted a series of studies to test their hypothesis. In the first study, 399 participants who identified as Democrats read about a fictional candidate named Sam Becker who was running for a seat in the House of Representatives. Some participants read that Becker held moderate views on climate change, while others read that he held extreme views. The researchers measured how much the participants felt their own attitudes on climate change were relevant to their identity.

The results suggests that as identity relevance increased, the participants reported having more extreme personal views on the issue. Those with high identity relevance showed a preference for the extreme version of Sam Becker and a dislike for the moderate version. This study provides initial evidence that the more someone sees an issue as a reflection of their character, the more they favor radical politicians.

The second study involved 349 participants and used a more complex choice task to see if these patterns held across different topics. Participants were shown pairs of candidates with varying ages, genders, and professional backgrounds. One candidate in each pair held a moderate position on a social issue, while the other held an extreme position.

The researchers tested five separate issues: abortion, gun control, immigration, climate change, and transgender rights. The data suggests that across all these topics, higher identity relevance predicted a greater likelihood of choosing the extreme candidate. Additionally, participants with high identity relevance reported being more receptive to hearing the views of the extreme candidate.

In the third study, the researchers aimed to see if they could change a person’s identity relevance by shifting their perception of what their political party valued. They recruited 584 Democrats and asked them to read a news article about the priorities of the Democratic National Committee. One group read that the party was prioritizing corn subsidies, a topic that is generally not a core identity issue for most voters.

The results suggests that when participants believed their party viewed corn subsidies as a priority, they began to see the issue as more relevant to their own identity. This shift in identity relevance led them to adopt more extreme personal views on the topic. Consequently, these participants showed a higher preference for candidates who supported radical changes to agricultural subsidies.

This experiment also allowed the researchers to rule out other factors that might influence candidate choice. They measured whether participants felt more certain, more moral, or more knowledgeable about the issue. The analysis provides evidence that identity relevance influences candidate choice primarily through its effect on attitude extremity rather than through these other psychological states.

The fourth study sought to prove that this effect can occur even when people have no factual information about a topic. The researchers presented 752 participants with a fictitious ballot initiative called Prop DW. The participants were told nothing about what the proposal would actually do.

Some participants were told their political party had taken a position on Prop DW, while others were told the party had no stance. Even without knowing the details of the policy, those who believed their party had a stance reported that Prop DW felt more identity-relevant. These individuals developed more extreme attitudes and favored candidates who took extreme positions on the made-up issue.

This finding suggests that the psychological pull toward extremity is not necessarily based on a deep understanding of policy. Instead, it seems to be a reaction to the social and personal significance assigned to the topic. It also suggests that people can form strong, radical opinions on matters they do not fully understand if they feel those matters define their social group.

Studies five and six moved away from group dynamics to see if individual reflection could trigger the same results. The researchers used a digital tool that allowed 514 participants to have a live conversation with a large language model. In one condition, the computer program was instructed to help participants reflect on how their views on corn subsidies related to their core values and sense of self.

This reflection process led to a measurable increase in identity relevance. Participants who reflected on their identity reported a higher desire for clarity, which means they wanted their opinions to be certain and distinct. This desire for clarity pushed them toward more extreme views and a higher probability of choosing an extreme candidate.

The final study involving 807 participants replicated this effect with a more rigorous comparison group. In this version, the control group also discussed corn subsidies with the language model but was not prompted to think about their personal identity. The results provides evidence that only the participants who specifically linked the issue to their identity showed a significant shift toward extremity.

The researchers note that this effect was symmetric across political parties. Both Democrats and Republicans showed the same pattern of moving toward extreme candidates when an issue felt relevant to their identity. This suggests that the psychological mechanism is a general feature of human behavior rather than a trait specific to one side of the political aisle.

“Across six studies with over 3,000 participants, we found that the more people see their political attitudes as tied to identity, the more likely they are to choose extreme, versus moderate, candidates,” Hussein told PsyPost. “The more central fighting climate change felt to the identity of participants, the more they liked the extreme Sam and the more they disliked the moderate Sam. Put simply, identity relevance increased liking of extreme candidates but decreased liking of moderate ones.”

“These results were remarkably robust. Across studies we tested a range of issues including climate change, abortion, immigration, transgender rights, gun control, and corn subsidies . We even created a fictitious issue (“Prop DW”) that participants had no information about. Across issues, we found that when we framed the issue as central to their identity, people formed more extreme views on it and then preferred extreme candidates who promised bolder action. Even on a made-up issue, identity relevance pushed people toward extremes.”

“These results were also robust regardless of how we talked about candidate extremity,” Hussein continued. “In addition to having candidates describe themselves as extreme, we also signaled extremity in different ways. In some studies, the candidates endorsed different policies, some that were moderate and others that were extreme.”

“In other studies, we held the policy constant but changed the level of action that candidates supported (e.g., increasing a subsidy by a small amount compared to a large amount). Lastly, in some studies, we explicitly labeled candidates as ‘moderate’ or ‘extreme’ on an issue. Regardless of how candidate extremity was described to participants, the results held.”

But there are some potential misinterpretations and limitations to consider regarding this research. One limitation is that the studies were conducted within the specific political context of the United States. The American two-party system might encourage a greater need for distinct, polarized identities compared to countries with multiple competing parties.

Future research could explore whether these findings apply to people in other nations with different electoral structures. It would also be useful to investigate whether certain personality types are more prone to linking their identity to political issues. Some individuals may naturally seek more self-definition through their opinions than others.

Another direction for future study involves finding ways to decrease political tension. If identity relevance is a primary driver of the preference for extreme candidates, it suggests that finding ways to de-emphasize the personal significance of political stances might lead to more moderate dialogue. Interventions that help people feel secure in their identity without needing to hold radical opinions could potentially reduce social polarization.

“Politics has always been personal, but it’s becoming more identity-defining than ever,” Hussein said. “And when politics becomes identity-relevant, our research suggests that extremity gains in appeal. Illuminating this psychological process helps us understand today’s political landscape and provides a roadmap for how to change it. Our results suggest that if we can loosen the grip of identity on politics, the appeal of extreme candidates might start to wane.”

The study, “Why do people choose extreme candidates? The role of identity relevance,” was authored by Mohamed A. Hussein, Zakary L. Tormala, and S. Christian Wheeler.