A recent study published in the Journal of Personality and Social Psychology suggests that people consistently overestimate how much their political peers will judge them for changing their minds on polarizing issues. This inflated fear of rejection tends to make individuals hide their shifting views, which deprives the public discourse of diverse perspectives. The research provides evidence that the social penalty for political dissent within one’s own party is generally much milder than expected.
In highly polarized environments, people often treat political beliefs as strict markers of group loyalty. When an individual’s opinion evolves on a divisive topic like gun control or immigration, they face a difficult choice. They can voice their new perspective and risk being ostracized, or they can stay silent to protect their social standing.
Because humans have a deep, evolutionary need to belong to groups, they are highly sensitive to the threat of social rejection. This sensitivity could lead individuals to adopt a “better safe than sorry” mindset, causing them to expect a much harsher backlash than they will actually experience.
When people conceal their true thoughts based on these exaggerated fears, it creates a false illusion that everyone in a political group completely agrees with the party line. This illusion is known as pluralistic ignorance, a situation where a majority of group members privately reject a norm but assume everyone else accepts it. The scientists designed a series of studies to test whether these expectations of social punishment are systematically miscalibrated and to explore how this fear influences communication.
“Two trends really stood out to me. The first is that Americans are becoming increasingly afraid to speak their minds about politics — polling shows this fear has grown substantially over the past two decades. The second is that people tend to perceive their political parties as having a uniform set of beliefs, when in reality, private polling reveals much more diversity of opinion underneath the surface,” explained Trevor Spelman, a PhD student at the Kellogg School of Management at Northwestern University.
“This got me thinking: if people misperceive how uniform their party actually is, they might also be overestimating how much backlash they’d face for voicing a dissenting view. And if that’s happening, it could mean people are staying silent based on fears that are out of proportion with reality — which might have real consequences for the quality of political discourse within parties.”
To explore this dynamic, the researchers first conducted a pilot survey with 131 American partisans who had recently changed their minds on a political issue. The data showed that individuals who anticipated severe social penalties were more likely to self-censor their updated views.
To test the accuracy of these expectations, the researchers then conducted Study 1 with 500 Democrats and Republicans. Participants were divided into a predict group and a react group. The predictors imagined adopting the opposing political party’s view on abortion, immigration, or gun control, and then estimated how much another member of their own party would reject them. The reactors evaluated a hypothetical peer who had actually made that ideological shift.
The findings revealed a perception gap between the two groups. Predictors consistently expected more exclusion, criticism, and disrespect than the reactors actually reported feeling toward a dissenting peer.
Next, Study 2 examined whether this perception gap persists during live interactions with real financial stakes. The researchers paired 278 participants in live, anonymous text conversations. After a brief getting-to-know-you chat, predictors were told their partner would learn that they had slightly decreased their agreement with their party’s typical stance on a specific issue.
Predictors then guessed how their partner would react in a subsequent cooperative task. The reactors, meanwhile, were given the news of their partner’s slight ideological shift and were asked to make choices about the upcoming task.
The researchers found that predictors overestimated the behavioral rejection they would face. Specifically, predictors guessed their partner would choose to work with a new person 18.7 percent of the time, but reactors only abandoned their partner 7.9 percent of the time. Predictors also expected their partner to withhold more money in a bonus-sharing game than the reactors actually kept.
While the first two studies relied on hypothetical or slight belief changes, the researchers conducted Study 3 to investigate what happens when people genuinely alter their political stances. In the first phase, 494 participants wrote persuasive essays arguing for the opposing party’s view on a specific topic. This exercise successfully induced a genuine shift in beliefs for 147 of the participants.
These individuals then predicted how another party member would react to their newfound stance, and their estimates were compared against the actual reactions of an evaluating group. Once again, predictors anticipated harsher judgments than reactors delivered.
Two months later, the researchers recontacted 93 of the participants whose beliefs had changed. Half of the group was reminded of their previous ideological shift, while the other half received a generic reminder of their participation. Only the participants who were explicitly reminded of their belief change overestimated the social sanctions they would face. This suggests that the perception gap is driven by the immediate psychological awareness of the dissent, rather than a permanent personality trait.
“We were surprised by how strongly the effect held up across different contexts,” Spelman told PsyPost. “Whether participants were strangers or had just gotten to know each other through a live conversation, whether the belief change was hypothetical or real, and whether we measured reactions through survey items or actual financial decisions — the overestimation was consistently there.”
In Studies 4a, 4b, and 4c, which included 393, 282, and 596 participants respectively, the scientists sought to identify the underlying psychological mechanism. They compared expectations of rejection for shifting views on highly partisan topics against nonpartisan topics, such as postal service delivery schedules.
The gap between expected and actual rejection was larger for the partisan issues. To understand why, the scientists measured signal amplification bias. This is a psychological concept describing the human tendency to assume our actions send a much stronger message to observers than they actually do.
The researchers found that predictors expected their political dissent to be viewed as a massive betrayal of group loyalty. The reactors, however, did not view the belief change as a severe indicator of disloyalty. This exaggerated fear of appearing traitorous helped explain why predictors expected such intense social backlash.
“The average effect size across our studies was d = .87, which is considered large by conventional standards in psychology,” Spelman noted. “To put that in more concrete terms: in our studies, roughly 8 out of 10 people in the predictor role overestimated how much rejection they would face relative to what their partner actually reported.”
Finally, Study 5 tested a potential intervention to reduce this exaggerated fear. The scientists recruited 620 Democrats and Republicans who had genuinely changed their minds on a polarizing topic in the past year. Predictors were divided into two groups for a brief psychological exercise.
One group completed a loyalty-affirming exercise, where they listed three actions they had taken to support their political party. The other group listed three actions that went against their party. Afterward, all predictors estimated how much rejection they would face from a peer who had also changed their mind on a different topic.
The intervention was effective. Participants who reflected on their past loyalty anticipated less social rejection than those who reflected on their disloyalty. By affirming their secure standing within the group, participants felt less threatened by the prospect of sharing their dissenting views, leading to more accurate predictions of social acceptance.
“The main takeaway is that the fear of speaking up within your own political group is often worse than the reality,” Spelman said. “Across our studies, people consistently anticipated moderately harsh rejection for breaking from their party’s position on issues like abortion, gun control, and immigration. But the people actually doing the judging reported much milder reactions. We saw the same pattern emerge across survey, behavioral, and qualitative measures – predictors consistently and robustly overestimate how much other group members would socially punish them for expressing dissent.”
“This matters because when people self-censor dissenting views out of fear, they deprive the information environment of their point of view. When only conforming views are expressed, it can create a false impression that everyone in the party agrees. That false impression then makes the next person even more afraid to speak up. It’s a self-reinforcing cycle, and our findings suggest the entry point is miscalibrated expectations about how others will react.”
“That said, social backlash is real, and the experience of rejection is genuinely painful,” Spelman added. “We’re not saying there’s no cost to dissent — but these findings suggest that people tend to overestimate that cost.”
As with all research, there are some limitations. The studies were also conducted entirely within the United States during a period of intense political polarization. The social dynamics of ideological dissent might operate differently in other cultural contexts or political systems.
Additionally, the interactions in these experiments primarily took place between strangers or loose acquaintances. Disagreeing with close friends, family members, or colleagues carries different relational stakes, which might alter how people predict and experience rejection.
“The most important thing to emphasize is that we’re not saying social backlash doesn’t exist or that dissent is cost-free,” Spelman said. “The costs of dissent are real. What we’re showing is that people systematically overestimate those costs, and that this overestimation predicts self-censorship.”
Future research could explore how these interactions unfold in established relationships and organizational settings. The scientists also plan to develop and test broader interventions that can correct these social misperceptions on a larger scale. By helping individuals calibrate their expectations, they hope to encourage more authentic communication and foster a healthier public discourse.
“This paper is part of my broader dissertation work, which examines how the overestimation of social costs for dissent plays out not just between individuals but across groups and at the societal level,” Spelman told PsyPost. “When many people simultaneously overestimate backlash and self-censor, the cumulative effect can distort entire information environments — making political parties, organizations, and communities appear far more ideologically uniform than they actually are.”
“A major focus going forward is developing and testing interventions that can correct these miscalibrated expectations, increase people’s willingness to disclose dissenting views, and ultimately contribute to healthier public spheres and more representative discourse. The loyalty affirmation exercise in this paper is a promising starting point — simply reflecting on past demonstrations of group loyalty was enough to reduce overestimation significantly. But there’s much more to explore in terms of what kinds of interventions are most effective, how durable they are, and how they can be scaled beyond the lab into real-world contexts.”
“Healthy democratic discourse depends on people being willing to voice dissenting views within their own coalitions,” Spelman concluded. “When people systematically overestimate the social costs of doing so, we end up with a distorted picture of what people actually believe — and that undermines the quality of political deliberation within parties. Our hope is that this research can help people calibrate their expectations and feel more comfortable speaking up when they disagree, even on issues that feel risky.”
The study, “Overestimating the Social Costs of Political Belief Change,” was authored by Trevor Spelman, Abdo Elnakouri, Nour Kteily, and Eli J. Finkel.
