People routinely dismiss evidence that contradicts what they already believe, and the reason is not stupidity or stubbornness in any simple sense. Decades of research point to a collection of cognitive processes, rooted in emotion, identity, and even brain architecture, that make humans far better at defending their existing views than at updating them. The more interesting question, though, is what the research says about the handful of approaches that actually do get through.
Beliefs Are Identity, Not Inventory
The standard assumption is that people hold beliefs the way they hold items in a filing cabinet: new information comes in, the outdated file gets swapped out. But the evidence consistently shows that beliefs, especially on politically or socially charged topics, function more like badges of group membership. When you challenge someone’s stance on a contested issue, you are not just correcting a data point. You are, in their experience, questioning who they are and which people they belong to.
Researchers studying political judgment have found that a leader’s party affiliation shapes how constituents evaluate that leader’s actions far more than the actions themselves do. People tend to attribute noble motives to leaders from their own political side and selfish motives to leaders from the opposing side, regardless of what those leaders actually did. The pattern holds across trust, perceived patriotism, and willingness to see the leader represent the country internationally.1PubMed. Identity-motivated reasoning: Biased judgments regarding political leaders and their actions This is not a failure of logic. It is logic operating in service of a social goal: maintaining alignment with one’s group.
One framework for understanding this treats certain beliefs as social signals. Climate change denial, for instance, or particular religious convictions can be modeled not as conclusions people arrived at through weighing evidence but as displays that communicate tribal loyalty. The belief persists because it is doing a job: telling the people around you that you are one of them.2Wiley Online Library. A tribal mind: Beliefs that signal group identity or commitment When a fact threatens a belief that carries this kind of social weight, the fact is the thing that gets discarded, because the social cost of abandoning the belief is immediate and real, while the cost of being technically wrong feels abstract.
The Brain Treats Counterevidence Like a Threat
Neuroimaging research has put people inside brain scanners and then challenged their political beliefs with counterevidence. What happens in the brain is revealing. Challenges to deeply held political views triggered increased activity in the default mode network, a set of brain structures linked to self-reflection and internal thought, and decreased engagement with the outside world. In other words, when confronted with inconvenient facts, people literally turned inward.3Scientific Reports. Neural correlates of maintaining one’s political beliefs in the face of counterevidence
The same study found that participants who were more resistant to changing their minds showed stronger activation in the insula and the amygdala, brain regions tied to emotional processing and threat detection. People who did update their beliefs in response to evidence showed less activity in those areas. The implication is stark: for many people, encountering a fact that undermines a core belief triggers something closer to the brain’s danger response than its curiosity response. The emotional system fires up before the analytical one even gets a turn.
This emotional dimension helps explain something that frustrates anyone who has tried to win an argument with data. The issue is not that the other person cannot understand the evidence. It is that the evidence arrives wrapped in a signal that the brain reads as a social or psychological threat, and the threat response takes priority.
Why Reasoning Evolved to Win Arguments, Not Find Truth
One influential theory proposes that human reasoning did not evolve primarily to help individuals figure out what is true. Instead, it evolved to help people argue, persuade, and evaluate the arguments of others. Under this view, reasoning is fundamentally a social tool. It is adaptive because humans depend on communication and are constantly vulnerable to being misled, so the ability to construct and assess arguments was more useful than the ability to reason dispassionately in isolation.4PubMed. Why do humans reason? Arguments for an argumentative theory
This reframes the entire problem. If reasoning was designed by evolution to be a tool for social persuasion, then “confirmation bias” is not a bug. It is the system working as intended. You are built to find evidence that supports your position and poke holes in evidence that supports your opponent’s position, because that is what helped your ancestors maintain standing in their communities. The side effect is that you are remarkably bad at evaluating evidence neutrally when you have a stake in the outcome.
It also explains why people reason much better in groups than alone, but only under specific conditions. When everyone in the group already agrees, reasoning just becomes a machine for generating rationalizations. When there is genuine disagreement and people have to defend their positions to skeptics, the quality of reasoning improves. The system works best when used for what it was built for: adversarial exchange, not solitary reflection.
Cognitive Dissonance and the Art of Self-Justification
When people do encounter a fact that gets past their initial defenses, a second layer of protection kicks in. Cognitive dissonance, the discomfort that arises from holding two contradictory ideas at the same time, generates a strong motivation to resolve the conflict. But the resolution almost never involves accepting the uncomfortable new fact and discarding the old belief. Instead, people find ways to neutralize the threat: they minimize the importance of the new information, reinterpret it, or simply deny responsibility for the contradiction.5PubMed Central. A General Model of Dissonance Reduction: Unifying Past Accounts via an Emotion Regulation Perspective
Smoking provides a well-studied example. A longitudinal survey across four countries tracked how smokers rationalized their behavior over time. Smokers who never attempted to quit maintained the highest levels of rationalization, telling themselves things like “smoking is not as dangerous as people say” or “it helps me cope with stress.” When smokers did quit, their rationalizations dropped. But here is the telling part: among those who tried to quit and relapsed, the rationalizations bounced right back, particularly the ones about smoking being useful or functional.6Tobacco Control. Patterns of cognitive dissonance-reducing beliefs among smokers: a longitudinal analysis from the International Tobacco Control (ITC) Four Country Survey The beliefs were not based on an honest assessment of risk. They were psychological tools the brain deployed to keep behavior and belief in alignment.
This process happens so automatically that people rarely notice it. You do not decide to rationalize. The discomfort of the contradiction just fades as your mind quietly rearranges the furniture, and you walk away feeling like you were right all along.
The Illusion of Understanding
Another piece of the puzzle is that people systematically overestimate how well they understand things. Research on what is called the illusion of explanatory depth shows that people feel they understand complex topics with far greater precision and coherence than they actually do.7PubMed Central. The misunderstood limits of folk science: an illusion of explanatory depth Ask someone whether they understand how a toilet works or how a policy proposal would affect the economy, and they will say yes with confidence. Then ask them to explain the mechanism step by step, and the confidence collapses.
This illusion creates a specific obstacle to belief change. People resist new information partly because they believe they already have a thorough understanding of the topic, which makes the new information feel unnecessary or suspect. If you already “know” how something works, why would you need to update? The illusion protects existing beliefs by convincing you that they rest on a solid foundation when they are actually sitting on vague impressions and half-remembered fragments.8European Journal of Social Psychology. The illusion of explanatory depth and endorsement of conspiracy beliefs
Interestingly, the same research suggests a potential lever for change. When people are forced to explain a topic in detail and confront how shallow their understanding really is, they tend to become more moderate and more open to alternative views. The act of trying to explain punctures the illusion, and with it goes some of the certainty that was blocking new information.
The Backfire Effect Is Mostly a Myth
For years, one of the most widely cited ideas in this space was the “backfire effect,” the claim that correcting someone’s false belief would actually make them believe it more strongly. The idea took on a life of its own in popular science writing and became a common reason people gave for not even bothering to correct misinformation. But the research tells a very different story.
Multiple attempts to replicate the backfire effect have come up empty. One set of experiments specifically designed to test for it found that no corrected items produced a backfire compared to a control group that simply restated the original belief without correction. The researchers concluded that the effect’s poor replicability likely stems from measurement artifacts rather than from a real psychological phenomenon.9PubMed Central. The backfire effect after correcting misinformation is strongly associated with reliability
A broader review of the evidence reached a similar conclusion: corrections typically do increase accuracy, at least in the short term. The real problem is not that corrections backfire. It is that their effects tend to fade. Over time, the corrected belief drifts back toward the original misconception, especially when people continue to encounter messages from political leaders or media figures reinforcing the inaccurate version.10PubMed Central. Why the backfire effect does not explain the durability of political misperceptions The practical upshot is important: correcting misinformation is not futile, but a single correction is rarely enough when the person’s information environment keeps pushing them in the other direction.
What Actually Changes Minds
If the brain is wired to protect existing beliefs, especially identity-linked ones, then blunt fact-delivery is never going to be the most effective strategy. The interventions that do work tend to share a common thread: they find ways to lower the psychological threat before introducing the new information.
One approach that shows consistent results is self-affirmation, having people reflect on their core values before encountering a challenging message. In one experiment, participants who affirmed their values before being exposed to health messages showed more activity in brain regions associated with self-relevance, and they went on to actually change their behavior, increasing their physical activity levels more than a control group that did not go through the affirmation step.11PubMed Central. Self-affirmation alters the brain’s response to health messages and subsequent behavior change The logic is that when your sense of self-worth has just been reinforced, a threatening message feels less threatening. You can afford to take it seriously because your identity is not on the line.
Personal contact also matters, and the messenger’s identity can be decisive. A field experiment on attitudes toward gay equality found that face-to-face conversations with canvassers produced large initial shifts in opinion regardless of who the canvasser was. But only conversations with gay canvassers produced changes that persisted at three weeks, six weeks, and nine months. The effect even spread within households: people who lived with someone who had a conversation with a gay canvasser shifted their views too.12PubMed. When contact changes minds: an experiment on transmission of support for gay equality There is something about encountering a real person whose life is directly affected by the issue that bypasses the usual defenses in a way that statistics and arguments do not.
Narrative and storytelling offer yet another route. Stories capture attention and evoke emotion in ways that reduce resistance to a message.13PubMed. Strategic Storytelling: When Narratives Help Versus Hurt the Persuasive Power of Facts When you are absorbed in someone’s story, the analytical defenses that would normally kick in to scrutinize each claim take a back seat. This does not mean stories are a tool for manipulation. It means that wrapping factual information inside a human narrative can be a more effective delivery vehicle than presenting the same information as a bulleted list of evidence.
Inoculation Against Misinformation
Rather than trying to correct false beliefs after they have taken hold, a growing body of work focuses on prevention. The approach is modeled on vaccination: expose people to a weakened form of a misleading argument along with an explanation of why it is misleading, and they become more resistant to the full-strength version when they encounter it later.14The ANNALS of the American Academy of Political and Social Science. Psychological Inoculation against Misinformation: Current Evidence and Future Directions
This has been tested at real-world scale. Researchers developed short videos that teach people to recognize specific manipulation techniques used in misinformation, including emotionally charged language, false dichotomies, scapegoating, and personal attacks. Across multiple randomized studies involving thousands of participants and a large field study on YouTube with over 22,000 viewers, the videos improved people’s ability to distinguish trustworthy content from untrustworthy content and made them more careful about what they shared.15PubMed Central. Psychological inoculation improves resilience against misinformation on social media
The inoculation approach has a significant advantage over after-the-fact correction: it teaches a transferable skill rather than addressing one specific falsehood at a time. Once someone learns to spot emotionally manipulative framing, they can apply that knowledge to new misinformation they have never seen before. The challenge is durability. Like biological vaccines, the effects can weaken over time, which is why researchers are studying “booster” interventions designed to refresh the protective effect.16PubMed Central. Psychological booster shots targeting memory increase long-term resistance against misinformation
When Facts Do Get Through
None of this means that facts are powerless. The research on motivated reasoning in areas like climate change preference formation confirms that people reject information contradicting their beliefs, but the strength of this rejection varies a great deal depending on context.17Nature Climate Change. The evidence for motivated reasoning in climate change preference formation When the topic is not tied to identity, when the stakes are practical rather than tribal, people update their beliefs quite readily. You do not see motivated reasoning when someone learns their flight has been delayed, or that a recipe calls for a different oven temperature than they assumed. The machinery of belief defense activates primarily when the belief in question says something about who you are.
Even on identity-linked topics, the right kind of information can reduce polarization. One set of experiments found that providing people with an objective, nonpartisan breakdown of how their tax dollars were actually spent reduced political polarization on the topic of taxation. The effect disappeared, though, when participants were simultaneously asked how they would prefer their taxes to be spent, which re-engaged the identity and preference machinery.18CrossRef API. Can Information Decrease Political Polarization? Evidence From the U.S. Taxpayer Receipt The lesson is that facts can get through, but only when the delivery avoids activating the tribal alarm system.
The pattern across all the successful interventions is consistent. They do not try to overpower the brain’s defenses with bigger, louder evidence. They work around those defenses by lowering the emotional stakes, establishing trust through shared identity or personal connection, teaching critical thinking as a skill rather than delivering it as a correction, or catching people before a false belief has time to become part of who they are. The facts still matter. But the path they take to get from the outside world into someone’s actual beliefs runs through territory that has very little to do with logic and a great deal to do with emotion, identity, and the social world the person lives in.