Polarization, in its political and social sense, refers to the process by which a population divides into opposing groups whose beliefs, identities, or feelings toward each other grow increasingly distant. The term sounds simple, but researchers now recognize it as an umbrella covering at least two distinct phenomena that do not always move in lockstep: disagreements over policy and the raw emotional dislike people feel toward the other side. That distinction matters more than most public conversations acknowledge, because the remedies for each type look very different, and mistaking one for the other leads to interventions that miss the mark entirely.
Ideological Versus Affective Polarization
The most important thing to grasp about modern polarization research is that scholars split the concept into two broad categories. Ideological polarization is what most people picture first: citizens moving apart on actual policy questions like taxation, immigration, or climate regulation. Affective polarization is something else entirely. It describes the growing emotional hostility between partisan groups, the tendency to see the other side not merely as wrong but as threatening, immoral, or alien. A study spanning 153 countries between 2000 and 2021 treated these as “relatively distinct” phenomena with different downstream effects on political life, including how each type relates to corruption.
1Political Studies. Two Tales of Polarization? How Affective and Ideological Polarization Influence Political Corruption—A Panel Analysis of 153 Countries Between 2000 and 2021What makes this distinction especially striking is the evidence that the two types have not risen at the same rate. While affective polarization has grown sharply by most accounts, particularly in the United States, there is less evidence for a comparable surge in ideological polarization. On many specific issues, Americans’ and Europeans’ actual views have grown less divided over recent decades, not more.
2Electoral Studies. Ticking all the boxes? A comparative study of social sorting and affective polarizationIn other words, people increasingly hate the other team without necessarily disagreeing with them more on substance. That paradox is central to understanding why polarization feels so intense in daily life even when survey data on individual issues sometimes tells a muddier story.
Why Identity Drives Polarization More Than Issues Do
If ideological disagreement is not the main engine, what is? A large body of research points to identity. Political parties function as social groups, and humans are wired to favor their own group and view outsiders with suspicion. When researchers measured social identification with political parties directly, using psychological scales designed for group belonging, they found that the strength of someone’s partisan identity predicted their political attitudes and behavior better than traditional measures of how strongly they leaned left or right.
3Social Science Quarterly. Social Identity Theory and Party IdentificationThis identity-based framework helps explain a process called partisan sorting. Over the past few decades, political identity has become increasingly aligned with other identities: racial, religious, cultural, geographic. As those identities stack on top of one another, partisan identity strengthens, and the activism, bias, and anger that flow from strong group identities intensify, even when people’s actual positions on issues have not moved much.
4American Journal of Political Science. “I Disrespectfully Agree”: The Differential Effects of Partisan Sorting on Social and Issue PolarizationThis is worth sitting with: a nation can become angrier, more tribal, and more hostile across party lines without the average citizen actually changing their mind about health care or gun control. The emotional temperature rises because identity consolidation makes partisan labels feel more meaningful, not because the policy stakes have changed.
An Evolutionary Backdrop
The tendency to split into loyal factions is not a bug of modern politics. Humans evolved in an environment of intense competition between groups, and groups filled with loyal members outcompeted those that were not. That selective pressure shaped minds to be tribal, favoring in-group members and viewing outsiders with wariness. Modern politics happens to be one of the most visible arenas for this ancient coalitional instinct, and it triggers substantial cognitive biases as a result.
5Current Directions in Psychological Science. Tribalism Is Human NatureKnowing that tribalism is baked into human cognition does not excuse it, but it does reframe the question. Polarization is not simply a product of bad actors, lying politicians, or broken media. It draws on deep psychological tendencies that any society, in any era, is vulnerable to when the conditions are right.
The Perception Gap and False Polarization
One of the most consequential findings in polarization research is that people consistently overestimate how divided they actually are. This phenomenon, called false polarization, means that even where real polarization exists, the perceived version is substantially worse than reality. Partisans overestimate how extreme the other side’s views are, how much the other side dislikes them, and how dehumanizing the other side’s attitudes are.
6PubMed. False polarization: Cognitive mechanisms and potential solutionsA representative U.S. sample found that both Democrats and Republicans thought the other party’s prejudice and dehumanization toward them was roughly twice as strong as what members of that party actually reported. This negativity bias was consistent across demographic groups, though people with stronger ideological convictions were especially prone to it.
7PubMed Central. Exaggerated meta-perceptions predict intergroup hostility between American political partisansThe danger is that these inflated perceptions are not idle misunderstandings. They feed back into real behavior. People who believe the other side deeply despises them tend to want more social distance from those people and are more willing to support policies that harm the country overall if doing so advantages their own party.
8PubMed Central. Exaggerated meta-perceptions predict intergroup hostility between American political partisans Research on misperceived polarization describes these inaccurate beliefs about the other side as “overly pessimistic” and notes they directly worsen intergroup conflict.
9PubMed Central. Understanding and combating misperceived polarizationThree cognitive processes appear to drive false polarization: categorical thinking (lumping all opponents into a single extreme type), oversimplification (reducing complex positions to binary categories), and emotional amplification (letting strong feelings distort estimates of how the other side thinks).
10PubMed. False polarization: Cognitive mechanisms and potential solutionsSocial Media, Algorithms, and Echo Chambers
No discussion of polarization is complete without social media, though the picture here is more nuanced than the standard narrative suggests. Research confirms that people do cluster into like-minded groups online. A large-scale study of Facebook and Twitter found that homophilic clusters, groups where users primarily interact with people who share their views, dominate online interactions, and information spreads preferentially toward like-minded peers.
11PubMed Central. The echo chamber effect on social mediaAlgorithms play a role, but not always in the way people assume. Link recommendation features like “People you may know” tend to connect users with structurally similar people, those who share many mutual connections, rather than ideologically similar people. But the effect is the same: the resulting network topologies are ripe for opinion polarization. Echo chambers form not because users deliberately shield themselves from disagreement, but because the architecture of connection creates inadvertent information bubbles.
12PubMed Central. Link recommendation algorithms and dynamics of polarization in online social networksYet the research does not straightforwardly show that algorithms cause echo chambers, or that echo chambers cause polarization. A review of social and algorithmic mechanisms on digital media concluded that algorithms may actually weaken echo chambers in some cases by exposing people to views they disagree with. The bigger concern may be that digital media as a whole fuels perceived polarization by making extreme voices louder and hiding moderate majorities.
13PubMed Central. Social Drivers and Algorithmic Mechanisms on Digital MediaThere is one piece of causal evidence worth highlighting. An experiment used a large language model to rerank social media posts that expressed partisan animosity and antidemocratic attitudes. Decreasing or increasing exposure to that kind of content shifted how warmly participants felt toward the opposing party by more than two points on a 100-point scale, with no difference between party lines. That is a modest but real causal effect of algorithmic content ordering on affective polarization.
14PubMed. Reranking partisan animosity in algorithmic social media feeds alters affective polarizationContent that bashes the opposing side, rather than content that praises one’s own side, appears to drive the most engagement. An analysis of social media posts found that language about the out-group was the strongest predictor of “angry” reactions, the most popular reaction type across all datasets studied, while in-group language predicted “love” reactions.
15PubMed Central. Out-group animosity drives engagement on social mediaEconomic Roots of Political Division
Polarization does not emerge from psychology and media alone. Economic conditions matter. Cross-country data shows a positive and statistically significant association between income inequality and political polarization.
16PubMed Central. Income Inequality and Global Political Polarization: The Economic Origin of Political Polarization in the WorldA modeling study offered a compelling explanation for why: under conditions of rising inequality or economic decline, some members of a population benefit from adopting a risk-averse, in-group-favoring strategy. That strategy can spread rapidly through the whole population and persist even after the economic conditions that triggered it have reversed.
17PubMed Central. Polarization under rising inequality and economic declineThis finding is sobering because it suggests that polarization can become self-sustaining. An economic shock might plant the seed, but once the tribal strategy takes hold, it sticks around even when the economy recovers. Similarly, in developing democracies, electoral volatility creates uncertainty among political elites, who respond by staking out clear policy positions far from the center, feeding ideological polarization from the top down.
18Party Politics. Electoral volatility and political polarization in developing democracies: Evidence from Latin America, 1993–2016How Polarization Reshapes Daily Life
The effects of polarization extend beyond election-day behavior. Experimental evidence shows that polarization changes how people actually treat one another. Across a series of behavioral experiments with over 8,600 participants, researchers found a consistent pattern: in-group love showed up in how close people felt toward others who shared their political identity, but out-group hate showed up more powerfully in behavior, in whether people helped, harmed, or cooperated with someone from the other side.
19Management Science. Hate Trumps Love: The Impact of Political Polarization on Social PreferencesEven physical movement patterns may reflect partisan divisions. A study of U.S. county-level mobility found that while general political leaning did not affect where people traveled, the strength of that leaning did. People living in counties with homogeneous political populations were more likely to visit politically similar destinations, after controlling for economic and demographic factors.
20PubMed Central. Political polarization on the move: Analyzing geographical mobility between counties in the U.S.At its most damaging, polarization reaches a stage that researchers call “pernicious polarization,” where society splits into mutually distrustful “Us vs. Them” camps. Comparative evidence suggests this stage is not driven by any single social cleavage or institutional structure. It can emerge from religious, ethnic, class, or purely partisan divides, meaning no society is structurally immune.
21The ANNALS of the American Academy of Political and Social Science. Toward a Theory of Pernicious Polarization and How It Harms Democracies: Comparative Evidence and Possible RemediesWhat Happens in the Brain
Neuroscience research has begun to map how political orientation interacts with brain function. When people with liberal and conservative immigration attitudes watched identical political videos while undergoing brain imaging, their neural responses diverged in the dorsomedial prefrontal cortex, a region involved in interpreting narratives. This divergence intensified during moments that included risk-related and morally charged language.
22PubMed Central. Conservative and liberal attitudes drive polarized neural responses to political contentSeparate research found that economic conservatism predicted stronger connectivity between the amygdala and a nearby brain structure during periods of perceived threat.
23PubMed Central. Conservatism and the neural circuitry of threat: economic conservatism predicts greater amygdala–BNST connectivity during periods of threat vs safety A broader review of the neurological literature described evidence for a right-hemisphere “conservative complex” involving structures sensitive to negativity, threat, and avoidance.
24PubMed. A Neurology of the Conservative-Liberal Dimension of Political IdeologyNone of this means political views are hard-wired or that one orientation is more “natural” than another. But it does suggest that political content is not processed neutrally. Two people watching the same speech genuinely experience it differently at a neurological level, which helps explain why political conversations so often feel like the two sides are reacting to entirely different events.
Can Anything Reduce Polarization
If the causes are this deeply rooted, the natural question is whether anything works. The honest answer is that effective interventions exist, but their effects tend to be modest and often short-lived. A randomized controlled trial in post-Brexit Britain tested an intervention based on intergroup contact theory, where Leavers and Remainers engaged in structured dialogue. Participants who went through the intervention showed a meaningful increase in warmth toward the opposing side, reduced their tendency to attribute the other group’s positions to emotion or external manipulation, and became more willing to compromise. But these effects were observed only in the short term; longer-term follow-ups found no lasting difference.
25PubMed Central. A post-Brexit intergroup contact intervention reduces affective polarization between Leavers and Remainers short-termThe social media reranking experiment mentioned earlier offers a different angle. By reducing people’s exposure to posts expressing partisan hostility and antidemocratic attitudes, researchers produced a measurable shift in how participants felt about the other party.
26PubMed. Reranking partisan animosity in algorithmic social media feeds alters affective polarization That points to platform design as a lever, though a lever that technology companies would need to voluntarily pull against the grain of engagement-maximizing business models, since out-group hostility drives the most reactions.
Correcting false polarization is another promising route. If a large chunk of the problem is that people think the other side is more extreme and hostile than it really is, simply providing accurate information about what the other side actually believes could help. The challenge is getting that information to stick in an environment that rewards the exaggerated version.
Why Candidates Move to the Extremes
Classical political theory predicts that candidates in a two-party system should converge toward the center to capture the median voter. That prediction depends on voters having preferences that peak at a single point on the ideological spectrum. When voter preferences become bimodal, with peaks on both the left and the right rather than in the middle, that logic breaks down. A model examining this dynamic found that under conditions of polarized electorates, a relative cost of voting that deters indifferent centrists, and the presence of ideological third-party alternatives that pull extreme voters away, strategically minded candidates fail to converge to the center and may become more polarized than the electorate itself.
27Humanities and Social Sciences Communications. Polarization, abstention, and the median voter theoremThis creates a feedback loop. A polarized electorate incentivizes extreme candidates, whose campaigns further polarize the electorate, which incentivizes still more extreme candidates. Breaking that cycle would require structural changes to elections, such as reforms that make centrist candidates viable or increase turnout among moderate voters, rather than simply urging politicians to be more reasonable.