Othering is the process of perceiving or treating individuals and groups as fundamentally different from and inferior to one’s own group. It operates through categorization, stereotyping, and sometimes outright dehumanization, and its roots run deep enough in human psychology that researchers can trigger it in a laboratory with nothing more than an arbitrary label. The consequences ripple outward from reduced empathy for outsiders to measurable physiological damage in the people being othered, and the mechanisms are worth understanding because they are not always obvious, even to the people doing it.
Why Arbitrary Labels Are Enough
One of the most striking findings in social psychology is that people do not need meaningful differences to start treating others unequally. In the minimal group paradigm, participants are sorted into groups based on trivial criteria, sometimes as meaningless as a coin flip or a preference for one painter over another. Even under these bare conditions, people reliably favor their own group when distributing resources. In some cases, the drive to create a gap between ingroup and outgroup rewards was stronger than the drive to maximize the ingroup’s actual take, meaning people sacrificed real gains just to come out relatively ahead of the other group.1Current Opinion in Psychology. The Minimal Group Paradigm and its maximal impact in research on social categorization
This ingroup favoritism is often explained as a way people maintain a positive view of themselves through the groups they belong to. If your group is better than the other group, you get to feel better about yourself by association.2Journal of Experimental Social Psychology. Priming In-Group Favoritism: The Impact of Normative Scripts in the Minimal Group Paradigm And it does not appear to be driven purely by self-interest. In one experiment, participants who were told they would receive the maximum number of points regardless of how they allocated resources still discriminated against the outgroup just as much as participants whose own payoffs depended on the group. What mattered was how strongly they identified with their assigned group, not whether favoring it would benefit them personally.3Personality and Social Psychology Bulletin. Discrimination in the Minimal Group Paradigm: Social Identity or Self-Interest?
This means othering is not just something that happens when groups have genuine conflicts of interest over territory, jobs, or political power. The tendency to draw lines between “us” and “them” and to tilt the playing field can kick in with no tangible stakes at all. Real-world divisions, with their deep histories and genuine grievances, simply add fuel to a fire that barely needs a spark.
An Evolutionary Backdrop
If arbitrary categorization is this easy to trigger, one natural question is whether there is something deeper going on, something wired into the species. One hypothesis frames intergroup bias as an adaptive response to the threat of coalitional violence. Throughout human evolutionary history, groups of males formed alliances for raiding and defense. Being quick to identify who is and is not part of your coalition, and being psychologically prepared to treat outsiders as threats, would have carried survival advantages.4PubMed Central. Evolution and the psychology of intergroup conflict: the male warrior hypothesis
This does not mean othering is inevitable or morally neutral. Evolutionary accounts describe the conditions under which a tendency might have developed, not the conditions under which it must continue. But it helps explain why the impulse is so automatic and so resistant to simple instruction to be fair. The bias appears early, runs fast, and does not always announce itself in conscious thought.
How the Brain Responds to Outsiders
Neuroscience work has started to map what happens inside the brain when someone registers another person as an outgroup member. One line of research looks at empathic responses. When participants watched ingroup members experiencing sadness, their brain activity looked roughly the same as when they felt sad themselves. But when they watched outgroup members in the same situation, that empathic mirroring dropped off significantly.5Social Cognitive and Affective Neuroscience. Intergroup differences in the sharing of emotive states: neural evidence of an empathy gap In plain terms, other people’s pain registers less if your brain codes them as outsiders.
The encouraging part of this research is that the empathy gap is not fixed. Studies using learning interventions have shown that empathy-related brain responses for outgroup suffering can be boosted, suggesting the neural machinery for empathy is flexible and responsive to experience rather than permanently set at a lower level for anyone categorized as “them.”6PubMed Central. How learning shapes the empathic brain
The Role of Disgust in Dehumanization
Not all emotions contribute equally to othering. Disgust appears to play a uniquely powerful role. In one experiment, participants who were induced to feel disgusted (rather than sad or neutral) showed stronger implicit associations between an outgroup and animals, and between the ingroup and humanity. Sadness did not produce the same effect; the dehumanizing bump was specific to disgust.7Group Processes & Intergroup Relations. Disgust facilitates outgroup dehumanization
This has real-world echoes. In a Norwegian study, participants who read a newspaper article emphasizing allegedly poor hygiene among the Roma reported higher levels of disgust, which in turn predicted stronger dehumanization of Roma people and greater support for their deportation.8International Journal of Intercultural Relations. The effect of disgust-eliciting media portrayals on outgroup dehumanization and support of deportation in a Norwegian sample The pathway was clear: media content triggered disgust, disgust made it easier to see the outgroup as less than human, and dehumanization made harsh policy preferences feel more acceptable. This is one reason language about “filth,” “infestation,” and “contamination” shows up so reliably in propaganda targeting minority groups, and why scholars who study genocide have flagged biological metaphors as a recurring feature of the worst episodes of mass violence.9Journal of Historical Sociology. “Disease Incarnate”: Biopolitical Discourse and Genocidal Dehumanisation in the Age of Modernity
How Stereotypes Sort Groups Along Two Dimensions
People do not apply a single blanket stereotype to every outgroup. Research on how stereotype content works has found that groups tend to be judged along two main axes: how warm or likable they seem, and how competent or capable they seem. Many outgroups get “mixed” evaluations, rated high on one dimension and low on the other. Groups perceived as low-status but non-threatening tend to be stereotyped as warm but incompetent, which produces paternalistic attitudes like pity. Groups seen as high-status but competitive tend to be viewed as competent but cold, which produces envy. Groups rated low on both dimensions tend to elicit contempt.10PubMed. A model of (often mixed) stereotype content: competence and warmth respectively follow from perceived status and competition
This framework has held up across cultures. A study spanning seven European countries and three East Asian countries found the same two dimensions at work, and confirmed that mixed stereotypes are common across very different societies. High-status groups were consistently seen as competent, and competitive groups were consistently rated as lacking warmth.11PubMed Central. Stereotype content model across cultures: towards universal similarities and some differences The practical takeaway is that othering does not always look like hatred. It can look like pity, condescension, or envy. A group can be “othered” through the assumption that they are helpless and need guidance just as thoroughly as through the assumption that they are dangerous.
Language as the Delivery System
Metaphor is one of the most effective tools for encoding othering into everyday communication. A large-scale analysis of about 400,000 U.S. tweets about immigration identified seven recurring metaphorical frames, including descriptions of immigrants as water (floods, waves, pouring in) and as vermin. Dehumanizing metaphors appeared across the political spectrum but were used more frequently by conservative-leaning accounts, with the frequency varying depending on which specific metaphorical concept was invoked.12ACL Anthology. When People are Floods: Analyzing Dehumanizing Metaphors in Immigration Discourse with Large Language Models
What makes metaphor particularly effective is that it works below conscious awareness. Describing people as a “flood” does not require the speaker or the listener to consciously think “these people are not fully human.” The framing quietly imports the logic of natural disaster onto human beings: floods are uncontrollable, destructive, and must be contained. The audience absorbs the implication without necessarily registering that a comparison has been made at all.
Digital Platforms and the Amplification Loop
Social media has added a new layer to othering by creating environments that intensify ingroup-outgroup divisions at scale. Link recommendation algorithms that suggest connections based on structural similarity, meaning users who share many mutual contacts, tend to create network layouts where opinion groups become increasingly separated. The key insight from modeling research is that this polarization does not require people to actively seek out like-minded peers. The architecture of the platform creates echo chambers as a side effect of how it decides who to connect.13PubMed Central. Link recommendation algorithms and dynamics of polarization in online social networks
A recent large-scale experiment provided causal evidence for the other side of the equation: what happens when the content itself changes. By using an algorithm to rerank social media posts that expressed antidemocratic attitudes and hostility toward the opposing political party, researchers shifted out-party animosity by more than two points on a 100-point scale in both directions, and the effect was symmetric across political parties.14PubMed. Reranking partisan animosity in algorithmic social media feeds alters affective polarization Two points on a thermometer might sound small, but in a system serving hundreds of millions of users, even modest shifts in the average level of animosity translate to enormous aggregate changes in how people view and treat members of the other side.
Children and the Early Development of Social Categories
Social categorization appears early. Infants and toddlers are already sorting people by perceptual cues like race, gender, and language, and these sorting systems undergo important changes across childhood as children learn which categories their culture treats as meaningful, how to use those categories to predict behavior, and how their own identity fits into the system.15PubMed Central. The Development of Social Categorization The useful part of this process is genuine: children need to simplify a complex social world. But the same mechanism that helps a child figure out who speaks their language or who is likely to be friendly also lays the groundwork for stereotyping, prejudice, and discrimination.
This means that by the time children enter school, the cognitive architecture for othering is already in place. What differs across children is which groups they have learned to categorize as “other,” which depends heavily on the norms, language, and social structures they are exposed to. The implication is not that prejudice is set in stone by kindergarten, but that interventions aimed at reducing it need to start early to take advantage of a period when category boundaries are still being negotiated.
What Othering Does to the Body
Being on the receiving end of othering is not just a social or emotional experience. It leaves measurable marks on the body. Allostatic load is a composite index of physiological wear and tear that includes markers of cardiovascular function, inflammation, metabolism, and stress hormones. An integrative review found that nine studies showed a significant positive association between discrimination and allostatic load, across different types of discrimination including lifetime experiences, everyday encounters, and childhood racial discrimination.16PubMed Central. The impact of discrimination on allostatic load in adults: An integrative review of literature
The damage is specific and dose-dependent. A study of middle-aged adults found that everyday discrimination was associated with elevated lipid levels, while lifetime experiences of discrimination were linked to higher cardiovascular risk, increased inflammation, and worse metabolic markers. People reporting three or more lifetime experiences of discrimination showed worse numbers across nearly every system measured than those reporting fewer.17PubMed Central. Associations between multiple indicators of discrimination and allostatic load among middle-aged adults Childhood exposure may be especially harmful. Research with Indigenous populations found that the frequency of childhood racial discrimination explained a significant portion of the variation in adult allostatic load scores, suggesting that early-life othering can set health trajectories in motion that persist for decades.18PubMed. Childhood racial discrimination and adult allostatic load: The role of Indigenous cultural continuity in allostatic resiliency
These are not vague claims about stress being bad for you. They are specific, measurable changes in the body’s inflammatory, cardiovascular, and metabolic systems that track with the frequency and severity of discrimination experienced. The pathway runs from social exclusion through chronic stress activation to organ-level damage.
Mental Health and the Weight of Multiple Identities
The mental health effects of being othered are well documented, but they do not fall equally on everyone. People holding more than one marginalized identity, such as being both a racial minority and a sexual minority, report worse psychological wellbeing than those with a single marginalized identity. The relationship is mediated by greater minority stress and a lower sense of belonging.19Social and Personality Psychology Compass. Multiple Marginalized Identities, Minority Stress, and Mental Health
Intersectional approaches to studying discrimination have shown that looking at only one dimension of identity at a time can mask important patterns. In one study, the number of marginalized identities a person held and the frequency of discrimination they faced explained a substantial share of the variation in trauma-related stress and quality of life, above and beyond structural factors like income and education level.20PubMed Central. Marginalized identities, discrimination burden, and mental health: empirical exploration of an interpersonal-level approach to modeling intersectionality A national survey in England confirmed that intersectional analyses detected patterns in mental health service use and discrimination that were completely invisible when researchers examined social statuses one at a time.21PubMed Central. A comparison of single and intersectional social identities associated with discrimination and mental health service use: data from the 2014 Adult Psychiatric Morbidity Survey in England
People also attribute discrimination differently depending on which identities they hold. Race, age, gender, and socioeconomic status interact in complex ways that shape not just how much discrimination someone reports, but what they believe is driving it.22PubMed Central. The Intersections of Race, Gender, Age, and Socioeconomic Status: Implications for Reporting Discrimination and Attributions to Discrimination A low-income older woman of color may experience an interaction as racially motivated where a younger man from the same racial group reads a similar interaction as economic discrimination. These attribution differences matter because they shape how people cope, whom they trust, and whether they seek help.
When Othering Turns Inward
One of the less visible consequences of sustained othering is internalized stigma: the process by which people who are consistently treated as lesser begin to accept and absorb those negative evaluations. This is distinct from simply being aware that others hold prejudiced views. Internalized stigma involves a progression from recognizing the stereotype, to agreeing with it, to applying it to oneself.23PubMed. Internalized stigma and self-stigma in people diagnosed with a mental disorder. One concept or two? A scoping review The effects include withdrawal from social participation, reluctance to seek help, and diminished self-esteem.24PubMed Central. On the self-stigma of mental illness: stages, disclosure, and strategies for change
Research on mental illness stigma has documented this pattern extensively, but internalized othering is not limited to that context. It occurs wherever a devalued identity is persistent enough and culturally reinforced enough that the person carrying it starts to see themselves through the lens of the dominant group’s prejudice. It is, in effect, othering that no longer needs an external agent to operate.
How Polarization Erodes Social Trust
Othering is not just an interpersonal phenomenon. At scale, it corrodes the social infrastructure that allows diverse societies to function. Nationally representative data from the United States shows that perceived political polarization directly undermines people’s trust in each other. When people believe that members of their society no longer share common values, they become less likely to believe others can be trusted to do the right thing, which in turn reduces their willingness to cooperate for collective goals.25PubMed Central. Social Trust in Polarized Times: How Perceptions of Political Polarization Affect Americans’ Trust in Each Other
The mechanism here is worth pausing on because it is self-reinforcing. Othering across partisan lines reduces trust. Reduced trust makes cooperation harder. Failed cooperation becomes evidence that the other side cannot be trusted. And so the cycle accelerates. This is the civic-level version of the empathy gap described in brain-imaging studies: when another group feels sufficiently foreign, their pain matters less, their motives become suspect, and the shared ground that democratic institutions depend on narrows.
What Actually Reduces Othering
Given how deep the roots are, the question of what works to reduce othering is understandably important. The most robust evidence supports intergroup contact. A meta-analysis of contact interventions targeting ethnic prejudice found that they reliably improve attitudes toward outgroups, and that the effects persist over time. Both direct contact, like cooperative programs in schools or workplaces, and indirect contact, like extended contact through friends of friends or media exposure, showed positive results.26European Journal of Social Psychology. Can we really reduce ethnic prejudice outside the lab? A meta‐analysis of direct and indirect contact interventions
One cognitive strategy that has shown promise is recategorization: getting people to see members of an outgroup as part of a shared, overarching ingroup. Framing interactions in terms of a common identity, like being fellow students, coworkers, or citizens, can redirect the ingroup bias that drives othering so that it now includes the former outgroup. Research on this “common ingroup identity” approach has shown it can reduce bias, including weight-based bias, especially when the specific form of discrimination is made visible at the same time the shared identity is emphasized.27PubMed Central. Does a Common Ingroup Identity Reduce Weight Bias? Only When Weight Discrimination Is Salient The broader framework behind this work argues that by recategorizing “us” and “them” into “we,” the same psychological forces that create ingroup favoritism can be harnessed to work against prejudice rather than for it.28Journal of Social Issues. Understanding and Addressing Contemporary Racism: From Aversive Racism to the Common Ingroup Identity Model
Narrative fiction offers another route, and one that scales more easily than face-to-face contact. In two studies, participants who read an excerpt from a novel featuring a counterstereotypical Arab-Muslim woman rated Arab-Muslims lower on stereotypically negative traits and showed less negative attitudes post-reading. The key variable was not just whether they read the story, but how transported into it they felt. Readers who were more deeply absorbed showed the largest reductions in prejudice, suggesting that the immersive quality of good fiction does some of the same work as real-world contact by letting people temporarily inhabit another perspective.29Scientific Study of Literature. Transportation into literary fiction reduces prejudice against and increases empathy for Arab-Muslims
Scapegoating During Crises
If othering can be reduced by contact and recategorization, it can also be intensified by external shocks. Pandemics, economic downturns, and security threats have historically served as accelerants for scapegoating, in which anxiety and anger about an uncontrollable situation get redirected toward a visible outgroup. Research during the COVID-19 pandemic confirmed the theoretical expectation that pandemic threat would stoke xenophobic attitudes, though the specific ethnoracial boundaries around who got scapegoated varied depending on local context and political framing.30PubMed Central. Has Pandemic Threat Stoked Xenophobia? How COVID-19 Influences California Voters’ Attitudes toward Diversity and Immigration
The pattern is consistent enough across history to be practically useful as a warning sign. When a society is under collective stress, the psychological machinery for othering does not just continue operating at baseline. It ramps up, and the targets it selects are shaped by whichever narratives are available and politically convenient. Recognizing that pattern in real time is harder than recognizing it in retrospect, but the regularity of the cycle is itself a piece of knowledge worth holding onto.