The Facial Anomaly-Is-Bad Bias: A Fragile Stereotype
Facial differences are weaker social cues than sex and race.
Posted April 22, 2026 | Reviewed by Monica Vilhauer Ph.D.
Co-written by Dr. Soma Chaudhuri and Dr. Anjan Chatterjee.
We all categorize objects constantly. In a grocery store, we sort apples from oranges. Scrolling through our inbox, we file messages as urgent or ignorable. And when we encounter other people on the street, at a party, or in a waiting room, our mind does the same thing, sorting them into groups with remarkable speed. Social categorization is one way we navigate a world full of strangers, a kind of automatic mental filing system. This propensity shapes whom we notice, whom we remember, and how we behave toward them — often before we have exchanged a single word.
Among the most powerful triggers for this sorting are the features of a person's face. Sex , race, and age — sometimes called the "big three" of social perception — are encoded quickly when we look at someone. But what about unusual features? What if the person we are meeting has a noticeable facial scar? Does such a salient visual difference establish its own mental category, too, like sex or race?
The burden of looking different
People with visible facial differences like scars, burns, and palsies, experience real social costs. A robust body of research documents the "anomalous-is-bad" stereotype: people tend to judge people with atypical faces as less warm, less competent, and less morally trustworthy, despite not knowing anything about the person (Workman et al., 2021; Hartung et al., 2019). Although seemingly a learned and not innate bias (Workman et al., 2022), this pernicious stereotyping is automatic, largely unconscious , and widespread in the general population (Jamrozik et al., 2019).
How deeply does this bias run? The unfortunate prejudice suggests the human mind may treat anomalous faces as a social category in their own right, the way it does with men and women, or people of different races. If so, a scar might shape how someone is filed in memory , forming a lasting organization of a person's social identity .
The memory confusion game
To test this possibility, in a recent study, we (Chaudhuri, Bobrow, & Chatterjee, 2026) used a technique called the Who-Said-What paradigm (Taylor et al., 1978). Developed in the 1970s, this clever experimental design reveals how people spontaneously sort others into social groups — without ever asking them directly. Here is how our study worked. People viewed a series of eight faces on a screen. The faces varied by sex, race, and whether they had a visible facial burn. Each face was paired with one of three brief, uncontroversial first-person statements, like, "I have a pet dog," "I read poetry," or "I spend time on puzzles.” In this way, 24 unique face-statement pairs were generated and presented in random order. Participants were asked simply to pay attention to what was presented, with no hint that a memory test was coming. This omission ensured that whatever they later recalled reflected natural, spontaneous encoding rather than deliberate memorization. After a brief unrelated task, all eight faces appeared together on a grid, and each of the 24 statements was shown again one at a time. Participants had to decide which statement matched each face.
The key insight of this method is derived from the mistakes people make . If a person pays attention to a given social category, like sex, then they should be more likely to confuse the statements of two male speakers (attributing one man’s statement to another man) than to confuse the statements from a woman and a man. This kind of mistake means they had grouped the statements in memory by sex. Similarly, if they confuse one white speaker with another white person, we can infer they are grouping people by race. Errors reveal the underlying architecture of social memory — the categories our minds generate, often unconsciously, to sort people.
Five hundred people completed our online Who-Said-What task. Categorization by sex was most robust. People were more likely to confuse people of the same sex with one another — swapping one man’s statement for another man’s, or one woman’s for another woman’s. This result aligns with decades of research suggesting that sex is encoded rapidly and almost automatically when we see a face (Kurzban et al., 2001).
Race produced a weaker signal. People showed some tendency to confuse people of the same racial background, but these mistakes were modest —smaller than the effect of sex. This finding is also consistent with prior work (Kurzban et al., 2001), suggesting that racial categorization, while real, is more context-dependent than categorization by sex.
And facial scars? They barely registered as a category driving memory mistakes. People were not more likely to confuse two people with scars than to confuse a scarred and an unscarred person. We found only a small window in which scar-based grouping was evident: among the few people who categorized race weakly. Even then, the effect was faint, occurring in only about one in six people.
What this result means, and what it does not
Let us be clear about what this study does and does not tell us. It does not say that the "anomalous-is-bad" stereotype is harmless or exaggerated. The negative snap judgments people make about people with facial differences are well-documented and real. However, this bias operates differently from how we categorize sex or race.
When we see a person with a scar, we may form a negative impression in an instant, and that impression can indeed be unfair. But that scar does not appear to create a mental folder, the way "man" or "woman" might. Our brain does not seem to group scarred people into a social category. A scarred face can be judged negatively at the level of an individual without being encoded psychologically as a member of a social group. This finding suggests that the anomalous-is-bad stereotype has a shallower cognitive basis than the categories we form using sex and to some extent race.
Our social minds operate with a hierarchal structure. Some features, like sex, are encoded with remarkable speed and consistency. Others, like race, are present but flexible. And some features, despite being visually striking and socially consequential, do not signal sufficient social influence to make it into the brain's primary sorting list. Facial scars appear to fall into this last category.
For the many people who live with visible facial differences, this real-world implication of this finding is significant — and optimistic . While first impressions remain a challenge — the anomalous-is-bad stereotype is an obstacle that merits continued confrontation — these impressions appear not to automatically harden into deeply encoded social categories. The fragility of the facial- anomaly-is-bad bias suggests that it can be broken.
Chaudhuri, S., Bobrow, I., & Chatterjee, A. (2026). Who Do We Remember? Facial Anomalies, Race, and Sex in Social Categorization. In Behavioral Sciences (Vol. 16, Issue 3, p. 462). https://doi.org/10.3390/bs16030462
Hartung, F., Jamrozik, A., Rosen, M. E., Aguirre, G., Sarwer, D. B., & Chatterjee, A. (2019). Behavioural and Neural Responses to Facial Disfigurement. Scientific Reports , 9 (1), 8021. https://doi.org/10.1038/s41598-019-44408-8
Jamrozik, A., Oraa Ali, M., Sarwer, D. B., & Chatterjee, A. (2019). More than skin deep: Judgments of individuals with facial disfigurement. Psychology of Aesthetics, Creativity, and the Arts , 13 (1), 117. https://doi.org/10.1037/aca0000147
Kurzban, R., Tooby, J., & Cosmides, L. (2001). Can race be erased? Coalitional computation and social categorization. Proceedings of the National Academy of Sciences , 98 (26), 15387–15392. https://doi.org/10.1073/pnas.251541498
Taylor, S. E., Fiske, S. T., Etcoff, N. L., & Ruderman, A. J. (1978). Categorical and contextual bases of person memory and stereotyping. Journal of Personality and Social Psychology , 36 (7), 778–793. https://doi.org/10.1037/0022-3514.36.7.778
Workman, C. I., Humphries, S., Hartung, F., Aguirre, G. K., Kable, J. W., & Chatterjee, A. (2021). Morality is in the eye of the beholder: the neurocognitive basis of the “anomalous-is-bad” stereotype. Annals of the New York Academy of Sciences , 1494 (1), 3–17. https://doi.org/10.1111/nyas.14575
Workman, C. I., Smith, K. M., Apicella, C. L., & Chatterjee, A. (2022). Evidence against the “anomalous-is-bad” stereotype in Hadza hunter gatherers. Scientific reports , 12 (1), 8693. https://doi.org/10.1038/s41598-022-12440-w
Share this post Facebook Bluesky Linkedin Email
There was a problem adding your email address. Please try again.
By submitting your information you agree to the Psychology Today Terms & Conditions and Privacy Policy
Anjan Chatterjee, MD, FAAN, is Professor of Neurology, Psychology, and Architecture at the University of Pennsylvania Perelman School of Medicine.
Get the help you need from a therapist near you–a FREE service from Psychology Today.
This article is part of the Bringwise Psychology Journal — daily insights on human behavior, mental health, and personal growth.