Halo effect
Claim assessed: A visible positive trait, especially good looks, leads people to credit someone with other qualities, such as intelligence or honesty.
Published · how we work
In short
The halo effect means that a positive trait that stands out “lights up” the rest of the person: if someone is good-looking, we assume they are also intelligent, trustworthy or capable. It is one of the best-replicated ideas in social psychology. In a study with over 11,000 participants from 45 countries, faces judged more attractive were also judged more intelligent, trustworthy and responsible, in all 11 regions of the world. An experiment with beauty filters shows the link is causal: the same person, beautified, seems more intelligent. And another study shows the impression isn’t an accurate cue.
What it says
The origin. The term “halo” was first used by Thorndike (1920), according to Kordsmeyer et al. (2026): a trait that stands out influences how the others are judged, even when there is enough information to judge them separately. The most studied form since then is the beauty halo, summed up by Dion, Berscheid and Walster (1972) in the title “What is beautiful is good”. We describe both from the sources we read. Kordsmeyer et al. note, citing the literature, that the effect is largely automatic and unconscious.
All over the world. Batres and Shiramizu (2022) used the data collected by the Psychological Science Accelerator network: 11,570 participants from 45 countries, grouped into 11 regions, rated the same 120 faces from standard photographs with neutral expressions. Each participant rated a single trait, so those judging attractiveness were not the same people judging intelligence. The analysis code was preregistered. Faces judged more attractive were also rated more confident, emotionally stable, intelligent, responsible, sociable and trustworthy, for both male and female faces, in all 11 regions.
It is causal. A correlation across faces doesn’t show that looks are the cause: perhaps some faces really have something that also suggests intelligence. Gulati et al. (2024) tested this directly: 2,748 participants rated photographs of 462 people, either in the original version or after an AI beauty filter. Participants weren’t told that half the photographs were beautified. The filter raised perceived attractiveness, and the same people were also judged more intelligent, trustworthy, sociable and happy. An unexpected detail: after the filter, the link between looks and perceived intelligence weakened, as if the effect had a ceiling.
And across cultures. Kordsmeyer et al. (2026), in a preregistered study not yet peer-reviewed, asked raters from Brazil and Germany to judge 583 faces of Brazilians and Germans. The beauty halo was strong and largely the same in both countries, larger for health, social competence and familiarity than for intelligence, prosociality or social dominance.
The impression isn’t accurate. Talamas, Mavor and Perrett (2016) compared impressions of 100 students, judged from a photograph, with their actual grades. Perceived attractiveness did not predict the grades. When the influence of looks was taken out, impressions of how conscientious someone is became more accurate. The authors conclude that we are “blinded by beauty”: the halo makes us less accurate, not more perceptive.
Example
Two candidates with nearly identical CVs come to an interview. One is very good-looking and well dressed. Without realising it, the interviewer finds him more convincing, more organised and more intelligent and gives him higher marks on every criterion, although the conversation went equally well with both. The example is an illustration, not a study cited here.
How to apply it
The steps below are a practical way to apply it that we propose, based on the studies cited.
- Rate traits separately. The halo appears when an overall impression spills over every criterion. Score each criterion in turn, with concrete evidence for each.
- Remove the photo when it doesn’t matter. For CVs, papers or files, physical appearance says nothing about competence, according to Talamas et al., but it does influence judgement.
- Ask what you actually know. When someone seems “obviously” capable, look for the concrete information behind the impression, beyond how they look or speak.
- Watch out for filters. Digitally beautified photographs change not only how attractive someone seems, but also how intelligent or trustworthy.
Limits and nuances
- Measured on photographs, not decisions. The studies cited measure impressions of faces, not hiring, salaries or sentences. According to Batres and Shiramizu, studies with mock jurors find that attractive defendants are less often found guilty and receive lighter sentences, but we haven’t read those studies.
- Not every trait. The effect varies: in the Brazilian–German study it was smaller for intelligence and prosociality than for health and social competence, and in Batres and Shiramizu the link with dominance or caring was weaker and less consistent across regions.
- Gulati et al.’s raters were mostly white, from the UK and the US. The global study partly compensates, but it is correlational.
- We haven’t read the classic meta-analyses. The main syntheses on the beauty halo, by Eagly et al. (1991) and Langlois et al. (2000), are available only behind a paywall. The evidence level rests on the large 45-country replication and the filter experiment.
- What we read. We read in full Talamas et al. (2016) and Gulati et al. (2024), from PubMed Central, Batres and Shiramizu (2022), in a copy on gwern.net of the published version, and the preprint by Kordsmeyer et al. (2026), from OSF. Thorndike (1920) and Dion et al. (1972) we describe from these sources.
Sources
- Edward L. Thorndike (1920). A constant error in psychological ratings
- Karen Dion, Ellen Berscheid, Elaine Walster (1972). What is beautiful is good
- Sean N. Talamas, Kenneth I. Mavor, David I. Perrett (2016). Blinded by beauty: Attractiveness bias and accurate perceptions of academic performance
- Carlota Batres, Victor Shiramizu (2022). Examining the "attractiveness halo effect" across cultures
- Aditya Gulati, Marina Martínez-Garcia, Daniel Fernández, Miguel Angel Lozano, Bruno Lepri, Nuria Oliver (2024). What is beautiful is still good: The attractiveness halo effect in the era of beauty filters
- Tobias L. Kordsmeyer et al. (2026). A cross-cultural comparison of the attractiveness halo effect in men and women
See also: Confirmation bias, Fundamental attribution error, Pygmalion effect