Halo Effect: Definition, Examples and the Experiments Behind It

By the IQ Revealed Team · Updated November 2, 2026

The halo effect is the tendency for one good impression of a person, product or organisation to colour how you judge its other, unrelated qualities. Someone who is attractive, warm or confident gets rated as more intelligent, more competent and more trustworthy than the evidence supports. The psychologist Edward Thorndike named it in 1920. The striking part is not that first impressions matter, but that people rarely notice the spill-over happening, and that faces turn out to be a very poor guide to how intelligent someone actually is.

The halo effect at a glance

What happens: one salient trait sets a general impression, and the general impression then shapes judgements of every other trait.
The reverse: the horn effect, in which one negative trait drags everything else down.
Best-known form: the attractiveness halo, often summed up as “what is beautiful is good”.
Where it matters: grading, hiring, performance reviews, first dates, brands and food labels.
Family: it is one of the classic cognitive biases, closely related to confirmation bias.

Where the name comes from: Thorndike, 1920

Edward L. Thorndike studied the ratings that superiors gave the people under them, including army officers rating their soldiers on separate qualities such as physique, intelligence, leadership and character. If the qualities were judged independently, the ratings should have been only loosely related. Instead they moved together far too closely and too evenly. Thorndike concluded that raters were forming one overall sense of a man as rather good or rather poor, and then letting that feeling colour each specific judgement. He called it “a constant error in psychological ratings” and gave it the image that stuck: a halo that makes everything about the person look brighter.

The warm teacher, the cold teacher

The most elegant demonstration came in 1977. Richard Nisbett and Timothy Wilson showed students one of two videotaped interviews with the same college instructor, who spoke English with a European accent. In one he was warm and friendly; in the other, cold and distant. Everything else was identical.

Students who saw the warm version rated his appearance, mannerisms and accent as appealing; those who saw the cold version rated the very same features as irritating. Then came the telling part. The students were unaware of the influence. Those who saw the cold instructor insisted that their dislike of him had not affected their ratings of his features, and that it was the other way round: his irritating features had made them dislike him. They had the direction of cause exactly backwards. The halo works below awareness, which is why simply trying to be fair does not remove it.

Beauty: “what is beautiful is good”

In 1972 Karen Dion, Ellen Berscheid and Elaine Walster showed people photographs of faces that differed in attractiveness. The attractive faces were assumed to have more socially desirable personalities and to be headed for better lives, including more successful careers and marriages. The paper's title, “What is beautiful is good”, became the shorthand for the whole idea.

Two later meta-analyses refined it. In 1991 Alice Eagly and colleagues found the effect was real but moderate on average, and uneven. It was largest for social competence, intermediate for adjustment, potency and intellectual competence, and close to zero for integrity and concern for others. It also shrank when people were given other information about the person. In 2000 Judith Langlois and colleagues, combining 11 meta-analyses, found that attractive children and adults are judged and treated more positively even by people who know them, and also, honestly reported, that attractive people on average show somewhat more positive behaviours and traits. Part of that could itself be a result of being treated better.

Does looking smart mean being smart?

This is where the halo meets intelligence directly, and the answer is mostly no.

St Andrews, 2016. Sean Talamas, Kenneth Mavor and David Perrett had people rate standardised photographs of 100 university students. Attractiveness correlated at r = 0.81 with perceived intelligence but at r = 0.03 with actual academic performance. Judgements of conscientiousness were a better guide to real results than judgements of intelligence, and became more accurate once the influence of attractiveness was removed.
Across the lifespan, 2002. Leslie Zebrowitz and colleagues compared judgements of intelligence from facial photos with measured IQ. Judges did better than chance for children and at puberty, only marginally in middle adulthood, and no better than chance for adolescents and older adults. What accuracy there was could be explained by judges relying on attractiveness.
Prague, 2014. Karel Kleisner and colleagues at Charles University photographed 40 men and 40 women who had taken an IQ test. Raters judged the men's intelligence better than chance but not the women's, and the study found no relation between measured intelligence and either attractiveness or face shape.

Put together: people confidently read intelligence into faces, the reading is driven largely by attractiveness, and attractiveness is a weak or useless signal of measured ability. A face is not an IQ test. If you want to know what an actual test measures, see how IQ tests work and the signs of high intelligence that research does support.

Beauty filters: the halo, recalibrated

A 2024 study by researchers in Alicante, Castellón, Barcelona and Trento tested the halo on filtered selfies. Some 2,748 participants rated photographs of 462 people, either as taken or after an AI beauty filter. The same individuals received significantly higher ratings for attractiveness and also for intelligence and trustworthiness once filtered. The authors also found the halo was weaker among the beautified faces, and raised ethical concerns about how widely such filters are used.

Beyond people: brands, labels and food

The halo is not only about faces. In 2015 Patrik Sörqvist and colleagues at the University of Gävle in Sweden found that people rated identical bananas, grapes and raisins as tasting better when they were labelled as eco-friendly, and judged them healthier and worth paying more for. With plain water, the label no longer changed taste ratings, but it still raised willingness to pay and beliefs about health and even mental performance benefits. Marketers speak of a health halo and a brand halo: a reputation earned in one product or one quality is lent to the next.

In the classroom and at work

In a 2023 experiment at the University of Hamburg, Fabian Schmidt, Aurelia Kaiser and Jan Retelsdorf asked 107 participants to grade the same student in two subjects. The first description showed weak, average or strong work; the second was always average. The quality of the first piece significantly predicted the grade given for the second, even after controlling for the participants' and the student's gender. The authors concluded that teacher training should cover the sources of the halo effect.

The same mechanism runs through job interviews and annual reviews: a polished manner, a prestigious university or one outstanding project can lift ratings of unrelated skills, while one awkward moment can sink them.

The horn effect: the halo in reverse

A single negative trait can work the other way. A cold manner, an unfamiliar accent, a typo in a cover letter or an untidy appearance can drag down judgements of competence, honesty or intelligence that have nothing to do with it. The 1977 experiment showed both directions with one man: warmth made his accent charming, coldness made the same accent grating. A 2006 study by Alison Griffin and Judith Langlois, comparing faces of low, medium and high attractiveness, found that most often being unattractive was a disadvantage rather than being attractive an advantage: the horn may be stronger than the halo.

Halo, confirmation and first impressions

Confirmation bias keeps a halo alive: once you have decided someone is impressive, you notice and remember the evidence that fits.
The availability heuristic is about how easily examples come to mind; the halo is about one impression spreading to unrelated traits. They can compound when the memorable example is also the flattering one.
Cognitive dissonance helps explain why a halo resists contrary evidence: admitting that a person or brand you admire got something badly wrong is uncomfortable.

Does a high IQ protect you?

There is no evidence that it does. The students in the 1977 experiment were university undergraduates and still misjudged the direction of their own reasoning. More broadly, Keith Stanovich and Richard West reported in 2008 that many classic thinking biases were largely unrelated to cognitive ability. The halo effect itself was not among the biases they tested, so the honest summary is that a high IQ has not been shown to protect against it. Our pages on critical thinking and the Dunning-Kruger effect look at other ways capable people misjudge.

How to reduce the halo effect

Judge one quality at a time. Rate every candidate, essay or supplier on one criterion before moving to the next, instead of rating each one on everything at once.
Fix the criteria in advance. Write down what good looks like for each quality before you meet the person or see the product.
Remove what should not count. Anonymous marking and CVs without photographs take the most common halo triggers out of view.
Delay the overall verdict. Add up the separate ratings first; form the general impression last.
Ask for the evidence per trait. “What makes me think she is a good analyst?” often reveals that the answer is “she presents well”.
Do not rely on awareness. People in the warm and cold teacher study were sure they had not been influenced. Change the procedure rather than trusting good intentions.

Trust and scope notes

This page summarises published research for general information, with figures taken from the studies listed below. The IQ Revealed test is an online reasoning test of 40 original items for adults aged 18 and over; it does not measure the halo effect or any other thinking bias, and it does not judge anyone by appearance. Our guide to how IQ tests work explains what an online test can and cannot tell you.

IQ Revealed is not affiliated with, endorsed by or connected to the researchers, universities, journals or publishers named on this page. Names are used only to identify the research being described.

Frequently asked questions

What is the halo effect?＋

The halo effect is the tendency for one good impression of a person, product or organisation to spill over into judgements of its other, unrelated qualities. Someone who is attractive, warm or confident is also rated as more intelligent, competent or honest than the evidence supports. The psychologist Edward Thorndike named it in 1920 after finding that officers' ratings of their soldiers' separate qualities ran together far more than they should.

What is an example of the halo effect?＋

In a 2016 study at the University of St Andrews, people rated photographs of 100 students. How attractive a face looked correlated at r = 0.81 with how intelligent it was judged to be, but only at r = 0.03 with the students' real academic results. Everyday examples follow the same pattern: a charming job candidate is assumed to be capable, a well-liked brand's new product is assumed to be good, and a food labelled organic or vegetarian is assumed to be healthier or lower in calories.

What is the difference between the halo effect and the horn effect?＋

They are the same bias pointing in opposite directions. In the halo effect, one positive trait makes everything else about a person look better. In the horn effect, sometimes called the reverse halo, one negative trait, such as a cold manner or a scruffy appearance, drags down judgements of everything else. Thorndike already noted both directions in 1920, and the warm-versus-cold teacher experiment of 1977 showed both at once.

Who discovered the halo effect?＋

Edward L. Thorndike, the American psychologist, gave it its name in his 1920 paper “A constant error in psychological ratings” in the Journal of Applied Psychology. He found that when superiors rated the same people on several different qualities, the ratings were far too closely related, as if each rater formed one general impression and coloured every specific judgement with it.

Are attractive people really seen as smarter?＋

Yes, on average, though the effect is smaller than the phrase “what is beautiful is good” suggests. A 1991 meta-analysis found that attractive people are credited with more intellectual competence than unattractive people, a moderate difference, with the largest gap for social skills and almost none for honesty or concern for others. Whether they are actually smarter is a different question: studies that measured intelligence directly found little or no link with how attractive a face looked.

How can you avoid the halo effect?＋

Separate the qualities you are judging. Rate one quality for every candidate, essay or option before moving on to the next, use written criteria fixed in advance, and remove information that should not matter, such as names or photographs, where you can. Postpone your overall verdict until the separate ratings are in. Knowing about the bias is not enough on its own: in the 1977 experiment, people were confident it had not affected them.