What makes a face beautiful, and does a “beautiful nose” exist? Popular answers range from ancient golden ratios to social media trends. The research is more modest and in some places more surprising: people agree on faces more than you might expect, but they disagree about noses. This article separates what has been tested from what is repeated, and ends with what it means if you are thinking about changing your own profile.

Key takeaways

  • Meta-analyses find that people agree, within and across cultures, on who is attractive; averageness, symmetry and sexual dimorphism are the traits most often linked to it.
  • Preferences about the nose itself vary: a survey in 50 countries found that regions and surgeons differ from the general public on width, tip and projection.
  • No source supports a single ideal nose, and a hospital leaflet says outright that there is no perfect shape and any change has to suit the rest of the face.

Is there a universal beautiful face?

Partly. A meta-analytic review by Langlois and colleagues in Psychological Bulletin (2000) pooled 11 meta-analyses and concluded that raters agree about who is and is not attractive, both within and across cultures. The same review found that attractive people are judged and treated more positively, which helps explain why appearance carries so much weight in a decision about surgery.

Which features drive that agreement? In an Annual Review of Psychology article, Gillian Rhodes reviewed the evidence and reported that averageness, symmetry and sexual dimorphism are attractive in male and female faces and across cultures. Averageness means a face close to the typical face of a population. Sexual dimorphism means features that differ between the sexes. Rhodes also notes two explanations in the literature, that these preferences signal health or mate quality, or that they are by-products of how the brain processes faces, and argues that both may play a part.

These are tendencies from rating studies. They do not describe any single person, and they say nothing about expression, personality or the effect of context, which no photograph captures.

What about the golden ratio?

Online beauty guides often claim that a face is beautiful when it matches the golden ratio (about 1.618). I did not find a reliable source that tested this claim on faces and confirmed it.

What I did find is a set of four experiments by Pallett and colleagues in Vision Research (2010). They asked people to compare faces whose features were identical but arranged differently. Faces were rated most attractive when the vertical distance between the eyes and mouth was about 36% of the face length and the distance between the eyes about 46% of the face width. The authors describe these as “new” ratios, and they match those of an average face. That fits with the averageness finding above, and it is a different number from the classic golden ratio. It is one set of experiments, with limits on how far it applies to real, moving faces. Our guide to facial proportion and harmony goes into measurements further.

How old is the idea of enhancing the face?

Very old. The Victoria and Albert Museum holds a glass kohl tube from Egypt, dated to the New Kingdom (about 1392 to 1334 BC), and its gallery label says the vessel was intended to hold cosmetic eye paint. Grooming and adornment of the face were already organised crafts more than three thousand years ago.

Popular histories of beauty usually go on to list what Greeks, Romans, medieval Europeans and Victorians admired. I did not find a single checked source for those summaries, so I leave them out rather than repeat them. For the history of surgery on the nose itself, see our guide to the history of rhinoplasty, which limits itself to dates that sources support.

Is there an ideal nose?

The best-tested answer is no, and preferences differ by place and by who is asked. Broer and colleagues surveyed more than 13,000 plastic surgeons and lay people in 50 countries using computer-altered images of one nose, with a response rate of 9.6%. They published the results in 2012 in Plastic and Reconstructive Surgery:

  • Plastic surgeons from Latin America and the Caribbean overall preferred smaller and narrower noses with more projecting tips, lips and chins, and the general public in that region showed similar trends.
  • Plastic surgeons and the general public differed significantly. Surgeons preferred wider nasal roots and tips and, in combination, more projected dorsi, tips, lips and chins.
  • The authors state that ideal measurements suggested in the literature “might not apply on a cross-cultural basis”.

The survey used images of a single model and had a low response rate, so it shows that opinions differ rather than which opinion is right. It also means that “ideal angle” tables you see online, such as numbers for the nose-to-lip angle, should be read with care. I did not verify those figures against a primary source, so I have not repeated them.

A hospital leaflet from West Suffolk NHS Foundation Trust (issued 2019, review date 2022) puts it plainly: there is no perfect shape to the nose and any alteration has to fit and suit the rest of the face. The same leaflet warns that people who believe cosmetic surgery will change their lives are often disappointed.

What does this mean if you are planning surgery?

Three practical points follow from the research.

First, agree on a goal that fits your own face. Our guide to nose types shows how features such as the bridge, tip and width combine, and our piece on natural-looking rhinoplasty explains what surgeons mean by that term.

Second, a preferred look can reflect trends in your region, your age group or your social feed, as much as your own taste. Bring several reference images rather than one, and ask your surgeon what they see in your own anatomy that would limit or support each.

Third, cultural and ethnic features are part of a face. If you want to keep them, say so early. Our overview of ethnic rhinoplasty covers how this conversation can go, and simulation before a nose job explains why a morphed image is a poor promise of a result.

Sources

All pages were opened and checked on 10 October 2026.