What makes a face attractive? What science measures—and what it cannot decide

Editorial responsibility: My Best Version. About MBV · Sources, review and corrections (Portuguese).

English guides · About MBV · Português

A face does not become attractive by following one proportion. Research links attractiveness ratings to features such as facial shape, contrast, and skin appearance, but the strength of those associations depends on the images, observers, and method. Individual preferences also matter. The findings provide useful questions rather than a definitive score for a person.¹,²,⁶

How is attractiveness studied?

Researchers may ask several observers to rate photographs, modify one feature while keeping others as consistent as possible, or measure multiple features in real images and examine their associations with ratings. These designs answer different questions: they describe judgments, test a visual manipulation, or try to disentangle features that normally occur together.¹,²,³

A correlation between a feature and ratings does not mean changing that feature in a person will produce the same effect. Procedures introduce recovery, risks, and individual tissue responses. A group preference is not an individualized aesthetic prescription.

Symmetry and proportions: is there a formula?

Symmetry describes approximate similarity between the sides of a face. The literature has examined symmetry, average shape, and other features, but their relationships with attractiveness depend on the research design. “More symmetrical means more beautiful” is an oversimplification.¹

In 2025, Lee and colleagues examined shape symmetry, averageness, and sexual dimorphism within the same images. Averageness predicted ratings for male and female faces, and femininity contributed in the female images. Symmetry did not make a significant independent contribution in that dataset. This does not establish that symmetry never matters; it illustrates how an apparent relationship can change when other features are considered simultaneously.²

An average shape is a statistical reference within the study’s image set. It is not a universal proportion, and changing the reference set changes the average.

Contrast around the eyes and mouth

Contrast refers to differences in appearance between features and surrounding skin. Studies have explored relative luminance and lip colour, showing that perception is influenced by more than geometry alone. The findings depend on the particular manipulations and observers; they do not prescribe a makeup style or lip colour for everyone.³,⁴

Explore lip shapes and proportions to distinguish contour from visible height and projection. These features can contribute together without being interchangeable.

Skin and visual attention

Fink and colleagues investigated attention to variations in the distribution of female facial skin colour. The study links aspects of visible skin appearance to attention and judgments within its design; it does not mean that a photograph can diagnose health.⁵ A naturally textured face is not a failed version of a filtered one.

See facial harmony for the relationship between surface appearance and facial shape. Uniformity within the skin is different from a hierarchy of skin tones.

Why do preferences differ?

Germine and colleagues examined individual aesthetic preferences and found that differences were associated largely with non-shared environmental experiences in their sample. The analysis estimates sources of variation; it does not calculate an individual’s taste or establish the true beauty of a face.⁶ Familiarity and personal experience help explain why two observers can notice similar features and reach different impressions.

Does attractiveness reliably reveal health?

No. Foo and colleagues examined attractiveness, perceived health, and biological health measures in the same study. Associations between appearance and the health indicators measured were weak or inconsistent and varied across groups and measures. “Looks healthy in this photograph” must be distinguished from clinical health assessment.⁷

Photography changes the comparison

Ward and colleagues used a mathematical model to examine short-distance photographic distortion. Distance can change the apparent relationship between the nose and facial width; a selfie is not a direct three-dimensional measurement.⁸ See nose shapes for practical distinctions between length, projection, and tip rotation.

Use an analysis critically

Ask what is being observed, how it was measured, in which population, and with what limitations. A visual illustration can clarify an idea, but the research evidence comes from the cited studies—not from the AI-generated portraits.

MBV’s visual analysis is illustrative and does not diagnose conditions or recommend treatment. Read about our approach, explore dark circles, or return to the English guides.

Frequently asked questions

Is there a universal ideal facial proportion?

The studies discussed here do not provide a single equation that defines beauty for every person. Their conclusions must be read within their samples and methods.¹,²

Does an AI before-and-after prove that a procedure works?

No. A synthetic image explains a visual possibility. It is not evidence of efficacy, safety, durability, or an individual clinical outcome.

References

1. Little AC, Jones BC, DeBruine LM. Facial attractiveness: evolutionary based research. Philos Trans R Soc Lond B Biol Sci. 2011;366(1571):1638–1659. doi: 10.1098/rstb.2010.0404. Full text.

2. Lee P, Li J, Rafiee Y, Jones BC, Shiramizu VKM. Further evidence that averageness and femininity, rather than symmetry and masculinity, predict facial attractiveness judgments. Sci Rep. 2025;15:5498. doi: 10.1038/s41598-025-86974-0. Full text.

3. Russell R. Sex, beauty, and the relative luminance of facial features. Perception. 2003;32(9):1093–1107. doi: 10.1068/p5101.

4. Stephen ID, McKeegan AM. Lip colour affects perceived sex typicality and attractiveness of human faces. Perception. 2010;39(8):1104–1110. doi: 10.1068/p6730.

5. Fink B, Matts PJ, Klingenberg H, Kuntze S, Weege B, Grammer K. Visual attention to variation in female facial skin color distribution. J Cosmet Dermatol. 2008;7(2):155–161. doi: 10.1111/j.1473-2165.2008.00382.x.

6. Germine L, Russell R, Bronstad PM, Blokland GAM, Smoller JW, Kwok H, et al. Individual aesthetic preferences for faces are shaped mostly by environments, not genes. Curr Biol. 2015;25(20):2684–2689. doi: 10.1016/j.cub.2015.08.048. Full text.

7. Foo YZ, Simmons LW, Rhodes G. Predictors of facial attractiveness and health in humans. Sci Rep. 2017;7:39731. doi: 10.1038/srep39731. Full text.

8. Ward B, Ward M, Fried O, Paskhover B. Nasal distortion in short-distance photographs: the selfie effect. JAMA Facial Plast Surg. 2018;20(4):333–335. doi: 10.1001/jamafacial.2018.0009. Full text.

Educational information. This English edition is an editorial adaptation of the Portuguese guide, retaining the cited sources and their limitations. External specialist review has not been attributed. For corrections, contact contato@mybestversion.com.br.