Key findings
- 01The first behavioural study linked a higher fWHR in men to more aggressive behaviour, in a laboratory task and in ice hockey penalty minutes.2
- 02Two meta-analyses find a positive link with aggression or threat behaviour in men. One calls it small. The other puts it at r = .16.3, 4
- 03The difference between men and women is very small in pooled data (d = .11 across 10,853 people). One study that used three measurement methods found none.4, 5
- 04In 137,163 people fWHR was not substantially linked to any of 55 self-report measures of personality and behaviour.6
- 05Pooled ratings do not show that a higher fWHR looks better. Faces with a higher fWHR were judged less attractive (r = -.26).4
- 06The number depends on the photo. The same people measured a smaller fWHR in portraits taken with a shorter focal length.7
Section 1
What fWHR is
The facial width-to-height ratio is the width of the face divided by the height of the upper face.4 Width is taken across the cheekbones. Height runs from the brow to the upper lip. A wider, shorter midface gives a higher number.
fWHR has drawn a great deal of research attention since it was proposed as a visible sign of aggression, dominance and testosterone in men.5 Those claims now circulate well beyond the research that produced them. This review sets out what the studies found, how large the effects are and where later work failed to find them.
| Claim | What the studies report | Evidence |
|---|---|---|
| Men with a higher fWHR behave more aggressively | Two meta-analyses find a positive link. It is small. One estimate for threat behaviour is r = .16.3, 4 | Consistent |
| People read a higher fWHR as more threatening and dominant | Yes in pooled data. Higher fWHR faces were judged more threatening (r = .46) and more dominant (r = .20).4 | Consistent |
| Men have a higher fWHR than women | A pooled estimate across 10,853 people is d = .11. A study using photographs, 3D scans and direct measurement found no difference.4, 5 | Mixed |
| fWHR reflects testosterone | One study reported that facial width was associated with testosterone in men. A study of adolescents found that fWHR did not follow testosterone through puberty.8, 9 | Mixed |
| fWHR predicts personality and everyday behaviour | No. In 137,163 people it was not substantially linked to any of 55 self-report measures.6 | Not supported |
| A higher fWHR looks more attractive | Pooled ratings point the other way. Higher fWHR faces were judged less attractive (r = -.26).4 | Not supported |
| fWHR is a fixed property of a face | No. It was smaller in photos taken with a shorter focal length. It also rises with body mass index.4, 5, 7 | Not supported |
Consistent: several independent studies point the same way. Mixed: studies point in different directions. Limited: one or two studies, not yet repeated. Not supported: tested, with no effect found or one too small to matter.
Section 2
Where the idea came from
The measurement comes from a study of skulls and facial growth. Weston and colleagues reported that the height of the upper face, between the lip and the brow, develops differently in males and females at puberty. The difference was not explained by body size.1
Carré and McCormick then asked whether the ratio was linked to behaviour.2 Their paper reports three studies.
- In the first, men had a larger fWHR than women, scored higher on trait dominance and were more reactively aggressive in a laboratory task. Differences in fWHR predicted reactive aggression in men and accounted for 15% of the variance. They did not predict it in women.
- In the second, fWHR in male university hockey players was positively related to penalty minutes per game over a season and accounted for 29% of the variance.
- In the third, the same relation held in male professional hockey players and accounted for 9% of the variance.
The authors suggested that fWHR may be an honest signal of a tendency to aggressive behaviour. Many studies of fWHR and behaviour have been published since.3, 6 The estimates in this first paper, 9% to 29% of the variance in behaviour, are far larger than the pooled estimates that followed.
Section 3
What pooled data show
Single studies disagree, so the more useful evidence is the two meta-analyses published in 2015.
Haselhuhn and colleagues pooled the existing work on fWHR and aggression in men. They note that some studies had failed to repeat the original finding. Their result was a small but significant positive relationship.3
Geniole and colleagues ran a wider set of meta-analyses.4 Their figures:
- fWHR was larger in men than in women: d = .11 across 10,853 people.
- fWHR predicted threat behaviour in men: r = .16 across 4,603 people.
- fWHR predicted dominance behaviour in both sexes: r = .12 across 948 people.
- fWHR was related to body mass index: r = .31 across 2,506 people.
- Observers judged higher fWHR faces as more threatening (r = .46), more dominant (r = .20) and more masculine in men (r = .35).
- Observers judged higher fWHR faces as less attractive (r = -.26), most of all when women made the judgement.
Two things follow. The link with behaviour is real in pooled data and it is small. A correlation of .16 means fWHR accounts for under 3% of the differences in threat behaviour between men. The link with perception is several times stronger. People respond to a wide, short face more reliably than that face predicts what its owner does.
The authors add a caution about their own perception results. Few of the faces used as stimuli were female or older adult faces.
Section 4
Large samples and null results
Self-reported behaviour in 137,163 people
Kosinski tested the behavioural claims in a very large sample.6 The paper describes the earlier studies as mostly laboratory based and low powered. It compared fWHR with 55 established psychometric scales in 137,163 participants. The scales covered intelligence, the five-factor model of personality, impulsiveness, sense of fairness and satisfaction with life, among others.
fWHR was not substantially linked to any of them. The measures were self-reported and not observed behaviour, which limits how far the result reaches. It still questions whether the laboratory findings carry over to people in general.
The sex difference
Kramer and colleagues measured fWHR in three white European samples by three methods: 2D photographs, 3D scans and direct anthropometry.5 Measuring the same people several ways gave high agreement between methods. The study found no evidence of a difference between men and women in the face, although it had enough statistical power to detect one. It did find a link between fWHR and body mass index in both sexes.
Set beside the pooled estimate of d = .11, the sex difference in living faces is somewhere between very small and absent.4 Observers do read a higher fWHR as more masculine. The measured difference between the sexes is far smaller than that impression suggests.
Section 5
Testosterone
The proposed mechanism behind fWHR is testosterone at puberty. The evidence for it is thin and divided.
One study reported that facial width was associated with testosterone levels in men.8 We could not obtain an indexed abstract for that paper, so we report only what its title states.
Hodges-Simeon and colleagues tested the idea during adolescence, the period when facial growth is set.9 They studied adolescent males from the Tsimane population and set out five predictions that should hold if testosterone shapes fWHR at that age.
- fWHR should rise with age as testosterone rises.
- fWHR should reflect testosterone in adolescence.
- The growth of fWHR should speed up in step with the pubertal rise in testosterone.
- fWHR should correlate with testosterone once age and other factors are controlled.
- fWHR should be strongly tied to other testosterone-dependent traits, such as strength and voice pitch.
Only the fourth was observed.
Three other facial ratios in the same study did track age, testosterone and testosterone-dependent traits. The authors concluded that their results question fWHR as a signal of pubertal testosterone.
Section 6
Why one face gives different numbers
fWHR is a ratio of two distances on a photo. Anything that changes the photo changes the ratio.
The camera
Třebický and colleagues took three portraits of each of 45 people under standard conditions, changing only the focal length: 50, 85 and 105 mm.7 Shorter focal lengths produced faces with a smaller fWHR. Overall face shape changed as well. The authors point to perspective distortion as the likely cause. Our review of camera distance covers that effect in full.
Body mass
fWHR rises with body mass index. The pooled correlation is r = .31.4 The three-method study found the same link in both sexes.5 A change in weight can therefore move fWHR without any change in bone.
The method
The reassuring result is that careful methods agree. Photographs, 3D scans and direct measurement gave closely matching values for the same people.5 A photo can measure fWHR well. It has to be a standard photo.
LooksMeter measurements
LooksMeter measures fWHR as cheekbone width divided by the height from mid-brow to the upper lip. Across 1,404 faces in our facial measurement norms study the median was 1.63. The middle half of faces fell between 1.54 and 1.73. 11.8% fell inside the band of 1.8 to 2.1 that is often quoted for the ratio.
Camera distance was not controlled in that dataset and many of the photos are likely to be selfies. Given the focal length result above, that may pull our figures down. We have not tested it. The practical point is that a figure quoted from studio photographs and a figure measured from a selfie are not on the same scale.
Section 7
Reading your own fWHR
- It is a real proportion of your face in that photo. It describes how wide the midface is for its height.
- It says very little about character. The pooled link with behaviour is under 3% of the variance and the largest study found no substantial link with any self-report measure.
- It is not an attractiveness score. In pooled ratings a higher fWHR was judged less attractive, not more.
- It is not a testosterone reading. The studies disagree and the most direct test in adolescents did not support the idea.
- It moves with the photo and with body weight. Compare two readings only when the photos were taken the same way.
Section 8
Limits of this review
- This is a narrative review of selected peer-reviewed studies. It is not a systematic review and it may have missed relevant work.
- Each study is summarised from its published abstract. One paper had no indexed abstract and is reported by its title alone.
- Most fWHR research has studied men. The meta-analysis notes that female and older faces are scarce in the perception studies.
- Correlations describe groups. They do not predict what any one person is like.
- The LooksMeter figures come from self-selected users and single photos, with no demographic split. The norms study sets out those limits in full.
References
- 1Weston EM, Friday AE, Liò P. Biometric evidence that sexual selection has shaped the hominin face. PLoS ONE. 2007;2(8):e710. doi:10.1371/journal.pone.0000710 · PubMed
- 2Carré JM, McCormick CM. In your face: facial metrics predict aggressive behaviour in the laboratory and in varsity and professional hockey players. Proceedings of the Royal Society B: Biological Sciences. 2008;275(1651):2651-2656. doi:10.1098/rspb.2008.0873 · PubMed
- 3Haselhuhn MP, Ormiston ME, Wong EM. Men's facial width-to-height ratio predicts aggression: a meta-analysis. PLOS ONE. 2015;10(4):e0122637. doi:10.1371/journal.pone.0122637 · PubMed
- 4Geniole SN, Denson TF, Dixson BJ, Carré JM, McCormick CM. Evidence from meta-analyses of the facial width-to-height ratio as an evolved cue of threat. PLOS ONE. 2015;10(7):e0132726. doi:10.1371/journal.pone.0132726 · PubMed
- 5Kramer RSS, Jones AL, Ward R. A lack of sexual dimorphism in width-to-height ratio in white European faces using 2D photographs, 3D scans, and anthropometry. PLoS ONE. 2012;7(8):e42705. doi:10.1371/journal.pone.0042705 · PubMed
- 6Kosinski M. Facial width-to-height ratio does not predict self-reported behavioral tendencies. Psychological Science. 2017;28(11):1675-1682. doi:10.1177/0956797617716929 · PubMed
- 7Třebický V, Fialová J, Kleisner K, Havlíček J. Focal length affects depicted shape and perception of facial images. PLOS ONE. 2016;11(2):e0149313. doi:10.1371/journal.pone.0149313 · PubMed
- 8Lefevre CE, Lewis GJ, Perrett DI, Penke L. Telling facial metrics: facial width is associated with testosterone levels in men. Evolution and Human Behavior. 2013;34(4):273-279. doi:10.1016/j.evolhumbehav.2013.03.005
- 9Hodges-Simeon CR, Hanson Sobraske KN, Samore T, Gurven M, Gaulin SJC. Facial width-to-height ratio (fWHR) is not associated with adolescent testosterone levels. PLOS ONE. 2016;11(4):e0153083. doi:10.1371/journal.pone.0153083 · PubMed
Every reference was checked against Crossref before publication. Those indexed in PubMed were checked there as well. Findings are summarised in our own words. Read the original before relying on a number.
How to cite
LooksMeter Research. Facial width-to-height ratio (fWHR): what it predicts and what it does not. Version 1.1. 18 November 2026. https://www.looksmeter.com/research/fwhr-evidence
About this document
LooksMeter is a consumer appearance analysis product. This document is written by the LooksMeter team for general information. It has not been peer reviewed or reviewed by a medical professional. The measurements it describes are not clinically validated and are not a diagnosis or a medical device. See a qualified clinician about any health or surgical decision.
Corrections are welcome at [email protected]. Material changes are recorded by raising the version number above.
Continue