These grids of soft, symmetrical portraits went viral in 2011 under headlines promising the "average face" of women in 40 countries. Each portrait is not a real person but a blend of many photographs, digitally aligned and merged into one. The result is fascinating to look at, but it is also easy to read far more into it than the images can support. Here is how the composites were made, what face-averaging research actually shows, and why they should be enjoyed as a visual experiment rather than as a portrait of a nation.
Where the images came from
The composites were made with face-averaging software from FaceResearch.org, a website run by experimental psychologists in Scotland that offers online demonstrations of face research. When the images spread, many sites credited them to psychologists at the University of Aberdeen or the University of Glasgow; PetaPixel, one of the first outlets to share the full set in February 2011, traced it to a blog project called "Make an Average", which had applied the Face Research tools to photos of women from each country.
Two important details were never published: where the source photographs came from and how many faces went into each country's composite. The full chart below also carries a "via 9GAG" watermark, a reminder that it reached most people through meme sites rather than a scientific paper.
The 40 composites
The full chart covers Germany, France, England, the Netherlands, Mexico, Peru, Argentina, Brazil, the Philippines, Cambodia, Vietnam, Thailand, Greece, Italy, Spain, Israel, China (Han), Korea, Japan, Mongolia, Ukraine, Sweden, Poland, Russia, Serbia, the Czech Republic, Hungary, Romania, South India, India, Burma, Samoa, Turkey, Lebanon, Afghanistan, Iran, Chad/Cameroon, Ethiopia, South Africa and West Africa. Some labels are countries, others are regions or groups, such as "South India" alongside "India", "Chad / Cameroon", "West Africa" and "China (Han)", which already shows that the set was assembled informally rather than designed as a systematic survey.
How face averaging works
The idea is older than photography software. In the late 1800s the British scientist Sir Francis Galton layered photographic portraits on a single plate to produce "composite" faces. His versions were blurry because the features did not line up.
Modern tools solve that problem. Key points on each face, such as the corners of the eyes and mouth, the tip of the nose and the jawline, are marked. The software calculates the average position of every point, warps each photo to that shared shape and only then blends the colours. Because the shape is averaged before the images are merged, the result looks sharp and lifelike, even though it belongs to nobody.
Why the faces all look so attractive
Many viewers noticed that almost every composite looks pretty, smooth-skinned and young. That is not a coincidence. Psychology research since the 1990s has repeatedly found that averaged faces tend to be rated as more attractive than most of the individual faces that make them up. Averaging removes asymmetries, blemishes, shadows and unusual features, leaving a face that looks clear and balanced.
The photo selection matters too. If the source pictures were mostly of young women, or of women who had chosen to post photos online, the composite will reflect that group rather than the female population of a country as a whole.
What the composites cannot tell you
- They are not a national standard. Countries such as India, Brazil, Russia or South Africa contain enormous diversity. A single blended face hides almost all of it.
- The sample is unknown. Without knowing how many photos were used, from where, and of whom, the images cannot be treated as data about a population.
- Lighting and style leak in. Hair, make-up, camera quality and backgrounds all shape the final look, which is why some composites have halos of blended hair or tinted skin.
- Nobody looks like this. An average is a mathematical construction. No real woman's face is "wrong" for differing from it.
Why they still fascinate us
Even with their limits, the composites are a striking reminder of how much people around the world have in common. Line the 40 faces up together and the similarities are as noticeable as the differences: the same proportions, the same calm expression, the same human face appearing again and again in slightly different variations. Seen that way, the project is less about measuring nations and more about how alike we all are.
Sources: PetaPixel: Average faces of women in 40 countries (2011); PetaPixel: The average faces of women around the world; Fstoppers: The average faces of women around the world; composites made with software from FaceResearch.org; chart as circulated via 9GAG.
Related:- This is What Americans Will Look like by 2050 – and it’s Stunning
- 16 Children & Their Bedrooms From Around the World…
- Twenty Archive Photographs and the Context Behind Their Captions
- John Lennon’s Imagine in Comic Form
- A Short Comic About Love and Everyday Care
- This 14-Year-Old Boy Just Wrote The Most Important Poem Of The 21st Century
They're all the same.
ReplyDeleteThose who live closely with neighborhood countries can identify easily all faces of the contries. But those who live far from other countries will be difficult to identify faces from those far countries. I can identify easily the japanese, thai, india, chinese, koreans people, but find difficulities to identify the Europe people: france, germany, scott, UK.
ReplyDeleteYou said countries and you put West Africa, Is West Africa a country?
ReplyDeleteUh....the average woman in any country is not as pretty or as young as these are.
ReplyDelete