The most attractive face in a crowd is often the least unusual one.
It sounds backwards. We tend to imagine beauty as something rare and striking, a face unlike any other. Yet decades of research point the opposite way. The faces people rate as most appealing tend to sit close to the mathematical middle of all the faces around them, and that is one of the odder facts about how we see each other.
A 1990 experiment turned this strange idea into a measurement.
Judith Langlois and Lori Roggman digitised nearly two hundred faces and used a computer to blend them, pixel by pixel, into composites of 2, 4, 8, 16 and 32 people. Then they asked hundreds of judges to rate everything for attractiveness. The blended faces beat almost every individual face that went into them, and the more faces were mixed together, the more attractive the result became. Their paper carried a title as blunt as the finding, “attractive faces are only average.”
The word average here does not mean plain.
This is where people misread the study. Averaging many faces does not produce a forgettable one, it produces a smooth, balanced, oddly luminous face that looks better than the ordinary faces it came from. The blend quietly cancels out the quirks, the crooked features and the marks of a hard week, leaving a kind of statistical ideal that no single person actually wears.
Symmetry rides along for a related reason.
Averaged faces are also more symmetrical, and symmetry has its own pull on the eye. A face whose halves match closely is generally rated as more attractive than one that leans or tilts. The two qualities travel together, though researchers still debate exactly how much each one contributes on its own.
Two explanations compete, and both may be true.
One is evolutionary. Averageness and symmetry may act as rough signals of healthy development, a body that grew without too much disruption, so the preference could be an old instinct dressed as taste. The other is about the brain itself. We process a face faster when it sits near our mental prototype, and that ease of processing feels pleasant, so what is familiar to compute reads as beautiful to look at.
The twist is that perfectly average is not quite the peak.
Later work complicated the tidy story. The very most attractive faces are not the plain average but the average nudged slightly in certain directions, a little more of this, a little less of that. Beauty, it seems, lives near the centre but not exactly on it, close to the norm with a small, deliberate exaggeration.
For LOOKS, this is a lesson in how perception really works.
The eye is not chasing novelty for its own sake, it is quietly measuring every face against an internal template built from all the faces it has ever seen. Beauty, in this light, is less a fixed object and more a relationship between a face and its crowd. Understanding that is not about ranking people. It is about seeing that the standard we call attractive was written, statistically, by everyone at once.
Sources: Langlois and Roggman, “Attractive Faces Are Only Average,” Psychological Science (1990); Overview of averageness and facial attractiveness research.