Why do almost-human faces creep us out, and is the uncanny valley even real?

Nobody agrees on a single reason, and some researchers doubt the effect is one thing at all. The uncanny valley hypothesis predicts that an entity appearing almost human will elicit uncanny or eerie feelings in viewers. The robotics professor Masahiro Mori introduced the concept in 1970, in his book Bukimi No Tani. As he described it, as a robot is made more human the emotional response grows more positive, until it becomes almost human, at which point the response quickly becomes strong revulsion; as the robot becomes less distinguishable from a person, the response turns positive again.

Animation ran into it. According to the roboticist Dario Floreano, the baby character Billy in Pixar's 1988 short Tin Toy provoked negative audience reactions that first caused the movie industry to consider the concept seriously. Final Fantasy: The Spirits Within (2001), one of the first photorealistic computer-animated features, provoked negative reactions from some viewers because of its near-realistic yet imperfect human characters, and several reviewers of The Polar Express (2004) called its animation eerie.

One explanation points to mismatch: the valley is often described as arising when something that looks mechanical moves like a human, or something that looks human moves mechanically. In an fMRI study of an android, researchers saw evidence of mismatch or perceptual conflict: the brain lit up when the android's human-like appearance and its robotic motion did not compute.

But the science is unsettled. A number of theories have been proposed for the mechanism, though support for each is inconsistent, and critics question whether the valley exists as a single phenomenon at all.