I saw this one linked from a Slashdot story on a planned photorealistic CGI TV series. Japanese roboticist Doctor Masahiro Mori discovered what he calls the Uncanny Valley.
In short, a human-like animal or machine becomes more likeable as it becomes more similar to humans (at first). Past a certain point, however, the likability takes a dive before going back up again (hence the “valley&rdquo).
This chasm &mdashj; the uncanny valley of Doctor Mori’s thesis — represents the point at which a person observing the creature or object in question sees something that is nearly human, but just enough off-kilter to seem eerie or disquieting. The first peak, moreover, is where that same individual would see something that is human enough to arouse some empathy, yet at the same time is clearly enough not human to avoid the sense of wrongness. The slope leading up to this first peak is a province of relative emotional detachment — affection, perhaps, but rarely more than that. […]
So, an industrial robot or even an android are generally perceived positively (they’re on the up-slope before the valley), but a moving corpse would garner a negative reaction (this would be deep in the valley, as it’s just-human-enough to be eerie). If this doesn’t make sense yet, check out the graphs on the site — those should clear things up.
And, that’s why talking animals aren’t threatening — they’re far enough from being human yet their slight-humanness raises their likability without falling into the valley.
These literary tricks have become mainstays, for they work exceedingly well. Public-service advertisements of the eighties and nineties featured an animated anthropomorphic hound in a trench-coat, encouraging cooperation with police agencies and the formation of neighborhood watch programs. With this character, the creators avoided using a human of any recognizable ethnic type, which might have impaired the effectiveness of the campaign in some areas. […]