In “Navigating a social world with robot partners: A quantitative cartography of the Uncanny Valley”, Mather and Reichling argue that there is a hypothetical uncanny valley (UV) in which imperfect human-likeness evokes feelings of dislike and hatred. They demonstrate this by researching the likability of 80 real-word robot faces and complementary digitally edited ones.

The study confirms that the uncanny valley influences people’s implicit decisions concerning the trustworthiness of robots and that these decisions depend on the same subtle cues and facial expressions that humans use to judge other humans.

Moreover, the evidence they produce suggests that, although some elements of traditional social psychology govern human-robot social interaction, the effects of an uncanny valley might diminish the impact of modern android designs.

I’m not so into robots, let alone ~ humanoid ~ robots, but research on how humans interact with them will undoubtedly be used to inform a new generation of technology designers. Whether it’s tamagotshis or personal assistants, artificially intelligent robots are bound to be produced in humanoid form, if for no other reason than to satisfy humanity’s desire to turn science fiction into reality. How we interact with those humanoids – whether they are friend or foe – could depend on psychological factors, like the uncanny valley, that are not yet well understood.