The concept of the Uncanny Valley has its origins in early 20th century psychology. German psychologist Ernst Jentsch’s 1906 paper, titled “On the Psychology of the Uncanny“, explores ideas discussed by Sigmund Freud on childhood fears, such as the fear of dolls coming to life, known as pediophobia. More recently, roboticist Masahiro Mori proposed the theory that the more human-like robots become, the closer they get to a critical point where positive acceptance by humans quickly turns to revulsion. The “valley” concept derives from Mori’s chart of the phenomenon, which shows a downward dip in the data along a y axis which rates “familiarity” against “human likeness” (x axis).
Here’s a discussion on the Uncanny Valley (UV) and how the theory impacts on recent media by cognitive psychologist Karl MacDorman (Parts 2-7 can be found by searching YouTube or clicking its “More from:” link:
The recent box-office success of James Cameron‘s film Avatar prompted a lot of commentary on how sophisticated 3D animation has brought us ever closer to the UV precipice. Some CGI production houses, such as Image Metrics have made it their mission to cross the Uncanny Valley by producing graphic animations of human faces that are hard to distinguish from the real thing. Meet IM’s Emily:
Roboticists such as David Hanson have also taken up the challenge of bridging the Valley. His robot torsos of Einstein and SF writer Phillip K. Dick are “uncannily” real. Digital media artist Peggy Weil discusses the UV and the concept of the “blurring test” a kind of “reverse Turing test“, in Fora.TV’s Picnic series, here.
For me, these artefacts are certainly getting us closer to the other side of the Valley, but we’re not quite there yet. If nothing else, they force us to consider the very concept of humanity and what it means to be human. So we’re close, but no cigar.