The Language of New Media, Chapter Four
The “Uncanny Valley” is a phenomenon introduced in the 70’s that describes an inevitable reality of technology: we will create something so realistic and humanlike that it will cause us to recoil and shudder. This will eventually occur most predominantly in the field of robotics but the foundations can be seen in the past 20 years in breakthroughs of computer generated movie special effects. Manovich goes into great detail in chapter four of The Language of New Media about how media and technology have evolved over the centuries to produce photorealistic art that aspires to trick the human eye. An unintended and unanticipated consequence of this pursuit is a sense of “wrongness” on our part: there’s a line that can be crossed when something is just a little too real.
Frank Pollick at the University of Glasgow has gathered research and offered some ideas about the uncanny valley in his paper “In Search of the Uncanny Valley” :
The primary evidence to support its existence comes from research by MacDorman and Ishiguro (MacDorman & Ishiguro, 2006) that explored observers reactions to facial morphs from a mechanistic robot – to a human looking robot – to an actual human. What they found was that at the boundary of the mechanistic robot and the human looking robot there was a rise in judgments of the eeriness of the display that was consistent with judgments of the morph being seen as less human.
Creating the illusion of the “real world” seems to be the goal when constructing virtual worlds and simulated environments, but one thing a computer may not account for is that the real world is not pristine; it has dirt, grime and character. For a while, CG artists were producing scenes that were too clean and had to learn how to gracefully “degrade” the results. Manovich states:
Typical images produced with 3-D computer graphics still appear unnaturally clean, sharp and geometric looking. Their limitations especially stand out when juxtaposed with a normal photograph. Thus one of the landmark achievements of Jurassic Park was the seamless integration of film footage of real scenes with computer-simulated objects. To achieve this integration, computer-generated images had to be degraded; their perfection had to be diluted to match the imperfection of film’s graininess.
I love movies that promise spectacular effects, and like many others, I’m immediately taken out of the experience when I see an effect or other CG that doesn’t look real or like it doesn’t fit into the scene. I can appreciate the advancement of technology, though, and know that bigger and better advancements towards photorealism are on the horizon. The uncanny valley is a ways off but the illusions of modern media are getting more refined and believable every day. I think I want to keep them on the screen though; I’m not thrilled about hyperrealistic robots getting in my face.