"The uncanny valley" - Rory Hamilton at Lightning UX
You surely can't go wrong with a lightning talk that prompts a Twitter response stating: "Zombies make bad UX & monkeys are racist". That was one consequence of Rory Hamilton's talk at Lightning UX last week about "the uncanny valley", which is the basis for my final set of notes from the event, having previously blogged about the talks by Harry Brignull, Alex Horstmann, Cennydd Bowles, Boon Chew, Jonathan Kahn, Lucy Spence and Tyler Tate.
"The uncanny valley" - Rory Hamilton
"The uncanny valley" is a reference to some work by Japanese roboticist Masahiro Mori on how human beings react to robots. He defined a sliding scale which showed how comfortable people felt around them. At one end of the comfort scale was "a healthy human being", and at the other end, the robot most likely to elicit a "meh!" was an industrial production line robot. Somewhere in the middle was a C3PO-type, which was humanoid but still clearly robotic.
What Mori's theory identified was that there is then an area where robots closely resemble humans or human behaviour, but something about their movements or failure to completely mimic human interaction makes them "creepy". In this "uncanny valley" Alex placed fairground automata, and zombies, which, whilst not technically robots, definitely creep most people out.
The reason Rory referenced the theory is that as a service designer, one constant challenge is how to scale customer service in a way that doesn't fall into the trap of irritating people. He cited keypad driven automated customer service over the telephone as one area where despite trying to imitate a natural tone of voice, the instructions invariably grate on customers. Likewise, my personal bugbear is train station announcements where a cut'n'pasted pre-recorded set of fragmented vocal recordings issues an apology to me. Another notable item to go into the "uncanny valley" equivalent of Room 101 was Microsoft's Clippy.
Since I've ended up blogging these notes after also seeing Relly Annett-Baker talking at London IA on Wednesday, I can note a thread they had in common. Rory suggested that Innocent Smoothie packaging was just about tipping over the edge into "the uncanny valley" as a product that tried desperately to ingratiate itself with you. Relly also cited Innocent, suggesting that one of the worst things about not having an authentic voice was when banks wanted to re-brand themselves in an Innocent style. She wants banks to feel secure and responsible in their personality, not kooky.
Next...
That is the end of my round-up of my notes from Lightning UX. Next week I'll have my take on Dan Lockton's presentation "Change, work with, ignore, fail?" at London IA.
When Mori suggests that robots resemble humans but fail to mimic human behaviour makes them creepy, he is just affirming the fact that i personally think is correct; robots are more of zombies. You look at their movements and general behaviour, they really are zombies! There is no way robots can ever resemble humans. Period.
human beings love robots because the robots make their job easier. now let the robot take their job from the human being then ask the human being how they feel about the robot.
"Somewhere in the middle was a C3PO-type, which was humanoid but still clearly robotic."
But was it fluent in over six million forms of communication!!! :)