Robot Acceptance
Can the Uncanny Valley be extended beyond aesthetics?
Background
The Uncanny valley is a way to understand why we think some robots are adorable and some are creepy. In general we like robots with human features. The more human features, the more we like the robot. However, the Uncanny Valley is the exception to this rule. In the valley, robots
that look almost human, but not quite perfect. The robot's appearance is said to be eerie.
Challenges
Uncanny valley previously only applies to aesthetics. Could the theory of the Uncanny Valley be extended beyond the appearance of the robot? To extend the theory would require a deeper understanding of the specific emotions driving the Uncanny Valley reaction. Additionally, this project required a measure of humanness. How could we accurately decide how human a specific task was?
Actions
-Fear and Disgust were discovered to be the two driving emotions for the Uncanny Valley. Scales were created from the psychological literature focused on these emotions.
-To prevent aesthetics from playing a role in the emotional reaction of participants, images were not used. Instead, each robot was given a rich description focused on the task the robot is used to complete, not its appearance.
-Finally, a scale created to measure the level of humanness we project onto other humans who are members of different social groups was used to measure the humanness projected onto these robots.
Outcome
In general, participants disliked robots completing tasks considered to be human-like. However, the relationship was not linear. Instead, a valley in likability was created when a task was slightly human. The graph is similar to the traditional Uncanny Valley if the traditional graph was flipped 180 degrees. This project won a National Top Paper award.
