The future of domestic and service robots will look more like Roomba, and less like, well, less like humans. While we might find a mechanical face cute when it appears spontaneously from the headlights and grill of a car, we get totally creeped out by robots that ape human form. A new study, published in the International Journal of Social Robotics, tells us why.
We know that human facsimiles make us uncomfortable, whether it’s a humanoid robot, or ventriloquist’s dummy, or a dimly lit warehouse full of naked showroom mannequins: This is what’s called the “uncanny valley.” The closer it is to a human–but not a human–the more uncomfortable they make us. Why? The reason, revealed by authors Francesco Ferrari, Maria Paola Paladino, and Jolanda Jetten, is that they threaten our identities.
The idea is that when robots either look like us, the similarity “blurs category boundaries, undermining human uniqueness.” In the study, participants were shown pictures of three kinds of robot: Robots that looked nothing like humans, just automatic machines; humanoid robots that walked on two legs and “somewhat” resembled humans; and androids, the robotic versions of people that are a staple of science fiction.
While they were looking at photos, the researchers asked the participants about the robots’ potential for human damage. Not surprisingly, the human-mimicking androids scored worst.
“Robot human-likeness directly increases the perception of robot as a source of danger to humans and their identity,” says the report. “The more the robot’s appearance resembles that of a real person, the more the boundaries between humans and machines are perceived to be blurred.”
In this case, perception is everything, and the study got similar results to a Japanese study where people were shown videos of androids “in everyday situations.” In that study, say the authors, “Many comments referred to the fear that these robots would be used for evil, and that their presence would threaten human relations.”
The problem may be that, as these robots get ever more human-like, they force us to reconsider what it is to be human, “forcing us to redefine ourselves, and humanness in general.” That’s not to say we realize this and sit pondering it. Quite the opposite. The creepiness stems more from an instinct that something isn’t right.
These findings are consistent with the idea that worries and concerns about the impact on human identity of highly human-like social robots are related to the fact that these robots look so similar to humans that they can be mistaken to be one of us.
It’s worth remembering that the study was conducted by showing the participants still photographs, and not video or even live robots, which might cause an even stronger reaction. Another point to consider is how uncomfortable we might feel giving orders to our robots when they look just like us. It’s one thing to shout at and punch your computer when it crashes just before you save that PowerPoint presentation. It’s another to kick and insult a robot servant that resembles a real human being.
And that’s what makes studies like this important. As the world’s population ages, we will become more reliant on robots to care for the elderly. Knowing how to build those robots so they don’t scare the bejesus out of their wards might become a big design problem in the 21st century.