It's time for our mostly daily Kotaku science post! Today we take a look at an important scientific study that has found that babies can't tell human beings from humanoid robots.
I firmly believe that any scientific study that brings together babies and robots is worthy of our attention. This is mainly because it tickles me to death (but not the babies!) that some scientists somewhere were just sitting around one day and decided, "Hey! We should see if babies can tell Dave from a robot."
This is what I imagine goes on at most if not all science-based social occasions. I'm probably wrong, but I don't really want to know.
In this case, the scientists were researchers at the University of Washington's Institute for Learning and Brain Sciences, and it wasn't Dave, but researcher Rechele Brooks.
Researchers took a sample of 64 18-month old babies and stuck them in a room with Rechele and a remote-controlled humanoid robot. Note that the babies were tested individually, and were sitting in their parents' laps the entire time, because parents are rightfully scared of leaving their children alone with strange robots, and robots (and researchers) are terrified of being alone with 64 babies at once.
Then Rechele and the robot would put on a skit. Rechele would ask the robot scientific questions, such as "Where is your tummy?" The robot would react by pointing at its torso approximately where its tummy would be if it had one.
During the skit, the babies would look back and forth between the human and the robot. Why? Because even an 18-month-old baby knows where a belly is, and marvels over the fact that a grown adult scientist does not.
After the skit is completed, Rechele leaves the baby (and parent) alone with the robot. The robot would then make a beeping noise, shifting about to get the babies attention, before focusing its lifeless eyes on a nearby toy.
13 out of 18 babies would follow the robot's gaze. Scientists believe this is because the children see the robot as a sentient being instead of a cold, unfeeling metal construct, like a toaster or Dick Cheney.
I'm no scientist, but I feel the baby-robot interaction is something slightly different. Without its human companion, the robot beings to feel self-conscious, letting out a whistling sigh and rocking nervously before averting its gaze. The baby, sensing the awkward moment, pretends to be interested in whatever the robot is looking at.
Again I'm no scientist.
In the control portion of the experiment, the robots and babies were left alone without the benefit of an introductory skit. The babies stared at the robot. The robot looked away. All but 3 of 18 babies continued staring.
What have we learned? Babies can be easily fooled into believing a robot is a human being as long as they're provided with a floor show first. Or, more scientifically, socialization plays an important part in a baby's acceptance of robots.
So why did the scientists do this? Oh come on, you'd do the same thing if you had access to a robot and 64 babies.
In New Study, Babies Think A Silvery Robot Is Human, As Long As It Acts Friendly [Popular Science]