When I attended Design Night on robotics, there was a discussion as to whether or not robots would ever have emotions. The audience was split about 50-50 in terms of answering yes and no. Last night I watched the movie, Life of Pi, via Netflix. In my opinion, it contained a related conversation between a father, Sanosh, and his young son, Pi, who almost lost an arm while attempting to feed a tiger by hand:
Sanosh Patel: You think tiger is your friend. He is an animal, not a playmate.
Pi Patel: Animals have souls. I have seen it in their eyes.
Sanosh Patel: When you look into his eyes, you are seeing your own emotions reflected back at you.
And there it is — why robots will one day have emotions. Robots will mimic physical characteristics that we humans will interpret as emotions. They will make smiles or frowns with fake lips, have wide-open or droopy mechanical eyes, and perhaps even tears. In the same way Pi believed that animals have souls from what he perceived in the tiger's eyes, people will project emotions on to robots.
As was presented on Design Night, researchers discussed the following:
Panelist: How would you feel if someone took a baseball bat to your refrigerator?
Respondent: I would not like it, but not so bad.
Panelist: How would you feel if someone took a baseball bat to your robot?
Respondent: I think I would feel the same.
Panelist: Then explain this. We had people play with a robot dog. After a short while, we gave them a bat and asked them to destroy the dog. They could not do it.
They say communication is defined by the receiver, not the sender. Beauty is in the eye of the beholder. I guess it's the same with emotion. People will see physical characteristics that they associate with human emotions and believe that the sender has these emotions when in fact it is they who are projecting the emotions on to the sender. Whoa! In the Design Night example, people believed the mechanical dog had feelings because it acted like a dog even though it was the equivalent of a blender or toaster.
Let me guess. He's angry?
At the start of the Design Night, Adviser to the CEO/CTO, Jonathan Knowles, introduced the panelists and also recognized Dr. Don Greenberg who was in the audience. Dr. Greenberg is a professor at Cornell University and pioneer in computer graphics. Back in the early days had anyone asked him, or anyone else for that matter, if we'd ever see computer graphics that were so good that one could not distinguish them from a photograph, the answers would not have been split 50-50, but 100% no. But just look at visually striking movies, such as Life of Pi, today. No tiger was harmed in the making of the movie because the tiger was computer-generated. Robots are indeed in their infancy, but just give them time. One day we'll say "They're Gr-r-reat!" One day we might not be able to distinguish them from people — much like computer renderings and photographs.
Getting emotional is alive in the lab.