Empathizing with Human-like Robots

James Coan, a psychology professor at the Univ. of Virginia, claims empathy is hardwired into the human brain. Given enough time and familiarity, you begin associating friends, spouses and lovers with yourself. A lover’s pain becomes your pain; their joy, your joy, and so forth.

Using electroencephalography (EEG) studies on 15 adults, Japanese researchers found neurophysiological evidence of humans’ ability to empathize with robots in perceived painful situations. Albeit, there are still some differences when it comes to human-human empathy versus human-robot.

Study participants were shown a series of photographs showing human and robot hands cutting fruit with knives. In four photos, the human and robot hands are positioned in such a manner it appears as if they’ll cut their fingers.

“The ascending phase of P3 (350 to 500 msec after the stimulus presentation) showed a positive shift in the observer for a human in pain in comparison with no-pain condition, but not for a robot in perceived pain. Then, the difference between empathy toward humans and robots disappeared in the descending phase of P3 (500 to 650 msec),” said Michiteru Kitazaki, an associate professor at Toyohashi Univ. of Technology.

The collaborative study between researchers at Toyohashi Univ.’s Dept. of Information Science and Engineering and Kyoto Univ.’s Dept. of Psychology was published in Scientific Reports.

The researchers credit the delayed reaction in human-robot empathy to humans’ inability to take on the perspective of a robot. “The difference between the human painful hand and the robot painful hand may be explained by the unnaturalness of robot hands cut by knives,” the researchers write.  

Later, they continue, “The pain discrimination was easier for the human-hand stimuli than for the robot-hand stimuli, possibly because the robot stimuli had low color contrast and the robot hand was larger than the human hand, although the knives were identical.”    

As robots become more prevalent in human society, it is imperative for them to understand human social interactions. Matthew Howard, a robotics lecturer at Kings College London, told Al Jazeera naturalizing robot behavior to be more human-like will allow humans to predict robot behavior and therefore be safer around it.

Selmer Bringsjord, a professor at Rensselaer Polytechnic Institute and director of the Rensselaer AI and Reasoning (RAIR) Lab, is programming robots to detect physical changes. Though they may not be able to detect pain when hit with a hammer, they can be programmed to manifest behaviors associated with pain, he told R&D Magazine.

Sympathetic human-friendly robots may be just around the corner.