“Hey, buddy. You seem a little down, how about I order you a cheeseburger?” In the future, you could find yourself responding to that kind of a pep talk from your smartphone: a new device created by the bright minds at the Massachusetts Institute of Technology have developed a device that can detect emotions by reading wireless signals bouncing off a person.
Researchers in MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) have created “EQ-Radio,” a piece of technology that measures subtle changes in breathing and heart rhythms to detect how someone is feeling.
EQ-Radio is 87% accurate at telling if someone is excited, happy, angry, or sad, CSAIL researchers say, noting that unlike this device, existing methods of detection emotions rely on audiovisual cues or sensors attached to someone’s body, but those things are inconvenient to wear and can become inaccurate if they’re knocked out of position.
EQ-Radio sends wireless signals that bounce off a person’s body and then back to the device. It then uses “beat-extraction algorithms” to break those signals down into individual heartbeats, and analyze any variations in heartbeat intervals that could determine their levels of arousal and positive affect.
Once it has those measurements, it can detect emotion: signals correlating to low arousal and negative affect are more likely to mean that the person is sad, while someone with signals correlating to high arousal and positive affect get tagged as excited. While correlations vary from person to person, researchers say they’re consistent enough to detect emotions with 70% accuracy without having measured the person’s heartbeat first.
MIT professor and project lead Dina Katabi sees the system being used in anything from entertainment to consumer behavior and health care. Movie studios could test the audience’s reactions in real time, or your smart home might take your mood into account when adjusting the light in the house.
“Our work shows that wireless signals can capture information about human behavior that is not always visible to the naked eye,” says Katabi, who co-wrote a paper on the topic with PhD students Mingmin Zhao and Fadel Adib. “We believe that our results could pave the way for future technologies that could help monitor and diagnose conditions like depression and anxiety.”