Researchers at the Massachusetts Institute of Technology (MIT) have developed what they’re calling “EQ-radio,” a device that can wirelessly detect a person’s emotions. The institute’s Computer Science and Artificial Intelligence Lab (CSAIL) built a machine that is able to detect emotions using radio frequency waves. Researchers bounced the waves off test-subjects in order to monitor their heart and breathing rates.
That data was then run through an algorithm designed to identify various physical indicators connected to emotions. The emotions were categorized under four categories: sadness, anger, pleasure, and joy.
Researchers set test-subjects three to four feet away from the machine and instructed them to recall a memory that elicited a particular emotion. This established what the CSAIL team called a “ground truth,” or a baseline emotion the EQ-radio could use to measure emotional states, according to MIT News.
That “ground-truth” could be compared with a subject’s emotional state at a different time. When the radio monitored a subject whose baseline it had previously taken, it could detect the user’s emotion with 87 percent accuracy. Even when the machine had not read a specific person’s baseline, it could use the information measured in other subjects to detect emotion with 70 percent accuracy.
Other machines used to detect emotion such as the electrocardiograph, or ECG, require electrodes to be placed on a user’s body and can be inaccurate over time if the electrodes move. The MIT team found that heart-beat was the best determiner of emotion, followed by breathing. The team had to decipher between chest movements caused by breathing and those caused by heart-beat, a difficult task in that movement from breathing is much greater than those from heart-beat.
MIT professor Dina Katabi, who worked as project leader on the EQ-Radio imagines that the technology could be put to use in the fields of advertising, entertainment, and healthcare. A movie studio could use the device to track viewer’s reactions to a film while health care professionals could detect emotions such as depression or anxiety.
In the future, the technology could be used in so-called “smart homes” or homes built with wireless technology to monitor a person’s emotional state and make changes such as adjusting the thermostat.
Professor Katabi has a spin-off company called Emerald, which makes a device designed to predict falls in elderly people. Emerald uses a 3D sensor to detect a person’s movements and analyze factors such as gait to determine risk of falling. The EQ-radio could be used to detect the emotions that occur just before a person falls, Engadget reported.
The CSAIL team is set to present their technology next month at MobiCom, the Association of Computing Machinery’s International Conference on Mobile Computing and Networking. For more CDA News, follow our tweets on Twitter and like us on Facebook.
By Peter Segall