The world of wearable MedTech and smart clothing is, right now, a functional one. We have smart watches that can read all your biological signs to guide you on your way to peak fitness. Other devices can diagnose illnesses through samples of your bodily fluids or assess the quality of last night’s sleep.
Our social lives, however, despite the absorption of social media into the general consciousness, remain forever dictated by our oftentimes messy and confused, but very human, emotions. Mostly, the best you can do to help solve social problems is to visit a counsellor. But thanks to one team at the University of Southern California, we should all be excited about the potential to change this in the future.
Using the latest AI and machine learning methods, graduate researchers Adela Timmons and Theodora Chaspari have created an algorithm that diagnoses emotional states based upon various physiological factors observed in subjects. These include not only heart rate, sweat levels and frequency of movement, but also the very content of the words users choose in conversation.
What’s the ultimate goal of this research?
“We visualize an application [that can] track a user’s relationship status each day. It would provide indications of how good or how bad relationships would be on a specific day. And beyond that, it could provide feedback or maybe online intervention,” Chaspari told NBC.
To ‘train’ their algorithm, the USC team recruited 34 college-age couples and gave them an unobtrusive smartwatch to wear throughout a single day. Out of those 34 couples, 19 experienced some kind of interpersonal conflict during the 24-hour period of the study. The watches listened to the participants’ conversations and recorded what they said – as well as their physical state.
Afterwards, the team collated their data and found that using keyword triggers and biometrics, they could correctly identify an argument or conflict 82% of the time. Using this method, they ‘taught’ their algorithm to recognise a personal conflict, in real-time, with a 79% success rate.
Imagine a smart wearable device that could see the subtle signals indicating you or your partner (or co-worker, family member or anyone else) was nearing anger or annoyance before you do? The impact that could have on our relationships would be immeasurable – and with research like this ongoing, it could soon be a reality.