AI offers insights into conversations through physiology alone – Zoo House News

Engineers at the University of Cincinnati say the technology might not be far off. They trained a computer — using data from wearable technology that measures breathing, heart rate and sweat — to identify the type of conversation two people were having based solely on their physiological responses.

Researchers studied a phenomenon in which people’s heart rate, breathing and other autonomic nervous system responses are synchronized when they speak or work together. This effect, known as physiological synchrony, is stronger when two people are deeply engaged in a conversation or working closely together on a task.

“Physiological synchrony is also evident when people talk on Zoom,” said study co-author Vesna Novak, associate professor of electrical engineering in UC’s College of Engineering and Applied Science.

In experiments with human participants, the computer was able to distinguish four different conversation scenarios with an accuracy of up to 75%. The study is one of the first of its kind to train artificial intelligence to recognize aspects of a conversation based solely on the participants’ physiology.

The study was published in the journal IEEE Transactions on Affective Computing.

Lead author and UC graduate student Iman Chatterjee said a computer can give you honest feedback about your date — or yourself.

“The computer could tell if you’re a bore,” Chatterjee said. “A modified version of our system could measure how interested one person is in the conversation, how compatible you two are, and how engaged the other person is in the conversation.”

Chatterjee said the physiological synchrony is likely an evolutionary adaptation. Humans have evolved to share and collaborate with one another, which manifests itself even at a subconscious level, he said.

“It’s definitely no coincidence,” he said. “We only notice physiological synchrony when we measure it, but it probably creates a better level of coordination.”

Studies have shown that physiological synchrony can predict how well two people will work together to complete a task. The degree of synchronicity also correlates with how much empathy a patient perceives in a therapist or how committed students feel to their teachers.

“You could probably use our system to determine which people in an organization work better together in a group and which are inherently antagonistic,” Chatterjee said.

This aspect of affective computing holds tremendous potential for providing real-time feedback for educators, therapists, or even people with autism, Novak said.

“There are many potential applications in this area. We have seen that it is designed to look for implicit bias. They may not even be aware of these prejudices,” Novak said.