A new device created at MIT using Artificial Intelligence can detect emotion, which could help people clear up confusion or untangle mixed signals, particularly in high-stakes conversations such as job interviews or salary negotiations. It could also help people, such as those with Asperger's or social anxiety, who have trouble interpreting facial expressions or social cues. 

 Researchers from the university’s Computer Science and AI Lab and the Institute of Medical Engineering and Science partnered with Samsung’s innovation center to create a wearable band that monitors vital signs and speech patterns to predict if a conversation is happy, sad, or neutral.

Read: Understanding Another Person's Emotion Signals Similarity, And May Make You Find Them More Attractive

Tuka Alhanai and Mohammad Mahdi Ghassemi, who conducted the study, asked 31 subjects to tell a story while wearing a Samsung Simband. The band measured heart rate, blood pressure, blood flow and skin temperature in addition to acquiring data and transcripts to gauge tone, pitch, energy and vocabulary.

Following the conversations, each of which were several minutes, the duo trained two algorithms. The first denoted if an overall conversation was happy or sad while the second classified the exchanges as positive, negative or netural in five-second increments.

According to MIT, the band associated sadder stories with cues like increased cardiovascular activity, while energetic, varied speech patterns identified happier conversations.

“It’s quite remarkable that a machine could approximate how we humans perceive these interactions, without significant input from us as researchers,” Alhanai said in a statement.

Read: Can High Levels Of Negative, Unpleasant Emotions Cause Brain Damage?

However, the system is not yet ready for mass production. The Simband performed modestly with an accuracy rated at 17.9 percent above chance, which is 7.5 percent higher than the baseline, according to MIT. Alhanai and Ghassemi chalk this up to the small sample size and believe this could serve as a starting point for larger-scale tests.

Though the technology still needs work, it represents an impressive first step for understanding emotions in real-time, which as the authors point out in their paper, is particularly important during difficult conversations. 

See Also:

Skin Cancer Detection 2017: Artificial Intelligence Algorithm Can Diagnose Carcinomas, Others, As Well As Doctors

Sex On Your Period: Menstrual Discs For Women Tout 'Mess-Free' Sex, Without Toxic Shock Syndrome Risk