New Software Reads Human Emotion Better than Humans

What’s the Latest Development?

If someone asked you to imitate a frustrated person, you would probably not begin your impression by smiling, but new behavioral research out of MIT demonstrates that a smile is a common expression of frustration. In a study, individuals were recorded while completing tasks meant to solicit responses of happiness and frustration. In the latter case, people filled out an online form only to have their information erased upon clicking the “submit” button. When strangers were shown photos of the smiles and asked to tell which was “happy” and which was “frustrated”, they performed no better than chance, suggesting we may not judge the emotion of others with as much accuracy as we would like. 

What’s the Big Idea?

MIT researchers then created a computer program which could differentiate between frustrated and happy smiles better than humans. Jeffrey Cohn, a professor of psychology at the University of Pittsburgh who was not involved in the research, said the work was especially interesting because of how it integrated psychology, computer vision, speech processing and machine learning to generate new knowledge. “The research could pave the way for computers that better assess the emotional states of their users and respond accordingly. It could also help train those who have difficulty interpreting expressions, such as people with autism, to more accurately gauge the expressions they see.”

Photo credit: Shutterstock.com