Machines Can See Through That Poker Face

The most surprising and impactful new stories delivered to your inbox every week, for free.

When we try to mask our emotions, we never put on a perfect poker face. There are little “microexpressions,” but catching them when they happen is tough. Law enforcement officials have tried training agents to spot these cracks in other people’s emotional masks, but there’s been much debate whether some people have the sensory and cognitive skills to be trained in such a fine area.

Researchers now believe they’ve developed an artificial intelligence capable of catching and dissecting these microexpressions.

Singularity University’s Vivek Wadhwa details emerging technologies — including artificial intelligence — that we have to be smart about implementing.

Scientists say that machine vision has improved to the point where the computers outperform humans in areas of object and facial recognition. But being able to see the subtleties of expressions — the difference between a smile and a smirk — has also come a long way. Xiaobai Li at the University of Oulu in Finland and his team believe they’ve <a href="http://arxiv.org/abs/1511.00423">developed</a> a machine that rivals human capabilities in spotting and recognizing these microexpressions. The barrier here was having a dataset to teach the computer in the first place. Machine learning requires a large database of information to work off of and getting such a niche database worth of microexpressions doesn’t sound like the easiest of tasks. The <a href="http://www.technologyreview.com/view/543501/machine-vision-algorithm-learns-to-recognize-hidden-facial-expressions/?utm_campaign=socialsync&utm_medium=social-post&utm_source=facebook">MIT Technology Review</a> agrees, writing that “much previous work has focused on posed expressions, but various psychologists have pointed out the limitations of this method, not least of which is that microexpressions look significantly different to posed expressions.” However, Li and his team apparently tackled this by showing 20 participants a series of emotional videos. But incentive to not show emotion was given by the researchers telling participants they would have to fill out a long survey about any emotions they displayed while watching the videos. Tricky. <rebelmouse-image iden="18405118" original_size="616x403" expand=1></rebelmouse-image> “Our method is the first system that has ever been tested on a hard, spontaneous microexpression data set, containing natural microexpressions,” the team <a href="http://arxiv.org/abs/1511.00423">said</a>. “It outperforms humans at microexpression recognition by a significant margin, and performs comparably to humans at the combined microexpression spotting and recognition task.” The possibilities of this technology could extend beyond law enforcement and psychology. There could be a Google Glass-type device for emotion sensing. However, I wonder if a computer taking over the ability to sense emotion would cripple some areas of our brains, making us incapable of recognizing emotions without it. I can’t help but make a comparison to the GPS — a great piece of technology, but one that we’ve come to rely on so much that we’re <a href="https://www.bostonglobe.com/ideas/2013/08/17/our-brains-pay-price-for-gps/d2Tnvo4hiWjuybid5UhQVO/story.html">incapable of finding our way without it now</a>. *** Natalie has been writing professionally for about 6 years. After graduating from Ithaca College with a degree in Feature Writing, she snagged a job at PCMag.com where she had the opportunity to review all the latest consumer gadgets. Since then she has become a writer for hire, freelancing for various websites. In her spare time, you may find her riding her motorcycle, reading YA novels, hiking, or playing video games. Follow her on Twitter: @nat_schumaker Photo Credit: <a href="http://www.gettyimages.com/search/2/image?artist=Ethan%20Miller&family=editorial" id="photographer">Ethan Miller</a> / Getty Staff

The most surprising and impactful new stories delivered to your inbox every week, for free.