Computers are able to detect whether a human expression of pain is real or fake better than humans can, a new study reveals. Fascinating, isn't it? The question is, does this discovery have any practical use?
The research team — from the University of California-San Diego and the University of Toronto in Canada — write that social species including humans have evolved to externally show valuable expressions in social contexts. This includes pain and emotions, of course. While it's extremely useful to have an idea of what other humans are feeling just by looking at them, we have also obviously evolved to lie. Body language and facial expressions are an integral part of a successful lie.
Deceiving Humans Is Easier Than Lying To Computers
"Humans can simulate facial expressions and fake emotions well enough to deceive most observers," senior author Professor Kang Lee from the University of Toronto says.
Though humans can fake facial expressions very successfully, real emotions and sensations are controlled by a different motor pathway in the brain than fake emotions. The subcortical extrapyramidal motor system controls spontaneous facial expressions, while the cortical pyramidal motor system is in charge of voluntary (fake!) facial expressions.
Experiments the research team carried out showed that humans aren't able to tell the difference between fake pain and real pain very well at all — they wouldn't do better if they picked at random, and even after receiving special training, they can only distinguish real pain from fake pain 55 percent of the time.
The computer system the team developed is able to detect the origin of the facial expression, and figure out whether the expression was involuntary or spontaneous. This gives the computer system an 85 percent success rate.
Computers are better at interpreting human feelings than humans are, in other words.
Marian Bartlett, lead author from University of California-San Diego, put it this way: "The computer system managed to detect distinctive dynamic features of facial expressions that people missed. Human observers just aren't very good at telling real from faked expressions of pain."
Practical Applications Of The Facial Expression Detection System
Further research is now needed to find out if "over regularity" is a feature common to fake facial expressions. If the answer proves to be yes, the computer system the study team developed could have many practical applications in the future — in security, psychological evaluations, medicine and law.
She went on to speculate about possible uses for the computer system: "In addition, our computer-vision system can be applied to detect states in which the human face may provide important clues as to health, physiology, emotion or thought, such as drivers' expressions of sleepiness, students' expressions of attention and comprehension of lectures, or responses to treatment of affective disorders."