The human face is capable of being manipulated into so many different muscle combinations that the odds of correctly identifying one expression by chance are the same as winning the Powerball jackpot twice. Yet in a feat of statistical peculiarity, humans can reliably recognize a swath of basic emotions across cultures. This capability has been a hint to some scientists that there must be a shared evolutionary origin for our ability to express and understand emotion-conveying expressions.
In a recent study published in Psychological Inquiry, researchers claim that expressions developed as a byproduct of behavior and experience. Furthermore, authors Daniel Lee and Adam Anderson argue that the ability to read and interpret emotions can be distilled down to one very important part of human physiology: the eyes. The need to use our peepers in certain situations, they reason, may have started as a physical need that eventually became associated with certain social situations. The eyes may not exactly be the windows to the soul, but they do help us get a good sense of how others are feeling.
“This starts at the function origins of expressions theorized by Darwin, and it seems to be a human tendency to co-opt those evolved features for social function,” says Lee. “What we also find is that it’s not so much specific expressions but continuous dimensions of expression appearance. It’s easier to talk about discrete expressions but there are lots of variations, such as degree of eye opening, that communicate a wider range of mental states.”
Lee and Anderson believe that eyes widen or narrow to infer emotional states, and that these physical movements stem from the essential optical functions that eyes perform. They came to this conclusion by studying 28 participants as they looked at a total of 600 facial expressions. They were able to look only at the eyes of these expressions and were then asked to choose one of four emotional states provided below. After the participants turned in their surveys, the researchers analyzed which specific eye features were consistently related to particular mental states — such as the curve of an eyebrow or a wrinkle around the nose.
They found that eye expression was consistently the best clue for choosing the correct facial expressions. Furthermore, it was either eye-widening or eye-narrowing that stemmed into selecting specific emotions. Eye-narrowing was often, and accurately, associated with emotions like hate, suspicion, aggressiveness, and contempt. Meanwhile, eye-widening was associated with expressions that conveyed states like awe, interest, and anticipation.
Why widening and narrowing play an important role here, the researchers say it suggests that eyes have an “older utility for physical information” and a “younger utility for social information transmission.” Narrowed eyes enhance the ability of eyes to discriminate and discern what’s happening and may be why narrowed eyes became associated with scrutinizing emotions like disgust and suspicion.
The authors also made other interesting observations from their expression test, which anyone can take here and see how their perceptiveness (by the standards of this test) compares to others who’ve taken it. Results from this test show that women on average score better at correctly identifying emotions based on eyes, and they appropriately say they are more confident at reading emotions than men are. When it comes to identifying individual emotions, women are better at detecting emotions like being shocked, scared, and guilty, while men are better at recognizing desire and hostility.
While other researchers have posited that emotions are something that can’t be universally expressed and recognized, Lee believes that his research would likely hold up globally.
“Our prediction is, yes, it would hold up across cultures — there are plenty variations across cultures, but also a great deal of similarity,” says Lee. “This goes back to Darwin and our facial expressions having evolved for function, rather than arbitrary symbols. In other words, eye widening will enhance light sensitivity wherever you go in the world, and we predict have been socially co-opted for communicating mental states of sensitivity.”
Lee says that while the purpose of this research was to gain further insight into being human, he would be interested in seeing how these findings could be applied to computational facial expression coding. This, in turn, could be used to further machine learning and aid engineers designing robots that can recognize human emotions.