Can You Tell Someone Is Conservative Just By Looking At Their Face?
By now, everyone has seen Staten Island Mafia thug Michael "Mikey Suits" Grimm (R-NY) in action at the State of the Union address. You may have missed Sean Hannity though. He was there as a guest of right-wing kook Louie Gohmert (R-TX). And, in fact, Hannity has been threatening to leave New York and move down to Texas, where more people appreciate his anti-democratic leanings-- and where rich people don't pay taxes. When someone on Facebook asked him if he'd consider running for office in Texas, he said, “The answer is yes, I’d think about it. It would either be in Texas or Florida… How does Texas have a $12 billion surplus when they’re in debt up to their eyeballs in Albany?” Yes, Texas' and Florida’s lack of income tax are appealing lures. And speaking of eyeballs, can you tell from looking into Sean's eyeballs and into Mikey Suits' eyeballs-- or even at their faces in general-- that each one is an extreme right wing ideologue? In the academic book written by John Hibbing, Kevin Smith and John Alford, Predisposed, they make a compelling case that it's possible to tell a person's politics just by looking at their face. They start with a simple question: Why are faces so revealing?
If political temperament is biologically based, it makes sense that it is being broadcast by faces. Our faces are constantly, without any conscious input or even awareness, beaming to the world information about our feelings and social intent. Faces are the visual Twitter accounts of our nervous systems, able to distribute information about psychological states quickly and succinctly, and to many people at the same time. At least since Darwin, researchers have recognized that the face provides a universal means of human social communication. We can quickly and accurately assess someone's psychological state-- whether they are happy, sad, ticked off, surprised-- with a glance at his or her face. This form of social communication is so fundamental to human nature that psychophysiologists argue that "without [facial expressions] individuals do not communicate, do not affiliate, do not proliferate, do not interact-- in short, are not social." Indeed, faces are said to "leak" our internal psychological states; we involuntarily smile, frown or wrinkle our noses when we feel joy, disapproval or disgust. Humans may not all speak the same tongue but we are all universally fluent in face. We are also pretty good at detecting facial fibbing. We can usually tell, for example, when a smile expresses real joy or is just being faked for social consumption. Of course, there are people who are really good at faking it (actors, for example), but most people find faking impossible to do convincingly.
Faces do more than provide a way to communicate our feelings about our brother-in-law. They declare membership in socially meaningful groups. Some of this is intuitive. Faces, for example, make it easy to classify someone into a particular gender, racial or age group. Studies also show that people can accurately predict an individual's sexual orientation and even religious affiliation using only facial information. They are able to do this after looking at a face for only a fraction of a second. And the declaration of social affiliations apparent from our faces also includes political orientation.
Some of these studies suggest that people divine political orientation from faces by perceiving them as more or less powerful or socially superior. At least one study finds people think conservative faces look more intelligent. These are purely perceptual judgments, though, and do not actually measure anything about the face. We were interested in exactly what it is physiologically about faces that signals political orientation, and we suspected it might have something to do with the degree of emotional expressivity in a face. One of the aspects of personality known to separate liberals and conservatives (or at least partisan affiliations) is expressivity. For example, two psychologists, James Gross and Oliver John, developed a sort of personality test called the "Berkeley Expressivity Questionnaire" that is designed to measure individual-level variation in emotional expressivity. Democrats tend to score higher than Republicans on this set of items.
If facial expressions are known to be a primary and largely sub-threshold means of signaling emotional states, then it follows that Democrats (liberals) will also tend to have relatively expressive faces. We tested this hypothesis using a technique known as electromyography (EMG), a fancy term for putting sensors on the skin to measure the electrical activity picked up by muscle contractions. The specific muscle measured was the corrugator supercilii, found between the eyebrows. Its job is to furrow the brow. Even if we are not aware the muscle has moved, negative emotions like digest, anger, and fear tend to activate the corrugator; positive emotions tend to make it relax. Corrugator activation or deactivation helps to create the facial expressions associated with many of our primary emotions. Accordingly, we prevailed upon a group of adult subjects to tell us their ideological leanings and later we measured their facial expressivity by the extent of their corrugator activity in response to a variety of positive and negative stimuli. Our hypothesis was that liberals would be more facially expressive than conservatives and that turned out to be half true. Like other EMG studies, we found females, regardless of political persuasion, to be more facially expressive than males. Unlike any other EMG study, we found liberal males to be emotionally expressive at pretty much the same level as females. The most distinctive group by far was conservative males. While corrugator activation in response to the images was significant for everyone else, for conservative males it didn't budge.
Perhaps people are able to discern personality traits and therefore political orientations from images (most of the studies use pictures of males) because stoic, less expressive faces (think Clint Eastwood) signal traits associated with conservatism and sensitive, more expressive faces (think Alan Alda) signal traits associated with liberalism. Certainly these signals are not 100 percent accurate, but they do permit quick judgments that appear to be right more often than they are wrong. This conclusion is supported by another study that took "liberal" and "conservative" faces and created avatars that exaggerated facial features and expressions. The liberal avatar was smiling, with a relaxed corrugator; the conservative avatar had less of a smile and even looked bait frowny. Evidence that political temperaments are instantiated in our biology is found not just in individual-leve variation in our brains or the internal wiring of our automatic nervous systems. Quite literally, politics is also on our faces.