10/08/2013 10:40 am ET Updated Dec 08, 2013

The Face(s) of the GOP. Who Do You See?


I've worked in Washington, D.C. for decades, so I have witnessed a fair number of political logjams, even a few government shutdowns. So I'm not quick to panic when the two parties' leaders stubbornly stake out what are seemingly irreconcilable positions. But I confess that listening to House Republicans this time around -- especially but not only the Tea Party zealots -- is making me nervous. This is not just the usual posturing and brinkmanship. I really think they perceive a different reality than the rest of us.

Is that possible? Can people be so biased by their political attitudes that they look out and see a different world, a world where up is down and black is white? I look at Tea Party Republicans and I see incompetence and selfishness and cruelty. Is it possible that others look at the same faces and see trustworthiness and benevolence?

Coincidentally, I came across a new study this week that argues just that. Ohio State University psychological scientists Russell Fazio, Alison Young and Kyle Ratner used techniques from psychophysics to test the idea that our political attitudes can shape our most basic perceptions of human faces -- including our perceptions of trustworthiness. Specifically, they used a technique known as "reverse correlation image classification." That's a mouthful of jargon, but basically this technique allows scientists to obtain an approximation of people's internal representation of faces. Put another way, it's a way of peering inside people's minds to see what they see.

The scientists decided to study the face of Mitt Romney, failed Republican candidate for president in 2012. Ohio was a battleground state in that election, and the two parties spent more than $150 million on TV ads in the run-up there. The city of Columbus -- where Ohio State is located -- was inundated with an estimated 40,000 ads during the final months of the campaign. As a result, the scientists reasoned, Columbus residents should have thoroughly learned and embedded representations of the candidates' physical features in their minds.

Fazio and his colleagues wanted to look inside the minds of Ohioans, to see if they all saw the same Mitt Romney -- or if their memories of his familiar face were influenced by their politics. So they recruited a group of Ohio State students, some of whom supported and voted for Romney, some of whom did not. They asked them to categorize permutations of the Republican candidate's familiar, smiling face, both right before the November 6 election, and immediately after. These estimations of the volunteers' mental representation were then rated by independent judges, to see if they varied in any systematic and meaningful way.

They did. One would expect something as concrete and well-known as a presidential candidate's face, in the midst of a campaign, to be an unvarying, truthful image, free of bias. But that was not the case in this study, which will be published in a future issue of the journal Psychological Science. Indeed, Romney supporters saw a much more trustworthy and benevolent candidate than did Romney detractors. Notably, volunteers were not asked about Romney's trustworthiness or goodness; one would expect self-reports to be biased by attitudes. These were their internal, unfiltered perceptions of a human face.

As a practical matter, these findings suggest that citizens are more cognitively biased that previously known. Not only do voters interpret information to fit their opinions, they construct and live in a political world where they literally see political figures as different people with different traits. It's a sobering thought for citizens of our increasingly polarized nation.

Follow Wray Herbert's writing on psychological science on Twitter at @wrayherbert.