Tess Watson / Flickr / Composite Image

In the animal kingdom, individuals prove their dominance with displays of power, stature, strength, and aggression. We humans like to think we’re more evolved than to look for those traits in a leader—that instead of physiological dominance, ideas and reason are what guide our decisions.

We like to think that, but “we are not so separate from the rest of the animal kingdom as we might want to believe,”  says Casey Klofstad, who studies the intersection of biology and political science at the University of Miami. “We are indeed influenced—on some level—by subtle cues that are biologically determined.”

In particular, Klofstad studies the way people react to political candidates’ voices. He finds, consistently, that voters prefer politicians whose voices have a deeper tone.

“Men with lower-pitched voices are perceived as stronger and more attractive,” he says. “For women, it’s a little more complicated: Women with lower-pitched voices are seen as more confident and stronger. But women with higher voices are seen as more attractive.” For both males and females, going deep appears to be a winning strategy.

Klofstad demonstrated these findings in two recent papers published in the journals Political Psychology and PLOS ONE.

In the PLOS paper, he manipulated male and female recordings of the sentence “I urge you to vote for me this November” to range from a low baritone to a high soprano. Four hundred participants then took a listen to either the male or the female voices, and were asked which voices they thought were stronger, more competent, and older. They were also asked which voice they thought was more electable. The deeper voices won. “The preference for leaders with lower voices more likely reflects [correlations] between voice quality and leadership capability that were relevant at some earlier time in human evolutionary or cultural history,” the paper concludes.

In the Political Psychology paper, Klofstad wanted to see if any of these effects play out in the real world. He took recordings of 796 candidates in the 2012 House races and analyzed the pitch of their voices. He then compared that data with election results. The average voice pitch among winning male and female candidates was, overall, lower than the losers, as you can see in the chart below.

Even controlling for other factors that might explain the outcome of the elections—party affiliation of the candidate, district makeup, spending, incumbency—the difference voice pitch makes is still statistically significant.

This pattern broke down when Klofstad specifically analyzed data on races where males and females were up against each other. “When men and women face each other, men with lower voices were actually disadvantaged,” Klofstad says. It could be that men with deep voices may seem too aggressive next to a woman, but he has not tested that hypothesis in any experiment.

This research shouldn’t suggest that pundits should give up trying to call elections based on demographics, polling, and campaign issues. Those are more powerful indicators of how people will vote than voice pitch. It’s just that “on top of that are these very thin signals, these very thin, impressionistic judgments that we make,” Klofstad says. “And we may or may not be aware of them.”

It’s not just voice. We, subliminally, prefer a whole host of physical traits in our leaders. Those characteristics include:

A competent face. A famous 2005 study found that people can predict the outcomes of elections—to a degree greater than just pure chance—by looking at candidates’ faces. What was it about the faces? Faces that were deemed to be more “competent” were more likely to win. A competent face is the opposite of a baby face: square jaw, big eyes, fierce stare.

A tall frame. A 2012 analysis of all presidential elections (excluding those where the candidates were the same height) found that “candidates that were taller than their opponents received more popular votes,” though they were not necessarily more likely to win elections. “Taller presidents were also more likely to be reelected. In addition, presidents were, on average, much taller than men from the same birth cohort.”  

A happy face. “Just smil­ing in cam­paign pho­tos can sig­ni­fic­antly af­fect elec­tion out­comes,” a 2012 pa­per found in an ana­lys­is of 958 can­did­ates in Ja­pan­ese and Aus­trali­an elec­tions.

There’s no good reason why someone with a lower voice, or a taller frame, or a better smile would make a better candidate. Lower voices may indicate higher levels of testosterone—and therefore higher levels of aggression, and increased physicality—but our democracy needs sound thinkers, not brutes.

Humans in the election booth aren’t the completely rational beings the philosophers of yore said we should aspire to be. Keep that in mind while watching the presidential primary season unfold. Why do we really like one candidate over another: Is it because of their ideas, or because of what our guts are telling us?

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.