Verified by 
The next U.S. general election will be held in about 14 months. At the moment, no fewer than 22 hopefuls have declared themselves as candidates for president. That unusually large number will dwindle, of course. By Halloween, some candidates will be the political equivalent of a ghost. Others will fade in the stretch. Only one will declare victory in the early hours of November 9, 2016.
Trying to predict the winner of an election so far in the future is a fool’s errand, of course, but regular readers of psychology journals know that the winner will almost certainly look more competent than the runner-up, and most of the also-rans. Researchers have found that political candidates who look more competent than their rivals are more likely to win elections. This finding has been replicated in Western nations like France and Switzerland, as well as the United States.
Consider the results of a new study by psychologist Jinkyung Na and his colleagues at the University of Texas at Dallas and Seoul National University in South Korea.
Na and his team asked university students in both countries to examine pairs of photos. Each pair included the winner and runner-up in an election held in either South Korea or the U.S. (Well-known politicians who were recognized in a pretest were excluded from the main study.)
Each participant indicated which person in a pair looked more competent. Based solely on these judgments, American participants correctly predicted the outcomes of 69% of U.S. elections, but only 49% of Korean elections. (Someone flipping a coin would have been correct 50% of the time.)
Amazingly, the performance of Korean participants was nearly identical: They correctly predicted the outcomes of 67% of U.S. elections but only 44% of Korean elections.
Let me say that again in a slightly different way: University students in Korea looked at the faces of two unknown political opponents in the U.S. and chose the winner about two-thirds of the time, yet they couldn’t do the same thing for elections in their own country.
What’s going on here? Were the American and Korean participants on the same page when it came to judging competence from a person’s face?
Yes, they were.
Na and his team observed very high levels of agreement for in-group and out-group faces. So the fundamental mystery remains: Why can information about perceived competence predict election outcomes in the United States but not South Korea?
One possibility is that Koreans are more knowledgeable voters—and more knowledgeable voters are less likely to be influenced by superficial cues like facial appearance.
Another possibility stems from an analysis of cultural values and beliefs. Na and his colleagues argue that, in independent cultures like the United States and other Western societies, individuals are defined by their internal attributes or characteristics. A person possesses various traits, and one’s behavior is determined to a large extent by those traits. Warm people interact warmly with others, and competent people perform capably at school and work. Within this belief system, it makes sense to vote for the candidate who looks most competent. In interdependent cultures like South Korea and many other non-Western societies, individuals are embedded in networks of social relations, and one’s actions (such as voting) are motivated by social obligation rather than dispositional traits.
Within this belief system, it makes little sense to vote for a candidate simply because he or she looks more competent than other candidates. According to Professor Na, Korean voters rely heavily on information about candidates’ social affiliations and connections. A candidate’s hometown, for example, can be a crucial factor in Korean elections.
In any case, American voters would be well advised to learn as much about the current slate of candidates as they can. After all, looks can be deceiving.
Na, J., Kim, S., Oh, H., Choi, I., & O’Toole. (2015). Competence judgments based on facial appearance are better predictors of American elections than of Korean elections. Psychological Science, 26(7), 1107-1113.
