Let’s examine some ways that President Trump has behaved unpleasantly, shall we? He has been caught on tape making misogynistic comments and bragging about sexual assault; he circulated “alternative facts” exaggerating the turnout for his presidential inauguration; and he frequently bullies opponents and calls them names on social media. In short, his public persona is unorthodox for U.S. president, in a way that was already clear during his 2016 campaign. Nonetheless, he was elected. Republican voters continue to support him. And he might very well be reelected.

Pundits and commentators often suggest that Donald Trump was elected despite negative coverage. Voters, according to conventional wisdom, want leaders who are likable — the kind of person you could have a beer with. In this view, if Donald Trump gets reelected, his voters must be closing their eyes to or not believing the coverage.

Our research suggests that this conventional wisdom is likely wrong. Over the past few years, we have conducted empirical studies showing that some voters favor political leaders not despite but because of their unappealing personalities. Specifically, when voters feel threatened, they reach out to leaders who have a dominant personality, who get what they want through intimidation and fear.

Trump’s personality thus might be the very thing that ensures his reelection — especially if he can convince the U.S. public that the world is a dangerous place.

How we did our research

In our recent study published in the American Journal of Political Science, we fielded an online survey to a nationally representative sample of Danes. The survey was conducted in January 2015, and in total 1,524 Danes participated.

In the survey, the participants were presented with one of two personality descriptions. None of the descriptions had anything to do with politics. Instead, they described a fictional man named Christian, a very common Danish name.

In one description, Christian was a nice guy, always willing to help his friends and played board games for fun. The other Christian had a dominant personality; he liked to win and getting other people to follow his will.

We asked our participants how sympathetic they found Christian. Unsurprisingly, the nice version scored higher. We then asked them to imagine that Christian was running for office and asked them to rate his competence. Again, the nice version came out on top.

We waited a week and then contacted our participants again. We were able to reach 76 percent of the original sample. We did not provide them with any more information about Christian but asked them to rely on their memory. For a random half of our participants, we simply repeated the question about Christian's competence as a politician. Again, we found that nice Christian was rated as more competent than dominant Christian.

For the other half, however, we presented a short vignette describing increasing tensions between Denmark and Russia in the Arctic region. Results changed dramatically. Suddenly, in the face of conflict, the unappealing but dominant Christian was rated as significantly more competent than the appealing and nice Christian.

People worried about war and conflict want a dominant leader

It wasn’t that these participants had forgotten Christian’s disagreeable disposition. Rather, our results show that when we primed them to think of international conflict, they ceased to consider his dominant personality as negative and instead perceived it as positive. Specifically, we had asked our participants whether they perceived Christian as dominant in the first survey. And the more dominant they viewed Christian in the first survey, the more competent they found him in the second survey when primed with conflict.

The study was done in Denmark. But there are reasons to believe that the results apply to the United States. We have found similar results in a number of other countries, including the United States, Ukraine, Denmark and Poland. In these studies, people are consistently more likely to prefer leaders with dominant facial expressions in times of conflict.

Why do voters prefer a dominant leader when they feel threatened?

Our results suggest that people reach out to dominant politicians when they feel threatened by conflict. Why?

Our research suggests that for the sake of winning against other groups, people willingly promote individuals they would otherwise find unappealing. For example, we have found that threats from other social groups trigger preferences for dominant leaders more than risks from natural disasters.

We have also found that people who are oriented toward wanting their own group to dominate are more likely to prefer dominant leaders. Other research shows that such sentiments are much more pronounced among conservatives than liberals. This might explain why such leaders especially appeal to conservative voters.

Finally, we have found that anger rather than fear in the face of conflict is what triggers preferences for dominant leaders. When people reach for dominant leaders, they do not ask for protection. They ask for aggression.

Some evidence suggests that these results also extend to women leaders. Thus, when primed with war scenarios, people prefer dominant-looking leaders, independently of whether these leaders are men or women.

Donald Trump and the politics of conflict

Our research suggests that voters scan for and store information about the personalities of politicians. When conflict erupts and voters feel threatened, they often prefer politicians with more dominant personalities.

Just now, at a time of increasing polarization and unease in the Middle East, many U.S. voters do indeed feel threatened. For example, research by Diana Mutz showed that a 2016 vote for Trump reflected “increasing anxiety among high-status groups” that their status was being challenged by minority ethnic and racial groups.

If you happen to be a politician with a dominant personality, it would be in your electoral interest to promote such feelings of threat. Trump seems aware of this. One of his Instagram posts during the 2016 presidential election was, essentially, a concise summary of our research: “The world is a dangerous place. We need a tough, strong leader” was the actual message.

Will this strategy of conflict be used — and useful — in the upcoming election? Heated escalations of conflicts on social media over the impeachment process seems inevitable. More significant conflicts, like the current skirmish with Iran, can serve similar purposes.

Dominance is an asset only if you stir up conflict.

Michael Bang Petersen (@M_B_Petersen) is professor of political science at Aarhus University, Denmark.

Lasse Laustsen (@LasseLaustsen) is associate professor of political science at Aarhus University, Denmark.