In a key moment of the final 2016 Trump-Clinton presidential debate, Donald Trump turned to a question regarding Russian president Vladimir Putin:

“He has no respect for her,” Trump said, pointing at Hillary Clinton. “Putin, from everything I see, has no respect for this person.”

The two debaters then drilled down to try and gain a more nuanced understanding of the difficult policy issues involved. Clinton said,

“Are you suggesting that the aggressive approach I propose would actually fail to deter Russian expansionism?”

To which Trump responded,

“No, I certainly agree that it would deter Russian expansionism; it’s just that it would also serve to destabilize the ...”

Just kidding. That’s not at all what happened. Actually each side aimed to attack and defeat the other. Clinton really said,

“Well, that’s because he’d rather have a puppet as president of the United States.”

To which Trump retorted,

“You’re the puppet!”

Episodes like this one have become such a staple of contemporary political discourse that it is easy to forget how radically different they are from disputes we often have in ordinary life. Consider a couple of friends trying to decide on a restaurant for dinner. One might say, “Let’s try the new Indian restaurant tonight. I haven’t had Indian for months.” To which another replies, “You know, I saw that place is getting poor reviews. Let’s grab some pizza instead?” “Good to know—pizza it is,” says the first. Each comes in with an opinion. They begin a discussion in which each presents an argument, then listens to the other’s argument, and then they both move toward an agreement. This kind of dialogue happens all the time. In our research, which involves cognitive psychology and experimental philosophy, we refer to it as “arguing to learn.”

But as political polarization increases in the U.S., the kind of antagonistic exchange exemplified by the Trump-Clinton debate is occurring with increasing frequency—not just among policy makers but among us all. In interactions such as these, people may provide arguments for their views, but neither side is genuinely interested in learning from the other. Instead the real aim is to “score points,” in other words, to defeat the other side in a competitive activity. Conversations on Twitter, Facebook and even YouTube comment sections have become powerful symbols of what the combativeness of political discourse looks like these days. We refer to this kind of discussion as “arguing to win.”

The divergence of Americans’ ideology is accompanied by an animosity for those across the aisle. Polls have shown that partisan liberals and conservatives associate with one another less frequently, have unfavorable views of the opposing party, and would even be unhappy if a family member married someone from the other side. At the same time, the rise of social media has revolutionized how information is consumed—news is often personalized to one’s political preferences. Rival perspectives can be completely shut out from one’s self-created media bubble. Making matters worse, outrage-inducing content is more likely to spread on these platforms, creating a breeding ground for clickbait headlines and fake news. This toxic online environment is very likely driving Americans further apart and fostering unproductive exchanges.

In this time of rising polarization, an important question has arisen about the psychological effects of arguing to win. What happens in our minds—and to our minds—when we find ourselves conversing in a way that simply aims to defeat an opponent? Our research has explored this question using experimental methods, and we have found that the distinction between different modes of argument has some surprisingly far-reaching effects. Not only does it change people’s way of thinking about the debate and the people on the opposing side, but it also has a more fundamental effect on our way of understanding the very issue under discussion.

Are we objectivists or relativists?

The question of moral and political objectivity is notoriously thorny, one that philosophers have been debating for millennia. Still, the core of the question is easy enough to grasp by considering a few hypothetical conversations. Consider a debate about a perfectly straightforward question in science or mathematics. Suppose two friends are working together on a problem and find themselves disagreeing about the solution:

Mary: The cube root of 2,197 is 13.

Susan: No, the cube root of 2,197 is 14.

People observing this conflict might not know which answer is correct. Yet they might be entirely sure that there is a single objectively correct answer. This is not just a matter of opinion—there is a fact of the matter, and anyone who has an alternative view is simply mistaken.

Now consider a different kind of scenario. Suppose these two friends decide to take a break for lunch and find themselves disagreeing about what to put on their bagels:

Mary: Veggie cream cheese is really tasty.

Susan: No, veggie cream cheese is not tasty at all. It is completely disgusting.

In this example, observers might take up another attitude: Even if two people have opposite opinions, it could be that neither is incorrect. It seems that there is no objective truth of the matter.

With that in mind, think about what happens when people debate controversial questions about morally infused political topics. As our two friends are enjoying their lunch, suppose they wade into a heated political chat:

Mary: Abortion is morally wrong and should not be legal.

Susan: No, there is nothing wrong with abortion, and it should be perfectly legal.

The question we grapple with is how to understand this kind of debate. Is it like the math question, where there is an objectively right answer and anyone who says otherwise must be mistaken? Or is it more like a clash over a matter of taste, where there is no single right answer and people can have opposite opinions without either one being wrong?

In recent years work on this topic has expanded beyond the realm of philosophy and into psychology and cognitive science. Instead of relying on the intuitions of professional philosophers, researchers like us have begun gathering empirical evidence to understand how people actually think about these issues. Do people tend to think moral and political questions have objectively correct answers? Or do they have a more relativist view?

On the most basic level, the past decade or so of research has shown that the answer to this question is that it’s complicated. Some people are more objectivist; others are more relativist. That might seem obvious, but later studies explored the differences between people with these types of thinking. When participants are asked whether they would be willing to share an apartment with a roommate who holds opposing views on moral or political questions, objectivists are more inclined to say no. When participants are asked to sit down in a room next to a person who has opposing views, objectivists actually sit farther away. As University of Pennsylvania psychologist Geoffrey P. Goodwin once put it, people who hold an objectivist view tend to respond in a more “closed” fashion.

Why might this be? One straightforward possibility is that if you think there is an objectively correct answer, you may be drawn to conclude that everyone who holds the opposite view is simply incorrect and therefore not worth listening to. Thus, people’s view about objective moral truths could shape their approach to interacting with others. This is a plausible hypothesis and one worth investigating in further studies.

Yet we thought that there might be more to the story. In particular, we suspected there might be an effect in the opposite direction. Perhaps it’s not just that having objectivist views shapes your interactions with other people; perhaps your interactions with other people can actually shape the degree to which you hold objectivist views.

WINNING VS. LEARNING

To test this theory, we ran an experiment in which adults engaged in an online political conversation. Each participant logged on to a website and indicated their positions on a variety of controversial political topics, including abortion and gun rights. They were matched with another participant who held opposing views. The participants then engaged in an online conversation about a topic on which they disagreed.

Half of the participants were encouraged to argue to win. They were told that this would be a highly competitive exchange and that their goal should be to outperform the other person. The result was exactly the kind of communication one sees every day on social media. Here, for example, is a transcript from one of the actual conversations:

P1: I believe 100 percent in a woman's choice

P2: Abortion should be prohibited because it stops a beating heart

P1: Abortion is the law of the land, the land you live in

P2: The heart beats at 21 days its murder [sic]

The other half of participants were encouraged to argue to learn. They were told that this would be a very cooperative exchange and that they should try to learn as much as they could from their opponent. These conversations tended to have a quite different tone:

P3: I believe abortion is a right all women should possess. I do understand that some people choose to place certain determinants on when and why, but I think it should be for any reason before a certain time point in the pregnancy agreed upon by doctors, so as not to harm the mother.

P4: I believe that life begins at conception (sperm meeting egg), so abortion to me is the equivalent of murder.

P3: I can absolutely see that point. As a biologist, it is obvious from the first cell division that “life” is happening. But I do not think life is advanced enough to warrant abolishing abortion.

It is not all that surprising that these two sets of instructions led to such results. But would these exchanges in turn lead to different views about the very nature of the question being discussed? After the conversation was over, we asked participants whether they thought there was an objective truth about the topics they had just debated. Strikingly, these 15-minute exchanges actually shifted people’s views. Individuals were more objectivist after arguing to win than they were after arguing to learn. In other words, the social context of the discussion—how people frame the purpose of controversial discourse—actually changed their opinions on the deeply philosophical question about whether there is an objective truth at all.

These results naturally lead to another question that goes beyond what can be addressed through a scientific study. Which of these two modes of argument would be better to adopt when it comes to controversial political topics? At first, the answer seems straightforward. Who could fail to see that there is something deeply important about cooperative dialogue and something fundamentally counterproductive about sheer competition?

Although this simple answer may be right most of the time, there may also be cases in which things are not quite so clear-cut. Suppose we are engaged in a debate with a group of climate science skeptics. We could try to sit down together, listen to the arguments of the skeptics and do our best to learn from everything they have to say. But some might think that this approach is exactly the wrong one. There might not be anything to be gained by remaining open to ideas that contradict scientific consensus. Indeed, agreeing to partake in a cooperative dialogue might be an instance of what journalists call “false balance”—legitimizing an extreme outlier position that should not be weighed equally. Some would say that the best approach in this kind of case is to argue to win.

Of course, our studies cannot directly determine which mode of argument is “best.” And although plenty of evidence suggests that contemporary political discourse is becoming more combative and focused on winning, our findings do not elucidate why that change has occurred. Rather they provide an important new piece of information to consider: the mode of argument we engage in actually changes our understanding of the question itself.

The more we argue to win, the more we will feel that there is a single objectively correct answer and that all other answers are mistaken. Conversely, the more we argue to learn, the more we will feel that there is no single objective truth and different answers can be equally right.

So the next time you are deciding how to enter into an argument on Facebook about the controversial question of the day, remember that you are not just making a choice about how to interact with a person who holds the opposing view. You are also making a decision that will shape the way you—and others—think about whether the question itself has a correct answer.