Why should we stop using data to change people’s minds and use questions instead?

Hello, my name is Dan Ariely. I'm a professor of Psychology and Behavioral Economics at Duke University, and I want to share some quick thoughts about using questions as a tool to get people to change their minds.

Usually when we try to convince people of a specific viewpoint, we think that we should present them with evidence. But, when we think of the number of times we have tried that approach successfully, we all must admit that we have a very low success rate. Consider how many times in the last three years you have had a debate with somebody and after presenting them with a compelling amount of solid evidence they said, “You know what, you're correct, I'm changing my opinion.” Or consider how many times somebody has presented you with solid evidence during a debate and at the end you said to them, “You know what, you're right. I'm completely changing my opinion.”

The technique I want to propose here is asking questions instead of presenting information. This means that instead of attacking people we meet them where they are and say, “Help me understand your position.”

To start, imagine that we have a range of people: some of whom are our side, some who are in the middle, and some who are on the opposite side from us. Let's start with the people in the middle:

?

Using questions with “people in the middle:”

When it comes to most aspects in life, our confidence in how much we know is vastly greater than our actual knowledge. So, one approach to encourage people to reconsider their perspective is to change their confidence level. Not their knowledge, which is our first instinct, but their confidence. We call this the illusion of explanatory depth.

One way I demonstrate this concept is to show people a picture of flush toilet and ask them how well they understand how a flush toilet works. People generally say that they understand it very well. When I ask them to rate their understanding on a scale from zero to ten, they generally rate their knowledge on the high end of the scale. But when I say “Luckily for you, I have all the pieces of a flush toilet here. Why don't you try to assemble it?” What happens? People fail. Almost nobody can assemble it. And then I ask them again how well they understand the flush toilet? And as you might suspect, their confidence drops substantially.

How does this pertain to the domain of anti-Semitism, Israel vs. Palestine, and October 7th?? When people ask to talk with me about these topics, I start by making it clear that I’m not arguing with them and instead that I will ask questions about their proposed solutions. In my experience, the discussion generally starts with people saying: “I want an immediate ceasefire and I want the IDF to withdraw immediately from Gaza." To this I say, “I can't grant this to you, but for the sake of argument, let's say I could. Now what?” People say, “Well, it's over. I got what I wanted.” At this point, I say, “No, no, it's not over. We're just starting.”? What I do next is probe further into their perspective.? Next, I say, “Please help me out with the details. Do we let Hamas continue ruling Gaza?” And almost everybody says “Yes, they were elected until the next election.” Next, I say. “Do we let them arm again?” My discussion partners usually say, “I guess if they're the ruling party of Gaza, they can arm again.” Next, I say, “You know, they are highly connected to Iran. Do we allow them to have nuclear power?” At this point everyone says no. My follow-up questions are then, “What about the Air Force? What about the Navy? What happens if on April 17, 2024, they plan to repeat the same thing they did on October 7, 2023? What then? What do you propose? What should Israel and the IDF do then? What would you recommend?” Realize that I am not attacking them but instead asking them for help with their solution, at this point people almost always say something along the lines of, “It's more complicated than I thought and I’m not so sure what my solution would be.”

And that’s what we want, right? The best we can initially hope for is to get people to have reduced confidence and be more open to other information and to think more deeply about the topic.

As another example of this same approach, we used the illusion of explanatory depth with election misinformation. We asked people, to help us understand the election process including how elections are tallied and then asked them to explain how they were stolen. During this process they realized they didn’t know or understand. Their confidence weakened. They did not concede that Biden won the election, but their confidence was reduced.

A second approach for interacting with people in the middle, is to use questions to transpose them into our position and with our perspective. How? Usually, people on the other side of an issue have their perspective. And we have our perspective. How do we get people to view things from our perspective? It turns out that asking questions is a very good way to do this. Imagine I make the following statement “Israeli mothers whose children are hostages in Gaza, are unable to think about anything other than their kids for even a minute.” And let's say this phrase is factual. How impactful would it be to the person listening to it?? Not very.

Now imagine that instead of making this statement, I ask people, “How many minutes in the day do you think that a mother whose child is a hostage in Gaza, is able to not think about her kid?” And then I wait an answer.? By changing the approach from a statement to a question, I've forced people to think about the question from the perspective of the mother, and to take her perspective to formulate an answer to the question. With this questioning approach people are much more moved. The point is that questions are good to both reduce over-confidence and to help people consider the issue from the other perspective.

?

Using Questions with “people who are fully against our perspective:”

Now let’s discuss people with hardcore opposition to our views. These are not people we can realistically convince. These are people, who would be happy if Israel was destroyed. In this case, we basically need to slow them down. Here questions are also a very good strategy. We did this with US elections. There are many people in the US, specifically Republicans who think that the 2020 elections were stolen. We basically took people on X (Twitter) who were distributing misinformation, and asked them questions. One at a time, we gave them many different articles that questioned their perspective in all kinds of ways, and asked them to explain their take about the information presented in the articles.

Why did we do this? It was a way to make them work harder to keep the discussion going. They had to read the papers. They had to digest it, understand it, and respond. And every time they responded, we just pushed more and more articles in their direction, asked more questions, and requested more responses. If we just posted that they were dumb or wrong, they could have just ignored or dismissed us, but when we asked questions—publicly—it was not simple for them to appear dismissive and unserious in front of their followers. Over time they slowed down or moved to different topics.

In summary, the main point is that fighting misinformation with information is somewhat ineffective. All of us have tried this for years and we know it doesn't work. It is time to try other approaches and one approach is to use questions instead.

What questions will you ask going forward?

My links:

My latest book: https://danariely.com/

Bureaucracy teachings: CAB - Center for Advanced Bureaucracy - Home ( centerforbureaucracy.com )

Substack: Dan Ariely Looks at Life | Substack