Wharton marketing professor Jonah Berger discusses his published study, “Topography of Thought,” which was co-authored with Olivier Toubia, business professor at Columbia Business School. The paper examines how someone’s writing style can be indicative of their future success, and where generative AI might come into the picture.

Read an edited transcript below.

Transcript

Studying How Successful People Think

Angie Basiouny: Give us an overview of this paper by way of explaining your title. What is topography of thought?

Jonah Berger: I don’t have to tell you that we all use language all the time. We write emails, make presentations, and submit job applications. We use language all the time. And language, in some sense, is a fingerprint. It reveals or reflects things about the people who produce it. You can predict how extraverted someone is, for example, based on the words they use.

But beyond the individual words people use, might the pattern of ideas they put out there, the way they organize their ideas, reveal something interesting about them and their likelihood of future success?”

I think it’s important to talk about what I mean about the pattern of ideas. When someone talks about something, they can cover a small amount of ground or a large amount of ground. If you ask someone about their work history, for example, they can talk about a variety of things they’ve done or a smaller set of things they’ve done. They can cover a lot of ground or a little bit of ground.

If you want to use an analogy here, you can almost think about going for a run. Someone can go for a run and go all the way around the city, or they can go for the same number of miles, but just go around the block a number of times. In both cases, they did the same distance, but they covered more ground in one than the other. So, one way we express ideas is the amount of ground we cover. We cover more ground with our ideas, or less. We can talk about more ideas, more topics, more themes, more things that are disparate from one another, or things that are related to one another.

But it’s not just that. It’s also the speed with which we move between adjoining ideas. Imagine a movie. It can cover more or less ground across the course of the movie, but it can also move faster or slower between ideas. If a movie has one scene, for example, that’s at the beginning of a wedding, if the next scene is the later part in a wedding, that’s related to the first part. Different things may happen, but it’s pretty closely related. But if you jump from a wedding to an action scene, that would be really different sorts of ideas. They’re not very related. They’re moving further across those two points.

We wondered if these two ideas, how much ground someone covers, and how quickly they move between ideas, might tell us something about their likelihood of future success.

Basiouny: You’re not talking about the length of the piece of writing. You’re not talking about whether it’s a 500-word essay or 1,000-word essay. It’s really about how they use that space to move through their ideas. Correct?

Berger: Yes, great point. It’s not about the length. It’s about the ground covered. Are they covering a lot of ground in their hundred or thousand words, whatever it is? Or are they covering less ground? Someone talking about their vacation could share 1,000 words, but they could use those words to talk just about the food they ate, or also to talk about the sights they saw and the places they visited.  The former would cover less ground while the latter would cover more.

Basiouny: How did you go about studying this?

Berger: The same ideas can be applied to any type of content, but here we looked at college applications. In other work we’ve looked at online reviews, and in related work we looked at books, movies, and TV shows.

We took 40,000 college application essays from a variety of folks who were applying to school, and we looked at what they wrote and the topography of thought of what they wrote — how much ground they covered in that essay. Again, similar length, but how much ground they covered, and how quickly they moved between ideas. And we looked at their future success. How well did they do in school? What was their GPA once they got there? We were interested in seeing whether, not just the individual words they use, but the way they express ideas might that reveal something or predict how likely they are to be successful in the future.

The Essay Writing Style That’s Linked to Greater Success

Basiouny: What did you find?

Berger: We found two very important things. First, covering more ground, that notion of covering a broad range of things in the same amount of length, was linked to greater success. But doing so while moving rather slowly, was also important.

Think about the numbers arrayed on a circular clock. You could cover a lot of ground by moving in a circle, (e.g., going from 12 to 1 to 2 to 3 to 4) or by traversing the same ground but going from like 12 to 6 to 1 to 7 to 3 to 11. You’ve covered the same amount of ground, but you’ve taken a much longer route between each individual point.

What we found is that folks who are successful in school are able to blend these two things that might seem mutually exclusive. It might seem like covering a lot of ground requires moving really quickly between points to get there. But folks that end up doing well in school figure out a way to cover that ground really efficiently. They’re able to do so by moving slowly between these points, and they don’t have to take a lot of big jumps along the way.

Basiouny: In this paper, you controlled for some socioeconomic factors. Can you talk about that a little bit?

Berger: Yes, so someone could wonder, “OK, so you’ve found that people did well in school, and you’re using writing as a way to indicate how they think. But does it indicate something else?” Maybe it’s just that people who do better on the SAT also have higher grades and also tend to write a certain way. Maybe it’s that people who have parents who are more educated tend to write a certain way and also do better in school. Or maybe people who have parents who are more educated can afford to pay for an essay consultant who helps them write a certain way and also helps them do better in school.

So, we controlled for a variety of different things. We controlled for what they wrote about. Maybe certain types of people tend to write about certain types of things, rather than other types of things. Maybe it’s not about how they write, it’s about what they wrote about — the topics or themes they discussed. No, it wasn’t that.

Maybe it’s parents’ education. No, we controlled for that. Maybe it’s SAT scores. No, we can control for that. What this suggests is that the topography of thought goes beyond things related to just socioeconomic factors or family background. It’s not just that people who might have had wealthier families, for example, tend to write a certain way or have application consultants and do better in school because they get tutoring. No, it really suggests that writing reveals something about the way we think, which can reveal or predict our likelihood of being successful in the future.

Implications Beyond College Essay Writing

Basiouny: There are other critical forms of writing that we do every day in business, like cover letters, resumes, a press release, communications to the C-suite. Can you take this research and translate it into a business context?

Berger: What I find fascinating about these ideas is yes, we looked at the case of college application essays, but it doesn’t have to be only about application essays. These same ideas should apply more broadly to a variety of contexts, whether it’s a cover letter that someone writes, whether it’s an online review that someone puts together, whether it’s the emails they write at the office — all these things provide insight into who people are and what they’re likely to do in the future. I think on a previous episode that you had me on, I talked about a paper I loved recently, where they can tell whether someone is going to default on the loan or not by the language they use in their application. Similarly, you can predict whether someone is going to get promoted or fired or leave a job for a better opportunity elsewhere based on the language they use in their email.

Most of this work that I just mentioned is using individual words, but I think what our work suggests beyond the individual words someone used, you can get insight into who they are, how they think, and how well they’re going to do in the future, based on the pattern of ideas that they have or their topography of thought.

Basiouny: As a manager, it gives you an indication of how they might move through their physical work or their knowledge work, right?

Berger: Yes, one thing we’re looking at right now is, as people learn more, does that change the way their topography of thought looks? Obviously, as we gain more knowledge in a given domain, we may talk differently. We may think about ideas differently. One thought we have is, “Hey, if people who are able to cover a lot of ground really efficiently by moving slowly between points, how did they get there? Are they naturally that way?” Probably not. They may have gained more knowledge along the way that allows them to represent their ideas differently.

One thing we’re doing right now is looking at online forums where people write multiple reviews over time. Someone, for example, might write hundreds of wine reviews over the years. They’ve learned more about wine years later. We’re looking at how do they represent ideas differently as they gain knowledge? And that may help us understand why people who represent ideas certain ways end up doing better.

Humans Still Write Better than ChatGPT and Gen AI — For Now

Basiouny: ChatGPT and artificial intelligence have entered the conversation. People have access to these free tools that can help them perfect their cover letters and written business communication. How does this change things? We can’t really tell how good someone’s topography of thought is if they have an AI-assisted piece of writing. What do we do?

Berger: Yes, so I’d say a couple of things. I agree with much of what you said, except one word. I’m not sure they allow you to “perfect” your writing. At least at the moment. What they do is allow you to write something pretty good quickly and easily. You give it a prompt, and it produces content that’s pretty interesting, does a pretty good job of doing something that might have been difficult for you to do.

And to a degree it uses your own content somehow. Like you could say, “Take my CV and use it to put together a cover letter based on my past experiences.” So that is based on you, and someone else’s might look different, to the degree that their CV is different, but I wouldn’t say it necessarily perfects anything. At least at the moment.

Certainly, tools like ChatGPT and others have made the production of content much easier, and I can imagine a time down the road where we do use them for many tasks, rather than writing ourselves. But I still think there is a lot to be understood about how language reflects the people that produce it and how to write more effective content, based on understanding what makes language impactful.